Expert | Impartial | Innovative

The Chalk aquifer of the

Wessex Basin

Research Report RR/11/02

 BRITISH GEOLOGICAL SURVEY

The National Grid and other RESEARCH REPORT RR/11/02

Ordnance Survey data © Crown
Copyright and database rights
2017. Ordnance Survey Licence
No. 100021290 EUL.

Key words
Chalk, Hydrogeology Wessex.

Front cover
River Test near Whitechurch,
Hampshire (P775341). The Chalk aquifer of the
Bibliographic reference
A llen, D J (editor). 2017. The
Wessex Basin
Chalk aquifer of the Wessex
Basin. British Geological Survey
Research Report No. RR/11/02.
118pp. D J Allen2 and E J Crane1 (editors)
Your use of any information
provided by the British Geological
Survey (BGS) is at your own risk.
Contributors
Neither BGS nor UK Research
and Innovation (UKRI) gives D J Allen2, J P Bloomfield1, D K Buckley2, J E Cunningham3, P M Hopson1,
any warranty, condition or
representation as to the quality,
P D Merrin3, A J Newell1, P Shand3, E L Tribe3, and A G Vivanco1
accuracy or completeness of the
information or its suitability for
any use or purpose. All implied
conditions relating to the quality
of suitability of the information,
and all liabilities arising from
the supply of the information
(including any liability arising in
1
British Geological Survey, 2 Formerly British Geological Survey, retired,
negligence) are excluded to the
3
Formerly British Geological Survey
fullest extent permitted by law.

Copyright in materials derived

from the British Geological
Survey’s work is owned by
UK Research and Innovation
(UKRI) and/or the authority that
commissioned the work. You may
not copy or adapt this publication
without first obtaining permission.
Contact the BGS Intellectual
Property Rights Section, British
Geological Survey, Keyworth,
e-mail [email protected]. You may
quote extracts of a reasonable
length without prior permission,
provided a full acknowledgement
is given of the source of the
extract.

Maps and diagrams in this book

use topography based on Ordnance
Survey mapping.

Some features are based on spatial

data licensed from the Centre for
Ecology and Hydrogeology © CEH.

ISBN 978 0 85272 642 6

© UKRI 2019. All rights reserved. Keyworth, Nottingham British Geological Survey 2019
BRITISH GEOLOGICAL SURVEY British Geological Survey offices

The full range of our publications is available from BGS shops at Environmental Science Centre, Keyworth, Nottingham
Nottingham, Edinburgh, London and Cardiff (Welsh publications NG12 5GG
only) see contact details below or shop online at www.geologyshop. Tel 0115 936 3100
com
BGS Central Enquires Desk
The London Information Office also maintains a reference Tel 0115 936 3143
collection of BGS publications, including maps, for consultation. e-mail: [email protected]

We publish an annual catalogue of our maps and other publications; BGS Sales
this catalogue is available online or from any of the BGS shops. Tel 0115 936 3241
email [email protected]
The British Geological Survey carries out the geological survey of
Great Britain and Northern Ireland (the latter as an agency service The Lyell Centre, Research Avenue South, Edinburgh
for the government of Northern Ireland), and of the surrounding EH14 4AP
continental shelf, as well as basic research projects. It also Tel 0131 667 1000
undertakes programmes of technical aid in geology in developing e-mail: [email protected]
countries.
Natural History Museum, Cromwell Road, London
The British Geological Survey is a component body of UK SW7 5BD
Research and Innovation (UKRI). Tel 020 7589 4090
Tel 020 7942 5344/5
email: [email protected]

Cardiff University, Main Building, Park Place, Cardiff

CF10 3AT
Tel 029 2167 4280

Maclean Building, Crowmarsh Gifford, Wallingford,

OX10 8BB
Tel 01491 838800

Geological Survey of Northern Ireland, Dundonald House,

Upper Newtownards Road, Ballymiscaw, Belfast, BT4 3SB
Tel 028 9038 8462
www.bgs.ac.uk/gsni/

Natural Environment Research Council, Polaris House,

North Star Avenue, Swindon, Wiltshire SN2 1EU
Tel 01793 411500
www.nerc.ac.uk

UK Research and Innovation, Polaris House, Swindon

SN2 1FL
Tel 01793 444000
www.ukri.org

Website: www.bgs.ac.uk
Shop online at: www.geologyshop.com

ii
Preface

This study was undertaken as part of the National Ground- Agency and academic institutions. The information has been
water Survey (NGS), which aims to provide a compre- reviewed and summarised from a variety of standpoints;
hensive description of the major British aquifers and their geological, hydrogeological and hydrochemical. Given
groundwater resources. This includes the characterisation the large area covered by the aquifer across the basin the
of the physical and chemical properties and processes that approach taken has been to provide both basinwide overviews
govern groundwater flow and pollutant transport and attenu- and catchment-specific discussions. The outcome of such
ation, in both the unsaturated and saturated zones. The NGS an approach has resulted in a report which encompasses a
has commenced with a study of regions of the Chalk aquifer. broad range of aspects of the hydrogeology of the aquifer at
Although much work has been undertaken on the Chalk a variety of scales.
Group of the Wessex Basin, there is a need for current
The British Geological Survey has collaborated with the
knowledge to be brought together into a convenient format
Environment Agency, and a number of water companies.
that is in the public domain. The study aims to summarise
available hydrogeological information and current under- An Advisory Panel comprising members of the above
standing of the aquifer. This report is intended as a source of organisations ensured that the work was relevant to them, as
reference that will benefit and interest researchers, students well as contributing much valuable information and acting as
and those in the water industry (suppliers, regulators and reviewers. Work on this report commenced in the early 2000s
consultants) as well as the general public. and was mainly complete by 2011. Staff changes delayed
In order to meet the objectives of the study, information publication. The maps and data presented thus reflect that
has been collected from a variety of sources, including period and more up-to-date versions may now be available
British Geological Survey data, publicly available literature, elsewhere. Minor revisions of the management chapter were
information from the water industry and the Environment undertaken in 2015.

iii
Acknowledgments

This study has relied on the work of a number of current In addition we would like to thank Roger Harrington
and former members of BGS staff in addition to the named (Bournemouth and West Hampshire Water), and Mike
contributors. In particular the geological basis of the study Hedges (Portsmouth Water) for useful and fruitful discus-
has incorporated the work of D T Aldiss, C M Barton, K A sions.
Booth, C R Bristow, D J Evans, A R Farrant, and P M Hop- We would like to acknowledge the Environment Agency
son. The work was also helped substantially by the members (Southern and South West Regions), Portsmouth Water,
of the Project Advisory Board, which comprised the follow- Southern Water and Wessex Water for the provision of hy-
ing: drochemical data. Digital terrain model and river data used
in the report were sourced from the Centre for Ecology and
Giles Bryan (Environment Agency)*
Hydrology (CEH) (Morris and Flavin, 1990, 1994; Moore et
Tina Dijkstal (Southern Water Technology Group)
al., 1994). Other data are referenced in the text.
Jim Grundy (Environment Agency)
Maps and illustrations were prepared by J E Cunningham,
Alison Matthews (Environment Agency)*
P Sapey, P Lappage and I Longhurst. Page-setting was
Mike Packman (Southern Water)
by P Sapey, D Rayner and A Hill. Additional editing was
Paul Shaw (Environment Agency)*
undertaken by J E Thomas.
Paul Stanfield (Wessex Water Services)*
Carole Sharratt has supported the production of this and
(Those asterisked also provided written contributions to the many other reports with patience and supreme levels of
report). organisation.

iv
Contents

Preface iii FIGURES

Acknowledgements iv
1.1 Geology of the Wessex Basin 2
Executive summary ix 1.2 Topography of the Wessex Basin 3
1 Introduction 1 1.3 Land cover map for the study area (showing
1.1 The Wessex Basin Chalk aquifer 1 aggregate 1 km grid-square averages) 4
1.2 Topography and land use 1 1.4 1961–90 gridded rainfall data for the Wessex Basin
1.3 Climate 4 (data supplied by CEH, Wallingford) 5
1.4 Drainage 4 1.5 River network of the Wessex Basin 5
1.5 Demography 12 1.6 River Frome and River Piddle catchment area 6
1.6 Water usage and the importance of 1.7 River Stour and River Allen catchment area 7
groundwater 12 1.8 River Avon catchment area 8
1.9 River Test catchment area 9
2 Geology 13
1.10 River Itchen catchment area 10
2.1 Introduction 13
1.11 East Hampshire and Isle of Wight catchment areas 11
2.2 Chalk Group lithostratigraphy 17
2.1 Geology of the Wessex Basin showing location of
2.3 Palaeogene strata 21
cross-sections 13
2.4 Quaternary 22
2.2 Geological cross-sections through the Wessex
3 Groundwater flow in the Chalk aquifer 24 Basin 14
3.1 Hydraulic properties of the aquifer 24 2.3 Geophysical logs showing the stratigraphical
3.2 Geological controls on aquifer properties 27 subdivision of the Chalk, Wessex Basin 16
3.3 The aquifer in the water cycle 31 2.4 Chalk thickness (metres) in the Wessex Basin 17
4 Catchment hydrogeology 53 2.5 Gamma ray and resistivity logs showing expansion
4.1 South Dorset coast 53 of Chalk Group units from the Wessex Basin to Sus-
4.2 Frome and Piddle 53 sex (after Bristow et al., 1997) 18
4.3 Stour and Allen 57 2.6 Distribution of hardgrounds in the Chalk Rock
4.4 Avon 58 Member in the Wessex Basin (after Bromley and
4.5 Test and Itchen 63 Gale, 1982) 20
4.6 East Hampshire 67 2.7 Palaeogene correlation using gamma-ray logs,
4.7 Isle of Wight 68 Wareham Basin, Dorset 21
2.8 West–east section through the Palaeogene deposits
5 Hydrogeochemistry 70 of the Wareham Basin based on correlation of
5.1 Introduction 70 borehole gamma-ray logs 22
5.2 Previous hydrochemical studies 70 2.9 Map of superficial deposits in the Wessex Basin 23
5.3 Hydrochemical characteristics of the 3.1 Porosity profile for the Totford Borehole 24
unsaturated zone 71 3.2 Groundwater levels and structure for the Upper Itchen
5.4 Hydrochemical characteristics of catchment (after Giles and Lowings, 1990) 26
groundwaters 71 3.3 A schematic illustration of the variation of open
5.5 Age of the groundwater 75 fractures with depth in the Chalk (after Allen et al.,
5.6 Down-gradient evolution of Chalk 1997) 27
groundwater 76 3.4 The distribution of solution features in the Chalk of the
5.7 Spatial variations in Chalk groundwaters in Dorset heathlands (after Sperling et al., 1977) 29
the Wessex Basin 78 3.5 Relationship between Palaeogene minor aquifers in
5.8 Hydrochemistry and surface – groundwater the Wessex Basin (not to scale) (from Jones et al.,
interactions, Bourne case history 80 2000) 30
6 Groundwater management 90 3.6 Recharge mechanisms associated with clay-with-
6.1 Introduction 90 flints (from Klinck et al., 1998) 33
6.2 Groundwater resource issues and 3.7 West Woodyates hydrograph (Source: BGS
pressures 91 National Groundwater Level Archive) 34
6.3 Groundwater management strategies 94 3.8 Groundwater levels in the Chalk outcrop of the
6.4 Groundwater pollution and protection 98 Wessex Basin (IGS 1979a, 1979b) 35
6.5 Nitrate vulnerable zones 102 3.9 Unsaturated zone thickness in the Chalk outcrop of
the Wessex Basin 36
7 References 103
3.10 Geophysical logs showing water inflows from
Appendix List of boreholes 109 Seaford, Lewes Nodular and New Pit Chalk

v
 Formation, Figheldean Borehole, River Avon 4.1 Groundwater levels and topography at Lulworth
catchment (courtesy of Environment Agency) 37 (after Institute of Geological Sciences, 1979a) 53
3.11 Geophysical logs showing concentrated water inflow 4.2 Geological map of Wareham area 56
above a hard band in the Seaford Chalk, Tidworth 4.3 River Bourne location map, and groundwater
North borehole, River Bourne catchment (data kindly contours for 27th January 1994 (adapted from
provided by of Thames Water in 2002) 38 Environment Agency and Water Management
3.12 West to east cross-section showing the relationship of Consultants, 2004) 60
the Chalk Group units to topography and the position 4.4 Sketch geological cross-section of River Bourne
of the summer water table relative to the Chalk showing the general relationship of the units within
Rock hardgrounds, Fonthill Bishop area, Wiltshire the Chalk Group (adapted from Environment Agency
(courtesy of Environment Agency) 39 and Water Management Consultants, 2004) 61
3.13 Scale cross-section north to south showing the 5.1 Location of samples used in the present study 71
relationship of the Chalk Group units and the water 5.2 Piper diagram showing the relative proportions of major
inflow horizons, northern end of River Bourne cations and anions in Wessex Basin groundwaters 74
catchment (courtesy of Environment Agency and 5.3 Box plot showing ranges in concentration of major
Thames Water) 40 elements in groundwaters of the Wessex Basin
3.14 Scale cross-section north-east to south-west using Chalk 75
digitised 1970s logging data to correlate Chalk Group 5.4 Cumulative probability plot showing the ranges in
units to show water inflows, Candover Scheme concentration of major elements in groundwaters of
boreholes, Hampshire 42 the Wessex Basin Chalk 75
3.15 Pumped fluid and flowmeter logging and packer 5.5 Box plot showing the ranges in concentration of
test measurements used to identify horizons of fluid minor and trace elements in groundwaters of the
movement, Chalk aquifer, BGS Totford Borehole, Wessex Basin Chalk 76
Hampshire 43 5.6 Cumulative probability plot showing the ranges in
3.16 Comparisons of caliper logs and main water inflow concentration of major elements in groundwaters of
horizons, depth below surface plot, Candover Scheme the Wessex Basin Chalk 76
boreholes 44 5.7 Interstitial water chemistry, temperature and SEC
3.17 Scale cross-section Wareham area, based on profiles from the research borehole at West Lulworth
geophysical log data showing main water inflow (after Edmunds et al., 2002) 77
horizons in the Portsdown ChalkFormation 45 5.8 Depth variations in groundwaters of the confined
3.18 Fluid temperature logs of selected Wessex Basin aquifer in the Wareham area of the Wessex Basin 78
boreholes with thick Palaeogene cover displaying 5.9 Temporal variations in hydrochemical parameters
local cooling within casing indicating groundwater in the Litton Cheney spring [SY 550 908] (after
movement within sandier layers of the Palaeogene Edmunds et al., 2002) 79
strata 46 5.10 Temporal variations in hydrochemical parameters
3.19 South Winterbourne with locations of a number from the West Houghton Borehole [ST 823 045]
observation points, showing flow to the 12 km (after Edmunds et al., 2002) 79
mark (observation points courtesy of Environment 5.11 Major and minor element characteristics of Chalk
Agency) 46 groundwaters along a line of section between
3.20 South Winterbourne flow signature — Frome Powerstock and Wareham (after Edmunds et al.,
catchment (courtesy of Environment Agency) 47 2002) 80
3.21 River Till flow signature — Wylye (Avon) catchment 5.12 Hydrochemical characteristics of Chalk groundwaters
(courtesy of Environment Agency) 47 along a line of section between Powerstock and
3.22 Chitterne Brook flow signature — Wylye (Avon) Wareham (after Edmunds et al., 2002) 81
catchment (courtesy of Environment Agency) 47 5.13 Regional plot for electrical conductance in the
3.23 Bere Stream flow signature – Piddle catchment Wessex Basin Chalk 82
(courtesy of Environment Agency) 47 5.14 a) Regional plot for Cl in the Wessex Basin Chalk,
3.24 Devil’s Brook flow signature — Piddle catchment b) Regional plot for Na in the Wessex Basin Chalk 83
(courtesy of Environment Agency) 47 5.15 a) Regional plot for Ca in the Wessex Basin Chalk,
3.25 North Winterbourne flow signature — Stour catchment b) Regional plot for Mg in the Wessex Basin Chalk 84
(courtesy of Environment Agency) 48 5.16 a) Regional plot for F in the Wessex Basin Chalk,
3.26 River Bourne flow signature – Avon catchment b) Regional plot for Sr in the Wessex Basin Chalk 85
(courtesy of Environment Agency) 48 5.17 Regional plot for NO3 –N in the Wessex Basin
3.27 Winterbourne behaviour of Devil’s Brook, Bere Stream Chalk 86
and North Winterbourne in relation to geology 48 5.18 Conceptual diagram of the Chalk aquifer of Dorset
3.28 Geology of the Upper Piddle catchment 49 highlighting the main geochemical processes controlling
3.29 Flow accretion profiles for the River Piddle (using water quality (after Edmunds et al., 2002) 86
flow gaugings by Wessex Water Authority 1986) 5.19 Bourne hydrochemistry sampling sites 87
after Howden et al., 2004 50 5.20 Piper plot showing samples of the River Bourne and
3.30 a) South House stream flow (normalised by catchment groundwaters collected in the study 87
area), b) Little Puddle stream flow (normalised by 5.21 Hydrochemical measurements and major element
catchment area), c) Flow difference (normalised) concentrations for surface waters and groundwaters in
between South House and Little Puddle, d) Flow the Bourne catchment 88
difference between South House and Little Puddle vs 5.22 Selected major, minor and trace element
Little Puddle flow (normalised) (after Howden et al., concentrations for surface waters and groundwaters in
2004) 52 the Bourne catchment 89

vi
6.1 Water supply map of the Wessex Basin (boundaries formations present in the Bourne and Nine Mile
approximate) 92 catchments (adapted from Farrant et al., 2001) 62
6.2 Examples of protected sites in Wessex 93 4.5 Gauging station data from selected sites on the River
6.3 CAMS boundaries in Wessex 96 Test and its tributaries (source: Marsh and Lees,
6.4 Mean nitrate (as mg nitrogen from nitrate per litre) 2003) 65
in Chalk groundwater in the Wessex Basin for 4.6 Gauging station data from selected sites on the River
consecutive five-year periods since 1976 (after Roy et Itchen and its tributaries (source: Marsh and Lees,
al., 2007) 100 2003) 65
6.5 Groundwater vulnerability map for the Wessex 4.7 Water balance calculation for the Test catchment
Basin 101 (data from Entec, 2005) 66
4.8 Water balance calculation for the Itchen catchment
(data from Entec, 2005) 67
TABLES 4.9 Gauging station data from selected sites on rivers in
east Hampshire (source: Marsh and Lees, 2003) 68
1.1 Long-term monthly average (1971 to 2000) climate 4.10 Gauging station data from selected sites on rivers
data recorded by the Met Office for four sites across the in the Isle of Wight (source: Marsh and Lees,
Wessex Basin (www.met-office.gov.uk) 4 2003) 69
1.2 Some of the main towns and cities of Wessex and 5.1 Summary of chemical analyses of Chalk
their populations based on the 2001 Census groundwaters from the Wessex Basin 72
(www.statistics.gov) 12 5.2 Summary of trace element analyses of Chalk
2.1 Chalk Group stratigraphy of southern England (after groundwaters from the Wessex Basin 73
Bristow et al., 1997) 15 5.3 Chemistry of deep interstitial waters from the Lulworth
3.1 Geological evolution of the Chalk aquifer in the Borehole (from Edmunds et al., 2002) 77
Wessex Basin and its hydrogeological implications 32 5.4 Selected historical nitrate and chloride data from BGS
3.2 Hydrogeological controls on springs in the Piddle borehole archives 80
catchment (adapted from Marcus Hodges, 1999 and 6.1 Definition of CAMS ‘resource availability status’
Howden et al., 2004) 51 classifications 95
3.3 Summary of main hydrogeological controls on 6.2 CAMS areas for the Wessex Basin Chalk 97
springs in the Piddle valley 52
4.1 Gauging station data from selected sites on the rivers
Frome and Piddle (source: Marsh and Lees, 2003) 54 BOXES
4.2 Gauging station data from selected sites on the rivers
Stour and Allen (source: Marsh and Lees, 2003) 57 1 Chalk composition 15
4.3 Gauging station data from selected sites on the River 2 Geophysical properties of the Chalk Group 15
Avon and its tributaries (source: Marsh and Lees, 3 Mineralogy of the Chalk Group 70
2003) 59 4 Main water legislation 91
4.4 Summary of the principal geological and
hydrogeological characteristics of the Chalk Group

vii
viii
Executive summary

The geological structure known as the Wessex Basin and southern parts of the Wessex region and are sepa-
occupies a substantial area in southern England and contains rated from the Chalk by a major hiatus of some 15 mil-
deposits ranging from Permian to Cretaceous in age. The lion years; however, the two highly contrasting units are
basin stretches from Devon eastwards to the Weald, flanks largely structurally conformable, both being folded dur-
the southern side of the London Platform and extends south ing the main Miocene inversion event. The major hiatus
under the English Channel. Within this larger geological between the Chalk and the Palaeogene was a time of region-
structure, the onshore Chalk of the Wessex region covers al uplift, sea level fall and subaerial erosion of the exposed
large parts of the counties of Hampshire, Wiltshire and Chalk surface under subtropical climatic conditions.
Dorset and stretches from the Weymouth region in the The aquifer properties of the Chalk in the Wessex Region
west to near Chichester in the east and from the south coast are determined both by its intergranular (matrix) hydraulic
towards Newbury in the north, with an outcrop area of properties and by the nature and distribution of fractures.
approximately 4650 km2. The porosity of Chalk varies between about five per cent
The area of study is characterised mainly by Chalk and 45 per cent and depends on stratigraphy. For example,
downland that dominates over a core of forest and heath terrain the upper part of the White Chalk Subgroup of southern
associated with the outcrop of the Palaeogene sediments. England has an average porosity of 39 per cent, the lower
The Downs, which reach elevations of around 270 m, are part of the White Chalk Subgroup an average porosity of
characterised by undulating topography consisting of steep- 28 per cent and the Grey Chalk Subgroup a porosity of 23
sided escarpments with long, gentle dip slopes and rounded per cent. Local contrasts in porosity are due to variations
hills. The climate is typically mild and rainfall over most of in the nature of the original sedimentation: generally, lower
the region averages between 750 and 900 mm annually. porosities are associated with more clay-rich chalks and
The Wessex region is drained by a number of rivers, the with nodular and hardground seams. In addition, porosity
most significant of which generally have headwaters in tends to reduce with depth. Despite high values of porosity,
the Chalk uplands around the edge of the basin; these then the storage characteristics of the Chalk aquifer are poor
flow across the Chalk and Palaeogene deposits towards the as a result of small pore sizes and low values of specific
south coast, discharging along the coastline between Poole yield are normal. The hydraulic conductivity of the Chalk
Harbour and the Solent. From west to east the main depends principally on the occurrence of fractures, the
catchments are those of the Frome and Piddle, Stour and matrix values being very low (of the order of 10–3 m d–1) as
Allen, Avon, Test and Itchen. In addition, smaller rivers a result of the small pore throat sizes. In general terms, the
drain the Chalk of East Hampshire and the Isle of Wight. highest transmissivities and storage coefficients are found
Water usage varies across the region but includes in valleys, with lower values towards interfluves. Hydraulic
agriculture, public water supply, spray irrigation, industry, conductivity also rapidly decreases with depth.
private domestic supply, gravel washing, fish farming Pumping test data suggest that the aquifer properties of
and watercress production. Most of the licences are for the Chalk vary widely in the Wessex region, with transmis-
agricultural practises such as farming and spray irrigation, sivity values ranging from virtually zero to several thou-
but these are for relatively small volumes compared to public sands of square metres per day and with storage coeffi-
supply. Much of the water used is obtained from groundwater cients ranging from the order of 10–5 to around 0.06. Mean
sources, which principally utilise the Chalk aquifer. values of transmissivity from pumping tests are around
The distribution of the Chalk in the region results from 1500 m2 d–1 for much of the basin (with a bias to valley sites),
two main tectonic stages in its history. During the Permian falling to around 200 m2 d–1 in south Dorset (probably as a re-
to Cretaceous, basins were subsiding and infilling with sult of the local tectonic hardening of the Chalk). Groundwater
sediment under a regime of regional crustal extension and, models have used transmissivities that commonly vary from a
apart from a few local variations, the Cretaceous Chalk was few hundred square metres per day on interfluves to several
deposited as a uniform blanket across the central parts of the thousand square metres per day in valleys. Mean values of
Wessex Basin. During the Cenozoic, crustal extension was storage coefficients from pumping tests were of the order of
replaced by south to north-directed compression related to 0.005. Storage coefficients used in models have tended to be
the Alpine mountain building event. Sedimentary rocks that larger, with values ranging from 0.003 to 0.05, with the higher
had accumulated within the basins were pushed upwards as values characterising Chalk at shallow depths in valleys.
the movement on bounding faults changed from normal to The significant variability in the aquifer properties of
reverse and compression led to the development of east– the Chalk reflects a diverse range of matrix and fracture
west trending folds. The geographical distribution, geol- characteristics and the wide range of geological and
ogical structure and geomorphology of the Chalk is largely a hydrogeological processes that have contributed to the
function of Cenozoic compression and folding. The exposed development of the aquifer. For example, in the Wessex
Chalk uplands to the north of the Palaeogene lowlands of region, where the Chalk has been affected by Alpine
the Hampshire Basin are broadly anticlinal in structure, tectonics, there is a greater structural complexity in the
while beneath the Palaeogene deposits of the Hampshire aquifer than in other regions of England. Large folds and
Basin, the Chalk forms a syncline. faults are present, which affect the hydrogeology of the
Palaeogene strata, characterised by alternating sand region, while smaller structural features exert more subtle
and clay formations, cover significant areas of the central influences on the local hydrogeology. Although the area

ix
was not glaciated, it was significantly affected by periglacial perennial head in summer and commensurate wetting from
processes and clay-with-flints deposits cover large parts of the perennial head in winter, but exhibit much more complex
the aquifer on higher ground. Both the periglacial processes behaviour. For example, it is not uncommon for streamflow
and acidic drainage from Palaeogene cover rocks have to increase from the stream head, then to decrease for a
contributed to the local development of the aquifer, leading section (and in extreme cases to dry completely) before
to the development of karst features in some areas. increasing again further down. Some streams exhibit even
Recharge to the aquifer through the unsaturated zone may more complicated behaviour with, periodically, a succession
have both vertical and horizontal flow components and may of flowing and dry sections above the perennial head. There
involve both the Chalk matrix and fractures. The recharge are a number of causes for this complex behaviour, which
is influenced by the presence of permeable or impermeable can include abstraction and lithological effects.
horizons within the Chalk and the nature of overlying The major element chemistry of the Chalk groundwaters
Palaeogene and Quaternary deposits. The heterogeneous is controlled by the carbonate system and the groundwaters
nature of these deposits means that impermeable units in the outcrop areas are of Ca-HCO3 type. Mixing with
may promote run-off, whilst the more permeable units remnant formation water (probably within the chalk matrix)
may allow direct recharge to occur. However, the presence has led to an increase in salinity in the deepest, oldest waters.
of impermeable deposits may also concentrate flow from Ion exchange of Na for Ca in these waters has led to a trend
certain areas and hence locally increase the amount of intermediate between that of Na-HCO3 and Na-Cl types.
recharge to the underlying aquifer. Long-term average The largest changes in water chemistry are found where the
annual modelled recharge values in the basin vary from aquifer becomes confined beneath the overlying Palaeogene
around 200 mm to around 500 mm. In general, modelled sediments. A redox boundary occurs close to this boundary
values in the Bourne, Upper Avon and Test catchments are and beyond this zone the groundwaters are reducing and
around 250 to 350 mm per annum (a-1). contain low concentrations of nitrate. The reducing nature
The principal groundwater flow mechanism in the of the system leads to an increase in Fe and Mn. There are
saturated zone is along fractures, which are often enlarged by limited data on trace metals over much of the aquifer but
solution, rather than through the Chalk matrix. Subsurface the available information indicates that concentrations are
evidence for the location of groundwater flow horizons generally low, a notable exception being fluoride which
and their relationship with aquifer lithology is obtained generally exceeds the EU-MAC for drinking water in
primarily from the interpretation of geophysical borehole confined parts of the aquifer. The large chemical differences
logs. Formation and fluid logging of boreholes penetrating between the unconfined and confined groundwaters in the
the Chalk shows that water normally flows into a borehole Wessex region are consistent with differences in residence
at only a few specific points and these are often associated time and suggest that they may be controlled by different
with certain stratigraphical horizons (which are commonly flow systems.
located at flint, marl and hardground surfaces) at shallow Groundwater management is very important in the
depths. Wessex region, where groundwater from the Chalk aquifer
While the locations of flow horizons often appear to be represents a major source of water supply (and often forms
controlled by lithologically distinct stratigraphical horizons a large percentage of river water). There are a number
such as Chalk hardgrounds, the hydraulic and chemical of issues that are important to the management of the
influence on their development is such that the main flow groundwater resources of the region: for example increasing
horizons generally develop within a few tens of metres of demand, protection of habitats, the problems of drought and
the Chalk surface. This is because the Chalk is soluble and conversely of groundwater flooding, and, in the longer term,
groundwater tends to develop permeability pathways as the effects of climate change.
it moves towards surface discharge points. In the Wessex Urbanisation, industrial growth and intensive agricultural
region, groundwater flow is generally inwards from the activity can have a serious impact on groundwater quality.
edges of the basin towards the coast and is therefore broadly The rate of groundwater movement through the Chalk is
in the direction of dip. Generally, hydraulic gradients are generally slow and the effects of historical pollution may take
shallower than the dip of the strata and groundwater flow in a significant time to appear. Conversely, where the Chalk is
the shallow, solution-enlarged fracture systems will tend to karstic, pollutants such as pathogens may move rapidly from
be transferred from lower to higher stratigraphical horizons the ground surface to groundwater sources. In the Wessex
as it moves down a catchment. region, initial characterisation for the Water Framework
Interaction between surface streams and groundwater Directive showed that almost all of the groundwater bodies
is very common across the Wessex region and most of the in the Chalk aquifer in the eastern part of the basin were ‘at
surface watercourses are fed substantially by groundwater, risk’ as a result of diffuse pollution pressures (principally
with baseflow indices commonly high, of the order of 0.8 nutrients and pesticides) and groundwater bodies in the
or more. The low specific yield of the Chalk results in western part were classified as ‘probably at risk’.
winterbourne behaviour (involving streams drying in their A thorough understanding of the hydrogeology of the
upper reaches during summer) being very common in the Chalk of the Wessex region and the influences on Chalk
Chalk groundwater-fed streams. However in the Wessex groundwater is therefore essential for effective resource
region many streams do not follow the classic winterbourne management. It is hoped that this report will fulfil part of
model of a simple gradational drying down to the stream’s that need.

x
1 Introduction

1.1 THE WESSEX BASIN CHALK AQUIFER 1.2 TOPOGRAPHY AND LAND USE

The geological structure known as the Wessex Basin occupies The Wessex region rises from sea level to heights
a substantial area in southern England and contains deposits approximating 300 m above Ordnance Datum (AOD)
ranging from Permian to Cretaceous in age. The basin extends around the basin periphery (Figure 1.2). The area of
from Devon eastwards to the Weald, flanks the southern side study is characterised mainly by Chalk 1downland, which
of the London Platform and extends south under the English predominates around the edges of the basin and forest,
Channel. Within this larger geological structure, the onshore and heath terrain associated with the southern and central
Chalk Group of the Wessex Basin covers large parts of the outcrop of the Palaeogene sediments. Figure 1.3 illustrates
counties of Hampshire, Wiltshire and Dorset and stretches the main types of land cover across the region. The Chalk
from the Weymouth region in the west to beyond Portsmouth forms a broad belt of country running from the coast of south
in the east and from the south coast towards Newbury in the Dorset through the North Dorset Downs to Salisbury Plain
north (Figure 1.1). The area covered by the Chalk Group at to the north of Salisbury. East of Salisbury, the Chalk covers
outcrop is approximately 4650 km2. a large area between Winchester and Basingstoke, with the
The term Wessex Basin is used in two ways in this outcrop extending eastward, to the north of Portsmouth.
Research Report. In a geographical/hydrogeological sense it The Downs are characterised by undulating topography
is the region of south central England drained by the generally consisting of steep-sided escarpments with long, gently
southward and eastward flowing rivers of Dorset, east sloping dip slopes and rounded hills. In the west, the
Wiltshire, Hampshire and the Isle of Wight. In a geological landscape consists of rounded hills that are intersected by a
sense the Wessex Basin structure is more extensive, covering complex system of valleys and combes. In south Dorset, the
much of southern England south of the Variscan Front and Downs rise to a maximum elevation of 274 m at Bulbarrow
including strata of Permian to Cretaceous age. A younger Hill and over 220 m on Cranborne Chase (to the west and
Hampshire Basin is a more restricted, eroded remnant founded north of Blandford Forum respectively), with somewhat
on Palaeogene strata of east Dorset, southern Hampshire and lower elevations over Salisbury Plain to the north-east of
northern Isle of Wight. Salisbury. In the east of the Wessex Basin, the Downs rise
For the purposes of this report, the Chalk aquifer of to a maximum elevation of 271 m at Butser Hill to the north
the Wessex Basin is delineated by a number of natural of Portsmouth.
boundaries within the total area underlain by the Chalk The Isle of Wight is topographically dominated in the
Group. The eastern limit of the aquifer is the groundwater centre by two offset, steep-sided, narrow, east–west trending
divide between Portsmouth and Chichester, which has been ridges connected by a broad central downland area, the region
taken as the western limit of the South Downs1 Chalk aquifer rising to elevations of 100–200 m. In the south of the island
(Jones and Robins, 1999). The northern limit is taken to be are the Southern Downs, characterised by steep landslipped
the surface water divide to the south of the Thames Valley, margins, which attain about 240 m to the north of Ventnor.
and the Vale of Pewsey. The western boundary is the edge The most notable habitats of the Wessex region include
of the outcrop and the southern limit is the line of outcrop chalk grasslands, chalk rivers and woodlands with smaller
of the Chalk Group along the Isle of Wight–Purbeck areas of meadow land and wetland habitats that support a
monocline. The sea forms the southern limit of the landward wide variety of associated species. The Chalk Downs were
aquifer system, although the Chalk continues to dip beneath formerly well wooded, but now only scattered plantations
the English Channel. The Isle of Wight, although isolated are to be found. The Downs form characteristic rounded
from the mainland, is part of the same basin and contains summits where they are uncultivated and have a springy turf
steeply dipping Chalk strata that are connected underneath of fescue grass with distinctive vegetation, including rare
the Solent to the mainland. orchids and fauna. Agricultural technological advances have
The base of the Chalk Group forms the base of the aquifer. now made it possible to cultivate all but the steepest slopes
However, the Upper Greensand lying directly below the allowing crops to grow on the land on which sheep used to
Chalk is often considered to be in hydraulic connection and graze. In the central and southern parts of the Wessex region
may affect the response of the overlying aquifer. Furthermore, there is an area of heathland and forest (associated with the
even though the majority of the units in the Chalk sequence Palaeogene sands and clays) that is generally lower lying
may contain groundwater and the thickness of the Chalk and flatter than the Chalk downland.
Group reaches 450 m in places, only the uppermost few In the western part of the region, in Dorset and Wiltshire,
tens of metres of the saturated Chalk in a given locality are the main land use is for agriculture, although this accounts
generally considered to constitute an aquifer in terms of a for only a small percentage of the counties’ employment. The
viable groundwater resource. Thus the Chalk aquifer is not main towns in south-east Dorset are Dorchester, Wareham
synonymous with the Chalk Group. The central-southern and Poole and the area’s industrial developments tend to be
portion of the Chalk in the basin is overlain by a sequence located around Poole Harbour. On the Chalk uplands large
of younger Palaeogene sediments: these strata are generally farms concentrate on crop production and dairy farming,
considered to be poor aquifers or non-aquifers, but are
important because they influence processes that occur on, and
within, the Chalk. 1
The name ‘down’ derives from the Saxon English ‘dun’ meaning ‘hill’.

1
 160

Basingstoke

Wells Andover

140

Salisbury
Winchester

120
Yeovil
Southampton

Portsmouth
100

Newport N
Dorchester Bournemouth

080
Weymouth
0 30 km

360 380 400 420 440 460

Key
Coastline
River
1: 50 000 sheet boundary
Traditional nomenclature for the Chalk
River catchment boundary
where undivided by modern mapping
Palaeogene
Portsdown Chalk Formation
Culver Chalk Formation
Newhaven Chalk Formation ’Upper Chalk’ Seaford and Newhaven Chalk Undivided
Seaford Chalk Formation on the Shaftesbury Sheet area
Lewes Nodular Chalk Formation
New Pit Chalk Formation
’Middle Chalk’
Holywell Nodular Chalk Formation Lower and Middle Chalk
Zig Zag Chalk Formation of the Isle of Wight
’Lower Chalk’
West Melbury Marly Chalk Formation
Selborne Group
Lower Greensand Group
Wealden Group
Purbeck Group
Jurassic and older strata

Figure 1.1 Geology of the Wessex Basin.

contains OS data © Crown copyright and database rights 2017 Ordnance Survey [100021290 EUL]. Use of this data is subject to terms and conditions.

while steeper land accommodates large flocks of sheep. an important role. In the Purbeck District, the Portland Stone
The Frome and Piddle catchments, for example, are and Purbeck Marble are renowned as building materials and
predominantly rural and modern agricultural methods have Dorset also has one of the last mines in the country working
converted previously grazed pasture to arable land. In these ball clay, which is important in the ceramics industry.
catchments, around 70 per cent of the area is farmed. On the Offshore and onshore oil and gas resources are exploited in
Palaeogene mixed sand and clay areas in south-east Dorset, south Dorset, particularly to the south of Poole Harbour. The
agricultural enterprises are varied. Pig and poultry production majority of west Dorset is considered an Area of Outstanding
is practiced, as is some horticulture, and forestry now plays Natural Beauty and tourism is a major industry in the area.

2
Figure 1.2 1
60
Topography of the
Wessex Basin. WILTSHIRE Basingstoke
contains OS data © Crown
copyright and database Salisbury Plain Andover
Wells
rights 2017 Ordnance Survey HAMPSHIRE
1
[100021290 EUL]. Use of this 40
data is subject to terms and Stockbridge
conditions. SOMERSET
Salisbury
Winchester
hase
eC
urn
nbo
1
20 Romsey
Cra Butser Hill
Fordingbridge

SUSSEX
Yeovil

WEST
Southampton
DORSET New Forest
Bulbarrow Hill Blandford Forum Ringwood

1
00
e Solent
Th Portsmouth
Poole Christchurch

Wareham Newport
N
Bournemouth
Dorchester
ISLE OF WIGHT
Isle of
0
80 Purbeck
Weymouth

0 50 km

0
60
3
60 3
80 00
4 4
20 4
40 60
4

Elevation (m)
0−25 50−75 100−150 200−250 Above 300 Rivers
25−50 75−00 150−200 250−300 County boundary

Approximately 75 per cent of the Hampshire Avon century, land was parcelled up into smaller fields, enclosed
area (Environment Agency, 1998a) is farmed, with cereal, by many of the banks, hedges and ditches visible today.
cattle and sheep farming the dominant activities. There Medieval land use still survives in places like Stockbridge
are also a number of sand and gravel extraction pits in the in the Test Valley, where the old water meadows by the river
Ringwood area and chalk and sandstone quarries in the are now common land.
vicinity of Salisbury. Industry is mainly light in nature and The valley of the Test has supported a variety of industries
situated in the towns; in addition there are several major in the past, such as tanning leather and milling flour, but
military establishments on Salisbury Plain. Tourism forms a the majority of these industries have declined resulting
significant industry, particularly in Salisbury, Fordingbridge in disused mill leats and backwaters. The river is used
and Christchurch, and the area is also of high archaeological extensively for trout and salmon fishing. While the catchment
and historical importance. Abandoned water meadows are is predominantly rural with significant agricultural activity,
characteristic of all the river valleys, resulting from past it includes the urban areas of Andover, Romsey and part of
agricultural practices. Southampton.
The New Forest area is predominantly rural and the The landscape of east Hampshire is characterised by
enclosed forest is classified as a Site of Special Scientific distinctive coastal and inland areas. The main urban and
Interest (SSSI) and includes Forestry Commission industrial zones lie along the coast, especially around
plantations for both soft and hard woods, and enclosures the low-lying areas. The coastal strip contains important
for agricultural small holdings. The main urban areas are conservation areas, such as broad estuaries with tidal mud
in the south-west and these include industries such as flats, wetland, marshland and open coastline, some of which
petrochemical and power generation. Tourism is also a have been designated as important conservation sites.
major industry in this area. Inland from the coastal zone lies rolling open countryside
The Test and Itchen valleys stretch north–south from the that is predominantly rural and used chiefly for agriculture.
Chalk downlands, on the Berkshire border, to the meadows Watercress production is an important industry in this area,
and the valleys of the rivers that discharge into Southampton accounting for some 70 per cent of the total production in
Water and the Solent. The topography of the catchments the UK. In addition to river valleys, there are also a number
comprises small hills and shallow valleys. Whilst the of dry valleys on the Chalk.
catchments are predominantly rural areas, with significant On the Isle of Wight, the Chalk downs are characterised by
agricultural activity, they also have an economy based on rolling pasture and arable land with pockets of unimproved
industrial and commercial communities, principally in and grassland on steeper areas. Heathland/acidic pasture can be
around the main towns. Agriculture has always played a found on the Greensand between the two ranges of Chalk
large part in shaping the local landscape that, historically, downs. Dairy farming, creating lush, irregular fields bounded
would have been covered with trees. In the thirteenth by mature hedgerows and coppiced woodland, dominate

3
Figure 1.3 Land
cover map for the
1
60

study area (showing

aggregate 1 km grid- !
Basingstoke
square averages). Andover
contains OS data © Crown ! Wells !
copyright and database 1
40
rights 2017 Ordnance
Survey [100021290 EUL].
Use of this data is subject to ! Salisbury
!
(
terms and conditions. Winchester

1
20

! Yeovil
! Southampton

1
!
00
Portsmouth

Dorchester !
! Bournemouth Newport !
N

0
80 ! Weymouth

0 50 km

0
60 60
3 3
80 00
4 4
20 40
4
60
4

Land Cover 2000 (CEH)

Woodland Rough pasture, marsh & heath Littoral sediment and rock Rivers
Improved grassland Fen, marsh, swamp, saltmarsh, bog Urban
Arable & horticulture Inland bare ground Inland water

the northern pastures. Along the north coast are numerous 1.4 DRAINAGE
harbours, creeks, salt marshes and tidal mudflats fringed by
woodland. The southern coastal plain is dominated by an 1.4.1 Drainage pattern
intensively managed arable farmland with large open fields
The Wessex Basin is drained by a number of rivers (Figure
and few trees. Land use is predominately rural with mixed
1.5), the most significant of which generally have headwaters
farming over most the area and intensive horticulture in the
in the Chalk uplands around the edge of the basin. These then
Eastern Yar valley. There is a small industrial sector, mainly
flow and converge over the Palaeogene deposits towards the
in aerospace, but tourism and the leisure industry dominate
south coast, discharging along the coastline between Poole
the island’s economy.
Harbour and The Solent.
In the south-west of the region the River Frome
(Figure 1.6) rises from springs in the Upper Greensand near
1.3 CLIMATE
the village of Evershot on the West Dorset Downs. The River
Temperatures in Wessex are typically mild, with averages
ranging from 1°C in winter to 22°C during the summer
(Table 1.1). Rainfall (Figure 1.4) varies from less than Table 1.1 Long-term average (1971 to 2000) climate data
700 mm per year in north-east Hampshire and along the coast recorded by the Met Office for four sites across the Wessex
near Portsmouth to over 1100 mm on the higher ground west Basin (www.metoffice.gov.uk).
of Blandford Forum in Dorset, although over most of the
region the annual average is between 750 and 900 mm. The Bognor Boscombe
highest rainfall is observed in the south-western part of the Yeovilton Everton
Regis Down
region in the upper parts of the Frome and Piddle catchments Max. temperature (°C)
where a rainfall average (Met Office data, 1969 to 1990) of July/Aug
21.7 20.7 20.8 21.7
1033 mm a–1 has been recorded. Other areas of higher rainfall
are in west Wiltshire and the eastern edge of the basin in Min. temperature (°C)
1.3 2.5 2.3 1.0
Hampshire. The driest areas tend to occur around the coastline February
to the south from Portsmouth towards Bournemouth. Other Daily sunshine (hours)
4.2 5.2 4.8 4.6
notable dry areas with rainfall less than 750 mm a–1 are in the Annual average
catchments of the River Avon and Bourne north of Salisbury
Annual rainfall (mm) 725 717 764 736
and in the upper parts of the Test catchment around Andover.

4
Figure 1.4 1961–90 1
60
gridded rainfall data
for the Wessex Basin Basingstoke
!
(
(data supplied by CEH, Andover
Wells
Wallingford). !
( !
(
contains OS data © Crown 1 Boscombe
40
Down
copyright and database
rights 2017 Ordnance
Survey [100021290 EUL]. ( Salisbury
!
!
(
Use of this data is subject to Winchester
Yeovilton
terms and conditions. 1
20

!
( Southampton
Yeovil !
(

Bognor
Regis
1
00 !
(
Everton
Portsmouth
Dorchester !
!
( Bournemouth !
( Newport
N
0
80
!
( Weymouth

0 50 km

0
60 3
60 3
80 4
00 4
20 40
4 4
60 4
80

Rainfall (mm) 30-year 650−700 750−800 850−900 Above 950 Climatological

average (1961−1990): 700−750 800−850 900−950 Rivers station

Frome and its tributaries form the southernmost major joined on the way by two small streams, the Sydling Water
Chalk river in England. From its source, the River Frome and the River Cerne, which flow from the north. Below
flows south across Chalk and is joined by the Wraxall Brook Dorchester it passes over the Palaeogene deposits and flows
north of Cattistock and then by the River Hooke at Maiden mainly eastwards. The Frome is soon joined by the South
Newton, both draining land to the west. From Maiden Winterbourne at West Stafford, flowing from the south-
Newton the Frome flows south-east to Dorchester, being west. After the confluence with the South Winterbourne the

Figure 1.5 River 1

60 rn A
von
network of the Wessex Chalk Easte
We
Basin. River catchments ster
n Av
on
Rivers
Bo

contains OS data © Crown Basingstoke

ur
ne

!
Ri
vu

copyright and database

let

e
Nin
ok

Wells R. ile Andover

Bro

rights 2017 Ordnance ! M !

rne

R.
tte

Survey [100021290 EUL]. 1

40
An
Chi

ton
R. Till

R. Dever
on

W
rn
R. Av

Use of this data is subject to

all
ou

R. W
Candover
o

ylye
B

Stream
pB
R.

est
ro

terms and conditions.

ok

R. T

Salisbury
Wa en
ter

Alre ton
Shre

dder Winchester
Che ream

R. Na !
(
on

St

!
ri

R. Dun
odd

R. Ebble
R.

R. Itchen
R. L
C ale

1
20 R. Blackwater
ok
Bro

Yeovil R. nd
on

dla
nks

Cra
Gu

R. Av

ne Ca
Strea age

!
ss

Mo

R.
R. Iweme

ag

am
m
en

eon ble

it
eB
ydd

Herm

!
t Stre
am
roo

Southampton R. Wallington
Tarr ook
R. L

R. H
k
Br

Lavan
ant

W R. R. Bea
R. M

rax Sto ulieu

all ur
R. A
Lyscombe Str

Br
o
Devils Brook

llen

o
1
00 k R.
Sydling Wate

North Mo !
Winterbourne ors
Stream

R. H R. Lymington
Bere

ook
Portsmouth
eam

e
R. Mude
R. C

R. F
rom
r
ern

e R.
Pid !
e

Dorchester ! R. dle
ourne
Fro
me Bournemouth ! Newport N
interb
S. W Tadnoll i n
Brook .W
R. W

R
Eastern Yar
0
80 R. Medina
ey

! Weymouth

0 50 km

0 3 3 4 4 4 4
60 60 80 00 20 40 60

5
 N

Evershot

Alton Pancras

Lys eam

Devil
Broo
Str
com
R. C
R.

North

’s
100 Cattistock
Ho

k
be
ern
Winterbourne

Sydling Wat
ok
e

am
e

Bere
Stre
Maiden Lower
Newton

er
R. Waterson
Fr Burleston
om Puddletown
e

Dorchester

R.
Pi
R.

dd
Fr

e l
th

om
Sou ourne
terb

e
Wi n Tadnoll
Brook in
W
R.
R. Wey

080
0 10 km Weymouth Bay

360 380

Refer to Fig 1. for key

Figure 1.6 River Frome and River Piddle catchment area (see Figure 1.1 for key).
contains OS data © Crown copyright and database rights 2017 Ordnance Survey [100021290 EUL]. Use of this data is subject to terms and conditions.

Frome flows over Palaeogene deposits eastward towards join the Stour (principally the Cale and Lydden which mainly
Wareham, being joined on the way by minor tributaries, the drain Oxford Clay) the tributaries forming a dense drainage
largest of which are the Tadnoll Brook and the River Win, network. Below Sturminster Newton the river flows mainly
which join from the south. The Frome finally discharges south-east towards Blandford Forum and then to Wimborne
into Poole Harbour at Wareham where it is joined by the Minster. In this reach, which is predominantly underlain
River Piddle. The middle and lower sections of the Frome by Chalk, there are fewer tributaries, and these include the
have braided river features, many of which are the relics Iwerne, Tarrant, North Winterbourne and finally the Allen,
of historical water meadow systems but also include flood which joins the Stour at Wimborne Minster. The River Allen,
relief channels as well as natural channels (National Rivers which drains a catchment of 177 km2 above Walford Mill
Authority, 1996). Gauging Station rises from seasonal Chalk springs near
The River Piddle rises at springs near Alton Pancras and Monkton Up Wimborne; the perennial head of the river is
initially flows south and then eastwards from Puddletown further downstream, at springs near to Wimborne St Giles,
towards Poole Harbour (Figure 1.6). There are three major from where it flows approximately 18 km to its confluence
tributaries; the Lyscombe Stream (which has a confluence with the Stour. From Wimborne to the sea, the Stour flows
with Piddle near Lower Waterston), The Devil’s Brook over Palaeogene deposits and shortly before reaching
(which joins the Piddle at Burleston) and the Bere Stream Christchurch the river is joined by the Moors River (known
(with a confluence with the Piddle at Warren). as the Crane in its upper reaches where it flows over Chalk).
The River Stour rises in the Upper Greensand of Blackmoor Finally, the Stour discharges into Christchurch Harbour.
Vale at Stourhead, a few kilometres north-west of Mere, and The River Avon tributaries, the Wylye and Bourne and
flows south-east to the sea at Christchurch (Figure 1.7). The confluent streams, drain a large area of Salisbury Plain. The
Stour drains a catchment of 1300 km2 (predominantly in source of the River Avon is in the Greensand of the Vale of
Dorset, but including parts of Somerset and Wiltshire) and Pewsey, however most of the flow is gained from the Chalk,
over the 96 km of its course it falls approximately 230 m either directly to the Avon or via its tributaries as the river flows
(Environment Agency, 1997). From its source the Stour in a southerly direction over Salisbury Plain (Figure 1.8).
initially flows south to Gillingham where it is joined by two Above Salisbury the river is called the Upper Avon and
tributaries, the River Lodden and Shreen Water, which drain above Upavon it is divided into two tributaries, the Eastern
the Kimmeridge Clay. As it flows further south, towards Avon and Western Avon. These streams are themselves
Sturminster Newton, the other clay-influenced tributaries formed from several smaller watercourses, making the true

6
 140

en
N
r e er
Sh at
Stourhead W

Mere

on
odd
Gillingham

R. L
R.
Cale

120

Sturminster Newton
Monkton Up R. C
Wimbourne ra
ne
R. en
dd Gu
Ly s
Br sage

Tarr k
Wimbourne St Giles

Bro
oo

rne
k

ant
o
e
R. Iw
Blandford Forum

R.
Al
le
R.

n
S to
ur R.
100 North M
oo
Winterbourne rs

0 10 km
Bournemouth
Poole Bay
360 380 400 420

Refer to Fig 1. for key

Figure 1.7 River Stour and River Allen catchment area (see Figure 1.1 for key).
contains OS data © Crown copyright and database rights 2017 Ordnance Survey [100021290 EUL]. Use of this data is subject to terms and conditions.

source of the Avon difficult to establish. Streams at Bishops by a number of tributary rivers that have their sources on
Cannings are generally considered to comprise the source of the Palaeogene deposits of the New Forest area: these can
the Western Avon, with the Eastern Avon rising near Easton contribute significant quantities of floodwaters. The Avon
Royal. From Upavon towards Salisbury, the Upper Avon finally drains into Christchurch Harbour, where it is joined by
meanders strongly over narrow meadowlands in a steeply the Stour and the Mude. The total catchment area of the Avon
confined valley and is joined from the east by the Nine Mile is approximately 1700 km2, with a total fall from Pewsey
River north of Amesbury. There are numerous dry valleys, to the sea of 110 m and with an average gradient south of
some of which contain springs. At Salisbury, there are Salisbury of 1:1000 (National Rivers Authority, 1994).
confluences with the rivers Bourne and Wylye, which drain The River Bourne (Figure 1.8) drains the Chalk of the
the Chalk of Salisbury Plain. eastern edge of Salisbury Plain, forming a valley between
The River Wylye (Figure 1.8), with a catchment area of the Avon to the west and the Test to the east. The river
around 460 km2, rises from Upper Greensand springs near flows entirely over Chalk bedrock, rising in the north from
Maiden Bradley. The Wylye is joined by two tributaries springs in the lower formations of the Chalk Group. The
that originate on the Chalk of Salisbury Plain: the Chitterne River Bourne normally rises in winter near to Burbage and
Brook, which joins at Codford St Mary and the River Till, then flows south and south-west for over 40 km towards its
with a confluence at Stapleford. The River Till has its winter confluence with the Avon at Salisbury. The river is perennial
source near Tilshead some 12 km from its Wylye confluence below major springs at Idmiston.
(Halcrow, 1996). To the east of the Avon Catchment, the River Test
The Wylye tributary is itself joined at Wilton, near to (Figure 1.9) drains a large portion of the northern and north
Salisbury, by the Nadder, a spring-fed river that drains the eastern area of the Wessex region. The River Test rises near
escarpment of the Chalk Downs of south Wiltshire and the the village of Ashe and flows south-west and then south
Kimmeridge Clay of the Wardour Vale. The River Ebble, a across youngest formations of the Chalk Group to discharge
small river, joins the Avon to the south of Salisbury and flows at Southampton. There are a number of tributaries. The Test
continuously downstream of Broad Chalke but behaves as a is joined by the Bourne Rivulet at Hurstbourne Priors and
bourne above this point. Below Salisbury the valley widens, further downstream by the spring-fed Dever, Anton, Wallop
with several watercourses in places, but there are few towns Brook and Somborne Brook. Downstream of Mottisfont, the
near the riverbanks. Below Fordingbridge the Avon is joined Test flows off the Chalk outcrop onto Palaeogene deposits.

7
 N
Bishops Canning

Kennet and Avon Canal

Burbage
160 Pewsey
Eastern Royal
W von
es e rn A
ter
nA East
vo
n
Upavon

Tilshead
e
in
N e
R. Mil

ok
ro
eB

e
rn

rn
itte

ou
140
Ch

B
Maiden Bradley Codford St Mary

W roo
R.
R. Til
Amesbury

al k
B
R. W

lo
ylye

p
Stapleford Idmiston
Wa en
ter

Stourhead
e
Shr

R. Nadder
n
ddo
R. Lo

Salisbury
R. Ebble
Broad Chalke

Avon
120
r
Rive

R.
C ra Fordingbridge
n e

0 10 km

100 R.
M
oo
rs
R. Mude

380 400 420

Christchurch Harbour
Refer to Fig 1. for key

Figure 1.8 River Avon catchment area (see Figure 1.1 for key).
contains OS data © Crown copyright and database rights 2017 Ordnance Survey [100021290 EUL]. Use of this data is subject to terms and conditions.

The River Dun joins the Test from the west just south of this the upper Itchen — the Candover Stream, the Cheriton
point, and flows across both London Clay and the Chalk, Stream (also called Tichbourne) and the Alre. In general, the
receiving water both via Chalk baseflow and surface run-off Candover Stream rises at Brown Candover and the Cheriton
from the clays. The River Blackwater joins the River Test Stream normally rises at springs just south of Cheriton,
south of Broadlands. Its catchment, founded on variable however, the sources of both these tributaries recede in dry
sand and clay strata of the Palaeogene, gives the stream a years (Entec, 2002). The Alre has three tributaries, two of
‘flashy’ flow regime. The Test is not truly a natural river in which now start in watercress beds and fish farms where the
its current form. Along most of its length, the river is split natural spring flow of the river is supplemented by artesian
into two or more channels with sluices to regulate flows. or pumped borehole discharge. The third rises at a spring
The River Itchen (Figure 1.10) rises on the Chalk near near Bishops Sutton which can run dry when groundwater
Alresford in the Hampshire Downs. Three tributaries feed levels are low.

8
 N
160

Bo
ur Basingstoke
ne
Ri
vu
let
Ashe

Hurstbourne Priors
ve e
Ri Mil
r
ne

Andover
Ni

R.
An
ton
140
W roo
al k
B
lo
p

est
R. T

Salisbury
Winchester
R. Dun
Mottisfont

R.
120 B la Broadlands 0 10km
ck
w
at
er

420 440 460

Refer to Fig 1. for key

Figure 1.9 River Test catchment area (see Figure 1.1 for key).
contains OS data © Crown copyright and database rights 2017 Ordnance Survey [100021290 EUL]. Use of this data is subject to terms and conditions.

From the confluence of the Cheriton, Alre and Candover and Hayling Island in the east to the Hamble estuary in the
streams near Alresford, the River Itchen flows initially west west. The River Meon is the longest and most significant
and then it turns to flow south through Winchester to the edge river in the area and is sourced from Chalk springs near
of the Chalk outcrop at Otterbourne. The river then passes East Meon. The river initially flows north-west, and then
over Palaeogene cover through Eastleigh and Southampton west before turning south to cut through the downs towards
towards its mouth on the north bank of Southampton the coast. The River Hamble rises near to the juction
Water. A further tributary, Monks Brook, which drains a between the Palaeogene strata and the Chalk outcrop, as
mainly urban area, joins the Itchen just downstream of the does the River Wallington and the two rivers to the east of
tidal limit at Woodmill. For the majority of its length, the Portsmouth, the Hermitage Stream and Lavant Stream. Dry
Itchen is divided between two or more separate channels valleys exist throughout the Chalk outcrop of the Hamble
that run parallel to each other. This is because of the and Wallington catchments where run-off only occurs
past use of the river to provide water for milling, to during exceptionally wet periods.
supply water meadows, and for the development of The two principal rivers of the Isle of Wight (Figure 1.11),
navigation. the Medina and the Eastern Yar, originate at the south-east
The area of the surface water catchment of the River Itch- end of the island. The River Medina flows northward through
en to Otterbourne is approximately 360 km2. However, the a gap in the central downs and the Eastern Yar flows in a
groundwater catchment extends significantly to the north- north/north-easterly direction, cutting through the Sandown
east (Entec, 2002) beneath the surface catchment of the Riv- Pericline at Brading. In addition, a number of smaller rivers
er Wey (part of the extended River Thames Catchment) east drain the region, often rising on the Palaeogene deposits and
of New Alresford. The larger size of the Chalk groundwater running to discharge at the coast.
catchment that feeds the upper Itchen is reflected in high
baseflows, particularly in the Alre. 1.4.2 Winterbourne behaviour
The most easterly catchments of the Wessex Basin lie in
Hampshire and are drained by the Rivers Meon, Hamble A winterbourne is a stream that is commonly dry in its upper
and Wallington, together with numerous other smaller reaches in summer, while flowing over its maximum length
tributary rivers that feed into the three large harbours of in winter or early spring. This behaviour occurs when the
Portsmouth, Langstone and Chichester and into the Solent stream is in good hydraulic continuity with an underlying
(Figures 1.5 and 1.11). The area extends from Emsworth aquifer and where the summer streamflow depends

9
 Basingstoke

140

Brown Candover

Candover
Stream
New Alresford
R. Itchen
Bishop’s

Cher m
Strea
Sutton

iton
Winchester
Cheriton

Otterbourne

M
Br onk
120 oo s
k
Eastleigh

Woodmill

0 10 km
Southampton

440 460 480

Refer to Fig 1. for key

Figure 1.10 River Itchen catchment area (see Figure 1.1 for key).
contains OS data © Crown copyright and database rights 2017 Ordnance Survey [100021290 EUL]. Use of this data is subject to terms and conditions.

largely (perhaps almost entirely) on the contribution from behaviour with, periodically, a succession of flowing and
groundwater. Thus as the water table in the aquifer rises and dry sections.
falls annually, so the stream lengthens and shortens along There are a number of possible causes for this complex
the river valley. Since, for a given amount of recharge, behaviour. In some cases a local lowering of the water
the annual variation in water table elevation is inversely table, and concomitant reversal of effluent (gaining)
proportional to the specific yield of an aquifer, the effect is streams to influent (losing) systems over certain reaches,
particularly noticeable in streams fed by an aquifer having may be caused by abstraction. Natural causes (principally
a low specific yield, such as the Chalk. Thus, in a classic lithological and karstic effects) are also likely to be
winterbourne, the head of the stream would be expected involved, such as the influence of hardgrounds (see Section
to migrate in a predictable and relatively smooth manner 3.2.2).
upstream from its perennial location as groundwater
levels rose, with the stream reaching its maximum length 1.4.3 The development of drainage and landscape in
as groundwater levels peaked, and would then return in a Wessex
predictable way to the perennial head position as the water
table fell. It would also be expected that flow downstream The drainage of the Wessex Basin originally consisted of a
from the migrating stream head would steadily increase as system of rivers flowing towards a longitudinal trunk stream
more groundwater discharge was accepted by the stream that is now dismembered but formerly flowed along the basin
from the aquifer. axis from west to east, discharging to the south of the Solent.
In fact, as is discussed in Chapter 3, many streams and This drainage system parallels that of the larger London
rivers in the region do not follow the simple winterbourne Basin as drained today by the Kennet–Thames. These two
model outlined above. In some rivers, for example the river systems, the former Frome–Solent system and the
Piddle, flows increase downstream from the stream head, modern Kennet–Thames, are separated by a longitudinal
then commonly decrease over a section of the stream, watershed that runs broadly from west to east across the
before increasing again further down. In extreme cases, northern edge of Wessex towards the Weald. Today, almost
such as the River Bourne, the river often flows for a all the Wessex region drainage is southwards towards the
section, then dries, sometimes for a considerable length, Frome–Solent and the English Channel (Figure 1.5). As a
then flows again further downstream (commonly at its consequence, most of the major rivers flow across all the
perennial head). Some streams exhibit even more complex structural complexities of the basin.

10
 C e o

East Meon N
120
er
Ri v ble
m
Ha

n
eo
rM
ve
Ri
on
Hamble lli ngt
er Wa
Estuary Ri v Emsworth

Portsmouth
Harbour Langstone
Harbour
Portsmouth Hayling
100
The Solent Island

River Medina

Newport
Brading

080 0 10 km

420 440 460 480

Refer to Fig 1. for key

Figure 1.11 East Hampshire and Isle of Wight catchment areas (see Figure 1.1 for key).
contains OS data © Crown copyright and database rights 2017 Ordnance Survey [100021290 EUL]. Use of this data is subject to terms and conditions.

The geomorphology of the chalklands of southern More recently, however, such a simple model of landscape
England has been extensively investigated, but there is evolution has been questioned (Jones, 1999) on a number of
still great uncertainty related to the development of the grounds. For example, a model involving the development
landscape, and hence the underlying aquifer, in these areas of anteconsequent streams rather than a ‘superimposition
(Jones, 1999). Wooldridge and Linton (1955) described the model’ is currently favoured. More importantly, from the
development of discordant drainage in the Wessex area as a viewpoint of aquifer development, instead of modelling
series of discrete events as follows: landscape development as a series of discrete events Jones
(1999) has emphasised the continuity of landscape (and by
1. During the Miocene, rivers developed on the post- inference aquifer) evolution. He also notes that geological and
Alpine folded surface of the Chalk and broadly reflected geomorphological events often occur as pulses rather than
the east–west trending fold and fault lineations in the distinct episodes, for example, pulsed uplift or sedimentation.
underling Chalk. Finally, the concept of structural compartmentalisation into
2. The Pliocene sea inundated the uplifted Wessex uplands morphotectonic regions has been developed and Jones (1999)
producing a peneplaned surface. speculates that in southern England there may be evidence
3. Following the Pliocene, the area was uplifted and for up to 200 m of Pleistocene differential uplift. This is
a superimposed drainage developed on the tilted important as it removes the need for uniformity in landscape
peneplaned surface. The drainage had a dominant (and hence aquifer) evolution over large regions. Part of the
southward direction and the distribution of rivers that cited evidence for compartmentalisation of chalklands into
developed on the peneplaned surface formed the basis morphotectonic regions is the work of Edmonds (1983) and
of the present day drainage pattern. Farrant (2008) who described the distribution and density of
4. As the superimposed drainage network developed, karst features. Edmonds suggested that karst is more prone to
local accommodation of the river system led to some development in areas where recent deposits are most
outgrowth of rivers into east–west tributaries. Trunk extensive (e.g. the London Basin) and where erosion surfaces
streams cut down into the Chalk and the water table fell. between the Chalk and overlying deposits are preserved,
As the streams continued to cut down into the Chalk and are least developed where superficial deposits are
some lateral tributaries and eventually some main trunk generally thin and where erosion surfaces are relatively
streams dried up from perennial head downwards, poorly preserved due to dissection. Jones (1999) notes that
leaving the system of rivers and dry valleys we see today. the Pewsey–London Platform inversion appears to be an

11
important structural divide separating the London Basin summer holiday season. The main towns of Ryde, Newport
from the Wessex Basin and may mark the boundary between and Cowes are located along the coast or rivers.
two areas with differential Pleistocene uplift.

1.6 WATER USAGE AND THE IMPORTANCE OF

1.5 DEMOGRAPHY GROUNDWATER
The demography of the Wessex region is largely controlled
by land use and, particularly along the coast, by the tourism Water usage varies across the region but includes
industry. There are a number of towns and cities with agriculture, public water supply, spray irrigation, industry,
populations exceeding 100 000 people (Table 1.2). private domestic supply, gravel washing, fish farming
The population of Dorset is 390 986 (2001 Census). Outside and watercress production. Most of the licences are for
the main towns of Bournemouth, Poole, and Weymouth, many agricultural practises such as farming and spray irrigation,
people live in small towns with populations of around 10 000. but these are for relatively small volumes compared to public
The largest agricultural centres of the district are centred around supply. For example in the Test and Itchen catchments,
Shaftesbury, Blandford Forum, Dorchester and Salisbury industry and agricultural uses combined made up only three
in Wiltshire. There has been substantial urban development per cent of the total volumes licensed, despite a large number
along the coastline in response to demands from the tourist of abstraction licences (Environment Agency, 1999a).
industry, concentrated in the four districts of Christchurch, A large proportion of the abstraction licences across the
Bournemouth, Poole and Weymouth and Portland. Wessex Basin are for groundwater, underlining the fact that
The population of Hampshire is 1 240 032 (2001 Census). the Chalk is the most important aquifer in Britain and in
The most significant centres of population are along the particular in southern and eastern England, supplying over
coast, principally Southampton, which is an important half the nation’s groundwater-abstracted drinking water. In
English Channel port, and Portsmouth, a major naval the Wessex region groundwater provides a major source of
base. Inland, important centres are Winchester, Andover, water used for both public and private supplies. For example
Basingstoke, Farnborough and Aldershot. in the Stour and Allen catchment, groundwater comprises 49
The resident population for the Isle for Wight is 132 719 per cent of the total annual licensed abstractions (Environment
(2001 Census). This figure approximately doubles in the Agency, 1997); similarly in the Avon catchment groundwater
makes up approximately 45 per cent of the total (Environment
Agency, 1998a). On the Isle of Wight, an area which has
Table 1.2 Some of the main towns and cities of historically suffered from water supply problems, 68 per
Wessex and their populations based on the 2001 cent of the total licensed abstractions are from groundwater
Census (www.statistics.gov). (Environment Agency, 2003). In the East Hampshire CAMS
area groundwater accounts for 98 per cent of all licensed
Location Population abstraction (Environment Agency, 2002a).
The majority of groundwater sources used for water
Weymouth and Portland 63 665 supply are sited along the river valleys in the Wessex Basin,
Bournemouth 163 441 both because of the shallower depth to the water table in
these locations and as a result of the tendency for the Chalk
Poole 138 229 aquifer to have higher values of transmissivity in valleys.
Salisbury 114 614 Groundwater is used widely across the basin for both private
and public water supplies, with the volumes abstracted
Basingstoke and Deane 152 583 for public water supply significantly exceeding other
abstractions. For example on the Isle of Wight 86 per cent
Winchester 107 213
of all licensed groundwater abstractions are for public water
Southampton 217 478 supply (Environment Agency, 2003). Of the groundwater
abstracted only 1.5 per cent is used for agriculture and 1.4
Portsmouth 186 704 per cent for industry. (Environment Agency, 1999b).

12
2 Geology

2.1 INTRODUCTION Alpine mountain building event. Sedimentary rocks that

had accumulated within the basins were pushed upwards
2.1.1 Basin structure as the movement on bounding faults changed from
The Wessex Basin comprises a series of structurally normal to reverse. Compression led to the development
controlled sedimentary basins and intrabasinal highs of east–west trending folds.
filled with strata of Permian to Cretaceous age and covers
southern England and adjacent offshore areas (Underhill The distribution of the chalk within the Wessex region and
and Stoneley, 1998). Onshore, the geographical region, the Palaeogene preserved in the Hampshire Basin syncline
also termed the Wessex Basin, covers an area of Hampshire are shown in Figure 2.1. The geographical distribution,
and Dorset together with parts of east Devon, Somerset and geological structure and geomorphology of the Chalk is
Wiltshire and this approximates to the ancient Kingdom of largely a function of Cenozoic compression and folding and
the West Saxons. The geological units present in this basin erosion up to the present day. The exposed Chalk uplands
were affected by later tectonic events of mid-Miocene to the north of the Palaeogene lowlands of the Hampshire
age. The present structural configuration results from two Basin are broadly anticlinal in structure. However, the
main tectonic stages in the history of the Wessex Basin detailed morphology of this area owes much to the
(Chadwick, 1993): existence of numerous subsidiary fold axes of Tertiary
age, and developed as reactivations of older structures,
1. During the Permian to Early Cretaceous, sub-basins were including the Wardour, Dean Hill and Portsdown anticlines,
subsiding and infilling with sediment under a regime of which are offset along a diagonal belt (Figures 2.1 and 2.2).
regional crustal extension. Initially driven by differential The Hampshire Basin syncline preserves a thick succession
block faulting, the subsidence became thermally driven of Palaeogene strata and overlies the similarly folded
and more regional during the Cretaceous. Apart from Chalk. The conspicuous variation in the outcrop width of
a few local variations, the Chalk was deposited as a the Chalk reflects structural control, with a narrow ridge
uniform blanket across the central parts of the basin. developed where dips approach vertical along the Purbeck–
2. During the Cenozoic, crustal extension was replaced Isle of Wight Monocline and broader cuestas formed where
by south to north-directed compression related to the regional dips are shallow.

Figure 2.1 Geology 1

60 VALE OF PEWSEY
Kings
clere
of the Wessex Basin Antic
line
showing the locations
SECTION 3

of cross-sections.
contains OS data © Crown
copyright and database
40
1
rights 2017 Ordnance
Survey [100021290 EUL]. SECTION 4
Use of this data is subject to cline Weald
ur Anti line Anticline
terms and conditions. Wardo nt S y n c e n ti c li n e
ottisfo nticlin orne A
VALE OF WARDOUR ury-M Hill A Somb
Alderb Dean
Winchester
1
20 Anticline
SECTION 2
SECTION 1

Bere
Fore
st Sy
Port nclin
sdo e
wn
HAMPSHIRE BASIN Ant
iclin
e
1
00

WAREHAM BASIN N
ight
Portland-W
0
80
Monocline

0 50 km

3
60 3
80 4
00 4
20 40
4 4
60 4
80

Palaeogene Lower Chalk

Isle of Wight only, Lower Chalk and Middle Chalk
Middle and Upper Chalk Pre-Chalk

13
 N SECTION 1 Wardour Anticline S

Hog’s Back
200
WAREHAM BASIN

0
Mere Fault

–200

Key
–400 Palaeogene

Middle and Upper Chalk

–600
Lower Chalk

–800
0 10 000 20 000 30 000 40 000 50 000 60 000 70 000 80 000

N SECTION 2 S

Dean Hill
200
Anticline HAMPSHIRE BASIN

–200

Key
–400 Palaeogene
Alderbury–Mottisfont
Middle and Upper Chalk Syncline
–600
Lower Chalk

–800
0 10 000 20 000 30 000 40 000 50 000 60 000 70 000 80 000

N SECTION 3 S

Winchester Portsdown
200
Anticline Anticline

HAMPSHIRE BASIN
0

–200

Key
–400 Palaeogene

Middle and Upper Chalk

–600
Lower Chalk

–800
0 10 000 20 000 30 000 40 000 50 000 60 000 70 000 80 000

W SECTION 4 E
Avon Test Itchen
200

–200

–400
Key

Middle and Upper Chalk

–600
Lower Chalk

–800
0 10 000 20 000 30 000 40 000 50 000 60 000 70 000 80 000 90 000 100 000

Figure 2.2 Geological cross-sections through the Wessex Basin (vertical axis metres AOD, horizontal axis metres).
For locations see Figure 2.1.

14
Table 2.1 Chalk stratigraphy of southern England (after
Bristow et al., 1997). BOX 1 CHALK COMPOSITION

Chalk is a white to greyish white, very fine-grained, micropo-

Traditional rous limestone. It is composed predominantly of coccoliths
Stage Southern England (microscopic, calcareous skeletal remains of planktonic algae),
southern England
Hopson (2005)
subdivisions * but coarse-grained, bioclastic material derived from bivalves,
echinoderms and other shelly invertebrates is also present, al-
Portsdown Chalk though this typically forms less than 10 per cent of the Chalk
Formation
(Hancock, 1993). Further lithological variation in the Chalk is
Campanian Culver Spetisbury C M produced by the inclusion of:
(pars) Chalk Tarrant
Fm Chalk Member Marly Chalk and marl seams. Marly Chalk is a mixture of
chalk and clay, while marl seams are more horizon-specific
concentrations of clay.
Newhaven
Upper Chalk

Chalk Nodular chalks and chalkstones. Chalks display a continu-

Formation ous range of lithification from discrete nodules of hard chalk
Santonian through to fully lithified chalkstone.
White Chalk Subgroup

Omission surfaces. Unlithified ‘discontinuity surfaces’ are

Seaford Chalk identified by contrasts in colour and texture between succes-
Formation sive sediments and are usually penetrated by Thalassinoides
Coniacian burrows.
Top Rock Lewes Nodular Hardgrounds. Fully lithified omission surfaces are called
Chalk Formation hardgrounds and are characterised by the presence of bor-
ings and encrusting organisms and green glauconitic or brown
Chalk Rock Chalk Rock Member phosphatic mineralisation. Surface morphology can be planar,
Spurious hummocky or convolute. Nodular chalks and hardgrounds are
Chalk Rock
products of early diagenetic cementation that took place just
Middle Chalk

Turonian New Pit

Chalk Formation beneath the seafloor and are generally associated with reduced
sedimentation rates or hiatuses. They are best developed in
Holywell Nodular
thin basin-margin successions in the northwest of the Wessex
Chalk Formation
(Melbourn Rock Member)
Basin.
Melbourn Rock
Plenus Marls (Plenus Marls Member) Flints. Flints form a continuum from nodules to more or less
U
continuous sheets. They also occur in a range of styles in-
Grey Chalk Subgroup

Zig Zag
Grey Chalk
Chalk
cluding paramoudra, Zoophycos, sheet flint and the common
thalassinid horn-flints.
Lower Chalk

Formation
Cenomanian M

Chalk Marl West Melbury Marly

L
Chalk Formation BOX 2 GEOPHYSICAL PROPERTIES OF THE
Glauconitic Marl
CHALK
(Glauconitic Marl Member)
Upper Albian
(pars)
Upper Greensand Upper Greensand A significant amount of geophysical logging has been un-
dertaken in the Wessex Basin, principally for oil and gas
* Traditional Chalk subdivisions after exploration, geothermal investigations and water boreholes.
Jukes-Browne and Hill (1903, 1904, for example)
It is possible to subdivide the Chalk lithostratigraphically
based on the gamma ray, sonic and resistivity log profiles.
The example shown in Figure 2.3 shows some of the main
2.1.2 Chalk Group properties.
• The clayey Grey Chalk Subgroup, below the Plenus
For nearly a century following Jukes-Browne and Hill (1903; Marls, has high gamma-ray values relative to the White
1904) the Chalk of southern England had a simple threefold Chalk Subgroup where the gamma-ray values are low and
lithostratigraphy with Lower, Middle and Upper divisions. uniform. The upward decreasing clay and silt content as the
Today the Chalk Group of southern England is divided into West Melbury Marly Chalk passes into the Zig Zag Chalk
nine laterally persistent formations that are defined on the is usually clearly expressed as a steady decrease in gamma-
basics of contrasting, subtle lithological variations and give ray values.
rise to broad topographical features in the landscape. The • Conspicuous peaks in the gamma-ray log occur at the
distribution of these formations can be further justified on Glauconitic Marl and Plenus Marls and, where well-
developed, the Chalk Rock Hardgrounds at the base of the
the basis of their included macro-and micropalaeontology Lewis Nodular Chalks. The thicker marl units can generally
and by characteristic downhole geophysical log traces be detected by a coincidence of the higher gamma-ray
(Bristow et al., 1997; Rawson et al., 2001; Hopson, 2005). values and lower sonic velocities.
The terms Lower, Middle and Upper Chalk are abandoned
in the most modern scheme (but see Bristow et al., 1997) but The Holywell Nodular and Lewis Nodular chalks have a sig-
nificantly higher sonic velocity than the other chalk formations.
are still used in the literature as a ‘convenient’ shorthand. In the White Chalk Subgroup there is generally an overall de-
In this report, while the new stratigraphical nomenclature crease in sonic velocity corresponding to the tendency of soft-
takes precedence, the old informal names are frequently er Chalks to occur towards the top. The curve is moderately
employed. This is necessary, principally because the serrated, however, reflecting the alternation of hard and soft
historical hydrogeological studies on which much of the chalks with interbedded marls and flints. In general, the resis-
report is based used the old nomenclature, and to a minor tivity curve follows that of the sonic log.
extent because a limited area still remains to be surveyed

15
Gamma ray and sonic logs from BGS WELLOG. Resistivity logs digitised at Wallingford from paper copies (arbitrary scale).

Gamma ray Short normal RES Sonic

Depth
(mOD) 0 (API) 100 Lithological subdivision 0 (Arb) .65 150 (usecs/ft) 75
SN1
0 (Arbitrary scale) .65
60 55.23

40

20

0
PALAEOGENE

−20

−40
−53
−60

−80
PORTSDOWN CHALK
FM
−100
−113
−120

SPETISBURY CHALK
CULVER CHALK FM

−140
MBR
−160 −165

−180
TARRANT CHALK
−200 MBR

−220 −225

−240
NEWHAVEN CHALK FM
−260
−266

−280
SEAFORD CHALK FM
−300
−315
−320

−340 LEWES NODULAR

CHALK FM
−360
−374
−380 −385 NEW PIT CHALK FM
HOLYWELL NODULAR
−400 CHALK FM
−410
−415 PLENUS MARLS MBR
−420

ZIG ZAG CHALK FM

−440
−455
−460 WEST MELBURY MARLY
−476. CHALK FM
−480 −477 GLAUCONITIC MARL MBR

UPPER GREENSAND
−500
FM
−512
−520

GAULT FM
−540
−553
−560

−580 CORALLIAN

−600 −603

Figure 2.3 Geophysical logs showing the stratigraphical subdivision of the Chalk, Wessex Basin.

16
using the new system. Table 2.1 shows the relationship 2.2 CHALK GROUP LITHOSTRATIGRAPHY
between the old and new terms. Where possible the new
stratigraphical divisions are shown on Figures 1.6 to 1.11. 2.2.1 Grey Chalk Subgroup
The principal subdivision into two subgroups is made at
the base of a prominent mudstone interval, the Plenus Marls The Grey Chalk Subgroup, with the exclusion of the Plenus
Member. Below the Plenus Marls, the Grey Chalk Subgroup Marls Member, is broadly equivalent to the Lower Chalk
(Lower Chalk) comprises thin argillaceous and marly grey of the traditional scheme. Where fully developed, the Grey
and greyish white chalks, while the overlying White Chalk Chalk Subgroup is divided into two formations, a lower,
Subgroup (Middle and Upper Chalk) comprises thick units West Melbury Marly Chalk Formation and an upper, Zig
of white and nodular chalks with or without flints (Table 2.1 Zag Chalk Formation. In the field, the boundary between
and Box 1). the two units is marked by a sharp negative break in slope
with the more resistant Zig Zag Chalk being relatively
upstanding compared with the more easily weathered, softer
2.1.3 Chalk thickness variation West Melbury Marly Chalk below.
The thickness of the Chalk Group varies depending on the Throughout most of the Wessex Basin, the base of
original sedimentation conditions and the scale of post- the Chalk Group is traditionally taken at the base of the
Cretaceous erosion. A maximum of 473 m of Chalk has been Glauconitic Marl. However, in south-west Wiltshire the
recorded beneath Palaeogene sediments in the Hampshire Basin hiatus is marked by glauconitic calcareous sandstones and
(in the Lymington Borehole [SZ 30 96]), however, even below the base of the Chalk Group is taken at the base of the marl-
this protective cover there is significant variation partly related poor Melbury Sandstone Member that is the local equivalent
to pre-Palaeogene erosion that removed the youngest Chalk of the Glauconitic Marl Member. Over the ‘Mid Dorset
formations. At outcrop to the north of the Hampshire Basin, the Swell’, the West Melbury Marly Chalk is absent and the
Chalk is generally less than 300 m thick, thinning progressively base of the Chalk Group is diachronous and coincides with
toward the primary escarpment as the erosion level cuts deeper the Cenomanian Basement Bed.
into the stratigraphy. This general pattern is complicated by In the west, the thickness of the Lower Chalk ranges from
thinning over anticlinal axes in the centre of the basin (e.g. the 26 m at Batcombe to over 70 m in the Salisbury area, but
Winchester Anticline), cumulating in the total removal of Chalk for the most part, the thickness is between 40 and 50 m.
over the Weald and Wardour anticlines (Figure 2.4) The formation is up to 90 m thick east of Winchester and
Comparison of geophysical logs (see Box 2) from deep up to 95 m thick on the scarp overlooking the Weald in the
boreholes demonstrates a general regional thinning of a Alresford area.
number of the constituent formations from Sussex north and
westward into Hampshire and Wiltshire, reflecting the move 2.2.1.1 West Melbury Marly Chalk Formation
from basinward thicker successions in the south to areas The West Melbury Marly Chalk consists of rhythmically
characterised by thinner ‘shelf’ deposition. Figure 2.5 shows layered alternations of off-white to creamy marls, greyish
the thickening of the Chalk units from the Wessex shelf to the Chalks and pale grey to brown limestones. The top of the
Sussex trough. member is taken at the top of the Tenuis Limestone where

Figure 2.4 Chalk

thickness (metres) in
1
60 0
the Wessex Basin. 100 0
contains OS data © Crown
200
copyright and database
rights 2017 Ordnance
Survey [100021290 EUL].
Weald
Use of this data is subject to 1
40 Anticline
terms and conditions. 300

200
Wardour Anticline 400
100
400 Winchester
20
1 Anticline

400

1
00

N
30 300
0

0
40

200 0 10200
0 400 0
0
80
100 0

0 50 km

60
3
80
3 4
00 4
20 40
4 4
60 4
80

Chalk thickness contours (m) Middle and Upper Chalk Pre-Chalk

Palaeogene Lower Chalk Isle of Wight only, Lower and Middle Chalk

17
 Wessex Basin logs South Downs logs
Marchwood 1[SU31SE227]
B (Horton Heath) Gamma ray SFL−Resistivity
A (Wareham)

Chalk
Gamma ray Short normal RES

division
Lithological 0 (API) 50 0 (ohm.m) 40
Depth Gamma ray LLD Resistivity subdivision
0 (API) 60 0 (Arb) .6

Chalk
division
(mbd) 0 (API) 50 0(ohm.m)100

– 20

– 40 PALAEOGENE
– 60

– 80 Horndean
[SU71SW62]
– 100 – 108
Padnall 16− inch RES
– 120 0 Arb 1
– 128
PORTSDOWN Padnall RES1
– 140 CHALK FM

Chalk division
0 0 Arb 1
– 160
– 168
– 244
– 180
– 193 SPETISBURY
– 200 – 204 CHALK MBR
Sompting [TQ10NE80]

Wessex Basin to Sussex (after Bristow et al., 1997).

– 220 – 220 – 293 – 66
Gamma ray 16− inch RES
Chalk

– 240
division

TARRANT
0 (API) 50 0 (ohm.m) 150
– 257 CHALK MBR

CULVER CHALK FM
– 260

– 280 – 280 – 38

18
– 140
– 300 NEWHAVEN – 370
CHALK FM
– 320 – 317
– 321 Base
– 400
– 340 Base Seaford
SEAFORD Base rd
CHALK FM Seafo – 204.6 Chalk – 106
– 360 – 358 Seaford
Chalk Chalk
– 370 – 438
– 380 Base
Seaford LEWES
– 400 Chalk
– 406 NODULAR
– 266 – 168
– 420 CHALK FM
– 490

Figure 2.5 Gamma ray and resistivity logs showing expansion of Chalk units from the
– 429
– 440 – 438
– 448 – 440 NEW PIT CHALK FM
HOLYWELL – 517 – 300 – 205
– 460 NODULAR CHALK
– 471. – 465 FM – 536 – 323 – 225.
– 475 – 470 PLENUS MARLS MBR – 539 – 326 – 230
– 480 Plenus Marls Plenus Marls Plenus Marls
– 500 ZIG ZAG CHALK FM
– 360 – 265
– 520 – 520 – 510 – 582
WEST MELBURY
Glauconitic MARLY CHALK FM
– 540 Marl – 533
– 549.5 UPPER GREENSAND Glauconitic – 614
– 560 FM
Marl
– 574
Glauconitic M – 326
– 567 arl
– 580 – 581 – 642

GAULT FM – 350
– 600

Logs are aligned on Plenus Marls horizon. Log depths are metres below logging datum (mbd). Key to lithology columns is shown for Borehole B.
this is present, or at the erosion surface beneath the Coast Pit Chalk includes the Glynde Marls, which are of great value
Bed of the overlying Zig Zag Chalk Formation. The West for identifying the top of the New Pit Chalk on geophysical
Melbury Marly Chalk is absent over the Mid Dorset Swell, logs. The New Pit Chalk is 10 to 16 m thick in the west and
where it is represented within a phosphatic, ammonite-rich up to 45 m in the east of the basin.
conglomerate (Cenomanian Basement Bed or ‘Bookham
Conglomerate’) at the base of the overlying Zig Zag Chalk. 2.2.2.3 Lewes Nodular Chalk Formation
The Lewes Nodular Chalk comprises hard to very hard
2.2.1.2 Zig Zag Chalk Formation nodular chalk and hardgrounds with interbedded soft to hard
The Zig Zag Chalk comprises soft to medium hard, greyish gritty chalks. The base of the Chalk Rock Member within
blocky chalk with marl–limestone rhythms in the lowest the Lewes Nodular Chalk is the traditional base of the Upper
part. The base of the formation is placed at the base of the Chalk. However, strong nodularity and flint seams occur
Cast Bed that overlies the preceding Tenuis Limestone in lower in the succession, in the thicker, basinal parts of the
full successions; the base of the Plenus Marls Member Wessex Basin, and the base of those units forms the base of
marks the top. The Plenus Marls are present along the the Lewes Nodular Chalk.
whole outcrop of the Lower Chalk. They range from 2 to The basal Lewes Nodular Chalk becomes condensed
6 m thick and consist of an alternating sequence of blocky, towards the western margin of the depositional basin and the
white chalks and silty marls up to 0.5 m thick (Bristow nodular chalks are replaced progressively by chalkstones,
et al., 1995). The marls give a high gamma-ray response mineralised hardgrounds and marl seams of the Chalk
and are readily recognisable on geophysical logs. Rock which are best developed in Wiltshire, Berkshire
The Zig Zag Chalk ranges from 20 m thick in south Dorset and the Chilterns (Bromley and Gale, 1982). Individual
to 75 m thick in the east of the basin around Winchester. The hardgrounds within the Chalk Rock are recognisable by
lower part of the Zig Zag Chalk was not deposited over the hardground surface morphology and colour and have been
Mid Dorset Swell and the formation rests unconformably named and correlated over large areas by Bromley and
on the pitted and phosphatised top of the Upper Greensand; Gale (1982). Hardgrounds at the base of the Chalk Rock
its base is marked by the Cenomanian Basement Bed (Ogbourne and Pewsey hardgrounds) are present across
(Drummond, 1970) on the coast, which is the equivalent much of the western Wessex Basin in Wiltshire, Dorset and
of the Bookham Conglomerate of the Shaftesbury district Hampshire, while those at the top (Fognam and Hitch Wood
(Bristow et al., 1995). hardgrounds) show a significant geographical shift to the
north and east (Figure 2.6).
The Hope Gap Hardground occurs 5 to 10 m above
2.2.2 White Chalk Subgroup
the top of the Chalk Rock; it is another well-developed
The White Chalk comprises white chalks and nodular hardground (the ‘Upper Rock’ of White, 1923), but not
chalks, with or without flints, divided into seven formations. so strongly mineralised and correlates with the Top Rock
It encompasses the traditional Middle and Upper Chalk with of East Anglia. The Top Rock/Hope Gap Hardground is
the addition of the Plenus Marls Member at the base. The succeeded by up to 30 m of hard, nodular flinty chalk and
Holywell Nodular Chalk and smooth chalks of the New chalkstone, locally porcellanous, but with some interbeds
Pit Chalk form a readily identifiable succession across the of firm, white chalk. The nodularity and hardness decrease
whole of southern England. These two formations, which upwards and there is a transitional zone of a metre or two
are generally flint free, are for the most part equivalent to with the overlying Seaford Chalk.
the Middle Chalk. The Lewes Nodular Chalk and younger The Lewes Nodular Chalk is about 25 to 40 m thick
smooth white chalk (Seaford to Portsdown) without over much of the Wessex Basin, but reaches a thickness of
significant flint and marl seams are equivalent to the Upper 50 m near Blandford Forum. In the Wytch Farm boreholes,
Chalk of the traditional scheme. the thickness varies from 20 to 26 m. In the Winchester,
In Dorset, the Middle Chalk maintains a fairly constant Fareham and Alresford districts, the Lewes Nodular Chalk
thickness of 25 m, thickening eastwards to 65 m east of is estimated to be 60 to 65 m thick. At the western end of the
Winchester and up to 80 m on the western scarp of the Weald. Isle of Wight, the thickness is about 40 m.
The Upper Chalk is thickest in the south of the Wessex Basin,
where it is up to 365 m thick. The general northwards and 2.2.2.4 Seaford Chalk Formation
north-eastwards thinning is mostly a result of pre-Palaeogene The Seaford Chalk is a firm, white, flinty chalk with few
erosion. The Upper Chalk forms two prominent outward- marl seams (except near the base), overlain by similar chalk
facing escarpments around much of the outcrop. with common marl seams, the Newhaven Chalk. At the type
locality in Sussex, the base of the Seaford Chalk is taken at
2.2.2.1 Holywell Nodular Chalk Formation the base of Shoreham Marl 2 (Mortimore and Wood, 1986).
The Holywell Nodular Chalk comprises hard nodular Conspicuous, semi-continuous bands of large nodular flints
chalks with abundant shell debris and greyish green wispy are a feature of this member. The member consists of at
(or flaser) marl seams separating the nodules. The Melbourn least 80 m of firm, white chalk with regular courses of flint
Rock Member and the Plenus Marls Member form the base nodules. Flints in the basal part of the member are typically
of the Holywell Nodular Chalk. The Melbourn Rock forms carious. The upper boundary is delimited upwards by the
a marked lithological contrast to the underlying Plenus entry of marl seams marking the base of the succeeding
Marls and is one of the easiest of the Chalk boundaries to Newhaven Chalk. In a limited area south of Dorchester,
map in the field and to recognise on geophysical logs. The however, the succession differs, perhaps reflecting the
Holywell Nodular Chalk thickens across the basin from continuing influence of the Mid Dorset Swell structure and
15 m at the western outcrop to 35 m in the east. the undivided Seaford and Newhaven Chalk, together with
part of the Tarrant Chalk, consists of an alternating sequence
2.2.2.2 New Pit Chalk Formation of hard, nodular chalk and firm, blocky chalk. In that area,
The New Pit Chalk comprises smooth, firm, white chalks the three units have not been separated and are mapped as
with marl seams. The upper part of the newly defined New one unit, the so-called Blandford Chalk.

19
Figure 2.6
Distribution of
hardgrounds in the
Chalk Rock in the
Wessex Basin (after Limit of Fognam
Hardground
Bromley and Gale,
1982.
contains OS data © Crown
copyright and database Limit of Hitch Wood
rights 2017 Ordnance Hardground
1
40
Survey [100021290 EUL].
Use of this data is subject Limit of Ogbourne
to terms and conditions. Hardground

Limit of Pewsey
Hardground

0
90 N

0 50 km

3 4 4
80 30 80

Hitch Wood Hardground Palaeogene Lower Chalk Isle of Wight only

Fognam Hardground Middle and Upper Chalk Pre-Chalk Lower and Middle Chalk
Pewsey Hardground
Ogbourne Hardground

2.2.2.5 Newhaven Chalk Formation Purbeck structure, the steeply dipping Tarrant Chalk is
The Newhaven Chalk is a unit of firm, white, predominantly extremely hard and has not been distinguished lithologically
marly chalk with widely spaced flints and common marl from the Newhaven Chalk below or the Spetisbury Chalk
seams. At the type locality at Seaford Head, Sussex, the above. South of Dorchester, the succession differs. There,
base is taken at the base of Buckle Marl 1 (Mortimore and the Tarrant Chalk consists of an alternating sequence of
Wood, 1986). This stratotype section, however, is situated hard, nodular and firm, blocky chalk that has not been
close to a positive feature where marl seams are either separated from the combined Seaford and Newhaven Chalk.
relatively poorly developed or may virtually disappear The thickness of the Tarrant Chalk is mostly in the range 50
(Mortimore and Wood, 1986). In trough areas, such as to 60 m, but in the east, in the Fareham district, the thickness
around Salisbury and Brighton, not only are marl seams is only 30 to 40 m. At outcrop at the western end of the Isle
better developed, but additional strong marl seams occur of Wight, the member is about 36 m thick, but appears to be
higher in the sequence (Mortimore, 1983; Mortimore and 57 m thick in the Wilmingham Borehole [SZ 36 87].
Wood, 1986). In such situations, the upper limit of the 2.2.2.7 Culver Chalk Formation (Spetisbury Chalk
Newhaven Chalk is sharply defined both in field sections Member)
and in boreholes. In the field, the base of the member is
taken at a negative feature break. Locally (western side of The Spetisbury Chalk consists of firm, white chalk with
the Dorchester district) the boundary between the Seaford large flints, including the tabular form, in the lower part and
and Newhaven Chalks is taken at a positive feature within Zoophycos flints in the higher part. The exact stratigraphical
the Marsupites testudinarius Zone (Westhead, 1992) even level of the feature-forming beds has not been established
though the base of the Newhaven Chalk is at a slightly in the type area, but, in Sussex, is inferred to occur at about
higher level than in the type area. In south Dorset and in the level of the Whitecliff Flint, i.e. at the boundary between
particular the coastal exposures, the Newhaven Chalk is the Sompting and Whitecliff Beds (Mortimore and Wood,
commonly nodular and has only weakly developed marl 1986; Mortimore, 1986). The Spetisbury Chalk is equivalent
seams. South of the Purbeck–Isle of Wight Monocline in to the upper part of the Culver Chalk (i.e. the Whitecliff
Dorset, the chalk is extremely hard. The Newhaven Chalk Beds of Mortimore and Wood, 1986). The member has a
ranges from 40 to 80 m thick. characteristic sonic velocity log signature (Figure 2.3),
which suggests that its top is marked by a hardground or
2.2.2.6 Culver Chalk Formation (Tarrant Chalk Member) prominent layer of flints. Both at outcrop and at depth across
The Tarrant Chalk comprises firm, white chalk with the Wessex Basin, the member is generally between 60 and
relatively widely spaced, large flint bands. Along the 70 m thick, but is only some 40 m thick in the east.

20
2.2.2.8 Portsdown Chalk Formation of either steep dip or faulting, its outcrop is narrow or
The Portsdown Chalk is the youngest member of the Upper absent. The formation consists predominantly of clays and
Chalk. It crops out from beneath Palaeogene strata in an sandy clays with minor interbedded sand. The sand content
arc across the central, south-western and southern parts of of the London Clay Formation generally increases upwards
the district. It consists of up to 90 m of soft to firm, white with the upper 30 to 50 m developed as a sequence of thinly
chalk; flints are sparse in the lower part of the member, interbedded sands and clays. The formation thins from
and there are some large intervals (exceeding three metres) around 100 m in the eastern part of Hampshire to around
without flints. Marl seams are common in the lower part of 50 m or less in the west. Around Wareham, in the west, the
the member, and it is presumably their relatively common London Clay rests directly on Chalk and the base of the
occurrence in the lower part of the sequence, associated group is developed as a sand overlain by reddened clays of
with softer chalks, that give rise to the steep scarp face that the West Park Farm Member that were previously mapped
characterises the Portsdown Chalk. as ‘Reading Beds’ (Figure 2.8).

2.3.4 Bracklesham Group

2.3 PALAEOGENE STRATA
The Bracklesham Group encompasses the strata between
Palaeogene strata cover significant areas of the Wessex the top of the London Clay and the base of the Barton Group.
Basin (Figure 2.1); for example in the synclinal Hampshire In the western half of the Hampshire Basin, the group is
Basin they comprise a layered succession, up to 652 m dominated by the thick sands of the Poole Formation and
thick, of weakly consolidated sands and clays, with minor Branksome Sand. The Poole Formation reaches up to 160 m
limestone, lignite and flint gravel. The Palaeogene deposits thick and in general consists of four separate sand members
are separated from the Chalk by a major time gap of some separated by clays. The Branksome Sand is around 70 m
5 to 15 million years, but the two highly contrasting units thick and consists of alternating clean sands and thinly
are largely structurally conformable, both being folded bedded sands and clays. The Bracklesham Group changes
during the main Miocene inversion event. The major time facies in the eastern half of the Hampshire Basin where it
gap between the Chalk and the Palaeogene was a period consists of thinly laminated sands and clays, glauconitic
of regional uplift, sea-level fall and subaerial erosion of shelly sand and fine-grained clayey sand subdivided into
the exposed Chalk surface under subtropical climatic the Wittering, Earnley, Marsh Farm and Selsey formations.
conditions. Basal Palaeogene deposits infill valleys and
solution holes cut into the upper surface of the Chalk.
The complexities of Palaeogene stratigraphy result from
deposition in fluctuating nonmarine and shallow-marine CHRISTCHURCH
environments during successive coastal transgression 0 GR 180

and regressions. The stratigraphy is best known from the Barton Clay
spectacular coastal outcrops at Alum Bay and Whitecliff
Bay on the Isle of Wight. Inland, exposures are few but Boscombe Sand

numerous boreholes, many of which have geophysical

logs, provide stratigraphical control. The gamma-ray log is
particularly effective at discriminating the alternating sand Branksome Sand
and clay formations that characterise the Palaeogene (e.g.
50 m
Figures 2.7, 2.8 and 3.18).

2.3.1 Thickness trends and structure WAREHAM

0 GR 55
The Hampshire Basin is an elliptical, highly asymmetrical, Oakdale
west–east trending syncline with a vertical southern limb
Poole Formation

Sand
and a gently dipping northern limb. Palaeogene deposits Member
Poole Formation
attain a maximum thickness of 652 m in the Sandhills Creekmoor
Clay
Borehole on the Isle of Wight [SZ 45 90] and thin Member
progressively toward the western and northern margins of Creekmoor
Sand
the outcrop as the present-day erosion level cuts deeper into Member
the stratigraphy. Folding associated with the Portsdown
London Clay

and Dean Hill anticlines complicates this general structural Christchurch

pattern. Member

West Park
Farm
2.3.2 Reading Formation Member London Clay
Formation
The Reading Formation reaches a maximum thickness of
80 m and is a highly variable succession of red-mottled
clays, sands and gravels resting on a glauconitic sandy base.
The formation thins to the west and is not present west of
Wareham, where the London Clay rests directly on the Chalk.
Reading Formation
27.5 km
Upper Chalk
2.3.3 London Clay Formation
The London Clay has a fairly wide outcrop on the northern Figure 2.7 Palaeogene correlation using gamma-ray
and western margins of the basin, but in the south, because logs, Wareham Basin, Dorset.

21
 100
West East

RIVER FROME
RIVER PIDDLE
50

0 79
0 55 0 100
OAKDALE SAND
0
Elevation (m)

CREEKMOOR CLAY

CREEKMOOR SAND
–50

LONDON CLAY

CHALK GROUP
WEST PARK
FARM MEMBER
–100

–150
0 5 000 10 000 15 000 20 000 25 000 30 000
Distance (m)

Figure 2.8 West–east section through the Palaeogene deposits of the Wareham Basin based on correlation of borehole
gamma ray logs.

2.3.5 Barton Group several points along the coastline and in evidence of intense
periglacial activity in the form of soliflucted sediments.
The Barton Group outcrops widely in the central part of the There is a considerable time gap, broadly corresponding
Hampshire Basin. Over most of its outcrop, the lower part of to the Neogene, between the deposition of the youngest
the group (Barton Clay) consists of glauconitic sandy clays Palaeogene strata and the oldest Quaternary deposits in
with a large molluscan fauna. In the upper part there is a the Wessex Basin. During this interval the Chalk and its
transition through the clayey fine-grained sands of the Chama Palaeogene cover was folded, uplifted and extensively
Sand to the relatively clay-free sands of the Becton Sand that eroded. Superficial deposits in the Wessex Basin are shown
forms the top of the Barton Group. The Barton Clay varies in in Figure 2.9.
thickness from about 50 to 90 m, and the combined thickness
of the Chama Sand and Becton Sand can vary from about 30
to 100 m (Edwards and Freshney, 1987). 2.4.1 River terrace deposits
River terrace deposits occur along the flanks of the major
2.3.6 Solent Group rivers. They typically parallel the channels of the present
rivers and fall only very gently downstream. The terrace
The Solent Group is around 130 m thick and is subdivided deposits are generally less than 5 m thick and consist
into three formations, the uppermost of which just extends dominantly of gravels made up from angular, with some
into the Oligocene. The Headon Hill Formation forms rounded, flints and sand. High-level terraces typically form
the base and is a 90 m thick succession of shelly sands, flat-topped or gently inclined caps to interfluves. Many
muds, marls, lignites and thin limestones. The overlying cannot be related to the modern drainage system, with
Bembridge Limestone Formation is a 8.5 m- thick sequence those bordering the Solent being used to support the former
of pale brown, freshwater limestones, marls and muds. The existence of an easterly flowing ‘Solent River’ which pre-
uppermost Bouldner Formation is a thick unit of clays and dates breaching of the Purbeck–Isle of Wight Monocline
silts, with occasional thin sands, marls and limestones up to and present-day marine flooding. High-level terraces have
34 m on the mainland and as much as 90 m is preserved on undergone weathering and degradation by crytobation and
the Isle of Wight. may have an appreciable clay content.

2.4 QUATERNARY 2.4.2 Head

Head is a heterogeneous group of superficial deposits that
The Quaternary Period saw many oscillations of climate. have accumulated by solifluction, hillwash and hillcreep.
During several of the colder episodes extensive ice sheets Essentially it is a very gravelly, silty, sandy clay to clayey
pushed southwards across England, but they did not reach sandy gravel, with variable proportions of coarser granular
the Wessex Basin, and so there are none of the till sheets material. The clasts are primarily of large nodular and
that elsewhere prove valuable in unravelling the Quaternary coarse gravel-sized flint. It is regarded as a periglacial
sequence. Instead, the Quaternary record is preserved deposit resulting from the solifluction of Chalk, Palaeogene
primarily in river terraces, in records of high sea levels at and clay-with-flints material.

22
Figure 2.9 1
60
Map of superficial
deposits in the Wessex
Basin. !
Andover Basingstoke
contains OS data © Crown ! Wells !
copyright and database 1
40
rights 2017 Ordnance
Survey [100021290 EUL].
Use of this data is subject
Salisbury
to terms and conditions. ! !
Winchester

1
20
Yeovil
!
!
Southampton

1
00 !
Portsmouth

!
Dorchester ! Bournemouth ! Newport N

0
80
! Weymouth

0 50 km

0
60 3
60 3
80 4
00 4
20 4
40 4
60 4
80

Alluvium Head Langley Silt Formation No superficial deposits

Marine deposits Clay-with -flints Peat Rivers
River terrace deposits Brickearth Tufa

2.4.3 Clay-with-flints forms the flat top to hills and long dip-slope spurs. Deposits
Clay-with-flints is primarily a deposit created by modification are estimated to be between 2 and 8 m thick, but will be much
of the original Palaeogene cover and solution of the underlying thicker over the solution pipes that may extend 10 m or more
Chalk. It is typically composed of clays and sandy clays into the underlying Chalk.
containing abundant flint nodules and pebbles. In general it

23
3 Groundwater flow in the Chalk aquifer

3.1 HYDRAULIC PROPERTIES OF THE AQUIFER and Alre catchments in Hampshire have been investigated in
detail, while information is lacking on the aquifer properties
3.1.1 Matrix characteristics of other areas, such as on Cranborne Chase and the Dorset
Downs. Much of the following discussion is based on the
Chalk is a predominantly soft, fine-grained limestone data and analysis in Allen et al. (1997).
formed from the shells of marine organisms and can be
composed of up to 99 per cent calcium carbonate. Other 3.1.2.1 South Dorset
shells, detrital quartz grains and chert nodules (in the form of The Chalk of south Dorset is the most deformed in England.
flint) contribute small amounts of silica to its composition. The Purbeck Monocline, which is the dominant structure in
Small proportions of clay minerals, glauconite and calcium the area, and associated faults and folds (which extend over
phosphate are also present. the outcrop of the Chalk) are significant in controlling the
Chalk porosity varies between about 5 per cent and 45 per aquifer properties of the area.
cent and is related to stratigraphy (Price, 1987; Bloomfield Laboratory work, including porosity tests, has shown that
et al., 1995). For example, the Upper Chalk of southern the Chalk porosity varies with stratigraphical dip, highly
England has an average porosity of 39 per cent, the Middle folded chalk having a lower porosity and intergranular per-
Chalk an average porosity of 28 per cent and the Lower meability than undisturbed chalk (Alexander, 1981). Mimran
Chalk a porosity of 23 per cent (Bloomfield et al., 1995). (1976) studied the structure of the Purbeck Monocline and
Changes in porosity are also reflected in differences in the suggested that up to 30 per cent compaction might have taken
intact dry densities of the Chalk. In southern England, the place, with a resulting porosity of only 5 per cent (compared
average dry density of the Upper Chalk is 1650 kg m–3; in the with average porosity values of 35 to 44 per cent). It is evi-
Middle Chalk it is 1910 kg m–3 and in the Lower Chalk it is dent therefore that folding has changed the chalk from a soft,
2080 kg m–3 (Bloomfield et al., 1995). Figure 3.1 highly porous material into harder limestone.
illustrates the reduction in average porosity with depth in In the Mimran (1976) study, it was observed that,
the Chalk at the Totford Borehole [SU 57 38] determined generally, where the Chalk has been tectonically hard-
by core sample analysis. At Totford, porosity reduces by ened, values of transmissivity and storage coefficient are
about 1.5 per cent every 10 m below ground level. Local reduced, suggesting that the fracture distribution within the
contrasts in porosity are due to variations in the nature of Chalk has also been altered. The mechanism for this is un-
the original sedimentation. Generally, lower porosities are known but is possibly a result of chemical and mechani-
associated with more clay-rich chalks and with nodular cal diagenetic processes closing the fractures. However,
and hardground bands such as the Melbourn Rock. faulting associated with the folded Chalk appears to have
However, the overall reduction in porosity with depth is
due to differences in the degree of diagenetic compaction,
with the older sediments subject to greater maximum 0
overburden and so to a greater degree of mechanical and
chemical compaction (Bloomfield, 1997). Upper Chalk
Despite its commonly high porosity, pore sizes in the Middle Chalk
Chalk are generally very small. Matrix (intergranular)
hydraulic conductivity is controlled by pore throat size and 20

this is typically less than one micron (Price et al., 1976). This
results in very small values of matrix hydraulic conductivity.
For example, an average value of hydraulic conductivity of
6.3 x 10–4 m d–1 was determined from measurements on over
Depth (m bGL)

40
900 UK Chalk samples, of which around 99 per cent had
values of less than 10–2 m d–1 (Allen et al., 1997).

3.1.2 Hydraulic conductivity, transmissivity and

storage 60

Investigations of aquifer properties in the Wessex Basin

have used many different techniques, for example pumping-
test interpretation, groundwater-recession curve analysis,
groundwater-model calibration, water-level data interpretation
80
and tracer-test analysis. Most of the following regional
discussion however is based on the results of pumping-test
analysis. Further interpretations from catchment groundwater
modelling are included in Chapter 4. 20 25 30 35 40 45 50
From a review of literature, it is notable that aquifer Porosity (%)

properties data are not distributed evenly over the Wessex

Basin. For example, parts of south Dorset and the Candover Figure 3.1 Porosity profile for the Totford Borehole.

24
a significant effect in increasing transmissivity locally. The rather more predictable. Groundwater levels are a reflection
faults have provided a preferential zone for rapid groundwa- of topography and transmissivity reverts to a more classical
ter flow, which has been further enhanced by dissolution. An model of low values in the interfluves and high values in
example of faulting increasing the aquifer transmissivity is the valleys. Transmissivity data from valley sites indicate
seen in the Lulworth area (Houston et al., 1986). values ranging from several hundred to over 3000 m2 d–1
In the Allen et al. (1997) study, pumping test data were and storage coefficients of 2 x 10–4 to 0.06 (Allen et al.,
collected from 28 locations in the south Dorset area. Thirty- 1997). A range of small pumping tests have been carried
eight pumping tests were recorded, giving 38 estimates of out on interfluve locations throughout the west of the south
transmissivity and 19 of the storage coefficient. Dorset areas (Alexander, 1981). Although these tests are
Transmissivity data varied from 0.8 to over 3000 m2 d–1, probably unreliable due to their short duration, they do give
with a geometric mean of 210 m2 d–1 and median of a general indication of the variation of aquifer parameters.
330 m2 d–1. Storage coefficient data varied from 9 x 10–5 to Transmissivity values ranging from 0.04 to 340 m2 d–1 have
0.064 with a geometric mean of 0.003 and median of 0.0039. been measured, although typically values were less than
The majority of the Chalk data were from the unconfined 50 m2 d–1.
area, although Pleistocene deposits in the outliers resulted in In the West Knighton area (south-east of Dorches-
confined conditions. The high storage values were believed ter), where the Chalk becomes confined by Palaeogene
to have been influenced by the Upper Greensand aquifer in deposits, high transmissivity values have been recorded
hydraulic continuity with the Chalk. (>2000 m2 d–1). Low hydraulic gradients are associated with
The effective base of the aquifer locally is taken to be the high transmissivity, and it is possible that the aggres-
the middle of the Upper Greensand, at the top of the sive water from the Palaeogene through-flow has enhanced
Exogyra Sandstone. This horizon is well cemented and has the solution of fractures and led to the development of high
a low permeability. The Chalk Marl (West Melbury Marley transmissivity. It is anticipated that transmissivity will re-
Chalk Formation and basal Zig Zag Chalk Formation) at duce with increasing degree of confinement.
the base of the Lower Chalk (Grey Chalk Subgroup) also
has low permeability (Robins and Lloyd, 1975; Alexander, 3.1.2.2 Cranborne Chase and the Dorset Downs
1981), but the lack of springs emerging at the top of the The Allen et al. (1997) study of aquifer properties obtained
marls implies a large component of vertical leakage to the data from 11 locations in the Cranborne Chase area.
Upper Greensand. On the scarp slopes, most of the springs Sixteen pumping tests were recorded, giving 16 estimates
emerge from within the Upper Greensand at the top of the of transmissivity and eight of the storage coefficient.
Exogyra Sandstone. The Chalk Rock and Melbourn Rock Transmissivity data varied from 0.2 to 20 000 m2 d–1, with a
can also be important flow horizons when near to the water geometric mean of 1600 m2 d–1 and median of 2800 m2 d–1.
table and minor flows can be detected when these horizons Storage coefficient data vary from 2 x 10–4 to 0.057 with a
are found at depth. As found in other chalkland areas, the geometric mean of 0.0044 and median of 0.0055. The area
groundwater does not appear to circulate deeply within the had very few data points to describe the aquifer properties
aquifer; rather, the most important flow horizons are within and this limited information is very biased towards high
the top 50 metres of the water table where fractures and values in the valleys, therefore the statistics should be
bedding planes have been enlarged by solution. However, treated with caution.
sea level changes during the Pleistocene induced large water The lack of hydrogeological data for Cranborne Chase
table fluctuations near the coastlines where up to 100 m of and the Dorset Downs makes it difficult to assign an effective
effective aquifer may now exist (Edmunds et al., 1992). aquifer thickness to the Chalk. As with other areas of the
Chalk, the most productive sections are probably in the top
Lulworth 50 m of the aquifer where the most significant fractures are
The transmissivity distribution in the Chalk to the east of the present.
Lulworth area is strongly controlled by geological structure. The distribution of the hydraulic properties of the Chalk
As a consequence of the tectonic hardening of the Chalk and appears to conform with the topographical model of high
associated reduction in intergranular porosity, permeability transmissivity and storage in the valleys, with lower values
and open fractures, the general transmissivity and storage on the interfluves. The unsaturated zone is quite thick in
of the Chalk has been reduced. Transmissivity values from this area and the water table is situated in the Middle/
pumping test sites generally vary from around 250 m2 d–1 to Lower Chalk (basal White Chalk/Grey Chalk subgroups).
about 1000 m2 d–1 with typical values from 300 to 500 m2 d–1 Solution-enhanced fracture development may therefore be
(Allen et al., 1997). The calculated storage coefficient for somewhat inhibited. It is believed that the transmissivity
various tests is very low, usually about 1 x 10–4. There are of the Chalk is better developed at the periphery of the
no pumping test data from the Purbeck Hills on top of the Palaeogene deposits where acidic run-off has developed
monocline. However, it is likely that the aquifer properties fracture arrays.
will be poor as a result of the steep inclination of the bedding Between Shaftesbury and Westbury, the Boyne Hollow
planes and the low water levels. Chert occurs intermittently at the top of the Upper Greensand,
In certain areas, where fracture zones have been enlarged beneath the Lower Chalk (Grey Chalk Subgroup) and can
by solution processes to form groundwater conduits, act as a horizontal drain to vertical leakage through the less
significantly enhanced transmissivity is observed. Pumping permeable marls.
tests in high transmissivity zones indicate values ranging 3.1.2.3 Salisbury Plain
from around 1500 m2 d–1 to in excess of 2500 m2 d–1 and
storage coefficients of 0.0025 to 0.04. In this area, transmissivity data collected by the Allen
et al. (1997) study varied from 50 to 8200 m2 d–1,
Frome and Piddle catchments with a geometric mean of 1400 m2 d–1 and median of
The Frome and Piddle catchments lie to the north of the 1600 m2 d–1. Storage coefficient data varied from 1 x
structurally complex Lulworth area and the aquifer becomes 10–4 to 0.05 with a geometric mean of 0.0052 and median

25
of 0.0099. Insufficient data were obtained to enable the Chalk in valleys but, together with storage coefficients,
relationship between transmissivity and specific capacity are thought to decrease away from the valley up onto the
for the area to be described. interfluves.
A study conducted by the Avon and Dorset River Geological structure may play both a direct and indirect
Authority and others (1973), as part of the Upper Wylye role on the development of aquifer properties. Some valleys
investigation, used borehole geophysics to investigate the develop along the axes of synclines or faults (Southern Water
aquifer properties. The results of this study showed that 90 Authority, 1979), and aquifer properties may develop there
per cent of the flow came from the top 47 m of a borehole as in the typical topographical model. Giles and Lowings
that included Middle as well as Upper Chalk (lower White (1990) suggested that higher yields develop along the axes
Chalk Subgroup). Again, as observed in other areas, the of denuded synclines rather than anticlines. This is evident
most productive section of the aquifer appeared to be the on Figure 3.2, where groundwater contours are plotted
uppermost 50 m or so. The Chalk Marl (lower Grey Chalk with the trends of major fold axes. Groundwater mounds,
Subgroup) appears to be unproductive and may define the indicating low transmissivity, are observed to correlate well
base of the Chalk aquifer. with anticlinal axes.
From the limited dataset analysed by Allen et al. (1997), Finally, the nature and extent of Palaeogene deposits will
it appears that boreholes located in the Lower Chalk (Grey also influence the development of aquifer properties, for
Chalk Subgroup) have a lower transmissivity than those example fracture apertures in the Chalk will be small under
located in the Upper and Middle Chalk (White Chalk significant Palaeogene overburden, resulting in lowered
Subgroup). Valley pumping tests in the Upper and Middle transmissivities.
Chalk yielded transmissivity values ranging from around In East Hampshire, across the groundwater divide
450 to nearly 7000 m2 d–1, with most results between 700 towards the Weald, the aquifer properties are poorer and
and 1000 m2 d–1. Pumping tests carried out in the Lower transmissivities are usually less than 500 m2 d–1. This reflects
Chalk and Upper Greensand indicate transmissivity values the stratigraphy, where groundwater flow is predominantly
ranging from around 100 m2 d–1 to 1500 m2 d–1 but most through the Lower Chalk, which is more clayey and less
results were within the range 100 to 300 m2 d–1. fractured than the Middle and Upper Chalk and has lower
Halcrow (1992) calculated transmissivity and storage transmissivity and storage potential. However, in the north on
coefficient as part of a modelling programme to investigate the scarp slope facing the London Basin, the aquifer properties
the effects of groundwater abstraction on river flows in the improve. Tests within valleys here show transmissivity and
Upper Hampshire Avon. From this study, the transmissivity storage coefficients that are comparable to dip slope valley
was estimated to range from 250 m2 d–1 in the interfluves sites on Upper Chalk (approximately 1000 m2 d–1).
to 2500 m2 d–1 in the valleys. The storage coefficient varied
from 0.001 over the interfluves to 0.15 in the valley bottoms.
3.1.2.4 Hampshire
Chalk 60 70 SU
Data from 29 locations in the Hampshire area were ELLISFIELD
obtained during the Allen et al. (1997) study. Many of these Upper Greensand
N 120
100
locations had been researched intensively during previous 60
Groundwater
contour (mAOD) 120
investigations. Consequently, 63 pumping tests had been Anticline 100
undertaken giving 63 estimates of transmissivity and 53
Syncline 140
of the storage coefficient. Transmissivity data varied from 0 5 km Axford
120

0.55 to 29 000 m2 d–1, with a geometric mean of 1600 m2 d–1 Bradley LASHAM
and median of 2600 m2 d–1. Measurements of the storage Wield
coefficient varied from 7 x 10–5 to 0.06 with a geometric 40 40
mean of 0.008 and median of 0.009.
MEDSTEAD
A number of studies in Hampshire have discussed the
ey

120
vertical variation of aquifer properties. These are summarised
R.W
r
Candove

in Section 3.2.1.
The distribution of aquifer properties in the Hampshire 90
10
0

Chalk is complex. Lithology is an important control and

80

60 70
110
10

the hydraulic properties of the aquifer are influenced A lr

e
0

tc
he
I

by the presence of hardgrounds, flint and marl bands n

and their effect upon fracturing. Generally, the Upper
Chalk is considered a better aquifer than the Middle h
C

30 30
and Lower Chalks, although solution can increase the
e rito

60
hydraulic properties of the Middle Chalk where it is near
n

the surface. The Lower Chalk is generally a poor aquifer FROXFIELD

due to its generally higher marl content and poor fracture
development. Where groundwater flow is predominantly
80

80

12

80
0
10

through the Lower Chalk, as in east Hampshire towards

0

100 n
the Weald, transmissivity is thought to be usually less than Meo UPPER
PRESHAW
GREENSAND
500 m2 d–1. There is evidence to suggest that the Chalk
80

60 100
and Upper Greensand in this area are not in hydraulic
120
continuity (Giles and Lowings, 1990). SU 60 70
40

Topography is also believed to be an important factor;

yields from the boreholes sited in the interfluve areas are
generally less than in the valleys. Transmissivity values Figure 3.2 Groundwater levels and structure for the
of 1000 m2 d–1 are considered to be common in the Upper Upper Itchen catchment (after Giles and Lowings, 1990).

26
3.1.2.5 Isle of Wight table, decreasing with depth by about an order of
magnitude to 10 –3 m d–1 at 80 m below the water table.
Little aquifer properties information exists for the Isle of The packer tests showed that zones of high hydraulic
Wight but can be expected to be strongly influenced by conductivity corresponded to fracture locations. Most
lithology, tectonically induced hardening and structural of the saturated thickness of the Chalk had relatively
setting. Pumping tests within the Chalk indicate low permeabilities, of the order of 10 –1 to 1 m d–1,
transmissivity values ranging from around 100 to over with only a few fractures in each borehole providing
300 m2 d–1 (Allen et al., 1997). The effective thickness of a significant contribution to the overall transmissivity.
the Upper Greensand aquifer is generally 20 to 30 m and the These fractures were restricted to the top 40 to 50 m of
hydraulic conductivity is approximately 1 m d–1. The Plenus the saturated zone.
Marls at the top of the Lower Chalk and also the chert beds 2. Geophysical logging at Brixton Deverill, Heytesbury and
at the top of the Upper Greensand are both thought to have Chitterne in the Upper Wylye (Avon and Dorset River
low hydraulic conductivities and may limit groundwater Authority et al., 1973; Allen et al., 1997) demonstrated
flow. that even though fracturing was recorded to depths of up
to 100 m below ground level, most of the flow (about 90
per cent) came from the top 47 m of the boreholes with
3.2 GEOLOGICAL CONTROLS ON AQUIFER the most productive features in the top 35 m.
PROPERTIES 3. Studies of artesian boreholes at watercress farms at
Alresford in east Hampshire indicate that the majority
As discussed above, the aquifer properties of the Chalk of of flow is focused in a narrow zone of 30 m near the top
the Wessex Basin are highly variable. They result from a of the boreholes (Headworth, 1978).
diverse variety of matrix and fracture characteristics and 4. As part of the Itchen augmentation scheme, a group
reflect the wide variety of geological and hydrogeological pumping test was analysed to determine aquifer properties
processes that have contributed to the development of the variation with drawdown. A strongly non-linear decrease
aquifer. In the Wessex Basin, the Chalk has been affected by in both transmissivity and storage coefficient was found
Alpine tectonics, leading to a greater structural complexity with increasing drawdown (Southern Water Authority,
in the aquifer than in other regions of England. Large folds 1979). Subsequent modelling of the shape and size of
and faults are present which affect the hydrogeology of cones of depression in this area by Headworth et al.
the region, while smaller structural features exert more (1982) suggested that the highest transmissivities were
subtle influences on the local hydrogeology. Although associated with a layer about 6 m thick just below the
the area was not glaciated, it was significantly affected water table.
by periglacial processes and clay-with-flints deposits
cover large parts of the aquifer on higher ground. Both the In summary, various field investigations (principally in the
periglacial processes and acidic drainage from Palaeogene Itchen catchment) have demonstrated the presence of a
cover rocks have contributed to the development of the vertical gradient in the hydraulic properties of the aquifer
aquifer and locally have led to the development of karst controlled principally by a few enlarged fractures that tend
features. to be developed near the water table.

3.2.1 Variation in aquifer properties with depth

A characteristic feature of the Chalk aquifer is that signifi-
cant permeability generally only exists near the top of the
Weathered mantle
aquifer. Deeper in the aquifer the frequency and effective
aperture of fractures reduces as a result of the increased Solution enhanced
overburden and general reduction in groundwater circula- fractures at palaeo
tion and hence opportunity for development of the aqui- position of water
table
fer. Enlarged fractures may be particularly concentrated in
the zone of water table fluctuation, and flow through these Solution enhanced
fractures provides a significant contribution to the overall fractures in zone of
water table
transmissivity of the aquifer. Figure 3.3 is a schematic il- fluctuation
lustration of the variation in fracture density and style with
depth.
Such trends in decreasing hydraulic conductivity with Decreasing
depth associated with changing fracture density and style frequency
have been recognised in a number of areas in the Wessex of open
fractures
Basin, particularly in areas where the dip of the Chalk
is relatively shallow and structural complexity does not
dominate the hydrogeology. The following are some
examples.
Hardground with
increased fracturing
1. Packer tests were performed on three boreholes in the
Frequency of
Itchen catchment in the Candover area at Abbotstone open fractures
[SU 55 34], Itchen Down Farm [SU 54 33] and
Totford [SU 56 38] along with laboratory permeability
measurements on core samples taken from the boreholes Figure 3.3 A schematic illustration of the variation of
(Price et al., 1977). The hydraulic conductivity of the open fractures with depth in the Chalk (after Allen et al.,
matrix was found to be about 10 –2 m d–1 at the water 1997).

27
 Vertical variations in hydraulic properties are not however flow horizon when it is close to the water table, (Hopson et
ubiquitous in the Wessex Basin. In areas where karstic al. 2008). The outcrop of the Stockbridge Rock Member,
development of the aquifer has taken place, the depth a thin bed of very hard porcellaneous chalk in the Seaford
distribution of aquifer properties is less predictable. In the Chalk Formation, determines the location of the perennial
Alre catchment in Hampshire, a neighbouring catchment to head of the Bowne, Wallop Brook, Pilhil Brook, and River
the Candover where vertical gradients in hydraulic properties Anton (Farrant et al. 2001; Bryan et al., 2004).
have been described, karst has developed at various levels in The Middle Chalk is much more lithologically variable
the Upper and Middle Chalk (Allen et al., 1997). Maximum and its poorer aquifer properties have been attributed
transmissivity from boreholes in this catchment is estimated to the greater frequency of marls (Southern Water
to be as high as 30 000 m2 d–1 and high transmissivity is Authority, 1979). The Plenus Marls Member (Holywell
associated with localised karstic flow in the Middle Chalk. Nodular Chalk Formation) may locally support perched
In areas of structural complexity the depth distribution water tables and may also locally confine flow in the
of aquifer properties is also less predictable. Where the Bourne catchment, while the hard nodular Melbourn
Chalk has been tectonically hardened (i.e. matrix and Rock Member supports flowing horizons (Environment
fracture porosity has been reduced due to syn- and post- Agency, 2001). The Lower Chalk (Grey Chalk Subgroup)
deformational compaction and cementation) and typically is generally less permeable than the overlying Middle
in areas where the dip of the bedding is steep (such as across and Upper Chalks (White Chalk Subgroup) and the Chalk
the Purbeck monocline) the depth distribution of fractures Marl at the base of the Lower Chalk is a particularly
is influenced by local tectonic structures. In such areas, low permeability formation (Robins and Lloyd, 1975;
fracturing associated with the release of overburden stresses Alexander, 1981) that may influence the drawdown
is less important. Under these conditions there is only a behaviour of pumped wells (Allen et al., 1997).
limited potential for porosity development in the zone of The most direct evidence of relationships between
water table fluctuation and flow is more likely to exploit lithology and the location of groundwater flow horizons is
and develop local tectonic structures. For example, rapid provided by the interpretation of geophysical logs. Examples
transport towards springs at Arish Mell [SY 85 80] in south of such evidence are provided in the section of this chapter
Dorset has been attributed to flow through solution-enlarged discussing flow in the saturated zone.
fracture zones associated with faulting. Regional structure can have an important influence
on the depth and outcrop location of flow horizons, and
evidence of this has been found in recent studies of the
3.2.2 Lithological controls
Bourne (Environment Agency, 2001) and the Itchen (Entec,
The matrix properties of the Chalk aquifer, such as porosity 2002) catchments. In addition, faults can provide important
and pore-throat size distribution, vary as a result of differences controls on spring locations (for example in the Piddle
in primary sedimentological characteristics and diagenetic catchment).
history (Scholle, 1977; Hancock, 1993; Bloomfield, 1997). Variations in Chalk lithology not only affect the aquifer’s
This in turn may affect the fracture characteristics of the hydraulic properties but may also influence the bourne
different Chalk formations so that lithology may be expected behaviour of some of the Chalk streams in the Wessex Basin.
to influence both matrix and fracture properties of the aquifer. Stretches of some bournes may become perched for parts
Mortimore et al. (1990) were able to identify characteristic of the year where they pass over relatively impermeable
fracture styles and frequencies for specific formations in formations and may lose flow where they pass over more
the Chalk of Sussex and these characteristics have been permeable formations. For example, in the Upper Piddle
related to aquifer potential (Jones and Robins, 1999). there is only a limited loss of flow from reaches underlain by
Similar correlations between lithology, fracture density and the relatively hard, nodular Holywell Chalk, whilst loss of
aquifer properties can broadly be made for the Chalk of the flow from perched reaches of the river underlain by the New
Wessex Basin, although less work has been done than in Pit Chalk ranges from about 2 to 13 l s–1 and appears to be
Sussex. As part of the Bourne and Nine Mile River study independent of flow. This loss is inferred to be controlled by
(Environment Agency, 2001), the spatial distribution of soft the hydraulic conductivity of the New Pit Chalk outcropping
or hard units within the Chalk was identified as the principal on the river bed (Marcus Hodges Environment, 1999). This
factor in the development of relatively low or high hydraulic aspect of river–aquifer interaction is further discussed later
conductivity zones within the catchments. Relatively hard in this chapter.
formations, such the Lewes Nodular Chalk Formation,
are typically pervasively fractured and potentially provide 3.2.3 The relationship between topography, drainage
pathways for groundwater flow and, at outcrop, they act as development and aquifer properties
a focus for spring lines. Relatively softer chalks commonly
deform more plastically and typically have fewer connected In southern England, relatively high transmissivities and
fractures. These formations are generally less hydraulically storage coefficients have been reported within valleys and
conductive and marl bands may act as semi-confining layers lower transmissivities and storage coefficients towards
impeding down-gradient groundwater flow. the interfluves so that aquifer properties generally reflect
In the Wessex Basin the Upper Chalk is typically highly topography (Allen et al., 1997). Detailed discussions of why
fractured and has numerous flints and marl bands that act transmissivity and storage coefficients may have developed
to focus flow leading to secondary porosity development. to broadly mirror topography can be found in Price (1987),
Secondary porosity development has been recorded in the Downing et al. (1993) and Price et al. (1993). Several
Bourne and Nine Mile River catchments on sheet flints important factors contribute to the topographical control on
(Culver and Seaford Chalk Formations) and on thicker marl areal distribution of aquifer properties:
seams (Newhaven Chalk Formation) (Environment Agency,
2001). Hardgrounds can also act as sites of localised flow. 1. Valleys in the Chalk, including dry valleys, may follow
For example, throughout much of the basin the Chalk Rock structural lineaments or lines of structural weakness
near the base of the Upper Chalk may act as a preferential and may be the locus of increased fracture density.

28
2. Erosion along valleys can act to reduce effective stress and in the Newbury area along the edge of Palaeogene. Both
and lead to the development of horizontal fractures at solution features and karst in the Chalk appear to be most
shallow depths (Price et al., 1993). common where the present land surface is close to the sub-
3. Concentration of groundwater flow towards discharge Palaeogene peneplain. Solution features and karst may be of
areas within the valley can contribute to the development local hydrogeological significance as they can indicate areas
of localised secondary fracture porosity and hence to where there may be particularly rapid or focused recharge
increased transmissivity and storage coefficients in the and rapid, highly localised movement of groundwater in the
valley (Price et al., 1993 and references therein). saturated zone.
4. Periglacial processes may further contribute to the Solution features at the land surface, such as buried and
development of enhanced hydraulic conductivity within subsidence sinkholes (or dolines), may reach densities of
the valleys due to repeated freezing and thawing and the the order of 100 per km2 in parts of Dorset (Sperling et al.,
opening of fractures in the top 20 to 30 m of the Chalk 1977). Buried sinkholes (Culshaw and Waltham, 1987) are
(Younger, 1989). typified by pipe or cone-like cavities in the underlying Chalk
but have no surface expression. The pipes are commonly
In the Wessex Basin, however, structural control on filled with flinty, gravelly clay derived from superficial
topography and hence hydraulic properties is much cover, usually clay-with-flints. Subsidence sinkholes are
less clear. Many of the rivers in the basin cut across the closed surface depressions, usually bowl-, pipe- or cone-
predominant structural grain (Wooldridge and Linton, 1955) shaped. They may be isolated features or may occur in
and the Chalk in south Dorset is extensively deformed so groups, sometimes coalescing into large, composite dolines.
that relationships between the geomorphology, hydrology Most are found in unconsolidated cover sediments up to
and hydrogeology of the area are complex and do not simply 10 m thick, such as the clay-with-flints, for example on
conform to the patterns observed in other chalkland areas the scarp crest east of South Tidworth, at Collingbourne
(Allen et al., 1997). Wood near Ludgershall and between Everleigh and Sidbury
Examples of discordant relations between surface drain- Hill. The location of these solution features depends on
age, topography and structure include: a range of variables including lithology, fracture style,
geomorphological setting, structure and occasionally even
1. central Hampshire where the two principal streams, the anthropogenic factors. The main control, however, is the
Itchen and the Test, flow southwards in a discordant geomorphological setting and proximity to the margin
fashion across anticlinal folds at Winchester and Stock- of Palaeogene cover. The highest densities of solution
bridge features are generally found along spring lines and valley
2. the headwaters of the Itchen where both the Candover floors — for example along the middle and upper reaches
Brook and the Meon flow southwards over east–west of the Bourne, between Collingbourne Kingston and North
trending fold structures (Giles and Lowings, 1990) Tidworth — and near the margins of Palaeogene cover. An
3. the River Avon where it cuts across the eastern end of example of the latter can be found between the rivers Piddle
the Warminster Anticline. and Frome in the Puddletown Heath, Southover Heath
and Culpepper’s Dish areas (Figure 3.4) where extensive
Despite these observations, there are areas where there is solution features have developed on the margins of Eocene
clearly structural control on topography and subsequent cover (Sperling et al., 1977). Although Allen et al. (1997)
development of enhanced aquifer properties due to valley note that these solution features could permit rapid recharge
erosion, concentration of groundwater flow towards the to the aquifer, there is as yet no direct evidence linking such
valley and periglacial modification. For example, east–west geomorphological features to zones of high transmissivity
trending anticlinal structures are associated with the Vale of in this area.
Wardour, Wylye Valley and the Vale of Pewsey, and aquifer Allen et al. (1997) considered that several factors could
properties in these areas generally reflect the topography account for the high degree of solution activity associated
with transmissivity and storage tending to be highest in with cover deposits. These are:
the valleys and lowest in the interfluves. This is taken to
reflect the areal distribution of enlarged fractures (Allen et
al., 1997). N Puddletown Pidd
le Valle
In Hampshire, some rivers and dry valleys have developed Heath y
along the axes of synclines and fault lineaments (Southern
Water Authority, 1979). Giles and Lowings (1990) reported Southover
Culpepper's
Dish
pumping test data indicating higher yields along the axes Heath
of denuded synclines than anticlines. They also investigated
Frome Valley
the relationship between groundwater levels and structure in 90 90

the Upper Itchen catchment and found a positive correlation,

with high groundwater levels associated with small anticlinal
features and lower groundwater levels associated with small
synclinal features (Figure 3.2). This was attributed to low
transmissivity values resulting from relatively poor fracture
development around the axes of the anticlines.
SY 75 80 85

Pleistocene and Recent Chalk 0 5 km

3.2.4 Karst
Eocene Approximate area of major hollow development
Solution features at the land surface are widespread in
the Wessex Basin and in some areas karstification of the
underlying Chalk aquifer is particularly prevalent, for Figure 3.4 The distribution of solution features in the
example in parts of Dorset, around Salisbury and Mottisfont Chalk of the Dorset heathlands (after Sperling et al., 1977).

29
• soils associated with Palaeogene deposits and clay-with- meability. Perched water tables have also frequently led to
flints tend to be quite acidic (Edmunds et al., 1992) the development of thin, irregular layers of iron-cemented
• chalk soils are generally permeable, but those associated sands. Their cementation would impede vertical leakage
with cover can be quite clayey and therefore concentrate and may reduce the scope for recharge to the Chalk.
run-off to discrete points The regional pattern of groundwater flow is from the edge
• as recharge waters drain through the cover, they remain of the basin towards the centre. Local groundwater flow may
undersaturated with respect to calcite until reaching the be hard to predict, however, due to the lateral discontinuity
Chalk surface, thus allowing the acidic recharge to be of the sandy beds. Perched water tables are common. The
channelled to discrete points. sand-dominated minor aquifers are interlayered with clay-
dominated aquitards. Many of the Palaeogene minor aquifers
The karstic features in the Chalk to the east of Dorchester will therefore be confined or semi-confined by overlying
may be the result of intense and localised solution activity clay beds. Lithological variations may also result in sandy
promoted by highly acidic conditions under heathland layers within a formation being bounded laterally as well as
vegetation (Sperling et al., 1977). In south Hampshire, vertically by clay layers. In particular, in the Wessex Basin,
significant hydraulic connections between swallow holes the laterally discontinuous aquifers within the London Clay
and springs in the Chalk were established, using groundwater Formation such as the Whitecliff, Durley, Nursling and
tracing techniques, by Atkinson and Smith (1974) and were Portsmouth sands, are confined by the surrounding clay beds.
considered to be the result of solution activity caused by Brickearth, plateau gravels or valley gravels overlie
rapid run-off from the Palaeogene cover enhancing the Palaeogene beds in some areas. Only the brickearth is likely
transmissivity of the Chalk aquifer. to inhibit recharge to the underlying minor aquifers.
Where the basal Palaeogene strata are arenaceous, the sands
3.2.5 Role of Palaeogene are in hydraulic continuity with the underlying Chalk, and the
combined deposits are considered to be a single aquifer. The
From a hydrogeological point of view, there are few proportion of sand in these formations decreases to the west.
regionally significant minor Palaeogene aquifers in the Where clay predominates, the Palaeogene deposits provide an
Wessex area. The outcrop of Palaeogene deposits is shown aquitard of regional extent, and a useful element of additional
in Figure 2.1. In Figure 3.5 (from Jones et al., 2000) the drainable storage overlying the Chalk aquifer. North of
principal sand-dominated units that comprise the minor Southampton, borehole yields from the Reading Formation
aquifers are shown in relation to the major clay-dominated can be up to 200 m3 d–1, and this is thought to be partly due to
aquitards/aquicludes. The great vertical and lateral variation upward recharge to the formation from the underlying Chalk.
of the more argillaceous units may also give rise to minor
aquifers where sand lenses are present.
In many areas, oxidation and percolation of groundwa- 3.2.6 Periglacial influences
ter has weathered sand-dominated sequences to depths of During the Quaternary, the Chalk of southern England
several tens of metres, which is likely to increase the per- remained largely free of ice sheets but was significantly affected

Figure 3.5 Bournemouth Southampton Portsmouth

Relationship between
Palaeogene minor Headon Formation
aquifers in the Wessex Becton Sand Formation
Basin (not to scale)
Barton Group

Chama Sand Formation

(from Jones et al.,
2000). Barton Clay Formation

Boscombe
Sand Formation
Bournemouth Group

Bracklesham Group

Branksome Sand Selsey Sand Formation

Formation

Marsh Farm Formation

Poole
Formation Earnley Sand Formation

Wittering Formation
Whitecliff
Sand
London Clay Nursling Sand Portsmouth
Formation Sand

Reading Formation

Chalk Not to scale

Predominant lithologies: Clay and silt Silty, fine sand Fine to coarse sand

30
by ground ice (periglaciation). The permanently frozen southern England (e.g. Wooldridge and Linton, 1955; Jones,
substrata were probably not seriously affected by periglacial 1999). The earlier model of landscape development was based
activity. Close to the surface, however, the active zone on a sequential interpretation of events, while the later model
underwent cycles of freezing and thawing that, over the (Jones, 1999) envisaged a more evolutionary interpretation
whole of the Chalk outcrop, have produced a weathered of events. Although, for ease of representation, the
mantle frequently 1 to 2.5 m thick consisting of broken, hydrogeological evolution of the Chalk aquifer is presented
rubbly chalk (Williams, 1987). In some places, just below in Table 3.1 as a series of discrete events, it is envisaged that
the subsoil, the Chalk is highly pulverised to putty chalk, many of the geological events or processes are likely to have
which is generally structureless chalk with irregular sized occurred over more extended periods of time and will have
blocks set in a soft to firm putty matrix. The main effect overlapped and/or will have had cyclic characteristics. For
is a significant increase in fracture density in the top few example, there have been pulses of tectonic activity within
metres of the Chalk. In some valleys, periglacial activity has and from the end of the Cretaceous through to the present
led to fractures opening up to a depth of up to 20 or 30 m day (Mortimore and Pomerol, 1991), and although the major
(Higginbottom and Fookes, 1971; Williams, 1987) and this phase of deformation was Miocene (Alpine compression)
may contribute to the development of enhanced permeability minor compressive events have continued to cause limited
in valleys. Many of the present-day dry valleys in the region uplift since the Miocene. Another example is the uplift and
are possibly remnants of active stream valleys, formed at a erosion prior to deposition of the Palaeogene — it is thought
time of different sea levels. The different sea levels may also that this uplift occurred as a series of pulsed events or
have resulted in the maintenance of the water table at various episodes (Jones, 1999) rather than one or two discrete events.
levels and so allowed the development of subkarstic horizons
of enhanced permeability throughout the vertical sequence
of the Chalk (see Figure 3.3). 3.3 THE AQUIFER IN THE WATER CYCLE

3.2.7 Role of drift – Pleistocene and Recent 3.3.1 Unsaturated zone and recharge
Although most of the Chalk aquifer of Hampshire is The water table in the Chalk aquifer generally follows, in
unconfined, a number of different types of Pleistocene a subdued manner, the surface expression of topography:
deposits are present over the Chalk outcrop in the Wessex the unsaturated zone is thicker under hills and thinner in
Basin, the main expressions of which include clay-with- the valley areas. Given the variation in topography over the
flints, plateau gravels/alluvium, brickearth and head Wessex Basin, the thickness of the unsaturated zone can
deposits. However, there has been little work undertaken to therefore vary considerably. In addition, annual variations
investigate the effect of the drift on the hydrogeology of the in water level can be significant.
Chalk in the Wessex region. Flow through the unsaturated zone in the Chalk takes
The clay-with-flints forms extensive deposits on the place either by fracture flow or matrix flow or a combination
Chalk of the Hampshire Downs and to a lesser extent on of both, and there are a number of lines of evidence for each.
Salisbury Plain, south Dorset and the Isle of Wight where Recharge to the aquifer through the unsaturated zone may
these deposits are prominent only on the higher ground. The comprise both vertical and horizontal flow components. The
deposits range from about 0.5 to 10 m in thickness with an dominating process of recharge is likely to be influenced
average of around 6 m (Klinck et al., 1998). It commonly by the presence of permeable or impermeable horizons
forms the fill material in solution pipes in the Chalk where within the Chalk. The overlying Palaeogene and Quaternary
it can be up to 15 m in thickness. Klinck et al. (1998) deposits also have an influence on the recharge process.
conducted a series of laboratory and field experiments The heterogeneous nature of these deposits means that
in order to characterise the porosity and permeability the impermeable units may cause run-off, whilst the
characteristics of the clay-with-flints in southern England. more permeable units may allow direct recharge to occur.
The deposits tend to show a trend of increasing porosity However, the presence of impermeable deposits may also
and decreasing hydraulic conductivity with depth, with an concentrate flow from certain areas and hence increase the
increase in hydraulic conductivity again near to the contact amount of recharge to the underlying aquifer.
with the Chalk. Infiltrometer measurements for Dorset It has been postulated by some researchers that recharge
gave a mean hydraulic conductivity of around 1 m d–1. through unsaturated Chalk takes place mainly via the
The high values at the surface were attributed to cracking; fractures rather than through the matrix itself. The rapid
the higher conductivity at the base due to the presence of response of water table levels to rainfall events seemed to
closely spaced subvertical shearing adjacent to the contact indicate this to be the case (Headworth, 1972). There is also
with the Chalk. Plateau gravels and alluvium are observed evidence from the bacterial contamination of groundwater
lining valley sides up to 15 m above the present valley floors (Maclean, 1969) that fissure flow can predominate in the
and are capped by the most recent river deposits present as unsaturated zone of the Chalk. However, the work of Smith
alluvium. et al. (1970) concluded that in fact only about 15 per cent of
recharge occurs through the fissures; the remaining 85 per
cent recharges the aquifer by a piston displacement model,
3.2.8 Summary of geological events and their i.e. water draining from above by gravity displaces water
implications for the hydrogeological evolution of the already in the intergranular rock matrix downwards. Daily
aquifer rainfall in excess of a few millimetres appears to be stored
It is clear that a wide range of geological events and processes temporarily in the soil and near-surface weathered Chalk.
have contributed to the evolution of the Chalk aquifer. These The delayed drainage is often at rates low enough to be
events and processes and their corresponding hydrogeological conducted by the matrix alone (Gardner et al., 1990).
implications are detailed in Table 3.1. Similar approaches A number of study sites have been developed in the UK
have been adopted previously to describe the links between to examine how recharge processes vary in different rock
geological events and the geomorphological evolution of types. One such site, at Bridget’s Farm near Winchester

31
Table 3.1 Geological evolution of the Chalk aquifer in the Wessex Basin and its hydrogeological implications.

Geological events Factors influencing chalk rock mass properties Hydrogeological implications/significance

Chalk deposition Primary lithological characteristics established, The primary lithological nature of the Chalk aquifer
e.g. ratio of carbonate to clay minerals, has implications for the gross hydrogeological
hardground and flint development characteristics of different Chalk formations,
particularly the development of hardgrounds
Chalk burial and Compaction and cementation. Initial physical The location of marl seams, flaser marls, nodular
diagenesis compaction and, for more deeply buried chalks, chalks and flints, including tabular flints, within the
chemical compaction. Lithification of the Chalk Chalk sequence may later affect the development
and flints and the development of burial joints as of flow heterogeneity at both catchment and local
burial and lithification progresses scales within the aquifer. Burial joints are the first
stage in the development of the dual-porosity nature
of the Chalk
Chalk uplift and erosion, Faulting, jointing and minor flexuring of the Further development of the dual porosity nature
and development of pre- lithified Chalk to accommodate uplift. Faulting of the Chalk with the formation of faults, uplift
Palaeogene drainage may be localised over structural lineaments in the joints and gentle folds. Erosion and possible pre-
pre-Cretaceous basement. Followed by karstic Palaeogene drainage development may lead to
weathering and erosion and possible development initial flushing of connate waters and possibly early
of first drainage network on the eroded Chalk development of enhanced (or secondary) porosity
under a subtropical climatic regime near the water table
Palaeogene deposition and Marine transgression of the eroded Chalk surface Infill of irregularities on the eroded Chalk surface
burial and deposition of a thick sequence of Palaeogene by Palaeogene deposits forms the starting point for
deposits. Chalk burial was accommodated by future karstification of this surface with implications
re-activation (inversion) of earlier uplift faults and for groundwater recharge. Further modification of
folds and the development of new burial joints. the dual porosity nature of the Chalk with possible
Potential additional chemical compaction of the inversion of faults, closure of existing joints and
deeper Chalk formations potential development of new burial joints
Progressive re-emergence Uplift of the Palaeogene associated with re- Further development of dual porosity. Limited
of the sub-Palaeogene activation of existing faults and renewed uplift flushing of connate waters in the Chalk where
surface jointing. Re-establishment of a consequent groundwater can percolate through Palaeogene
drainage pattern on the exposed Palaeogene cover may lead to additional development of
surface secondary porosity. Start of the removal of
Palaeogene deposits. Karstification of the Chalk
Alpine deformation and Extensive deformation and uplift of the Chalk Major structural controls on hydrogeology initiated.
development of Miocene and Palaeogene deposits with the formation Groundwater flow directions possibly controlled
drainage of tight folds and monoclines and associated by Miocene drainage network and associated zones
accommodating jointing and faulting. of groundwater discharge. Continued freshening of
Development of a Miocene drainage network the aquifer. Continued removal of the Palaeogene
associated with the uplifted landscape and start of deposits and karstification of the Chalk
the development of clay-with-flints
Post-Pliocene transgression Minor transgression and limited planation of the Initiation of the present-day drainage network and
and planation post Miocene land surface. Removal of some of hence control on present-day groundwater flow
the Palaeogene deposits. Continued development directions. Continued limited development of present-
of drainage network day karst with removal of Palaeogene cover
Quaternary Episodic pulses of periglaciation. Freeze-thaw and Major modification of the shallow aquifer with
periglaciation cryoturbation of the top few metres of the Chalk. formation of local putty chalks and thick weathered
Continued removal of the Palaeogene cover and zones in valleys. Colder groundwaters are much
development of clay-with-flints deposits during more chemically aggressive so this may be the
thawing episodes and development of head major phase of development of secondary porosity
deposits. Deposition of river gravels (river terrace and karst (see Ford and Williams, 1989). ‘Dry
deposits) valley development’ at times of relatively high sea-
level stands
Recent erosion and ground- Continued removal of the Palaeogene cover and Development of the aquifer in the zone of
water circulation formation of the clay-with-flints. Deposition of groundwater level fluctuation and in areas of recent
recent alluvium groundwater discharge near river valleys. Present-
day groundwater flow heterogeneity established
as sea levels fluctuate and present-day drainage
is established. Limited development of the Chalk
aquifer covered by Palaeogene

32
in Hampshire, has been extensively used for this purpose. The presence of clay-with-flints cover may affect
Wellings (1984a, b) concluded from the analysis of soil recharge to the underlying Chalk aquifer in a number of
physics data that water moves predominantly through the different ways. While the main constituent of the deposits is
fine pores of the Chalk matrix with only a minor component clay based, tending to inhibit recharge, there are lithological
through the fissures. variations, involving both silt and sand fractions. These
The concept of delayed recharge may be of importance. more permeable horizons, together with the contact areas
Regional studies of the Chalk aquifer in the Thames catchment between the clay matrix and the flints, are likely to act as
suggests that up to 50 per cent of recharge is still in transit conduits, facilitating groundwater movement through the
in the unsaturated zone when the water table peaks (Lerner, deposit. In addition, recharge through superficial cracks and
1997). This water drains to the saturated zone over the summer rapid marginal recharge via karstic features in the Chalk is
at a slower rate than groundwater discharges to rivers. possible. A study by Klinck et al. (1998) concluded that a
Darling and Bath (1988) analysed stable isotope data variety of recharge pathways were likely to be associated
of 2H and 18O values in the unsaturated Chalk from Fleam with clay-with-flints; these are illustrated in Figure 3.6.
Dyke in Cambridgeshire and at Bridget’s Farm. A number Groundwater recharge at the catchment scale has usually
of differences were observed between the vertical isotope been simulated using models based on soil moisture balance
profiles from the two sites. These differences were attributed techniques, such as the Penman-Grindley (Penman, 1948;
to higher infiltration rates in southern England (>400 mm a–1) Grindley, 1967) or Food and Agriculture Organisation (Food
than in eastern England (<200 mm a–1) and also to different and Agriculture Organisation, 1998) methods. Whilst this
infiltration mechanisms. It was considered that differences has been the commonest approach, other types of model have
in lithology, due to stratigraphical position in the Chalk been used to simulate recharge, for example rainfall run-off
sequence and postdepositional history, e.g. glaciation, could models (Moore and Bell, 2002). Generally, soil-moisture
impose a major influence on the recharge mechanisms. For balance models have been based on measured daily rainfall
example, Wellings et al. (1982) reported that the matrix data and estimates of potential evaporation from the UK
conductivity of Chalk at Bridget’s Farm is approximately five Meteorological Office Rainfall and Evaporation Calculation
times greater than at Fleam Dyke. However, the conductivity System (MORECS). MORECS uses the concept of a soil
of fractured Chalk is likely to be orders of magnitude greater ‘field capacity’ and only when this is full does recharge
than unfractured Chalk, and so these differences in matrix occur. In addition to other soil water balance components,
conductivity may be less significant depending on the relative MORECS calculates effective precipitation, equivalent
proportion of matrix/fracture flow. This last point highlights to recharge plus run-off, but because it uses a 40 km grid,
that the results of site-specific experiments on mechanisms generally only the potential evaporation values have been
of groundwater recharge should be generalised with caution. used in groundwater recharge modelling studies. During the
A study into the mechanisms of water storage and flow 1990s, the UK Meteorological Office’s Surface Exchange
in the unsaturated zone of the Chalk (Lewis et al., 1993) System (MOSES) was developed, which incorporates a
found that for the Lambourn and Kennet catchments in the more complex description of the soil store and an improved
Berkshire Downs, the volume of water flowing out of the representation of surface run-off and evaporation from the soil
catchment was substantially greater than could be accounted than MORECS. MOSES does not use the concept of a field
for based on the change in the level of the water table, capacity or soil moisture deficit. Whilst MOSES provides an
unless an unrealistic value of specific yield was assumed. improved process description of the soil water balance, it has
They concluded that the most likely explanation for this generally still been the case that potential evaporation values
discrepancy was the slow release of water due to drainage from MORECS have been used in catchment scale recharge
of the Chalk in the unsaturated zone. It was calculated that models developed as part of hydrogeological studies.
a drainage of around 0.25 to 0.30 per cent of the volume Values of recharge in the Wessex Basin have been
of rock in the unsaturated zone was required to account for estimated in a number of groundwater modelling studies.
the anomaly. Price et al. (2000) concluded that this storage For example, a comparison of groundwater models in and
property of the unsaturated zone could be used to account around the River Test catchment (Environment Agency,
for the delay in the reponse of the water table to recharge and 2004) showed long-term, average annual modelled recharge
that in areas where the water table is deep and there is a thick values varying from around 200 mm a–1, for example in
unsaturated zone, the delay could be of the order of weeks. parts of the Bourne catchment) to around 500 mm a–1 in the

Figure 3.6 Direct Preferential

Recharge mechanisms recharge flow through
Marginal through Recharge cracked
associated with recharge chalk through surface into
clay-with-flints (from from run-off window cracked sand filled
Klinck et al., 1998). to dolines
f
surface pipes
and -of
e run
karstified
r fac
chalk Su

CHALK
Preferential Intraflow and
flow through preferential flow in Sheared
sheared clay contact sandy lenses clay-with-
with chalk flints

33
south-east of the Itchen catchment. In general, modelled between maximum and minimum water levels tends to occur
values in the Bourne, Upper Avon and Test catchments were beneath the high ground, and is typically approximately
around 250 to 350 mm a–1. In a modelling study of the River between 15 and 20 m. In valleys, it tends to be reduced and
Allen, catchment estimates of recharge varied from 340 to is more generally only a few metres, although there are
400 mm a–1 (Groundwater Development Consultants, 1992). exceptions. For example, a borehole at Woodside [SU 33 56],
situated near the top of a winterbourne valley, recorded an
average annual fluctuation in water level of 14.7 m (Marsh
3.3.2 Groundwater levels in the Wessex Basin Chalk and Lees, 2003). Conversely, a monitoring well at Lower
The Environment Agency regularly monitors groundwater Wield Farm [SU 63 40] sited in an interfluve on the surface
levels in boreholes across the Chalk aquifer of Wessex at a water divide, recorded an annual average fluctuation in water
variety of depths depending on the planned use of the data. level of only 2.40 m (Marsh and Lees, 2003). These seasonal
A subset of these boreholes, termed index wells, which are fluctuations mean that piezometric surfaces are only valid for
considered to be representative of the aquifer, are selected datasets that have been collected simultaneously.
to contribute to the National Hydrological Monitoring The Wessex and Hampshire and the Isle of Wight
Programme, which provides information on the state of the hydrogeological maps (Institute of Geological Sciences,
nation’s water resources on a monthly basis. Many index wells 1979a; Institute of Geological Sciences, 1979b) cover the
have been used to record water level variations in the aquifer Chalk of the Wessex Basin and were compiled at a scale
for over 50 years, and the oldest in Wessex began recording of 1:100 000 by the Institute of Geological Sciences with
in 1894 at Compton House [SU 77 14] near Rowlands Castle. Wessex Water Authority and Southern Water Authority
Just outside the Wessex region, in the South Downs, is the respectively. They include contours for the Chalk piezometric
Chilgrove House Borehole [SU 83 14], which has the longest surface for September 1975 (Wessex) and October 1973
record of groundwater levels in Europe (and possibly the (Hampshire) — the division occurring roughly along the
world) with a continuous record from 1836 to the present day. line of longitude passing through Salisbury. These water
The water level in a well or borehole is a representation level contours are reproduced in Figure 3.8. The contours
of the average vertical hydraulic head in the formations that show that the general direction of regional flow is from
have been penetrated and are open to the borehole, therefore the edges of the basin towards the coast, with the contours
comparisons of water levels across an area may involve generally reflecting the topography in a subdued form.
hydraulic heads measured in different formations (or Where the Chalk is confined by Palaeogene strata, the Chalk
different units within a formation) and at different depths. potentiometric surface is not shown, but limited data from
As a result, the data need to be used with care where vertical boreholes indicate a gradient towards the coast.
head gradients are thought to be significant. However, on On a regional scale, the contours generally indicate the
a broad regional basis it is considered that chalk borehole expected discharge of groundwater to the main river systems
water levels may be used to represent general head variations (effluent) on their middle reaches, prior to flowing over the
and flow directions. Palaeogene deposits. However, in some cases (for example
The water table in the Chalk is usually a subdued reflection the Piddle and the Bourne) the contours for the upper parts
of the topography and fluctuates seasonally in response to of the catchment are more ambiguous or indicate influent
recharge, discharge and abstraction. Generally, groundwater behaviour in the late summer. Groundwater divides are clear
levels rise throughout the winter, with the highest levels between some catchments (for example between the Bourne
commonly occurring in January and February, and fall over and Test) but it is less obvious where (and sometimes if) they
the summer, reaching lows in August or September. Often exist in others (for example between the Frome and Piddle).
multiple maximum levels are observed during the winter Figure 3.9 shows the thickness of the unsaturated zone across
due to an uneven rainfall distribution, with several prolonged the Wessex Basin. This clearly shows the importance of rivers
periods of rain resulting in discrete recharge events. draining groundwater from the aquifer (i.e. thin unsaturated
An example of an index well hydrograph for the West zone) and the fact that many tens of metres of unsaturated zone
Woodyates Manor Borehole [SU 01 19] can be seen in thicknesses can be found in some interfluve areas.
Figure 3.7. This shows the typical form of a chalk hydrograph In south Dorset, annual groundwater level fluctuations
in the Wessex region, although the mean annual range for this are at their greatest in the South Winterbourne catchment,
well (25.89 m) is larger than for many of the other index wells where fluctuations of up to 18 m are observed in the upper
in Wessex as it is sited near an interfluve. The maximum annual reaches of the catchment, with a decrease in the amount
range for this well (between 1942 and 2000) was 36.20 m with of fluctuation downstream. The area around Lulworth also
a minimum range of 4.72 m (Marsh and Lees, 2003). The shows marked variations (Alexander, 1981).
peak in water level can clearly be seen to occur early in the As part of their study of the River Itchen for the Environment
year, with the minimum occurring at the end of the summer. Agency, Entec (2002) investigated groundwater level data
The maximum annual range from an index well in the Wessex and compared a range of hydrograph characteristics with
Basin Chalk was recorded as 38.10 m (Marsh and Lees, 2003) other factors such as ground level, unsaturated depth and
in a well at Compton House [SU 77 14]. The greatest difference distance to the nearest main surface water course. The

Figure 3.7 110

105
West Woodyates 100
hydrograph 95
Metres AOD

90
(Source — BGS 85
National Groundwater 80
75
Level Archive). 70
65
60
Jan 90 Jan 91 Jan 92 Jan 93 Jan 94 Jan 95 Jan 96 Jan 97 Jan 98 Jan 99

34
Figure 3.8
Basingstoke
Groundwater levels !
in the Chalk of the Andover
!
Wessex Basin (IGS
1979a, 1979b). 1
40
contains OS data © Crown
copyright and database Winchester
rights 2017 Ordnance ! Salisbury !
Survey [100021290 EUL].
Use of this data is subject to
terms and conditions.
1
20

Yeovil
!
Southampton

1
00 !
Portsmouth
! N
! ! Newport
Bournemouth
Dorchester

0
80
! Weymouth
0 50 km

3
60 3
80 4
00 4
20 4
40 60
4
80
4

Chalk groundwater 0−25 50−75 100−125 150−175 Rivers

levels (mAOD) 25−50 75−100 125−150 175−180

unsaturated zone thickness was found to vary across the capacity of an aquifer is exceeded. It can arise in two main
catchment generally between 0 and 60 m but could be as ways. One of the most common forms is associated with
much as 100 m. The Alre subcatchment showed only very river floodplains and occurs when river gravels become fully
limited variation in water levels with ground level and it saturated and groundwater is forced out into the floodplain.
was inferred that groundwater levels here were controlled The second type of groundwater flooding occurs when the
by distinct high transmissivity levels in the Chalk. water level in consolidated rock aquifers, such as the Chalk,
The study also found a weak correlation between the seasonal is so high that the water emerges at the surface as springs
amplitude of a groundwater level record and ground level, with and seepages. This second type is the cause of the majority
slightly greater water level fluctuations where ground levels are of groundwater flooding events in Wessex associated with
higher. This is as would be expected, as in areas of low elevation the Chalk aquifer, especially in the upper reaches of tributary
(valleys) the groundwater levels tend to be controlled by the streams. A significant difference from surface water flooding
streams and shows that there is good connectivity between the is that the events last for weeks or months, rather than days,
aquifer and the rivers, which act as drains. Fluctuations in the as the groundwater in storage is slowly released.
Candover catchment, however, were almost always less than In Hampshire, more than 100 villages and many roads
5 m. This is consistent with the observation of Rennie (1994) and fields were flooded for weeks and even months during
that there is little spatial variation in the degree of fissuring in the winter of 2000/01 (Hampshire Water Partnership, 2003).
the catchment. The Dever catchment also shows little annual Similar problems were experienced throughout the Wessex
fluctuation in groundwater levels and this is explained by the region and have occurred on several occasions in the past.
role of the River Dever acting as a drain. The winterbourne tributary streams of the Test are often
A number of individual hydrographs were also analysed areas where Chalk groundwater flooding occurs, due to the
and it was observed that generally boreholes near water source of the watercourse moving in response to groundwater
courses showed relatively flat hydrographs indicating good level fluctuations. In the winters of 1993/94 and 1994/95,
connectivity with the surface water system. Boreholes in groundwater levels were exceptionally high and many
or near synclines tend to show smoother hydrographs than winterbournes rose much further up their valleys than normal,
those near anticlines, as observed in the Alre catchment. In causing local flooding. In contrast, low winter rainfall in
addition, generally, summer low water levels do not vary 1991/92 and 1996 caused very low groundwater levels and
much from year to year, even in drought years. From this river flows in the following drought summers (Environment
it is inferred that water levels reach the base of the active Agency, 1999a). In some villages such as Hambledon, in
Chalk and then do not tend to drop any further. In summary, Hampshire, some form of groundwater flooding is frequently
analysis of groundwater level amplitudes and responses to experienced, due to the fluctuations of the water table.
recharge show no simple or consistent relationship with
factors such as unsaturated zone thickness, nature of the 3.3.3 Flow in the saturated zone
superficial deposits, or distance to major rivers, and suggests
that a combination of factors influence water levels. 3.3.3.1 The unconfined aquifer
The principal groundwater flow mechanism in the saturated
3.3.2.1 Groundwater flooding zone is along fractures, which are often enlarged by solution,
Groundwater flooding, also described as clear water flooding, rather than through the Chalk matrix. The hydraulic
is a natural phenomenon that occurs when the natural storage conductivity of the Chalk matrix is very low (of the order of

35
Figure 3.9
Unsaturated zone Basingstoke
thickness in the Chalk !
(
of the Wessex Basin. Andover
contains OS data © Crown !
(
copyright and database 1
40
rights 2017 Ordnance
Survey [100021290 EUL].
Use of this data is subject to
Salisbury Winchester
terms and conditions. !
( !
(

1
20

!
(
Southampton

1
00 !
(
Portsmouth
Dorchester
!
( !
( N
Bournemouth !
(
Newport

0
80
!
( Weymouth
0 50 km

3
60 3
80 00
4
20
4
40
4
60
4

Unsaturated zone Below 10 25−50 75−100 125−150 Rivers

thickness (m) 10−25 50−75 100−125 150−163

10–3 m d–1) and consequently makes a negligible contribution to Rock horizon. During the pumped logging the discharge
the overall transmissivity of the aquifer. Subsurface evidence rate was 48.5 m3 h–1 and the borehole exhibited a specific
for the location of groundwater flow horizons and their capacity of 78.2 m3 h–1 m–1, which was approximately ten
relationship with aquifer lithology is obtained primarily times higher than the specific capacity of the Figheldean
from the interpretation of geophysical borehole logs. Borehole, indicating that a well-developed karstic flow
Formation and fluid logging of boreholes penetrating the feature was probably responsible. A small inflow was also
Chalk shows that water normally flows into a borehole at present at 16.6 m depth, a short distance below the casing.
only a few specific points and that these inflows are often In this example, as with Figheldean, the main inflows are
associated with certain stratigraphical horizons (which are from groundwater circulation along fractures associated
commonly located at flint, marl and hardground surfaces) with particular lithological horizons and junctions between
at shallow depths. For example Figure 3.10 illustrates water contrasting lithologies, which have probably been enlarged
inflows in the Figheldean observation borehole in the River by solution as groundwater flows towards discharge outlets.
Avon catchment [SU 15 46] where logging was undertaken The groundwater circulating more rapidly along these
as part of a low-flow study of the river by the Environment horizons is cooler than the deeper groundwater circulating
Agency. Electrical resistivity logging showed that the more slowly at greater depth.
borehole penetrated the Seaford, Lewes Nodular and New Specific water inflows to boreholes from only a few
Pit Chalk, and fluid logging whilst pumping identified four solution features is an important characteristic of the Chalk
separate inflows: at 18 m depth (Seaford Chalk); at 31 m aquifer and is responsible for ‘breakaway’ features in yield-
(top of Lewes Nodular Chalk); at 50 m (at the top of the drawdown plots when seasonal water levels decline or
Chalk Rock hardgrounds), and at 62 m depth (top New Pit pumping takes water levels to, or below, such contributing
Chalk). The borehole was pumped at 38.5 m3 h–1 and the horizons. These horizons clearly represent the main drains
specific capacity was 8.83 m3 h–1 m–1. Flowmeter logging of the rock mass and are linked to the matrix storage by a
showed the main inflows were the shallowest, at 18 m and finer network of fractures. Optical and acoustic imaging of
at 31 m depth. Chalk boreholes by video-scanning survey is able to show
Sometimes the groundwater flow in the Chalk can all the nature of these solution features and the details of the
be concentrated at one particular horizon. For example fracture systems present. Optical imaging and caliper logging
Figure 3.11 is a composite plot of geophysical logs recorded also identifies some well-developed fissure horizons present
by the British Geological Survey in a borehole near Tidworth above normal water level in some boreholes. These became
in the River Bourne catchment. The borehole penetrates the locally important in sustaining groundwater flooding that was
same stratigraphical interval as the Figheldean Borehole, experienced in some Chalk regions during the exceptionally
but the pumped-fluid logging revealed that most (>80 wet winter of 2000 to 2001. These high-level fissure systems
per cent) of the inflow was coming from the surface of a may have developed during periods of higher water table in
hard band having high gamma-ray activity in the Seaford the Pleistocene and Palaeogene.
Chalk, thought to be the Stockbridge Rock Member, at In both the River Bourne and River Avon catchments, gamma
26 to 29 m depth, approximately 50 m above the Chalk ray and resistivity logging reveals a strong development of the

36
Well name: Figheldean obh Main water inflows
Depth Caliper NGAM Focused RES Fluid TEMP Fluid EC25 Downlog_LS ac-Flowrate
(mbd) 5 (in) 10 0 (API) 20 50 (ohm.m) 150 10.8 (DegC) 11.3 450 (uS/cm) 500 0 (m/min) 40 0 (m3/h) 40
Chalk subdivision
Fluid TEMPQ Fluid ECQ25 Fluid vel Q inflow
10.8 (DegC) 11.3 450 (uS/cm) 500 0 (m/min) 40 0 (m3/h) 10
SWL

PWL
Q/s = 7.8 m3/h/m

−10

PLASTIC CASING
−16.04

−20
SEAFORD CHALK

−30 −30.5

−40 LEWES NODULAR

CHALK

hardgrounds

−50

Chalk Rock
hardgrounds

−60 −60.03

NEW PIT CHALK

−70

Figure 3.10 Geophysical logs showing water inflows from Seaford, Lewes Nodular and New Pit Chalk,
Figheldean Borehole, River Avon catchment (courtesy of Environment Agency).

Chalk Rock hardgrounds near the base of the Lewes Nodular the induction resistivity profiles (Induction RES). The Chalk
Chalk (formerly marking the base of the Upper Chalk). These Rock horizons are identified by the gamma-ray (NGAM)
hardgrounds are characteristic of shallower water environments logs and up to three individual hardground developments are
of deposition near the margins of the Chalk basins and they indicated on the diagram.
represent sea-floor surfaces and areas of shallower water The static water level (SWL) shown represents the water
where the sea-bed sediments are condensed and have become level measured at the time of logging in late summer and
indurated. Logging shows at least three distinct zones of autumn. In July, the water table is at the top of the lowest
higher gamma-ray activity and higher resistivity occupying hardground in Musseldean Copse Borehole but further east,
an interval of 8 to 15 m which are present towards the base closer to local abstraction, it lies within the underlying New
of the Lewes Nodular Chalk (see for example Figures 3.10, Pit Chalk. In November, the water level in the Willoughby
3.11 and 3.12). Minerals such as glauconite and francolite, Hedge Borehole further west is at the top of the New Pit
associated with the shallow water and subaerial exposure, Chalk. These observations suggest that local abstraction is
are responsible for the enhanced natural gamma-ray activity. fed by flow along the Chalk Rock hardgrounds for at least
Away from the margins of the depositional basin in areas of part of the year and that when the water level drops below the
deeper water sediment, these hardground surfaces become Chalk Rock horizons, increasingly deeper inflows in the New
replaced by an expanded nodular Chalk sequence containing Pit and Holywell Nodular Chalk contribute to the abstraction.
thicker, well-developed marl seams. The hardground surfaces While the locations of flow horizons often appear to be
are important hydrogeologically because they tend to become controlled by lithologically distinct stratigraphical horizons
the focus for concentrated shallow groundwater flow (as in the such as the Chalk hardgrounds, the hydraulic and chemical
Tidworth example) and can have associated high yields. This influence on their development is such that the flow horizons
is probably a result of the lithological contrast and different develop within a few tens of metres of the Chalk surface.
fracturing styles compared with the adjacent chalk beds. Therefore, while flow may follow a particular horizon at
Geophysical logging by the British Geological Survey shallow depths, in dipping strata, once the layer reaches a
in several Chalk boreholes in the Fonthill Bishop area certain depth, it will cease to be associated with fractures and
of Wiltshire, on behalf of the South West Region of the flow will be transferred to a different stratigraphical level
Environment Agency, also identified strong development (although again perhaps associated with certain lithologies)
of the Chalk Rock hardgrounds and their importance in as it moves towards surface discharge points. In the Wessex
groundwater flow. Figure 3.12 illustrates a west–east cross- Basin, groundwater flow is commonly inwards from the
section through some of the boreholes located close to the edge of the basin towards the coast and is therefore broadly
interfluve of the Fonthill Brook catchment where the water in the direction of dip. Generally, hydraulic gradients are
level is relatively deep (80 to 100 m). The land surface shallower than the dip of the strata and therefore groundwater
reflects the relative hardness of the Chalk layers as indicated by flow in the shallow solution-enlarged fracture systems will

37
Well name: Tidworth exploration borehole
Depth Caliper NGAM Focused RES Sonic DT Fluid TEMP Fluid EC25 ac-FM3DQ
(mbd)
7.5 (in) 10 0 (API) 20 60 (ohm/m) 160 180 (usecs/ft) 50 9 (DegC) 10.5 500 (uS/cm) 600 0 (m3/h) 25
Chalk subdivision
Fluid TEMPQ Fluid ECQ 25 ac-FM4DQ
9 10.5 500 600 0 (m3/h) 50
0
SWL 2.98mbd
Q/s = 78 m3/h/m

SEAFORD CHALK
−10
BLANK CASING inflow
−15.46

−20

main inflow
Stockbridge Rock ?
−30

−40

−50

SEAFORD CHALK
−58.37
−60

LEWES
−70 NODULAR
CHALK

−80 Chalk Rock

Hardgrounds

−90 −90.38

NEW PIT CHALK

Figure 3.11 Geophysical logs showing concentrated water inflow above a hard band in the Seaford Chalk,
North Tidworth Borehole, River Bourne catchment (data kindly provided by Thames Water in 2002).

tend to be transferred from lower to higher stratigraphical The logging data were digitised by the BGS and assembled
horizons as it moves down a catchment. into a scale cross-section to show the relationship of the
An example of this is given in Figure 3.13, which shows Chalk units and their water inflows along a north-east–
a section down the River Bourne catchment. The section south-west line from Bradley to Itchen Down Farm in part
is aligned approximately north–south and runs from of the River Itchen catchment. The lithostratigraphical
the Aughton Borehole to the Tidworth Borehole via the interpretation shows that the boreholes all commence
Leckford Bridge Borehole, all of which are close to the in the Seaford Chalk and penetrate the Lewes Nodular
river, hence groundwater is at shallow depths. It is evident Chalk and the deepest just encounters the New Pit Chalk.
that the strata dip down the catchment more steeply than the The profiles are very similar and it is possible to identify
topographical gradient or the water table and that the dip is particular named horizons (for example the Seven Sisters
steeper in the north. Geophysical logging of the Aughton Flint band, Belle Tout Marl, Navigation Marl) in each
Borehole indicated flows at horizons in both the basal West borehole. In the lower part of the Lewes Nodular Chalk,
Melbury Marly Chalk and the underlying Upper Greensand. the Lewes Marl, Caburn Marl, Southerham and Glynde
Further down the catchment at Leckford Bridge, where the Marl can be identified by their prominent low resistivity.
Seaford Chalk crops out, logging indicated flows in the These horizons represent the deeper water equivalent of
Lewes Nodular Chalk and the lower part of the overlying the Chalk Rock hardgrounds seen further west.
Seaford Chalk. To the south at Tidworth, where the Lewes The water inflows are interpreted from the fluid log profiles
Nodular Chalk is deeper, little flow was obtained from this recorded whilst pumping (suffixed ‘Q’) and their positions
unit, and a hard layer, believed to be the Stockbridge Rock are depicted by the horizontal blue lines. It is evident that
Member in the Seaford Chalk, provided most of the flow. the main inflows are predominantly from the Seaford Chalk
Thus flow is controlled by suitable horizons near to the and frequently from the same stratigraphical horizon,
ground surface. notably the Seven Sisters Flint band and the Belle Tout beds,
Figure 3.14 illustrates geophysical logs that were which contain some hardgrounds. Figure 3.14 suggests
recorded in several of the Candover Scheme exploration that the inflows at and above the Seven Sisters Flint band
boreholes in the eastern part of the Basin in the 1970s (by probably contribute to the spring flows north-east of Totford.
the Water Research Centre (WRC) and the Instititute of Figure 3.15 shows the geophysical log measurements for the
Geological Sciences (IGS) using analogue equipment). BGS Totford Borehole in the middle of the section in more

38
 WEST EAST
Willoughby Hedge
[ST83SE62] Musseldean Copse, Fonthill Bishop
Depth Caliper Induction RES Fluid TEMP [ST93NW7]
(mOD) 5 (in) 10 2220 (cps) 2260 10.2(DegC)11.2
NGAM Fluid EC25 Depth MCAL NGAM Fluid TEMP
0 (API) 40 360(uS/cm)440 (mOD) 5 (in) 7.5 0 (API) 20 9.6 (DegC)10.6
Induction RES Fluid EC25
2220 (cps) 2260 430 (uS/cm) 490
220 218.00
220 Ground su
rface Fonthill Bushes
215.00 [ST93NE43]
213.43
211.72
210 Harder 210 Depth Caliper Induction RES Fluid TEMP
(mOD) 5 (in) 7.5 2220 (cps) 2260 2220(DegC)10.5
Grou HSGR Fluid EC25
Softer nd surfac 0 (cps) 20 440 (uS/cm) 450
200 200 e 200 200

Harder
191.25
190 190 190 189.02
Softer

180 Harder 180 180

170 170 170

Seaford
161.74 Chalk
160 Fissures 160 160
above Lewes
water Nodula
table r Chalk
Seaf
ord C
150 150 150 149.96 halk 150

39
Elevation (mOD)
Hardground 3
SWL 'Chalk Rock'
140.77 Lew
7 Nov '01 140 140 Hardground 2
es N
odul 140
SWL ar Ch
New P 5 July '01
alk
it Chalk Chalk Rock 1 131.08
130 130 130
127.73

120 119.59 120 120

New
Pit C
Holyw halk
ell Nod
110.36 ular Ch
110 PlenusMarls a lk 110 110 108.05
107.77 107.61

Zig Za SWL
g Chalk 9 July '01
100 100 100 Holy 100
96.85 95.92
95.21
well
Plenus Marls Nod
ular
Cha
90 90 lk 90
Zig Z
ag C
82.34 halk
80 Water inflow identified by fluid logs 80

70

0 2500 5000 7500 10000

Section distance (metres)

Figure 3.12 West to east cross-section showing the relationship of the Chalk units to topography and the position of the summer
water table relative to the Chalk Rock hardgrounds, Fonthill Bishop area, Wiltshire (courtesy of Environment Agency).
 NORTH SOUTH

175
Aughton Collingbourne Kingston Collingbourne Ducis Leckford Bridge Tidworth

Aughton (+138m)
ac-FMDQ
Chalk Rock exposed
150 Depth HSGR
in railway cutting Leckford Bridge (+121m)
(mOD) 5 (cps) 35 0 (%) 100
140 Depth ABH 32in RES Fluid ECQ 25
(mOD) 70 (ohm/m) 200 585 615
Focused RES Ac-FlowmeterQ Tidworth exploration BH 1 (+114m)
130 Ground su
rface 70 (ohm/m) 200 0 (%) 100 Depth Focused RES ac-Flowmeter-Q
125 SWL (mOD) 60 (ohm/m) 160 -5 (m3/h) 50
120 +119 120 120

e
Ch
S al
or
d
+113 Le
w +109

af k FM
es
110 N 110 110
od +107
ul
ar
100 100 N Ch 100 100 +97

40
ew al
+94 H k
ol Pi FM

Elevation (mOD)
yw tC
90 el ha 90 90 +88
+86 lN lk +85
od FM +84
+81

Zi
ul

g
ar
80 80 80

Za
Ch

g
75 al
k

Ch
a
FM

lk
70 Chalk 70

FM
Rock
Hardgrounds

M
60 60

end of River Bourne catchment (courtesy of Environment Agency and Thames Water).
W arl

U
p
es y C

pe
r
t M ha

G
ry
el lk F
50 50 50

re
e
bu M

ns
and
40

FM
Chalk Rock
Depth Hardgrounds
(mOD) 30
Water inflow horizon
25

2000 4000 6000 8000 10000

Figure 3.13 Scale cross section N-S showing the relationship of the Chalk units and the water inflow horizons, northern
Section distance (metres)
detail. The area-corrected (AC-flow meter) measurement waters are relatively warm, all suggest that they are part of
represents the cumulative flow rate to the pump near the top a slow-moving, poorly developed groundwater flow system.
of the borehole and the inflow histogram, processed from the The more rapid groundwater flows in this system originate
curve (shown in green), depicts the main inflows. The interval at the Chalk outcrop and discharge via the sandier layers in
packer-test measurements (shown in red) agree well with the the Palaeogene cover.
flow-meter profile, although the packer testing identified Where the Chalk aquifer is less deeply confined it is more
higher permeability within the Lewes Marl–Southerham productive; generally the highest borehole yields in the
Marl interval, not seen by the flow meter (perhaps because Wareham area are from the top 50 m of the Chalk in outcrop
the flow-meter measurement is head dependent). The flow areas, and from confined Chalk where the Palaeogene cover
meter however resolved important inflow through the is less than 70 m thick (Figure 3.17).
wellscreen slots near the top of the borehole, which was The section in Figure 3.17 illustrates the likely transfer
not selected as a packer-test interval. The matrix hydraulic of groundwater from the Chalk to the Palaeogene sediments
conductivity measurements made on the cores, shown in as it moves down flow lines from west to east and from
black, are uniformly low and emphasise the importance of north-west to south-east. The potentiometric surface is
the solution-enhanced fractures for groundwater flow in the above ground level over much of the area and has a gradient
Chalk. The flow meter log reveals that when pumped, the which is shallower than the dip of the Chalk. Groundwater
overwhelming bulk of the groundwater is obtained from less is considered to circulate within the sandier layers between
than 30 m depth (above about 50 m OD) in the borehole. more clayey layers in the Palaeogene sequence in lower
Figure 3.16 presents a comparison of caliper logs of parts of the basin. In the Arne area east of Wareham, where
selected Candover scheme boreholes plotted as depth below the Palaeogene sediments are worked for ball clay, the sandy
log datum (usually ground level or casing top close to layers have relatively high heads and shallow boreholes are
ground level). The logs represent the variation of borehole allowed to flow to release the pressure.
diameter with depth and the plot reveals that generally Groundwater circulating within the Palaeogene sedi-
there is enlarged diameter above 40 m depth in this area. ments can sometimes be recognised by its effect on fluid
The diameter enlargements may represent intervals of temperature logs recorded within the casing of newly drilled
softer chalk having higher porosity due to weathering and boreholes. Figure 3.18 illustrates fluid temperature profiles
Pleistocene effects, solution enlargement of fractures due recorded within relatively long blank casing (approximately
to concentrated groundwater flow at certain horizons, or a 100 m) against Palaeogene deposits overlying the Chalk
particular lithology, for example soft sponge beds or marly in the Raglington Farm Borehole at Row Ash [SU 54 13],
chalk. They are also influenced by the current and historical in the Blackhorse Farm Borehole at Bishops Waltham in
range of seasonal water level fluctuation, and particular Hampshire [SU 56 14], and in a borehole in the Wareham
effects of the drilling, though generally drilling effects are area. In the latter, the two fluid temperature profiles reveal
usually minor. The position of water inflows identified by elevated temperature against the casing interval (0 to 100 m)
pumped fluid and flow meter logging drawn on the plot show caused by the casing cement. In each of these boreholes,
they are all less than 60 m below surface and the majority are the cooling observed opposite the sandy (low gamma ray)
at less than 40 m depth, reflecting the hydraulic control on horizons signifies a more rapid circulation of shallow
the development of permeability. Their shallow depth is an groundwater through the clean, sandier layers, removing the
indication that current recharge inputs can be accommodated heat generated by the cement.
by flow within this shallow zone. It may be, in fact, that The borehole construction shown for the Raglington
current recharge could be dissipated within the top few Farm Borehole in Figure 3.18 also illustrates one of the
metres only and the deeper flow horizons indicated may have difficulties that can be faced when drilling into the Chalk
been developed during drought periods or during intervals in beneath Palaeogene strata where the top Chalk surface has
the past when base levels and sea level were lower. been deeply eroded and karstified. The casing was initially
landed in firm Chalk strata at 126 m depth (see gamma-
3.3.3.2 The confined aquifer ray log), only to encounter additional Palaeogene sediment
The foregoing discussion indicates that the nature of the filling conduits in the Chalk land surface below. A dropset
groundwater flow system in the Chalk is such that inflows casing had to be installed into firm Chalk from 126 to 143 m.
and better borehole yields are usually obtained at shallower Geophysical logging has demonstrated that in several
depths where there has been a greater natural concentration areas around Wareham the basal Palaeogene sediments are
of groundwater flow to develop the permeability. However, sandy (Figure 3.17). In areas of artesian overflow from the
in the Wareham5 area, where the Palaeogene sediments Chalk, artesian pressure can be encountered within sandy
overlying the Chalk can be more than 100 m thick, fluid basal layers of the Palaeogene sediments, and cause drilling
logging has identified isolated water inflows at much greater difficulties, before the Chalk is penetrated.
depths (-170 m OD in the Stoborough and Holton Heath Neither the Raglington Farm nor Blackhorse Farm
boreholes, and at -220 m OD in Wareham Common OBH2). boreholes had artesian flow, but they had poor yields. Both
Such deep inflows are not normally encountered in the boreholes were acidised after drilling to improve the yields
Chalk aquifer, but the facts that the yield from these deep and residual acid was present when they were logged (see
horizons is negligible (borehole specific capacities are less Fluid EC 25 measurements, Figure 3.18) and the quantity
than 1 m3 h–1 m–1), that groundwaters have the chemical and of acid remaining was an indication of little or no natural
isotopic characteristics of long residence time and that the groundwater flow taking place to disperse it. Yields were not
improved by the acidisation because there was no existing
fracture system present to be developed. At these locations,
5
A significant number of water exploration boreholes were drilled, tested the fluid temperature logging suggested that the current
and geophysically logged in the Dorchester-Wareham area of Dorset as groundwater flow was taking place mostly within the
part of the Wessex Water Wareham Groundwater Project in 1992–1995.
Subsequently boreholes at Lytchett Minster, north-east of Wareham, were
overlying Palaeogene sediments. In the Blackhorse Farm
selected for trialling of an aquifer storage and recovery (ASR) scheme and Borehole, cooling of the borehole fluid between 110 and
were also geophysically logged. 120 m depth identifies some groundwater circulation within

41
 150
SOUTH-WEST Bradley 2B NORTH-EAST
N

142500
Wield 3B

140000
Wield 3B Bradley 2bS
wield3bS RES 3bS TEMP RES Fluid TEMP
Totford obh 80 180 10.2 10.7 70 (ohm.m) 170 9.5 11.5

Northing
3bS Fluid EC TQ

137500
430 460 9.5 (DegC)11.5
LINE OF SECTION Flowmeter-Q uplog Fluid ECQ

Depth (mOD)
Depth (mOD)
Bradley 2B
25 125 350 750
Abbotstone 115.01 115

135000
Itchen DF2 110 110
Abbotstone 455000 457500 460000 462500 e
16RES 7/4/76 Fluid ECQ Easting rf ac
100 (ohm/m) 200 420 (uS/cm) 425
100 su 100 99.36 100 99.29
Fluid TQ

Depth
(mOD)
n d
9.8 (1/4/76) 10.8
BGS Totford Borehole rou
92.3 Gr G
90
ou Focused RES TEMPQ ac-Flowmeter 90 90
88.44 nd L
su 50 (ohm.m) 150 10.5 (DegC) 11.5 -5 (m3/h) 35
rf Fluid ECQ25 Inflow %
SW

Depth
(mOD)
Itchen Down 2 ace
525 (uS/cm) 550 0 40
Fluid ECQ IGS 80 80 80 80
16 RES 21/10/76
70 (ohm.m) 170 382 387 76.02 SS Flint
Fluid TQ Pump

Depth
(mOD)
10.5 (DegC) 10.7
70 70 70 Seaford Chalk FM 70 70

SS Flint Pump
63.16 64.02
BTM
60 60 60 60 60

42
SS Flint
Seaford
SS Flint
53.95 Chalk FM BTM

Elevation (mOD)
50.29
50 50 50 50 50 50
Pump BTM

BTM BTM
41.62
40 40 40 40 40 Nav HG
Lew
Pump lar es N
odu Nav HG odu
31.94 e s N FM lar C
31.0 halk Shoreham M Lewes M
30 30 Lew halk
C 30 FM 30 29.33 30
Pump ...pump... Caburn
Nav HG Marl

Figure 3.14 Scale cross-section north-east to south-west using digitised 1970s logging
Nav HG Lewes M
20 20 20 20 20
Southerham M
Nav HG acid

data to correlate Chalk units to show water inflows, Candover Scheme boreholes, Hampshire.
residual
Lewes Marl 10.89
10 10 10 CM 10 10
6.57
Southerham M

0.48
0 0 0
CM
Glynde Marl
Southerham M New
FM Pit C Water inflow horizon
-9.79 it Chalk halk
-10 New P -10 F M Minor inflow
Glynde Marl BTM – Belle Tout Marls
CM – Caburn Marl
Nav HG – Navigation hardgrounds
-20 SS Flint – Seven Sisters Flint

2500 5000 7500 10000 12500

Section distance (metres)

the top 10 m of the Chalk. Such cooling was not evident 1974) proved a direct connection and indicated turbulent
however in the Raglington Farm Borehole where the base of flow through a discrete conduit system. The velocity of
the Palaeogene was deeper (139 m). groundwater movement was calculated to be 2 km d–1.
This conduit system is located in the chalk at the northern
3.3.3.3 Springs margin of the Palaeogene outcrop and it is thought that it has
Many springs issue from the Chalk and may result from the developed due to the increased solution potential of run-off
occurrence of marly bands, which impede vertical water from the nearby or overlying Palaeogene sediments. More
flow, and hard, well-fractured rock bands along which than 70 per cent of the tracer dye was recovered from the
there can be considerable lateral flow. Springs determined test, and the majority of it arrived at one spring within a few
by lithology are common at the Chalk scarp foot and are days. This implies that flow occurred through a discrete set
often seasonal. Springs may also occur at other levels in the of conduits, rather than a diffuse network.
Upper/Middle Chalk (White Chalk), but they are usually
small and tend to dry up during the summer and autumn. 3.3.4 Surface water – groundwater interaction
Those springs that occur on the dip slope of the Chalk are
nearly always in the base of valleys and are of the overflow 3.3.4.1 Complex winterbourne behaviour
kind, i.e. where the water table intersects the surface. Winterbourne behaviour, resulting from annual variations
Identifiable perennial springs often form the perennial in groundwater levels, is common in streams in the Wessex
head of winterbourne streams, with intermittent springs Basin and is described in Chapter 1. However, across the
contributing to winter streamflows upstream. During Wessex Basin there are several rivers and streams where
extreme recharge events, such as occurred in the winter of the surface and groundwater interactions appear to be more
2000 to 2001, fissures normally above the winter water table complex than those implied by the simple winterbourne
become saturated and flow as springs at high elevations, model. During the dry summer months, a number of Chalk
causing groundwater flooding. streams and rivers tend to flow in their upper reaches and
The springs at Bedhampton [SU 707 064] form the then cease flowing for a distance along their course before
largest public supply spring source from the Chalk in the flowing again further downstream; in a few instances, more
UK. Rapid groundwater flow to the springs has been shown, complicated flow patterns are observed.
where a tracer study carried out between a series of stream Such erratic bourne behaviour has been periodically
sinks to the north of the Bere Forest Syncline and the springs monitored by the Environment Agency in a number of Chalk
5 km away to the south of the syncline (Atkinson and Smith, streams in the Wessex Basin. In general, this has involved

BGS Totford Borehole

Depth Caliper Focused RES TEMPQ ac-Flowmeter Packer test K

(mOD) 4 (in) 9 50 (ohm.m) 150 10.5 (DegC) 11.5 -5 (m3/h) 35 .01 (m/d) 1000
Totford Lithology
Fluid ECQ25 Inflow % + matrix K
525 (uS/cm) 550 0 40 .001 (m/d) 1000
80
BLANK CASING 10in
76.02
SWL

70 PLASTIC WELLSCREEN
(8in)
Seven Sisters Flint
64.02

60
SEAFORD CHALK

Belle Tout Beds (HG)

50

41.62 Shoreham Marl

40

Navigation HG
30
LEWES NODULAR CHALK

20 Lewes Marl

Bridgewick Marl

10
Caburn Marl

Southerham Marl
0.48
0
Glynde Marl
NEW PIT CHALK

Figure 3.15 Pumped fluid and flowmeter logging and packer test measurements used to identify horizons
of fluid movement, Chalk aquifer, BGS Totford Borehole, Hampshire.

43
Well name: Candover boreholes (depth below ground plots) Main water inflows identified by fluid logging

Logs compared as depth below datum

Depth Totford caliper Wield 3A caliper Wield 3B caliper Bradley 2A caliper Bradley 2B caliper Axford 1A caliper Axford 1B caliper
(mbd)
5 (in) 10 9 (in) 14 9 (in) 14 9 (in) 14 9 (in) 14 8 (in) 18 8 (in) 18
0
SW L

−10 SW L
SW L
SW L
SW L
−20 SW L SW L

−30

−40

−50

−60

−70

−80

−90

−100

Figure 3.16 Comparisons of caliper logs and main water inflow horizons,
depth below surface plot, Candover Scheme boreholes.

noting whether the stream is flowing or not at a number of tables — the actual signature of the South Winterbourne was
specified points upstream from a datum near to the perennial always more complex during the monitored period. This was
head, then repeating the survey at intervals using the same particularly the case in mid/late summer when a signature
observation points. The results are then plotted to produce a of the form no flow/flow/no flow/flow occurred downstream
‘bourne signature’ for the stream. from the maximum river-head position each year.
The South Winterbourne, a tributary of the River Frome, Figures 3.21 to 3.26 show the winterbourne signatures
has been monitored at flow observation points at 1 km of a number of other streams and rivers monitored by the
intervals along the length of the river, upstream of West Environment Agency in the Wessex Basin (all data plots
Stafford Bridge (Figure 3.19). During the first survey courtesy of Environment Agency, South West Region).
(January 1993) a record was made of whether the river was It can be seen that that the River Till (Figure 3.21) and the
flowing or not at each observation point and then the survey Chitterne Brook (Figure 3.22), both tributaries of the Wylye,
was repeated at intervals over a period of several years. The show the expected winterbourne behaviour for the majority
results are illustrated in Figure 3.20. For each date on which of the period over which they were monitored, with the
the survey was undertaken, the corresponding location was upstream limit of the rivers gradually moving downstream
shaded if the river was found to be flowing, or left blank if during the drier summer months. However, in May 1997
no flow was observed. The chart was then shaded or left the River Till flows over a short distance between the 8 and
blank in intermediate, unobserved periods to produce the 9 km observation points before the flows disappears again
smoothed record shown in Figure 3.20. and re-emerges further downstream. This behaviour is seen
The figure therefore provides a simple visual indication again further upstream in Jan/Feb 1998 and in the Chitterne
of the complexity of the flow characteristics of the river. For Brook in April 1997.
example, considering the year 1994, the chart shows that In the Piddle catchment, however, the Bere Stream
in January the river flowed constantly from a point 15 km (Figure 3.23) shows the opposite behaviour to that
upstream from the datum point. The head of the river then expected, producing a signature with an inverted U-shape.
declined, as would be predicted by the classic winterbourne Thus it flows all year round in its upper reaches but dries
model until, by mid July, it had reached approximately the up during the summer months in the downstream section,
13 km mark. However, between mid July and the end of and the point at which it stops flowing moves progressively
November, while the head of the river declined very slightly, upstream from the perennial head.
flow also ceased between around 9 km and 1 km above the The Devil’s Brook (Figure 3.24) (also a tributary of the
datum — a phenomenon not predicted by the conventional Piddle), the South Winterbourne (Figure 3.20) (in the adjacent
bourne flow model. Thus, whereas the conventional Frome catchment) and the North Winterbourne (Figure 3.25)
winterbourne model would be expected to result in an annual (close to the Bere stream, in the Stour catchment) all show a
bourne ‘signature’ with a roughly ‘u’ or ‘v’ shaped open form similar behaviour. Flow occurs downstream from the stream
annually — indicating a progressive drying of the stream to head and then dries up over a short section before the flow
its perennial head as a result of falling and then rising water resumes again further downstream. It should be noted, however,

44
 WEST EAST SOUTH-EAST

A D
B
NGAM Focused RES Fluid TEMP C NGAM Focused RES Fluid TEMP
100 Gamma ray 1C Focused RES2 Fluid TEMP
0 (API) 10 80 (ohm.m)120 11 (DegC) 11.5 0 (API) 30 40 (ohm/m) 65 13.8(DegC)17.8
0 (API) 20 85 (ohm.m)105 11.2 (C) 12.2 NGAM Focused RES Fluid TEMP1
Fluid ECQ1 25 Fluid EC 25
Fluid EC25 0 (API) 30 50 (ohm/m) 90 13.2 13.7
450 570 700 (uS/cm)1200
ac-Downlog 300(uS/cm)900 Wessex 16RES Fluid EC1 25
ac-Flowmeter

Depth (mOD)

Depth (mOD)
UFLO1Q ac Flowmeter 50 (ohm/m)100 645(uS/cm)745

Depth (mOD)
0 .8

DODDF9 Lithology
5 (m3/h) 20

Holton-H Lithology

Lithology

Lithology subdiv.
-5 (% total) 110 ac Flowmeter

Depth (mOD)
0 (%total) 120
40 40 40 40

20 20 19.9 20 20 Potentiometric surface

9.87
−0.75 Ground surface
0 0 0 0 0

−20 −20 −24.17 −20 −20

−29.58
−40 −40 −40

−59.18 Palaeogene
−60 −60 −60
−66.69

−80 −82.17 −80 −80

−87.76

showing main water inflow horizons in the Portsdown Chalk.

−97.43

45
−100 −100 −100 −100 −102.96
−113.34
−120 −120

Elevation (mOD)
Portsdown
−140 −140.3 −140
Chalk Fm

−160

Figure 3.17 Scale cross-section Wareham area, based on geophysical log data
Spetis −174.32
bury C −180
halk M
br
−200 −200
Cross-section in the Wareham area
−220

−240 −238.97

A B
−260
C

Northings (m)
LINE OF SECTION D

Main water inflow identified by fluid logging

−300
390000 400000 410000

90000
Eastings (m)

5000 10000 15000 20000

Section distance (metres)

 Raglington Farm Borehole, Blackhorse Farm Borehole, Wareham
Rowash Bishops Waltham
Depth

Hyde-Farm-Lith.
Caliper NGAM Fluid TEMP Caliper NGAM Fluid TEMP Gamma ray HS Fluid TEMP

Bishops-W-Lith.
(mbd)
6 (in) 9 0 (API) 50 11 (DegC) 16 7.5 (in) 10 0 (API) 80 12.5 (DegC) 15.5 0 (cps) 50 14 (17/9/92) 19
Raglington Farm
lithology
Fluid EC25 Fluid EC 25 TEMP 29/9/92
250 (uS/cm) 1250 500 (uS/cm) 2500 14 (DegC) 19
0
SWL
−10 Oakdale
Sand
−20

−30 Creekmoor
Clay
−40
PALAEOGENE
−50 Creekmoor
Sand
−60

−70

−80 LONDON
CLAY
BLANK CASING
−90

−100
SPETISBURY
−110 CHALK

−120

−130 PALAEOGENE
Dropset Steel
−140 Casing TARRANT
Dropset Steel CHALK
−150 Casing
−160 SPETISBURY CHALK

−170

−180
NEWHAVEN
CHALK
−190
TARRANT CHALK
−200

−210

−220

−230
NEWHAVEN CHALK acid residual
−240

Figure 3.18 Fluid temperature logs of selected Wessex Basin boreholes with thick Palaeogene cover displaying
local cooling within casing indicating groundwater movement within sandier layers of the Palaeogene strata.

that the upper reaches of the River Piddle exhibit sections 1986; Wessex Water Authority, 1988; National Rivers
which, while not necessarily drying, are influent (Howden, Authority, 1992a, 1995a; Halcrow, 1995; Stuart, 2000). A
2004). This is likely to be the case for other streams, and this study carried out by Marcus Hodges Environment (1999) on
simple graphical ‘signature’ approach can only be regarded the Upper Piddle found that although reducing the abstraction
as indicating the general nature of the streams — their true
complexity can only be revealed by accretion gauging.
The River Bourne (Figure 3.26) was looked at in a detailed
study carried out by the Environment Agency that examined N
low flows in both the River Bourne and the Nine Mile River Fro
(Environment Agency, 2001; Environment Agency and m e
13
Water Management Consultants, 2004). The Bourne 12
appears to dry up along most of its length during the dry 0
90 11 0
months except for a section 24 to 25 km from the datum 10 1
9 2
point (probably associated with discharge from a sewage 4 3
8
treatment works) and from 2 km to the datum. However 6
5
e
7 ourn
a section around 12 km from the datum was seen to be in terb
th W
particularly vulnerable to drying. The Agency also looked Sou
at flow accretion profiles and found these to show a similar 65
3 3
70
flow pattern to that recorded at the observation points. Bagshot, Barton &
The causes of this behaviour in these rivers and streams Thames Group Chalk formations
Bracklesham groups
appear to be varied. In some cases, a lowering of the water table
by local abstractions may result in effluent streams becoming Figure 3.19 South Winterbourne with locations of a
influent over certain reaches. For example, studies carried out number of observation points, showing flow to the 12 km
on the River Piddle, which also shows a loss of flow in its upper mark (observation points courtesy of Environment Agency).
reaches, have concluded that stream-bed leakage is thought to contains OS data © Crown copyright and database rights 2017 Ordnance
be increased by groundwater abstractions. (Mansell-Moullin, Survey [100021290 EUL]. Use of this data is subject to terms and

46
Distance upstream (km)
25 25
20 20
15 15
10 10
5 5
0 0
JFMAMJJASOND JFMAMJJASOND JFMAMJJASOND JFMAMJJASOND JFMAMJJASOND J FM AM J
1993 1994 1995 1996 1997 1998

Figure 3.20 South Winterbourne flow signature – Frome catchment (courtesy of Environment Agency).
Distance upstream (km)

20 20
15 15
10 10
5 5
0 0
JFMAMJJASONDJFMAMJJASONDJFMAMJJASONDJFMAMJJASOND JFMAMJJASOND JFMAMJJASOND JFMAMJJASOND
1993 1994 1995 1996 1997 1998 1999

Figure 3.21 River Till flow signature – Wylye (Avon) catchment (courtesy of Environment Agency).
Distance upstream (km)

No Information

No Information

No Information
20 20
15 15
10 10
5 5
0 0
JFMAMJJASOND JFMAMJJASOND JFMAMJJASOND JFMAMJJASOND JFMAMJJASOND JFMAMJJASOND JFMAMJJASOND
1993 1994 1995 1996 1997 1998 1999

Figure 3.22 Chitterne Brook flow signature – Wylye (Avon) catchment (courtesy of Environment Agency).
Distance upstream (km)

25 25
20 20
15 15
10 10
5 5
0 0
J F MA M J JA S O N D J F MA M J JA S O N D J F MA M J JA S O N D J F MA M J JA S O N D J F MA M J JA S O N D J F M AM
1993 1994 1995 1996 1997 1998

Figure 3.23 Bere Stream flow signature – Piddle catchment (courtesy of Environment Agency).

25 25
Distance upstream (km)

20 20
15 15
10 10
5 5
0 0
J F MA M J JA S O N D J F MA M J JA S O N D J F MA M J JA S O N D J F MA M J JA S O N D J F MA M J JA S O N D J F M AM
1993 1994 1995 1996 1997 1998

Figure 3.24 Devil’s Brook flow signature – Piddle catchment (courtesy of Environment Agency).

47
 25 25
Distance upstream (km)

20 20
15 15
10 10
?
5 5
? ?
0 0
J F M A M J J A S O N DJ F M A M J J A S O N DJ F M A M J J A S O N DJ F M A M J J A S O N DJ F M A M J J A S O N DJ F M A M
1993 1994 1995 1996 1997 1998

Figure 3.25 North Winterbourne flow signature – Stour catchment (courtesy of Environment Agency).
Distance upstream (km)

35 35
30 30
25 25
20 20
15 15
10 10
5 5
0 0
JFMAMJJASONDJFMAMJJASONDJFMAMJJASOND JFMAMJJASONDJFMAMJJASOND JFMAMJJASOND JFMAMJJASONDJFMAMJJASONDJFMAMJJA
1992 1993 1994 1995 1996 1997 1998 1999 2000

Figure 3.26 River Bourne flow signature – Avon catchment (courtesy of Environment Agency).

rate at the nearby Alton Pancras Borehole [ST 70 01]] resulted geology (in terms of Chalk lithology and structure) and surface
in an increase in flow in the Piddle, the additional flow recorded water flows. One such study concerns the Upper Piddle.
in the river was only 30 per cent of the amount anticipated
and that a significant proportion of the water was effectively 3.3.4.2 Case history — the Upper Piddle
‘lost’. They concluded that the ‘missing’ water was being The flow characteristics of the River Piddle do not follow
transmitted through the Middle Chalk rather than emerging the typical winterbourne behaviour discussed in Chapter 1.
as flow in the River Piddle. Thus there would appear to be a The hydrogeological processes operating in the Piddle
natural streamflow loss from the Piddle that is exacerbated by catchment have been the basis of several detailed studies
abstraction, a conclusion supported by the evidence of mid- and a number of useful conclusions can be drawn from
stream drying in the Piddle tributaries, described above. these with respect to the complex surface–groundwater
In the River Bourne study (Environment Agency, 2001) interactions in chalk streams and rivers.
historical references (Shore, 1894; Dewar, 1898) were examined
and it was concluded that the River Bourne has always been an
ephemeral stream and that similar flow characteristics to those 1
05
seen today were observed in the late 1800s. The relationship Devil's Brook
North Winterbourne
N
between groundwater and surface water behaviour in these Bere Stream
streams is therefore complex, with groundwater abstractions
involved in some cases, and not in others.
One cause of the complex winterbourne behaviour
could be related to the geological setting of the stream.
For example, general lithological variations might be a
Oc

sum
Dry mer
ca in s

1
00
contributory factor. Thus several of the streams illustrated
sio um

in
na m

in Figures 3.20 to 3.26 appear prone to drying during

er n
m i

lly er
m up

the summer after they flow over the Lewes Nodular

dri
su ies

es
Dr

Chalk/Seaford Chalk boundary and the lower section

up

of the Seaford Chalk. This can be observed in the South 0 2.5 km

Winterbourne, and in the Devil’s Brook, the Bere Stream
and the North Winterbourne (Figure 3.27). However, the
0
95
behaviour of other streams does not follow this pattern. 75
3
80
3
85
3

The River Till (Figure 3.21) dries up progressively along Bagshot, Barton and Bracklesham
its length over the Lewes Nodular Chalk and the Seaford groups Holywell Nodular Chalk Formation

Chalk in its upper reaches but flows continuously in the Thames Group Zig Zag Chalk Formation
Portsdown Formation Gault Formation
lower reaches over the Lewes Nodular Chalk. The Chitterne
Culver Chalk Formation Lower Greensand Group
Brook (Figure 3.22) dries up continuously along its length,
Newhaven Chalk Formation Kimmeridge Clay Formation
which includes the Seaford Chalk, the New Pit Chalk, the Seaford Chalk Formation Corallian Group
Holywell Nodular Chalk, the Zig Zag Chalk and the West Lewes Nodular Chalk Formation Bourne observation point
Melbury Chalk. The River Bourne (Figure 3.26) exhibits New Pit Chalk Formation Rivers
the complex behaviour mentioned above.
Thus it is apparent that bourne behaviour may not be Figure 3.27 Winterbourne behaviour of Devil’s Brook,
controlled by Chalk geology at the gross stratigraphical Bere Stream and North Winterbourne in relation to geology.
division level, however, detailed studies of particular streams contains OS data © Crown copyright and database rights 2017 Ordnance
in the Wessex Basin have shown definite correlations between Survey [100021290 EUL]. Use of this data is subject to terms and conditions.

48
Figure 3.28
Geology of the Upper

Bere
Barcombe
Piddle catchment. Plush

Stre
Holcombe Dairy

Lysco
Strea
contains OS data © Crown Expanded

D ev

am
copyright and database Alton Manor area

mbe
m

ils B
rights 2017 Ordnance
Sheep Dip Sheep Dip

roo
Survey [100021290 EUL].
Alton Pancras

k
Use of this data is subject to
Sheep Dip Cecily Bridge
terms and conditions.
Briantspuddle Warren
Morningwell
Binnegar
1
00
Baggs Mill
Piddletrenthide

South House

Riv
rPe
N

idd

Dev ok
Bro
le

Lys eam
Piddlehinton

il’s
Str
com
be
Little Puddle

Waterston
0
95
Tolpuddle

Puddletown

0 2.5 5 Km

3
70 3
75

Flow gauging station (permanent) Head Seaford Chalk Formation

Public supply borehole River terrace deposits, 6 Lewes Nodular Chalk Formation
Springs Solent Group New Pit Chalk Formation
Rivers Bagshot, Barton and Holywell Nodular Chalk Formation
Fault Bracklesham groups Zig Zag Chalk Formation
Surface water catchment boundary Thames Group Upper Greensand Formation
Alluvium Lambeth Group Gault Formation
Clay-with-flints Newhaven Chalk Formation Lower Greensand Group

The River Piddle is relatively small and rises at springs The development of a new lithostratigraphical frame-
in the north Dorset Downs close to the village of Alton work for the Chalk (Bristow et al., 1997; Hopson, 2005) has
Pancras, and flows south-east for approximately 40 km to enabled a much more detailed assessment of the influence
a common estuary with the River Frome before discharging of lithological and structural controls on groundwater flow
into the English Channel at Poole Harbour (Figures 3.28; to be carried out through the development of a more detailed
1.6). Its three main tributaries are the Lyscombe Stream, the understanding of the geology. Howden et al. (2004) uses the
Bere Stream and the Devil’s Brook. new interpretation of the Piddle catchment geology (Newell
Groundwater abstraction in the catchment is high, with et al., 2002) to investigate surface–groundwater interactions
a total maximum daily licensed quantity of 197 Ml day–1 in the river valley and to determine how these effect the
(Howden et al., 2004). The majority of groundwater distribution of surface water flows in the River Piddle.
abstractions (by licensed volume) are for public water
supply of which there are four in the catchment and two of Catchment geology
these are located adjacent to the River Piddle. Low flows The geology of the Upper Piddle is shown in Figures 1.6 and
in the river due to abstraction are exacerbated by natural 3.28. It is predominantly underlain by Chalk, with the river
conditions whereby the river becomes perched and leakage rising just to the south of the Chalk scarp at the junction of
occurs (Marcus Hodges Environment, 1999). the Upper Greensand and Lower Chalk. The Chalk dips to
The hydrogeology of the Upper Piddle catchment has the south by about 2 to 3 degrees and is cut by two sets of
been the subject of several investigations by the Environment faults oriented north-east to south-west and north-west to
Agency and Wessex Water to determine the potential effects south-east6. Fault throws are typically around 5 to 10 m and
on river flow of groundwater abstraction from a source at with a maximum throw of 25 m.
Alton Pancras (Figure 3.28). The flow characteristics of The dip of the Chalk combined with faulting and erosion
the Piddle are complex, with both influent and effluent has resulted in a complex sequence of outcropping Chalk
behaviour seen. Previous studies (Halcrow, 1995; Marcus units along the valley floor and sides. For example, faults
Hodges Environment, 1999; Howden et al., 2004) have cross the Upper Piddle at Sheep Dip and Morningwell
identified the importance of spring sources in supporting (Figure 3.28). The Zig Zag Chalk crops out in the valley
streamflow. floor down to Morningwell, with the Holywell Nodular

49
 Distance along the river (m) Throop (about 1 km east of Briantspuddle) and Cecily
0 5000 10 000 15 000 20 000 25 000 Bridge, with stream-flow depletion occurring between Pid-
4000 dletrenthide and Piddlehinton and also between Tolpuddle
11th June 1986 and Throop. It was concluded that stream-flow accretion oc-
3rd December 1986 curs both due to spring flow entering the river channel and at
26th February 1987 a confluence with a tributary (e.g. the Devil’s Brook).
High Flow Both studies (Marcus Hodges Environment, 1999;
3000 Howden et al., 2004) found a strong correlation between
16th July 1986 the locations of the springs feeding the Upper Piddle
10th September 1986 and their geological setting. Table 3.2 summarises their
Streamflow (ls−1)

7th October 1986

findings and suggests that both lithological and structural
Low Flow
controls by faulting play a role in determining the location
2000 of springs.
Loss of flow to river-bed leakage is a recognised
characteristic of rivers in the Piddle catchment according to
the Avon and Dorset River Authority (1970), which records
the Piddle as ‘going to ground’ during the late summer
1000 months. The loss of flow in the Upper Piddle is quantifiable
using two streamflow gauges on the leakage section: South
House and Little Puddle. Data from these flow gauging
stations have been used to demonstrate loss of streamflow
(Halcrow, 1995; Marcus Hodges Environment, 1999; Stuart,
0 2000) although no analysis or firm conclusions were drawn
from them.
Tolpuddle
Alton Manor Hatch

Cecily Bridge
Affpuddle
Briantspuddle
Throop
Piddletrenthide

Piddlehinton

Puddletown
Barcombe Springs

Howden et al. (2004) derived a leakage hydrograph

from the difference between the hydrographs from two
gauging stations (Little Puddle and South House). If the
contributing subcatchment areas to the gauging stations
are considered, a flow per unit catchment area can be
Gauging Locations
derived for each station (Figures 3.30a, b), and the flow
difference plotted (Figure 3.30c), which enables leakage
Figure 3.29 Flow accretion profiles for the River Piddle and leakage variation with flow characteristics to be seen
(using flow gaugings by Wessex Water Authority, 1986) more clearly. Figure 3.30c shows that flow is always
after Howden et al., 2004. greater upstream than downstream, although the difference
in flow is typically small. However, during the high flow in
1996 to 1997, stream flow at Little Puddle was depleted by
Chalk and the New Pit Chalk outcropping along the valley almost 50 per cent compared with South House. Figure 3.30d
sides. Downstream of the fault at Morningwell, the New Pit shows that the leakage characteristic tends to dominate
Chalk crops out in the valley floor, and further downstream at low flows but that, as flow in Little Puddle increases, it
by the Lewes Nodular Chalk. At White Lackington the becomes less significant. However, there are a number of
Seaford Chalk crops out in the valley floor, and further down exceptions. As there are no springs contributing to stream
the catchment outcrops of younger Chalk units are found, flow between South House and Little Puddle Howden et al.
i.e. Newhaven and Tarrant Chalk. (2004) consider that it is most likely that groundwater flow
carries water under the Little Puddle gauging station and
Geological controls on surface–groundwater interactions later emerges as springs further down the valley.
As part of a study of the Upper Piddle, examination of Howden et al. (2004) conclude that surface–groundwater
borehole water level data (Marcus Hodges Environment, interactions in the River Piddle valley corridor take two
1999) indicated that significant reaches of the river were forms – spring flow and stream leakage–with the river tending
frequently perched. The study concluded that, while to accrete from a series of discrete springs, between which
the entire length of the Upper Piddle is usually effluent stream-flow depletion may occur. Both these processes
(gaining) during the winter recharge period (of several are linked to the geology of the river valley and the main
weeks in most years), as groundwater levels subsequently hydrogeological controls on the generation of springs can
decline nearly all the river becomes perched, resulting be summarised as shown in Table 3.3. Thus the springs are
in influent behaviour and groundwater flow towards the controlled by lithology, or by a combination of lithology and
perennial head downstream. For the periods of the year geological structure, and Howden et al. (2004) point out that
when the river is perched, the sources of river flow are it is significant that the lithostratigraphical units supporting
several headwater springs. significant groundwater flow from springs are limited — i.e.
Howden et al. (2004) found that the longitudinal the Melbourn Rock within the Holywell Nodular Chalk or
streamflow profiles (Figure 3.29) showed a consist- horizons of the Tarrant, Spetisbury and Portsdown Chalks
ent streamflow accretion between Holcombe Dairy (up- where upper or lower surfaces of flint beds act as a focus
stream of Barcombe) and Alton Pancras (downstream of for groundwater flow. They also conclude that stream-flow
Alton Manor), Piddlehinton and Puddletown and between leakage is either due to a lack of contributing spring sources
coupled with permeable stream-bed geology or due to the
effects of geological structural features along the base of the
river valley. However, it is also possible that spring inflows
6
These orientations are consistent with fault orientation trends throughout
the Chalk of southern England and the Paris Basin (Bevan and Hancock,
and bypass flow may be possible in gravelly superficial
1986). deposits.

50
Table 3.2 Hydrogeological controls on springs in the Piddle catchment (adapted from Marcus Hodges
Environment, 1999 and Howden et al., 2004).

Location Observations
Holcombe Dairy Groundwater discharges from a faulted outcrop of Upper Greensand. Summer groundwater levels adjacent
to the river are below that of river-bed level.

Barcombe Flow gauging and water-level monitoring indicate Barcombe to be an area of groundwater discharge to the
River Piddle, controlled by a fault that crosses the river at this point in the Zig Zag Chalk. The fault runs
north-north-east to south-south-west and throws the Holywell Nodular Chalk against the Zig Zag Chalk.
The water source is thought to be the Holywell Nodular Chalk and that confined groundwater in the under-
lying Upper Greensand may contribute in the winter. Both sources use the fault as a preferential flow path.

Alton Manor The springs at Alton Manor are situated on the same fault line that crosses the river at Barcombe, however,
the greater flow is likely to be attributed to a larger recharge area. The location of the springs may be due
to a permeability contrast between the Zig Zag Chalk and the Holywell Nodular Chalk, although other
observations in the catchment suggest that the Zig Zag Chalk has moderate to good transmissivity. It is
more likely that plastic deformation of the Plenus Marls Member (Holywell Nodular Chalk) and the Zig
Zag Chalk along the fault line induces a hydraulic anisotropy and produces a low permeability barrier.

Sheep Dip This spring appears to be fault controlled, close to the Holywell Nodular Chalk and Zig Zag Chalk boundary.

Morningwell No ‘spring’ inflows are observed in the main river where a south-west to north-east trending fault crosses,
however, there is a spring at Morningwell in the western-side valley. The spring appears to be fault controlled
as water levels in two boreholes up and downstream of a pond filled by the spring can be up to 8 m different.
The fault throws the Holywell Nodular Chalk against the Lewes Nodular Chalk and is the same fault that
causes both the Alton Manor and Barcombe springs. The role of the fault in the main valley is uncertain
although gauging in 1997 either side of the fault suggests a loss of flow between the two stations. There is
also a spring at Plush [ST 71 01] on an eastern tributary to the Upper Piddle. This provides flow to the river
between Morningwell and South House.

Morningwell to South Loss of flow along this perched reach ranges from about 2 to 13 l s–1. It appears to be independent of flow,
House reach therefore it is presumably controlled by the hydraulic conductivity of the river bed (New Pit Chalk) and the
wetted area.

South House to Little Loss of flow along this perched reach ranges from about 2 to 13 l s–1 and appears to be independent of
Puddle Chalk units. It is sufficient for groundwater to move south through the aquifer except for a limited period
in the winter and when recharge is high.

Waterston ‘Perennial’ springs emerge at Waterston (former cress beds). Marcus Hodges Environment (1999)
concludes that these are associated with the contact between the Chalk and overlying Paleocene deposits
to the south that restrict southward groundwater flow. However, Howden et al. (2004) describe the source
of the springs as a fault which runs along the river and intersects a fracture in the Tarrant and/or Spetisbury
Chalk, which is a preferential groundwater flow path under confined head. In addition, ten artesian
boreholes in the vicinity exaggerate the contribution of these springs to streamflow.

Puddletown, Stafford There are no geological boundaries or structural features close to these spring sources. Each source issues
Park, Church Knapp from the Portsdown Chalk, which has a series of flint and marl bands and it is likely that these bands
and Athelhampton provide a number of preferential flow pathways which form springs at outcrop.

Cecily Bridge Springs are located close to the Chalk/Palaeogene boundary. A borehole at Cecily Bridge indicates
confined groundwater head in the underlying Chalk. Thus the spring flow is controlled by the magnitude of
confined groundwater head, caused by the limited ability of groundwater to flow beneath the Palaeogene
deposits.

Warren Heath Many springs issue on Warren Heath south-west of the river and form channels that flow into the Piddle. The
springs are located in the Poole Formation and correspond to boundaries between the Bradstone Sand and
Clay members and it is likely they comprise waters from Palaeogene aquifers.
Binnegar Springs occur where the Oakdale Clay crops out between sand layers of the Poole Formation. Confined
groundwater head in the underlying sands creates springs and the permeability contrast at the sand/clay
boundary causes springs at the base of the sands overlying the clay aquifer.

51
 a 6

4
mm d–1

0
1994 1995 1996 1997 1998 1999 2000 2001

b 6

4
mm d–1

0
1994 1995 1996 1997 1998 1999 2000 2001

c 0.5
+ve = gain in flow
0.0

−0.5
mm d–1

-ve = loss in flow

−1.0

−1.5

−2.0
1994 1995 1996 1997 1998 1999 2000 2001

d 0.5
Table 3.3 Summary of main hydrogeological controls on
springs in the Piddle valley.
0.0 Upper Greensand Confined groundwater head in the
Upper Greensand overlain by the Zig
Zag Chalk or faulting of an Upper
Flow difference (mm d–1)

Greensand surface outcrop against less

–0.5 permeable strata
Holywell Gravity drainage at the Holywell
Nodular Chalk Nodular–Zig Zag Chalk boundary or
–1.0
faulting of the Holywell Nodular Chalk
against less permeable strata
Portsdown Chalk The surface outcrop of marl and flint beds
Portsdown Chalk/ The West Park Farm strata are a
–1.5
West Park Farm confining layer overlying the Portsdown
boundary Chalk, creating some artesian springs
associated with this confinement
–2.0 Poole Formation Interbedded sands and clays of the
0 2 4 6 8 10 Poole Formation enable springlines to
Little Puddle flow (mm d–1) be formed due to groundwater head in
sand layers confined by overlying clays,
Figure 3.30 a) South House streamflow (normalised by or drainage from the base of sands
catchment area); b) Little Puddle streamflow overlying clay layers
(normalised by catchment area); c) Flow difference
(normalised) between South House and Little Puddle;
d) Flow difference between South House and Little Puddle
vs Little Puddle flow (normalised). (after Howden et al.,
2004).

52
4 Catchment hydrogeology

4.1 SOUTH DORSET COAST SY 80 82 84

Frome
4.1.1 Geology R.

South Dorset is dominated by the Abbotsbury–Ridgeway Wool

and Purbeck–Isle of Wight fault zones. Interpreted as two 86 86

en échelon fault segments offset by about 4 km in the
N
vicinity of the Poxwell Anticline and Chaldon Pericline,
20
the faults form the boundary between the thinner Jurassic Winfrith
rocks deposited over the South Dorset High and the thicker Newburgh Coombe
Jurassic rocks deposited offshore in the Portland–Wight Keynes
84 84
Basin. Both fault zones have suffered repeated episodes of 40

20
East Chaldon 30
re-activation, both in extension and compression. Miocene 20

10
compression led to a substantial reversal of movement along
the Purbeck Fault, reducing net normal displacements of 60
50

30
40
Permian to lower Cretaceous strata and causing net reverse East Lulworth
displacements at higher stratigraphical levels. The Chalk 82 82

and Cenozoic successions were folded and suffered local 60

50
reverse faulting, producing a major eastwards-plunging and

40
30
20
70
40 50 60
northwards-verging asymmetrical inversion structure, the 2 30
10 0

northern limb of which represents a zone of steep northerly West Lulworth

to vertical (and locally overturned) dips, which form the 80 0–30 metres aOD Arish 80
Mell
elevated topography of the Purbeck Hills. The structure is 30–76
Lulworth Cove
76–122
generally referred to as the Purbeck Disturbance, Purbeck– Above 122 metres
Isle of Wight Disturbance or Purbeck Monocline. Between 10 Groundwater contours (metres aOD)
0 1 2 Kms

Lulworth and Worbarrow Bay, it is also referred to as the measured August 1978, under static,
non-pumped conditions.
Lulworth Monocline. Along much of the length of the 80 82 84 SY
Purbeck Monocline, the Chalk and Cenozoic boundary,
previously interpreted as an unconformable or normally Figure 4.1 Groundwater levels and topography at
faulted boundary, represents a high-angle reverse fault. Lulworth (after Institute of Geological Sciences, 1979a).
contains OS data © Crown copyright and database rights 2017 Ordnance
Survey [100021290 EUL]. Use of this data is subject to terms and
4.1.2 Hydrogeology — Lulworth
An investigation by Wessex Water Authority into the water
resources of the Chalk of north Lulworth was conducted flow is in a seaward direction and appears in the form of spring
in the early 1970s. This work was continued by Alexander discharges and seepages direct to the sea, for example those
(1981). Houston et al. (1986) integrated surface geophysical on the coast at Arish Mell [SY 85 80] and West Lulworth
and downhole logging techniques to try to locate high [SY 82 80]. Tracer tests and pumping tests have proved local
transmissivity zones within the unconfined Chalk around rapid transport towards the Arish Mell springs (Alexander,
Lulworth. A resistivity survey was conducted in the South 1981). A combination of locally low hydraulic gradients,
Winterbourne valley to ascertain the depth of the Grey large water level fluctuations in response to recharge and
Chalk–Chalk Marl interface (a traditional boundary within drilling results suggest the existence of a narrow network of
the Zig Zag Chalk Formation) and therefore the variations discrete solution-enhanced fractures in this area. These high
in the effective aquifer thickness (Robins and Lloyd, 1975). transmissivity solution features probably developed after
As discussed in Chapter 3, the tectonic hardening of the the sea breached the Chalk aquifer during the Holocene.
Chalk has, in general, tended to reduce the intergranular Groundwater flow was reversed, flowing southward along
aquifer properties of the Chalk in this area. However, in the fracture zones and discharging at discrete points along
certain areas, fracture zones, enlarged by solution processes the cliff line (Houston et al., 1986).
to form groundwater conduits, have significantly enhanced
the transmissivity.
4.2 FROME AND PIDDLE
An investigation by the Wessex Water Authority into the
water resource potential of the Chalk north of Lulworth 4.2.1 Catchment geology
revealed a complex pattern of groundwater movement
(Institute of Geological Sciences, 1979a). The shape of The northern and western boundaries of the Frome and Pid-
the water table in the unconfined Chalk was found to be dle groundwater catchments are formed by the high Chalk
unrelated to the surface topography (Figure 4.1), which may downland, the strata dipping towards the centre of the Wes-
be attributable to the intense tectonic disturbance that has sex Basin. To the far south and south-east the catchment
affected the Chalk in the area. The study found that, contrary boundaries are formed by structural features in the geol-
to surface drainage patterns, a high proportion of groundwater ogy — the Isle of Wight Monocline.

53
 The Frome–Piddle catchment comprises three distinct monthly flow data for the Frome and its tributaries and
geological zones (Figure 1.6): the headwaters of the Frome suggested that, since the low flows at Dorchester are only
and Piddle cut into Jurassic and Lower Cretaceous strata; around 40 per cent of those at East Stoke, the summer
the middle reaches flow across Cretaceous chalklands, and groundwater contributions to flow below Dorchester
the lower reaches traverse the Palaeogene deposits of the (including those from the tributaries) are important. Aston
Wareham Basin before discharging into Poole Harbour. (1999) pointed out that this extra flow occurs over an
Jurassic strata are dominated by the thick mudstone area mainly underlain by Palaeogene deposits and shows
formations of the Fuller’s Earth, Frome Clay, Oxford Clay similar hydrograph characteristics to the Chalk-derived
and Kimmeridge Clay, separated by intervals of limestone flow at Dorchester. Mansell-Moullin (1994) presents
and sandstone. Unconformably overlying these strata are data indicating both gaining and losing sections of the
the Lower Greensand, Gault Clay, Upper Greensand and river during low flows in 1970 and suggests a variety of
Chalk. Much of this stratigraphy is contained within the reasons for the pattern including losses by groundwater
narrow outcrop of the primary Chalk escarpment and most abstractions, river diversions and river bed leakage, with
of the catchment is underlain by gently inclined dip slopes gains from groundwater discharge, tributary inflows and
in Upper Chalk. A second major unconformity separates the effluent returns. The variation in flow along the River
Chalk from the Palaeogene, which comprises basal clay-rich Frome thus may be complicated, particularly during low
strata of the Reading Formation and London Clay overlain flow periods.
by the predominantly sandy Poole Formation (Bracklesham The River Piddle drains a smaller area than does the Frome
Group). The geological structure is dominated by the (44 per cent of the Frome catchment area) and mean flows are
eastward-plunging Wareham Syncline, whose axis runs smaller (Table 4.1). The baseflow index at Baggs Mill of 0.89
approximately along the Frome valley and is the primary is similar to the Frome figure and indicates the significance
control on the distribution of strata and on the orientation of of Chalk groundwater flows to the river. The River Piddle
the drainage system. The syncline is strongly asymmetrical: exhibits complex longitudinal flow behaviour involving
on the northern margin, Chalk and Palaeogene strata dip upstream gaining reaches, then losing sections followed by
gently south-east, but along the southern margin, they are further gaining reaches downstream, and this is discussed in
near-vertical or locally overturned along the Purbeck–Isle Chapter 3 (surface water–groundwater interaction).
of White Monocline. This structure forms the elevated
topography of the Purbeck Hills, which delineate part of 4.2.3 Hydrogeology
the southern margin of the Frome–Piddle catchment. Within
the catchment, laterally extensive north-west to south-east 4.2.3.1 Previous studies
trending faults have been mapped along many of the major There have been a number of hydrogeological studies car-
river valleys. Superficial deposits cover a substantial area ried out on the Rivers Frome and Piddle and these are de-
(around 40 per cent) of the catchment and include clay-with- scribed briefly below.
flints, head, river terrace deposits and alluvium with minor In 1985, Wessex Water Authority applied for a 5 per cent
amounts of peat and marine or estuarine alluvium near Poole increase in the abstraction rate at Alton Pancras pumping
Harbour. Subsidence hollows (dolines) and stream sinkholes station, one of their sources in the Piddle catchment. As a
are a notable feature along the northern edge of the Wareham result of public concern about the possible effects of this
Basin between Dorchester in the west, to near Lytchett increase on the river flow, Wessex Water Authority carried
Minster in the east. Culpepper’s Dish is the largest single out a number of studies to investigate the relationship be-
doline in the area. tween increased abstraction quantities and river flow. These
studies did not find an adverse link between groundwater
abstraction and river flow.
4.2.2 Surface water flows
The National Rivers Authority instigated further research
The Frome catchment has several river-flow monitoring as part of the River Piddle Action Plan (National Rivers
sites. Data from those at East Stoke (near to the rivers out- Authority, 1992a) to examine the problems of low flows in
flow into Poole Harbour) and at Dorchester (prior to the con- the Piddle catchment. The resultant model, constructed by
fluence with the South Winterbourne) are given in Table 4.1. Halcrow (1995) suggested that the groundwater abstractions
The high baseflow indices of the River Frome at the two did affect flow in the Piddle and its tributaries. A number of
sites indicate a groundwater-dominated Chalk river with a remediation alternatives were offered as part of this study,
significant increase in flow between Dorchester and East which included augmentation and reductions in the abstrac-
Stoke. Mansell-Moullin (1994) examined annual minimum tion quantities.

Table 4.1 Gauging station data from selected sites on the rivers Frome and Piddle (source: Marsh and Lees, 2003).

Station name Grid Catchment Period Mean Mean flow Q10 Q95 Base
(River name) reference area (km2) of annual (m3 s–1) (m3 s–1) (m3 s–1) flow
record rainfall index
(mm)
East Stoke SY 86 86 414.4 1965–2000 1009 6.41 12.1 2.17 0.85
total (Frome)
Dorchester SY 70 90 206.0 1971–2000 1044 3.07 6.1 0.87 0.83
total (Frome)
Baggs Mill SY 91 87 183.1 1963–2000 977 2.42 4.8 0.78 0.89
(Piddle)

54
 The Environment Agency, South West Region com- outcrop. There are important Upper Greensand springs such
missioned Aspinwall and Company (1997a) to model the as at Maiden Newton and at the head of the River Cerne
catchments of the Rivers Frome and Piddle as part of the (Whitaker and Edwards, 1926). Although the Lower Chalk
determination and implementation of the Groundwater (Grey Chalk Subgroup) is considered to be an inferior
Protection Zones (GPZ) policy. The Chalk was modelled as aquifer compared with the Middle and Upper Chalks (White
a steady-state, single-layer system, which was found to be Chalk Subgroup), there is likely to be hydraulic continuity
one of the limitations of the model. A multilayered system between the Chalk and Upper Greensand.
was probably more realistic. Nevertheless, the model did Cross-catchment groundwater flow between the surface
represent the general pattern of groundwater flow observed water catchments of the Piddle and Frome has been
across the region. Some sites were, however, found to be postulated. For example, a number of groundwater-fed
sensitive to parameter variations during model calibration. cress farms, which are situated close to the Frome, may
The model used steady-state conditions and it was felt that have recharge areas that lie partly in the Piddle surface
transient conditions should be modelled to allow verification catchment to the north (Environment Agency, 1996). It has
of the steady-state model. This would have necessitated been hypothesised that the Frome accepts groundwater from
the capture of detailed seasonal variations in groundwater the Piddle catchment via bypass flow along preferential flow
abstraction and seasonal river baseflow. horizons, such as fractures, geological faults or possibly
Recent British Geological Survey mapping using the dissolution features.
revised Chalk stratigraphy has resulted in a detailed geo- The major hydrogeological characteristics of the Chalk in
logical map of most of the Frome and Piddle basin, (British the Frome and Piddle catchments can be divided into four
Geological Survey, 2001). The revised geology of the Chalk types (Howden, 2004), which are described below.
is detailed in Chapter 2.
In 1999 a NERC thematic programme, ‘Lowland Catch- 1. Zig Zag Chalk — acts as an aquitard/aquiclude due to
ment Research’ (LOCAR) was set up to develop the sci- its clayey and marly nature. The basal Glauconitic Marl
entific understanding required to implement the EU Water Member equivalent, or the West Melbury Marly Chalk
Framework Directive. This was achieved by undertaking Formation where it is present, may restrict vertical
detailed, interdisciplinary and integrated hydro-ecological groundwater movement between the Chalk and the
research, the aims of which were to: Upper Greensand and promote spring flow at the Chalk
–Upper Greensand boundary.
• measure and model the processes controlling the trans- 2. Hardgrounds (e.g. Melbourn Rock, Chalk Rock) — act
fer of water and pollutants through permeable catchment as preferential groundwater flow horizons and form
systems and their associated aquatic habitats, considering springs at outcrop.
various spatial and temporal scales, different land uses 3. Preferential flow horizons — develop within the top 50
and geomorphological settings to 60 m and are independent of the stratigraphy. Marl
• create an integrated model embodying all current and and flint beds create impermeable barriers and promote
future knowledge to aid in the understanding of the the development of groundwater flow along their upper
responses of lowland permeable catchment systems to and lower surfaces. At outcrop, these flow horizons
direct and indirect anthropogenic influences. form springs.
4. Structural features — add more complexity at outcrop
Three densely instrumented experimental catchments were and, if preferential flow paths are intersected by faults,
supported by LOCAR and were selected so that detailed ex- this may promote spring flow.
perimental programmes could build on relatively long-term
hydrological records of rainfall, river flow and groundwater The nature of the geological control on springs in the Upper
levels. The Frome and Piddle (as a combined catchment) Piddle is discussed in Chapter 3 (surface water–groundwa-
comprised one of these three catchments, the other two be- ter interaction).
ing the Pang–Lambourn in the west Berkshire Downs and The overlying Palaeogene deposits, although generally
the River Tern, a tributary of the River Severn in Shropshire. considered to be poor or non-aquifers, exert an important
influence on the hydrogeological behaviour of the catch-
4.2.3.2 Hydrogeology ment. For example, springs at Broadmayne, to the south-
The Chalk forms a major hydrogeological unit in the Frome east of Dorchester, and at Tincleton occur where the Chalk
and Piddle catchments and can be subdivided into an uncon- is covered by the basal part of the Palaeogene sequence.
fined and confined Chalk aquifer. In the unconfined Chalk, Springs at Empool Bottom rise through gravels in the Read-
the groundwater head variations recorded in boreholes gen- ing Formation and then flow at the contact with overlying
erally show the expected patterns of high levels during the clays (Mansell-Moullin, 1994).
winter–early spring and low levels during the drier summer The potentiometric surface maps for the Piddle and
months. Frome system show that the groundwater contours tend to
As discussed above, the flow regime of the Frome and follow topographical contours. Areas where the hydraulic
Piddle catchments is dominated by groundwater baseflow gradients are low tend to correspond with areas of Upper/
(around 80 to 90 per cent) but flow percentages from direct Middle Chalk outcrop. Areas where the hydraulic gradients
run-off are increased locally by outcrops of impermeable are steeper tend to correspond with areas of Lower Chalk
superficial deposits (e.g. clay-with-flints) (Howden, 2004). and Upper Greensand, because of the lower transmissivity
Faulting also exerts a control over the surface water drainage of the marl-rich Lower Chalk aquifer.
pattern; rivers and some streams are aligned, in part, along While there do not appear to have been any comprehensive
fault axes. The Upper Greensand, which varies from estimates of the water balance of the Frome catchment,
between approximately 20 and 45 m in thickness (Institute a preliminary estimate made by Paolillo (1969) gave
of Geological Sciences, 1979a), may also be considered to an excess of run-off over recharge of approximately 25
be an important aquifer in the catchment, although it only megalitres per day (Ml d–1). Paolillo suggested that this
crops out in the valley bottoms and to the west of the Chalk apparent imbalance is probably due to uncertainties in the

55
calculation, but speculated that some groundwater may flow pumping was stopped, full recovery of groundwater levels
from the adjacent Piddle catchment. took several years. The results from the field data and hy-
draulic modelling (Wessex Water, 1995; Birmingham Uni-
versity, 1997) indicated that natural recharge to the basin
4.2.4 Hydrogeology of the Chalk aquifer at Wareham
was very limited and that the aquifer was unsuitable for con-
Wareham lies at the eastern limit of the Frome and Piddle ventional abstraction.
catchments and in this area of Dorset the Chalk is confined Fluid logging of boreholes as part of the Wareham
by up to 300 m of Palaeogene clays, sands and gravels (Fig- Groundwater project (Buckley, 1996), indicated that the ma-
ure 4.2). This overburden confines the Chalk groundwater jority of water-bearing horizons are in the top 50 m of the
and as a result artesian conditions are observed at several Chalk aquifer. The potentiometric surface from the edge of
boreholes and in some, heads of up to 17 m above ground the Palaeogene outcrop to the basin centre at Wareham falls
level have been recorded. only by approximately 10 m, suggesting that the confined
Wessex Water first investigated the Chalk aquifer beneath Chalk aquifer cannot discharge readily to either the Palaeo-
Wareham from 1991 to 1995 to determine its suitability as a gene beds or across the Purbeck structure to the south.
resource for groundwater abstraction. It was found that the In 1996, a feasibility study was carried out to look into
quality of the Chalk groundwater in the area was slightly the potential of using the scheme for aquifer storage and
mineralised, and fluoride concentrations of up to 4 mg l–1 recovery (ASR) in the vicinity of Wareham. It was found
were recorded, which exceeds the drinking water standard that the Chalk aquifer in this area had the appropriate hy-
(currently 1.5 mg l–1 in England and Wales). The water was drogeological properties considered suitable for ASR. The
dated at between 8000 and 10 000 years old, with no history principle of ASR and the conclusions of the investigation
of abstraction to disturb the aquifer (Eastwood and Stan- are discussed in Chapter 6.
field, 2001). In 1994, a six month pumping test was car-
ried out and three boreholes were pumped at an average of 4.2.5 Karst in the Frome and Piddle catchments
12.5 Ml d–1. The effects of these abstractions were localised,
with high drawdowns around each of the boreholes but with The area of the Frome and Piddle catchment, on Briantspud-
no visible adverse environmental impacts on the surface dle Heath is noted as hosting the highest density of solution
or in the unconfined Chalk (Wessex Water, 1995). Values features in the UK (Edmonds, 1983; Waltham et al., 1997).
of transmissivity typically ranged from 200 to 500 m2 d–1 These solution features are particularly prevalent near the
and storage coefficients were of the order of 2 x 10–4. After feather-edges of the outcrop of Palaeogene strata, e.g. Cul-

Figure 4.2

River Avon
Geological map of
Wareham area.
Dev
Lyscream

Bere Stream

contains OS data © Crown Wimborne Minster

ils B
St

copyright and database

omb

1
00 No
Winterth
rook

rights 2017 Ordnance rbour

ne
e

Survey [100021290 EUL].

Use of this data is subject to
terms and conditions.

Riv Lytchett Minster

er
Pid Poole
dle

oll River
dn k From
Ta roo e
B i n
W Wareham
e r N
Riv

0
80

0 10 km
3
80 4
00

Solent Group Seaford Chalk Formation Lower Greensand Group

Bagshot, Barton and Bracklesham Lewes Nodular, Seaford, Newhaven, Wealdon Group
groups Culver and Portsdown Chalk Purbeck Limestone Group
formations (undifferentiated)
Thames Group Portland Group
Lewes Nodular Chalk Formation
Lambeth Group Kimmeridge Clay Formation
New Pit Chalk Formation
Portsdown Formation Corallian Group
Holywell Nodular Chalk Formation
Culver Chalk Formation Oxford Clay Formation
Zig Zag Chalk Formation
Newhaven Chalk Formation Rivers
Upper Greensand Formation
Seaford, Newhaven and Culver Surface water catchment
Chalk formations (undifferentiated) Gault Formation boundary

56
pepper’s Dish [SY 81 92]. The dolines occur in several large unconformable contact between the Chalk and the overlying
groups, covering a total area of around 16 km2 and a signifi- Palaeogene strikes north-east to south-west. The basal part of
cant characteristic is their high density, which reaches 157 the Palaeogene succession comprises clay-rich strata of the
per km2 in the largest cluster (Sperling et al., 1977). Reading and London Clay formations. This is overlain by the
The features described by Sperling et al. (1977) are predominantly sandy succession of the Bracklesham Group.
located between the Piddle and the Frome (Figure 3.4). Along the major river courses, much of the Palaeogene
Their origin is uncertain; it may be that these dolines are is concealed beneath alluvium flanked by gravelly river
the surface expressions of fractures that have been enhanced terrace deposits. The catchment is structurally simple with
by aggressive water percolation from the Palaeogene cover only minor folding affecting the overall south-east dip of the
into the Chalk, resulting in focused dissolution. Another strata. Surface faults can be divided into two main groups.
possibility is that these areas provide preferential flow The first, those with east–west or east-north-east trends,
horizons linking the Frome and Piddle catchments together. appears to have affected the thickness of Jurassic strata. The
Similarly, to the east at Bere Regis, a significant proportion second group trends mainly north-north-west, but includes
of the water use may be derived from the surface catchment associated north-north-east and north-east trending faults.
of the intermittent North Winterbourne chalk stream. The
difficulty in quantifying the transfer of water between 4.3.2 Surface water flows
adjacent catchments complicates water balance calculations
(Environment Agency, 1996). River flows in the upper Stour catchment are flashy, due
mainly to the presence of the impervious Jurassic strata in
the area, which promote surface run-off rather than recharge.
4.3 STOUR AND ALLEN At the Hammoon gauging station (Table 4.2), where flows
are measured before the Stour flows over the Chalk, the low
4.3.1 Catchment geology baseflow index reflects this high degree of run-off.
The flashy nature of the hydrograph is also shown
The Stour–Allen catchment includes outcrop geology ranging by the high ratio of Q10 to Q95 flows. Downstream of
in age from the Middle Jurassic Fuller’s Earth Formation to Hammoon, the Stour is augmented by substantial inflows
the Palaeogene Branksome Sand Formation. The regional from tributaries draining the Chalk, for example the River
structural dip is to the south-east such that the headwaters Allen where the high baseflow indices from gauging
of the Stour cut into Jurassic limestones and mudstones, the stations (Table 4.2) indicate the Chalk-fed, groundwater-
middle reaches flow across chalklands, and the lower reaches dominated nature of the river. The North Winterbourne
traverse the Palaeogene deposits of the Hampshire Basin rises near Winterbourne Houghton, tending to go to ground
before discharging into Christchurch Harbour (Figure 1.7). at Winterbourne Stickland. The bed also dries between
In contrast to the other river catchments of Wessex, Jurassic Winterbourne Kingston and Sturminster Marshall. At
strata underlie a large area of the Stour–Allen basin. They Throop, flows are measured near to the outflow from the
produce a varied topography directly related to the geology catchment, but upstream from the confluence with the Moors
with prominent ridges, in part fault controlled, of resistant River. Flows at Throop are substantial compared with those
Jurassic limestones rising above vales formed by the Frome from the Frome and Piddle and the baseflow index reflects
Clay, Oxford Clay and Kimmeridge Clay. The primary the mixture of surface and groundwater sources.
escarpment of the Chalk forms an irregular north-west
trending ridge and is primarily formed by a narrow outcrop of
West Melbury Marly to New Pit Chalk formations resting on 4.3.3 Hydrogeology
a platform of Upper Greensand and Gault. Descending from There appears to be little hydrogeological interpretation of
the primary escarpment is a long, gently inclined slope in the Stour catchment. One of the subcatchments, the Allen,
Lewes Nodular to Newhaven Chalk formations that underlies has, however, been addressed by several studies.
much of the catchment. The slope is generally marked A water resource study of the River Allen catchment was
by two smaller, heavily dissected subsidiary escarpments undertaken by Groundwater Development Consultants for
capped by the Tarrant and Spetisbury Chalk members of the the National Rivers Authority (Groundwater Development
Culver Chalk Formation (see Bristow et al., 1997, fig. 4). The Consultants, 1992). This work involved the development of

Table 4.2 Gauging station data from selected sites on the rivers Stour and Allen
(source: Marsh and Lees, 2003).

Station name Grid Catchment Period Mean Mean Q10 Q95 Base
(River name) reference area (km2) of annual annual (m3 s–1) (m3 s–1) flow
record rainfall rainfall index
(mm) (mm)
Throop SZ 11 95 1073.0 1973–2000 881 13.74 31.1 2.61 0.65
(Stour)
Hammoon ST 82 14 523.1 1968–2000 875 7.52 20.6 0.64 0.32
(Stour)
Walford Mill SU 00 00 176.5 1974–2000 879 1.91 4.5 0.30 0.91
(Allen)
Loverley SU 00 08 94.0 1970–2000 906 1.04 2.6 0.17 0.89
Mill (Allen)

57
an integrated surface water and groundwater numerical model groundwater was first abstracted in 1946 (at Stanbridge
of the Chalk aquifer and river system. The main objectives Mill), there has been a significant reduction in salmonid
were to evaluate the effects of groundwater abstraction on populations, especially over the last few years. A model was
catchment hydrology and to examine alternative catchment constructed to quantify this effect in terms of the reduction
management strategies. Hydrogeological interpretations on the Q95 value for flow in the River Allen. For the summer
suggested by the model included the following: months, discharge at the Q95 value was depleted by 55 per
cent. The National Rivers Authority, in conjunction with
1. The final calibration of the model produced good Bournemouth Water plc, used the results to formulate an
simulation of summer low flows and winter peak flows action plan for the River Allen. These negotiations resulted
at the two main gauging stations (Loverley Mill in the in a reduction in the current abstraction rates by 50 per cent.
middle area of the catchment and Walford Mill, to Perkins and Robertson (1980) reported on a series of
the north of Wimborne Minster). However, it was not pumping tests conducted at Shapwick pumping station in
possible to simulate observed flow data well for the the Stour valley. The study concluded that the river and river
Allen upstream of the confluence with the Gussage valley deposits play a major part in the groundwater flow
Brook. It was hypothesised that complex aquifer regime and that the Chalk is layered in terms of permeability,
conditions involving low transmissivity barriers and with a top zone 15 to 20 m thick being highly permeable
preferential flow directions may be responsible for the and in hydraulic continuity with the river via the river valley
poor predictions in this upstream area. deposits.
2. The use of low transmissivity zones (less than The total Chalk outcrop for the Stour catchment is
200 m2 d–1) in the interfluve areas of the catchment estimated at 472 km2. This, coupled with an estimated
model improved the flow simulation results from those annual average infiltration of between 440 and 490 mm, has
of the previous study (low transmissivity was found at given a theoretical groundwater resource of 582 Mm3 d–1
the only interfluve borehole which was tested). Values (Avon and Dorset River Authority, 1970). By contrast, the
of transmissivity in the range 1000 to 5000 m2 d–1 were theoretical groundwater resource for the combined Frome–
used for the valley areas (i.e. similar to those obtained Piddle catchment system was estimated to be larger, at
from valley site pumping tests). 705 Mm3 d–1, despite a smaller outcrop area of 435 km2. The
3. The introduction of a permeability change with depth reason for this is the higher rainfall, and hence recharge, in
in the upper parts of the catchment improved the the upper regions of the Frome–Piddle catchment.
simulation of groundwater levels in that area.
4. Analysis of hydrographs for the Allen catchment
have illustrated that both groundwater levels and river 4.4 AVON
flows respond rapidly to recharge, implying that the
aquifer provides little effective groundwater storage 4.4.1 Catchment geology
and is highly fissured. During the model calibration
procedure, a low unconfined storage coefficient of The Avon catchment (Figure 1.8) includes bedrock geological
0.25 per cent had to be used over most of the catchment. units ranging in age from the Late Jurassic Kimmeridge Clay
5. Simulation of the effects that the main abstraction Formation to the Palaeogene Barton Group. Jurassic strata
at Stanbridge have on the catchment concluded that, are restricted to the Vale of Wardour where the eastward
at around half of the licensed abstraction rate of flowing River Nadder cuts into Kimmeridge Clay brought to
25 Ml d–1, the pumping station is drawing water from crop by reverse movement on the Mere Fault. The primary
the river rather than from groundwater storage. At larger escarpment of Chalk has a highly irregular trace in the Avon
abstraction rates, increasing quantities of water are catchment with abrupt deflections into the fault-controlled
taken from groundwater storage, especially during the vales of Wardour and Pewsey, and the valleys of the rivers
summer period. This storage is refilled at the expense of Wylye and Ebble. As elsewhere in Wessex, the primary
river baseflow during winter months. Chalk escarpment is formed from a narrow outcrop of Lower
6. The estimated mean annual recharge varies from and Middle Chalk capped by an extensive sheet of Upper
340 mm in the south of the catchment to 400 mm in Chalk which regionally dips to the south-east beneath a
the north. cover of Palaeogene sediments. The regional south-easterly
dip is punctuated by several broadly west–east trending
Aspinwall and Company (1997b) modelled the catchments anticlines and synclines, which include the Palaeogene-cored
of the Rivers Allen and Crane using MODFLOW, as part of Alderbury–Mottisfont Syncline and the Dean Hill Anticline.
the same programme of work that modelled the catchments These are developed on linear east–west trending zones of
of the Rivers Frome and Piddle. faulted flexures (as with the Vale of Pewsey and the Vale
Greenaway (1995) studied the hydrogeology of the River of Wardour), which are related directly to the reversal of
Allen catchment and modelled the effect of groundwater underlying basal controlling normal faults (Chadwick, 1986).
abstraction on the surface water flows using MODFLOW. There are very few mappable faults at surface in the area.
The positions of the groundwater divides and evaluation
of the amount of recharge entering the confined Chalk 4.4.2 Surface water flows
were assessed. The model comprised a three-layer system
representing the Palaeogene, the effective Chalk aquifer The River Avon drains the largest catchment area in the
system and the low permeability Chalk beneath. Results Wessex Basin and the flows measured at Knapp Mill near
suggested that the regional flow is influenced by the vertical its mouth are the largest recorded in the basin (Table 4.3).
permeability of confining beds and that streams leaving the The high baseflow index at this site indicates the dominance
unconfined part of the aquifer form a focus for groundwater of Chalk-derived groundwater in the river flow.
discharge. The Avon is augmented by a number of important tribu-
The River Allen has also received attention from an taries and selected gauging station data for these are shown
ecological perspective by Johnson et al. (1995). Since in Table 4.3.

58
Table 4.3 Gauging station data from selected sites on the River Avon and its tributaries
(source: Marsh and Lees, 2003).

Station name Grid Catchment Period Mean annual Mean flow Q10 Q95 Base
(river name) reference area (km2) of rainfall (m3 s–1) (m3 s–1) (m3 s–1) flow
record (mm) index
Knapp Mill (Avon) SZ 15 94 1706.0 1975–2000 840 19.66 39.4 6.38 0.90
East Mills (Avon) SU 15 14 1477.8 1965–2000 835 15.24 28.8 5.50 0.91
Amesbury (Avon) SU 15 41 323.7 1965–2000 777 3.49 6.7 1.09 0.90
South Newton SU 08 34 445.4 1967–2000 860 4.09 8.6 1.15 0.90
(Wylye)
Norton Bavant ST 90 42 112.4 1971–2000 950 1.11 2.1 0.45 0.87
(Wylye)
Wilton (Nadder) SU 09 30 220.6 1966–2000 916 2.90 5.8 0.92 0.82
Laverstock SU 15 30 263.6 1965–2000 790 0.76 1.4 0.19 0.92
(Bourne)

The River Wylye exhibits intermittent flows from its down to the River Ebble) was undertaken by Halcrow for the
source to Kingston Deverill. Here, springs issue at the National Rivers Authority (Halcrow, 1992). The conceptual
junction of the Middle and Lower Chalk that only cease flowing model for the Hampshire Avon was later updated by the
in relatively dry years (Halcrow, 1996). From Kingston Deverill, Environment Agency and stakeholders between 1999 and
the Wylye flows to the north-east over Lower Chalk to Brixton 2006. A new model was developed for the Bourne and Nine
Deverill and in this section leakage can occur from the river. Mile Rivers in 2004 and the whole of the Hampshire Avon
At Hill Deverill, the Wylye flows over the Upper Greensand catchment in 2005/06. The purposes of these modelling
aquifer, from which springs contribute a substantial flow to the studies were to quantify the impact of abstraction on local
river. The river continues to flow on Upper Greensand, and to groundwater resources and to provide a framework for use
gain flow, to Bishopstrow (to the south-east of Warminster). in the management of the groundwater resources of the
Beyond Bishopstrow, the river flows again over Lower Chalk catchment. The understanding of the catchment hydrogeology
and in summer months flow losses to the aquifer can occur, was significantly improved following the remapping of
continuing as far downstream as Upton Lovell. Downstream of the geology by the British Geological Survey and the
Upton Lovell, flow accretion generally occurs and the Wylye is interpretation of this and other information by the Agency.
joined by the Chitterne Brook and the River Till. The Chitterne The main findings from this work are summarised below.
Brook dries seasonally just above the Wylye confluence. The The aquifer properties distribution for the model was
River Till flows continuously from a major spring located at estimated from pumping tests, packer tests and well yields
the junction between the Upper and Middle Chalk (between and inferred from hydrogeological maps. The distribution
the Lewes Nodular and New Pit Chalk formations), a short was further refined during calibration. The transmissivity
distance upstream from Berwick St James (Halcrow, 1996). distribution is thought to be largely controlled by the Chalk
Flow data for the Wylye at South Newton, before its con- lithology and structure. Hard bands near to the surface
fluence with the Nadder, and upstream at Norton Bavant, in- that crop out down the catchment have been shown to
dicate the substantial gain in flow between the two stations significantly control surface and groundwater movement.
and the very high contribution of baseflow to the total flow, These hard bands are often associated with the location of
typical of Chalk catchments. major springs, and sometimes the perennial head, and their
Flow data for the Nadder, from a gauging station at Wilton upstream outcrop may coincide with a location where a
(Table 4.3), indicates that, while principally spring fed, the stream or river loses water. Anticlinal structures may control
river also produces a relatively large flood flow, as a result groundwater movement, diverting groundwater across a
of the varied geology of its catchment area. Such effects are surface water catchment.
noticeable on the Avon downstream of Salisbury (Avon and Further enhancement of transmissivity has occurred
Dorset River Authority, 1970). in the major valley areas through solution weather-
Flow data from the Laverstock gauging station on the low- ing. Transmissivities were found to vary from 100 to
er reaches of the Bourne are shown in Table 4.3. The general 250 m2 d–1 in the West Melbury Marly Chalk and Zig Zag
characteristics of flow data are typical of a Chalk catchment. Chalk to 1000 to 3000 m2 d–1 in part of the Seaford Chalk
However, the nature of the flow in the Bourne is unusual (where significant hard rock bands occur below the water
above its perennial head at Idmiston: while flow commonly table). Transmissivities are thought to increase in major
occurs in the upper reaches of the river for several kilometres valleys by approximately 50 per cent (Environment Agency
downstream, as far as Collingbourne Ducis/Collingbourne and Water Management Consultants, 2004).
Kingston, the river often dries in its central section down to The storage coefficient varied from 0.005 in the Lower
the Idmiston perennial springs (Environment Agency, 2001). Chalk, 0.02 in the Middle and Upper Chalk, and 0.05 in the
The Nine Mile River is a winterbourne for much of its length. Upper Greensand. The model gave a good representation of
water levels and surface flows and was used in simulation
runs to help understand the relationship between pumping
4.4.3 Hydrogeology — Upper Avon
and stream flows. Typically the abstractions were found to
A groundwater modelling study of the Upper Hampshire have the greatest volumetric impact during wetter months,
Avon catchment (encompassing the catchments of tributaries when chalk storage was replenished — November to April.

59
However, the greatest impact as a percentage of river flow The Bourne and Nine Mile Rivers flow in a south-
occurred during summer months — May to September. westerly direction from their sources, to the east of
Flows in some river reaches were found to increase as a Salisbury Plain, to their confluences with the River Avon
result of abstractions artificially extending the catchment (Figure 1.8). The total surface catchment area of the rivers
area and the subsequent discharge of water. is approximately 204 km2 and the catchments are underlain
by Chalk. Both rivers are substantially groundwater fed and
both exhibit winterbourne behaviour. Figure 3.26 shows
4.4.4 Hydrogeology — Wylye
the winterbourne signature of the Bourne and it is evident
In the 1990s the National Rivers Authority commissioned that the flow system is complex. The perennial head of the
Halcrow to undertake a groundwater modelling study of river is at Idmiston (Figure 4.3) where substantial springs
the Wylye catchment in order to evaluate the effects of occur; upstream of this point flow is often limited south of
abstraction and to help with selection of appropriate flow- Collingbourne Ducis, with Cholderton experiencing the
alleviation schemes. The conceptual and numerical models greatest proportion of dry periods, whereas further north
were updated and renewed by the Environment Agency and the river often flows in its upper reaches. It is evident from
Wessex Water in 2004 to 2006. During model calibration historical literature that the complex flow behaviour of the
the following features of hydrogeological significance were river is natural and predates current abstractions.
found by Halcrow (1996): The potentiometry of the Bourne catchment is greatly
affected by its location between the major rivers Avon and
1. Modelled transmissivities varied from less than Test, both of which are at significantly lower elevations
500 m2 d–1 over interfluves to between 4000 and (around 40 m) than the Bourne (Environment Agency and
10 000 m2 d–1 in the river valleys, with a general Water Management Consultants, 2004). The result of this,
increase from the Upper Wylye to the middle and lower and further hydrogeological behaviour outlined below, is that
reaches of the Wylye. High values were also used along the Bourne groundwater catchment is significantly smaller
the River Till.
2. Modelled storage coefficients varied from less than
0.003, characterising interfluve areas, to up to 0.03 for
the Chalk in valleys and up to 0.1 characterising the Surface catchment Groundwater catchment N
Upper Greensand. Groundwater level (metres aOD)
3. Model-generated stream-flow hydrographs generally
compared well with field data. However, there was
some overestimation of peak and low flows at the South 60
Newton gauging station. This was considered to have
been caused either by movement of groundwater divides
(for example occasionally resulting in groundwater m
flow from the River Till catchment to the Avon) or 140
groundwater flow out of the Wylye catchment beneath
12

55
10 0 m

Collingbourne
the river bed and therefore not recorded by the river m
m

Ducis 120
gauge.
4. The model simulated the observed groundwater head Leckford
Bridge
distribution and behaviour at most locations, including
the reduction of the potentiometric surface below the 50
river bed in influent sections of the Upper Wylye, Tidworth
m

the dampened response and prolonged hydrograph m

0

0
10
10

recessions in the Upper Greensand outcrop and the

Northing

dampened response of Chalk water levels near to the

perennial sections of the river.
45

As part of the Upper Wylye investigation, Avon and Dorset m

80
River Authority et al. (1973) utilised geophysical logging
Cholderton
techniques at three locations: Brixten Deverill, Heytesbury
and Chitterne. At Chitterne, a borehole that penetrates the m
greater part of the Chalk was tested. The results of this 40 Boscombe 80
survey were used to determine the major flow horizons
within the Chalk: 90 per cent of the flow was found to enter Idmiston
the borehole from the top 47 m of the Chalk.
m
60

35
4.4.5 Hydrogeology — Bourne
60

The Environment Agency carried out a detailed investigation

of the River Bourne and Nine Mile River catchments
(Environment Agency, 2001, Environment Agency and Water 0 5 km
Management Consultants, 2004). The work was undertaken 30
SU 15 20 25
in response to Habitats Directive and UK Biodiversity Action
Plan requirements and to assess the impacts of abstraction on
the river. The aim of the study was to collect and analyse Figure 4.3 River Bourne location map, and groundwater
data for the two catchments in order to develop a detailed contours for 27 January 1994 (adapted from Environment
conceptual, and then numerical, model of the hydrogeology. Agency and Water Management Consultants, 2004).

60
 Figure 4.4 Sketch of geological cross-section of River
Bourne showing the general relationship of the units within
the Chalk Group (adapted from Environment Agency and
220
200
180
160
140
120
100

OD
(m)

80
60
40
20
Water Management Consultants, 2004).

WmCk West Melbury Marly Chalk Formation

than the surface water catchment. The majority of recharge

Upper Greensand Formation

in the upper reaches of the catchment diverges towards the
Avon and Test rivers and their tributaries, rather than towards
2 km

the Bourne river. In addition, the groundwater catchment

boundaries vary significantly seasonally. When groundwater
levels are at their highest, the river gains along its whole length,
groundwater flows converge on the river and the catchment
area is at its largest, although even under these conditions
the groundwater catchment is significantly narrower in its
0

central section than in its headwater or downstream sections,

as illustrated in Figure 4.3. When groundwater levels are low,
the river only gains from the perennial springs at Idmiston,
Idmiston

the groundwater catchment tends to become narrower and

moves to the west, suggesting that under these conditions
groundwater in much of the upper Bourne catchment
flows towards the Test and its tributaries (Environment Agency
Holywell Nodular Chalk Formation

and Water Management Consultants, 2004). Commonly, in

wet months, intermediate and complex conditions occur,
with the river gaining in the north and south, while losing,
Zig Zag Chalk Formation

and acting as a groundwater divide, in its central section.

This distinction between groundwaters in the upper part of
the catchment and those in the lower part is supported by
LeCk
SCk

hydrochemical evidence (Chapter 5).

Cholderton

The Bourne rises to the north of the northward-facing

primary Chalk scarp, which separates the Upper Greensand
lowlands of the Vale of Pewsey anticline from the gently
undulating southerly dipping Chalk downland to the
HCk

ZCk

south. A much dissected secondary Chalk escarpment is

represented by the ranges of hills bounding the Bourne
catchment. At the head of the catchment, the Chalk
dips steeply to the south-east and units from the West
Melbury Marly Chalk to the Lewes Nodular Chalk outcrop
successively downstream (Figure 4.4). In the centre of the
LeCk Lewes Nodular Chalk Formation

catchment, from Leckford Bridge–Tidworth to south of

Idmiston, the dip of the Chalk decreases and the Seaford
NpCk New Pit Chalk Formation
NCk

Chalk outcrops. Further south the dip steepens to the south-

east again. A number of small anticlines with approximately
Tidworth

east–west trends occur; they are seen in the flatter section of

the Chalk, for example near Cholderton and Boscombe.
Contributing to the Bourne and Nine Mile River study,
Rock Member
Stockbridge
Hard band
(assumed

the Chalk was remapped by the British Geological Survey

of EA)

(Farrant et al., 2000) using the new Chalk stratigraphical

nomenclature. This work enabled the links between catchment
geology, hydrology and hydrogeology to be investigated.
NpCk

One result of this was the description of the Chalk aquifer

units in terms of their geological and hydrogeological
characteristics. These are summarised in Table 4.4.
Collingbourne Ducis

k
HC
Rock Mbr

In addition to the general hydrogeological characteristics

Chalk

Newhaven Chalk Formation

of the identified Chalk stratigraphic units, certain features of

Seaford Chalk Formation

the geology were identified as being particularly important

in terms of groundwater and surface water flow in the
ZCk
Melbourn
Rock Mbr

catchment. For example, the spatial distribution of hard

rock bands, such as that at 28 to 36 m above the base of
Ck

the Seaford Chalk at Tidworth (potentially the base of the

200 Plenus Marls Mbr

m
W

Stockbridge Rock and described as ‘assumed Stockbridge

Rock’), the Chalk Rock, the Melbourn Rock and of softer
marls such as the Plenus Marl, were considered to have an
NCk

SCk

important influence both on groundwater movement and on

groundwater–surface water interaction.
In the upper Bourne catchment, the low-permeability Plenus
(m)
220

180
160
140
120
100
80
60
40
20
OD

Marls, which dip more steeply than the surface topography,

61
Table 4.4 Summary of the principal geological and hydrogeological characteristics of the Chalk Group formations present
in the Bourne and Nine Mile River catchments (adapted from Farrant et al., 2001, Environment Agency, 2001).

Subgroup Formation Geological characteristics Hydrogeological characteristics

Portsdown Only a few outcrops in the extreme No observations: this formation is poorly represented in the study area
Chalk south of the study area. The
formation consists of white flinty
chalk with common marl seams

Culver Massively bedded, soft, white Solution features may occur above sheet flints and along faults and master joints
Chalk chalk without marl seams but well-
developed nodular and semi-tabular
flints. Regular orthogonal joints

Newhaven Lithologically similar to the Seaford Dissolution may occur along thicker marl seams where groundwater is forced to
Chalk Chalk. Soft to medium hard, smooth, flow horizontally
white chalk with numerous marl
seams and flint bands. Flint bands
not as numerous as in the Seaford
Chalk. Well-developed conjugate
joint sets, dissipated along marl
seams
White Chalk Subgroup

Seaford Smooth, white chalk with abundant Solution features may occur above sheet flints and along faults and master joints.
Chalk nodular and semi-tabular flints. Hard rock bands 28 to 36 m above the base of the Seaford Chalk (identified by
Stockbridge Rock Member is a geophysical logging) provide main inflow horizons to boreholes that intersect
very hard, porcellaneous chalk. this horizon i.e. Tidworth Borehole. Extrapolation of this hard rock band north to
Massively bedded chalk with Leckford Bridge coincides with the location of major losing reach of river, and to the
regular orthogonal joints south, the main spring flows at Idmiston

Lewes Hard to very hard, nodular chalk. Joints often solution enlarged, some solution cavities along the top of sheet flints.
Nodular Formation includes the Chalk The Chalk Rock, ~ 12 m above the base of the formation, provides the main
Chalk Rock Member, joint sets open and inflow horizon at a number of boreholes
steeply inclined

New Pit Smooth, firm, white chalk, massively Dissolution may occur associated with tops of thicker marl seams where ground-
Chalk bedded with marl seams. Well- water is forced to flow horizontally
developed conjugate joints, dissipate
along marl seams

Holywell Hard, nodular chalk. Plenus Marls Perched water tables occur above the clay-rich Plenus Marls Member and flowing
Nodular consist of alternating blocky, white horizons recorded in the hard nodular Melbourn Rock and overlying nodular shelly
Chalk chalk and grey, silty marls. The unit. Plenus Marls may confine waters below and may back up groundwater flow
Melbourn Rock is a very hard, down dip. Spring line may occur at these locations, followed by river leakage into
nodular chalk up to 3 m thick the Melbourn Rock. Melbourn Rock may provide spring lines in winter period
when recharge exceeds capacity of this relatively high transmissive horizon

Zig Zag Medium hard, pale grey, blocky May form spring line at base of formation immediately above relatively low
Chalk chalk. Lower part with higher marl permeability West Melbury Chalk. Vertical jointing within the limestone bands
content feed water to the marl bands that may act to prevent further vertical infiltration
Grey Chalk Subgroup

and may back up water down dip. Limestone horizons may carry greater volumes
of water and provide reaches of effluent river flow where a further outflow
location down gradient is present
West Mel- Bedded, grey, marly chalk with Forms an aquitard between Upper Greensand and overlying Zig Zag Chalk due to
bury Marly thin limestones. Base marked by high clay content. Vertical jointing within the limestone bands feeds water to the
Chalk marl that may be transitional with marl bands that may act to prevent further vertical infiltration. Dip of the chalk to
the Upper Greensand the south-east may back up recharge from Upper Greensand and produce spring
lines. Relatively low rate of recharge through this formation

were considered to form a barrier to southerly groundwater forming the perennial head of the Bourne. Geological
movement, causing springs to occur up gradient and modelling shows that this horizon also crops out at the
promoting river flow. Further south, the outcrop of hard locations of the perennial heads of the River Anton and
rock bands such as the Melbourn Rock, Chalk Rock and Wallop Brook in the Test catchment. On this basis and on
hard rock bands in the Seaford Chalk are associated with the potentiometric evidence, the Environment Agency study
more influent (losing) behaviour of the river. In the centre postulated that large volumes of water are transmitted within
of the Bourne catchment, the assumed Stockbridge Rock the hardgrounds (particularly the assumed Stockbridge
and deeper hardgrounds are found at increasing depth, with Rock) from the upper parts of the Bourne catchment to the
the assumed Stockbridge Rock rising to crop out again at south-west to the Avon, south to the Bourne at Idmiston
Idmiston. This coincides with the occurrence of the springs and south-east to the River Test tributaries. Evidence

62
from borehole logging of the importance of these Chalk Lower and Middle Chalk to the surface in a catchment that
hardgrounds in controlling groundwater flow at moderate is predominantly Upper Chalk.
depths was discussed earlier in this report — for example, The Chalk in the northern and eastern parts of the catch-
see Figure 3.13. Later numerical modelling confirmed that ments is covered by extensive amounts of clay-with-flints,
this conceptual understanding is likely to be correct. particularly the area to the east and north-east of Alresford.
In addition to the postulated lithological controls on To the west of Alresford, these deposits are less extensive
groundwater flows in the Bourne catchment, the study and are present as coverings to interfluves. To the south,
also concluded that the synclinal and anticlinal structural around the market town of Romsey, Palaeogene gravels,
features played a part, so that, for example, the groundwater sands and clays occur.
potentiometric surface is affected not only by the relative There is evidence that a number of processes may have
topography of the Bourne and its neighbouring river valleys, significantly modified the rock mass properties of the
but also by the nature and geometry of the anticlinal and Chalk following lithification. During the Miocene Alpine
synclinal structures running across the valley and the depth orogeny, gentle folds were formed, which generally trend
to different rock units. east–west. In the earliest Tertiary, uplift of the Chalk
In addition, it was noted that trends in drainage directions caused some erosion prior to the deposition of the Reading
are influenced by dominant fracture set orientations as well Formation. The main period of subaerial erosion, however,
as topographical trends and regional geological dip. Many affecting both the Chalk and Palaeogene deposits, occurred
river reaches in the study area run parallel to the dominant in the Quaternary following the main uplift at the end of
north-west to south-east and east-north-east to west-south- the Palaeogene. It is inferred (Entec, 2002) that the bulk
west or subordinate north to south and east to west fracture Chalk fabric was modified during the Lower and Middle
trends. Pleistocene due to the development of a ‘partly evolved’
A distributed, time-variant numerical model of the groundwater system that was in place prior to the Devensian
Bourne and Nine Mile river catchments was developed glacial period. During the Devensian, periglacial action
during the Environment Agency study in order to explore further modified the rock mass properties of the valleys
the impacts of abstractions on groundwater flows to the where dissolution probably took place in unfrozen taliks.
rivers and on head distributions. It was found that the model During the late Cenozoic and Quaternary, erosion is
was most accurate when the saturated aquifer thickness was thought to have been controlled at least in part by structural
70 m and when vertical hydraulic conductivity was half the style, with some initial valley development occurring
horizontal value. The highest transmissivities used were along the axes of synclines such as those trending east–
along the middle section of the river valley, where values for west through the River Alre. Headward erosion from the
the Seaford Chalk of between 1000 m2 d–1 and 3000 m2 d–1, south caused river capture of east–west-trending rivers.
depending on groundwater level, were used. Stress release associated with erosion of river valleys is
thought to have caused the development and enlargement of
fractures in the valleys and allowed groundwater movement,
4.5 TEST AND ITCHEN particularly along bedding planes. It is assumed that most of
this fracture development in the valleys took place during
4.5.1 Catchment geology the Devensian and the presence of dolines in the Chalk close
to the edge of the Reading Formation is interpreted as being
The Test is a chalkland catchment (Figure 1.9) located on a
associated with this phase of fracture development (Entec,
broad, highly asymmetrical anticlinal structure (Figures 2.1
2002).
and 2.2). To the north, the Chalk dips steeply below the
The Reading Formation shows a large degree of variation
Palaeogene of the London Basin. The Upper Greensand is
in these catchments. In general, the Reading Formation is
brought to crop in the core of the anticline at Kingsclere.
sandier in the west with only localised sandy deposits in
To the south, the Chalk dips at a low angle beneath the
other areas. The predominance of clay deposits in the east
Palaeogene cover of the Hampshire Basin. Superimposed
means that the unit is not effectively connected to the Chalk.
on this long, southward-dipping limb are smaller, west–east
In the west, sand units are in hydraulic continuity with the
trending, lower-amplitude synclines and anticlines including
Chalk.
the Winchester–Kings Somborne Syncline, the Stockbridge
There is some debate about the degree of connectivity
Anticline and the Micheldever Syncline. In the south-west
between the Chalk and Upper Greensand. It appears that
of the catchment, the River Dun drains axially along the
they may be in continuity to the west of the River Test but
west–east trending Alderbury–Mottisfont Syncline, which
pumping test data shows that they do not appear to be linked
contains a core of Palaeogene strata.
in the Itchen catchment.
The Itchen catchment (Figure 1.10) includes solid
geological units ranging in age from the Cretaceous
Lower Chalk to the Palaeogene Bracklesham Group. The 4.5.2 Relationship between geology and hydrogeology
catchment lies on the broad chalk plain at the western end
of the Weald Anticline. As with the Test catchment to the A hydrogeological study of the River Itchen for the
north and west, the Chalk dips below the Palaeogene of Environment Agency (Entec, 2002), was undertaken as part
the London Basin. Likewise to the south, the southerly of a broader study to determine a sustainable management
dip of the Chalk takes it below the Palaeogene of the strategy for the river Itchen. This involved a data collation
Hampshire Basin. Superimposed on this regional structure exercise, a review of existing models for the Bourne and
are smaller west–east trending, lower amplitude synclines Nine Mile River, the River Kennet and the River Itchen,
and anticlines including the Winchester Anticline, the and the preparation of a conceptual model of the catchment,
Stockbridge Anticline and the Micheldever Syncline. In leading to a numerical model. This model was subsequently
general, these structures have steeper dips on their northern extended to include the River Test and the new Test and
limbs, probably reflecting the reactivation of deep-seated, Itchen model replaced the previous Itchen model (Entec,
Jurassic fault blocks. The Winchester Anticline brings 2005).

63
 The model used the following boundaries. Groundwater It is envisaged that storage is influenced by similar
flow appears to be bounded to the south with little or no factors to transmissivity and will show variations across
flow southward into the undeveloped confined Chalk. To the the catchment. It is thought that the lowest unconfined
north, the Kingsclere Anticline is a flow barrier, although specific yield values of around 0.5 per cent are found on the
there may be some flow northwards into the Thames Basin interfluves, beneath clay-with-flints cover and for the Middle
east of Kingsclere. The groundwater divide north of the and Lower Chalk (Entec, 2005). Most of the remaining areas
Candover and Dever catchments is relatively well defined are assumed to have unconfined specific yields of the order
by the lower-transmissivity Chalk associated with the of 1 per cent and may be locally increased up to about 10
Ellisfield, Lasham and Medstead groundwater mounds. To per cent close to rivers and in valley floors where saturated
the west and south-east, the Upper Chalk drains to the River permeable drift deposits are in good continuity with the
Avon and River Meon respectively. underlying Chalk.
Based on the work of Headworth (1978) on fissure in-
flow zones to artesian boreholes in the Alre catchment, that 4.5.3 Surface water flows
of Rennie (1994) who reviewed fracture zone elevations
in part of the Candover catchment, and the new British Neither the Test nor the Itchen are truly natural rivers in
Geological Survey mapping, it has been suggested that their current form. Along most of their lengths, the rivers
important fissure zones associated with tabular flinty bands are split into two or more channels with sluices to regulate
and possibly in the Seaford Chalk occur everywhere at a flows. The surface catchment of the River Test is some
fixed stratigraphical elevation postulated to occur generally three times larger than that of the River Itchen (1260 km2
between 60 and 80 m above the base of the Upper Chalk compared with 400 km2). Of the 400 km2 of River Itchen,
(Entec, 2002). The fissure zones are assumed to affect 360 km2 drains Chalk.
aquifer transmissivity and groundwater flow significantly Gauging data for selected stations on the Test and its
across substantial parts of the catchment; however, in tributaries are shown in Table 4.5. They clearly show the
addition, a number of other factors are considered important high baseflows contributed by the Chalk tributaries such
(Entec, 2002). as the Dever, Anton and Wallop Brook (all with baseflow
indices of approximately 0.96), whereas the groundwater
• The presence or absence of marls – the Lower and contribution to surface flows in a river such as the Blackwater
Middle Chalk Formations are generally marlier than that drains Palaeogene deposits is less significant (baseflow
the Upper Chalk and are less transmissive. index of 0.51).
• The development of more localised solution enhanced The three tributaries which feed the Upper Itchen (the
fracture zones in the Upper Chalk above and below the Candover, Cheriton and Alre) are all substantially derived
main zone. from the Chalk as shown by their high baseflow indices
• The nature of cover – thick, low permeability cover (Table 4.6); of the three, the Alre is volumetrically the most
reduces recharge, the underlying Chalk remains significant, with much of its flow from artesian boreholes and
undeveloped, and transmissivity remains relatively low. springs, which supply watercress beds. It has been suggested
• Periglacial influences – periglacial processes enhance (Entec, 2005) that the groundwater catchment of the Alre is
flow and transmissivity in present-day dry, ephemeral significantly larger than the surface water catchment.
and perennial river valleys, compared to interfluves. As the Itchen initially flows west, its flow increases
• Postglacial and recent groundwater flow – this en- substantially until it reaches 4.22 m3 s–1 on average at Easton
hances any fracture development and leads to higher (Table 4.6). As the river turns south further increases in flow
transmissivities around present-day rivers. are seen, which are mainly from groundwaters (as shown for
example by the very high baseflow index at Highbridge) but
Comparison of aquifer properties data, published in the including, further south, a modest increment from the surface
Aquifer Properties Manual (Allen et al., 1997), with the water draining Monks Brook. Near to its mouth, gauged
geographical location of axes of major folds in the study area flows average 5.40 m3 s–1 (Table 4.6), which is around half
showed that there appear to be some systematic relationships, the flow of the River Test, and about a quarter of that of the
as discussed in Chapter 3. In general, transmissivities are Avon. This is somewhat larger than that expected given the
higher along the synclinal axes than anticlinal axes and surface catchment area, but this is an underestimate of the
storage coefficients are higher near water courses. These groundwater catchment area.
observations are consistent with those of Giles and Lowings
(1990) who identified five groundwater mounds (at Ellisfield, 4.5.4 Groundwater flooding
Lasham, Medstead, Froxfield and Preshaw: see Figure 3.2)
that form groundwater catchment divides — all but one A characteristic of the Test and Itchen catchments is
related to anticlinal structures in the Chalk. groundwater flooding following high winter rainfall. There
The potential for developing secondary hydraulic was extensive groundwater flooding in both catchments in
connectivity increases from Lower to Middle to Upper the winter of 1994–95, 2000–01 and 2002–03. Chalk stream
Chalk with brittle ‘hardbands’ locally enhancing fracturing, heads migrated kilometres upstream of their usual positions
but the potential for fracture development reduces with depth and flooding affected roads, houses and businesses over a
below ground level. Consequently, the presence of Upper period of several weeks.
Chalk and/or hardgrounds in valleys may be important in
enhancing flow to abstraction wells (Entec, 2002). 4.5.5 Hydrogeology — Test
The effect of the postulated stratigraphically related fissure
zones on groundwater flows will be strongly influenced by Historically, there appears to have been relatively little
regional structure. In both the Test and Itchen catchments, it hydrogeological work performed solely within the Test
appears that synclinal axes combined with fissure zones can catchment. Headworth (1972) for example calculated a
concentrate flow in certain river valleys, notably the River specific yield (of about 0.033) for the catchment using
Alre and River Anton. recession curves from boreholes and river hydrographs.

64
Table 4.5 Gauging station data from selected sites on the River Test and tributaries
(source: Marsh and Lees, 2003).

Station name Grid Catchment Period Mean annual Mean flow Q10 Q95 Base
(river name) reference area (km2) of record rainfall (mm) (m3 s–1) (m3 s–1) (m3 s–1) flow index
Weston Colley SU 49 39 52.7 1979–1995 774 0.10 0.2 0.03 0.96
(Dever)
Fullerton SU 37 39 185.0 1975–1999 782 1.82 2.7 0.95 0.96
(Anton)
Broughton SU 31 33 53.6 1955–1999 803 0.35 0.8 0.02 0.94
(Wallop Brook)
Ower SU 32 17 104.7 1976–1999 865 0.85 2.1 0.15 0.51
(Blackwater)
Chilbolton (Test) SU 38 39 453.0 1989–2000 836 5.45 8.4 2.89 0.97
Broadlands (Test) SU 35 18 1040 1957–2000 819 11.01 16.7 5.76 0.94

Table 4.6 Gauging station data from selected sites on the River Itchen and its tributaries
(source: Marsh and Lees, 2003).

Station name Grid Catchment Period Mean annual Mean flow Q10 Q95 Base
(river name) reference area (km2) of record rainfall (mm) (m3 s–1) (m3 s–1) (m3 s–1) flow index
Alresford (Alre) SU 57 32 57.0 1970–1999 873 1.55 2.1 1.03 0.98

Sewards Bridge SU 57 32 75.1 1970–2000 909 0.64 1.0 0.27 0.97

(Cheriton)
Borough Bridge SU 56 32 71.2 1970–1999 825 0.52 0.8 0.27 0.96
(Candover)
Stoneham Lane SU 44 17 43.3 1987–2000 823 0.23 0.6 0.02 0.41
(Monks Brook)
Easton (Itchen) SU 51 32 236.8 1975–2000 871 4.22 6.0 2.66 0.98

Highbridge SU 46 21 360 1958–2000 855 5.30 7.8 2.92 0.96

(Itchen)
Riverside Park SU 44 15 415 1982–1999 837 5.40 8.8 2.83 0.92
(Itchen)

More recently, however, the modelling study undertaken for Analysis of flow data for the River Test shows that the
the Environment Agency (Entec, 2005) examined in detail catchments with the biggest flows for their catchment areas
the available information for the River Test. are the Upper Test to Chilbolton, the Pilhill Brook and the
Previous modelling on the Bourne Rivulet and Wallop River Anton and River Dun. The catchment to Chilbolton
Brook by Mott MacDonald made use of stepped contrasts follows a syncline, which may be effective in draining the
in hydraulic conductivity with depth to simulate seasonal area. The Anton and Pilhill Brook appear to have very high
flow profiles and groundwater level responses. The model flows considering their surface catchment areas and it has
also simulated low transmissivity in an anticlinal axis in the been inferred (Entec, 2005) that the groundwater catchment
Wallop Brook catchment. For this catchment, observations, is far larger than the surface water catchment.
both from groundwater hydrographs and the calibration The Stockbridge Rock Member may influence flows in
exercise, indicated that a two-layered model was required for the Anton, Pilhill Brook and the Test around Chilbolton and
the area. The upper layer had a high hydraulic conductivity Kings Somborne. In these locations there is generally strong
and thinned from the river towards the interfluves. The lower flow accretion which may be related to this hardground.
layer was given a low hydraulic conductivity. The rock unit also crops out under the River Bourne and it
Large cress beds in the Test catchment affect the has been suggested (Entec, 2005) that it may act as a drain,
natural flow regime of several Chalk tributaries. The cress moving water eastwards towards the Anton and Pilhill
beds make use of both artesian and pumped boreholes. Brook.
Abstraction can cause flows upstream of the beds to be
depleted whilst flows downstream are augmented by the 4.5.5.1 Water balance
cress-bed discharges. A water balance was carried out over the period 1991 to
Pumping tests have been carried out at several of the 1995 with a monthly time step and with results aggregated
major public water supply sources and give a range of to annual average and long-term average figures (Entec,
transmissivity and storage values. As many of these 2005). The result is shown in Table 4.7 and indicates that a
abstractions are in valleys, the aquifer parameters do not modest percentage (<4 per cent) of the available resource in
represent the full range of Chalk aquifer characteristics. the Test catchment is committed to abstraction.

65
4.5.6 Hydrogeology — Itchen Rennie (1994) reviewed fracture-zone elevations in
boreholes on a section along the Candover catchment.
In contrast to the Test, the Itchen catchment has been relatively
This work suggested that zones of preferentially enlarged
well studied. Headworth (1972) estimated the specific yield
fractures can be extrapolated between Totford, Axford and
of the Chalk in the catchment as about 0.034 using recession
Wield, approximately parallel with the falling elevation of
curves from boreholes and river hydrographs. Later, a
the river bed. TV logs of boreholes at Totford and Axford
scheme designed to augment the flow of the River Itchen
show that some of these zones are associated with tabular
during drought years enabled the nature of the aquifer to be
flint bands (see also the discussion based on borehole
studied in greater detail. Initially, the behaviour of artesian
geophysical work in the Candover catchment in Chapter 3).
boreholes located at watercress farms (most of which
Allen et al. (1997) summarised the results of a number
were near Alresford, to the south of the Candover valley
of modelling projects conducted in the Hampshire region.
with some in the River Dever catchment) was investigated
Several pieces of evidence, including the collected data,
(Headworth, 1978). This work concluded that the boreholes
were examined to give an indication of the lateral variation
penetrated Chalk with very high transmissivity in which
of aquifer properties in the Hampshire area. It was thought
upward leakage occurred through less permeable layers and
that transmissivity values of 1000 m2 d–1 are common in the
that groundwater flow was concentrated into a relatively
valleys, and that both transmissivity and storage coefficient
narrow zone of little more than 30 m thickness.
decrease up the interfluves. A layered aquifer with an
In the Candover valley, six production boreholes were
extensive high transmissivity zone is thought to exist in
drilled near the heads of various dry valleys. These were
the Candover catchment, with typical transmissivity values
then tested by pumping them individually, and as a group.
of 1000 to 3000 m2 d–1 and storage coefficient values of
A mathematical model was subsequently constructed for the
0.01 to 0.03. The neighbouring Alre catchment is thought
catchment (Southern Water Authority, 1979; Keating, 1978;
to have a discrete set of large diameter conduits, which, if
Headworth et al., 1982; Keating, 1982). However, the model
intersected, will give extremely high transmissivity values
did not give a good representation of the water levels or
(>5000 m2 d–1). However, if a borehole does not intersect
stream flows, so a lumped parameter model was developed.
the system, the yield is very low. Estimates of storage
The aquifer was assumed to have two layers: a shallow
coefficient from pumping tests in the Alre are lower than
layer with a transmissivity of 10 000 m2 d–1 and a storage
those calculated from modelling or river hydrographs. Along
coefficient of 0.05, and a deeper layer with transmissivity
the axes of anticlines, aquifer properties are considered to
of 1000 m2 d–1 and a storage coefficient of 0.01. The
be less well developed. Often anticlinal axes are associated
model gave good representations of the winterbourne and
with groundwater mounds that have a low transmissivity,
groundwater hydrographs.
possibly less than 100 m2 d–1.
In parallel with the river augmentation project, the BGS
In the upper Itchen catchment, the east–west folding can
carried out permeability tests using a variety of different
strongly affect the relationship between fissured zones and
techniques (Price et al., 1977; Price et al., 1993). Three
the surface and therefore the ease with which water can
boreholes were studied in the Candover catchment using
enter and leave the groundwater system. For example, the
packer tests, rock-core analysis and geophysical logs.
Alre follows the course of a syncline that dies out around
As an extension of the river augmentation scheme,
the confluence with the Itchen as anticlinal axes to the north
another tributary catchment of the Itchen, the Alre, was
and south converge. It has been suggested (Entec, 2002) that
investigated. A number of pilot boreholes were drilled
this structural configuration, in conjunction with the general
and four production boreholes completed. As with the
regional structure, assists both recharge and discharge
Candover scheme the boreholes were tested together and
processes in the aquifer. Recharge is helped by the cropping
individually (Southern Water Authority, 1984; Giles and
out of the main fracture zone in the upper parts of the Alre and
Lowings, 1990; Southern Science, 1991). Subsequently, a
Itchen catchments (locally enhanced by run-off from clay-
groundwater model of the Alre and Cheriton catchments
with-flints cover) and by enlargement of the groundwater
was developed (Irving, 1993). Although requiring
catchment due to an anticlinal structure. Discharge is aided
refinement, the model gave a good representation of
because the structural setting — principally the convergence
groundwater levels and stream flow.
of the anticlines and the dying out of the Alre syncline —
causes outcrops of the structurally controlled fracture zone to
occur in the bed of the River Itchen, downstream of the Alre.
Table 4.7 Water balance calculation for the Test This causes significant baseflow accretion in this section of
catchment (1991 to 1995) (data from Entec, 2005). the river during summer months. The high flows of the Alre
itself are in fact somewhat artificial as they are supported
River Test to Kimbridge
by discharge from artesian boreholes. These boreholes
penetrate the high-transmissivity fracture zone which lies
Area (km2) 957 below a low-permeability Chalk layer. In the absence of the
boreholes most of the groundwater flow would continue
Total available water (Ml d )
–1
1073 down-gradient, to appear in the Itchen further downstream,
probably where the fracture zone meets the Itchen river bed
Recharge (Ml d–1) 999 (381mm a–1)
(Entec, 2002). The very high transmissivity values found in
Discharges (Ml d–1) 27 the Alre catchment are assumed to be associated with this
Abstractions (Ml d ) –1
40 zone.
In the Candover catchment, groundwater flow appears
River flow (Ml d )
–1
960 to be concentrated in a series of individual horizons in
Baseflow (Ml d )
–1
912 the Chalk (Headworth, et al., 1982; Rennie, 1994). This is
illustrated by the stepped nature of flow hydrographs
Imbalance (Ml d )
–1
97
which is taken to indicate the drying of fracture horizons
which supply water to the river as groundwater levels fall.

66
Entec (2002) suggest that comparison of the elevations of The zones are narrow, usually between 0.5 and 1 m wide,
groundwater flow horizons between the Candover and the with parallel faces (Hopson, 2000). Between the Portsdown
Alre imply that the two are related. The Cheriton accretes Anticline and the main outcrop of the Chalk Group to the
at a fairly constant rate down its length. Flow is baseflow north is the Bere Forest Syncline, where the sequence of
dominated and the river responds quickly to recharge. Palaeogene strata is preserved (in places exceeding 100 m
Geophysical evidence suggests that fracturing in the aquifer in thickness). The Bere Forest Syncline continues to the east
is not related to the same stratigraphical horizons as that of Portsdown where it is termed the Chichester Syncline.
seen in the Candover and the Alre (Entec, 2002). Drift deposits occur overlying both the Chalk and
The Alre catchment is significantly more productive than Palaeogene rocks. Head and alluvium are found mainly in
those of the Candover and Cheriton rivers and it appears from the north–south dry valleys on the Chalk with terrace gravels
potentiometric data that the Alre groundwater catchment is in the Meon valley. Clay-with-flints is commonly located on
substantially larger than the surface catchment. hilltops above the valleys. In the area of the Bere Syncline,
The Candover and Cheriton rivers generally respond head is common overlying the London Clay. To the south
to recharge more rapidly than the Alre and are more of Portsdown, the Palaeogene and Chalk has an extensive
seasonably variable. This is attributed to the fact that the cover of river terrace, raised marine and tidal flat deposits
Alre is augmented by artesian flows. Higher flows in the with aeolian sands and silts.
Cheriton than the Candover occur at times of high ground-
water water levels, because the former drains a larger block 4.6.2 Surface water flows
of aquifer, despite having a similar surface catchment size.
In the East Hampshire area, many streams rise from springs
4.5.6.1 Water balance at the southern margin of the Chalk, where it is overlain
A water balance was carried out over the period 1991 to by Palaeogene sands and clays. The Hamble has a number
1995 with a monthly time step and with results aggregated of tributaries which originate at this Chalk/Palaeogene
to annual average and long-term average figures (Entec springline. These springs are unreliable in summer months,
2005). The results in Table 4.8 indicate that a much higher leading to low stream flows (Environment Agency, 1999c).
percentage (about 20 per cent) of available groundwater Most of the Hamble’s flow is derived from a relatively
is abstracted compared with the Test catchment (<4 per dense network of minor streams, which drain a wide area of
cent). superficial deposits, leading to a lower baseflow index than
would be expected for a purely chalk-fed river (Table 4.9).
The River Meon is sometimes intermittent upstream
4.6 EAST HAMPSHIRE of Warnford, where it is augmented by large springs
and discharges from cress beds. Where it is gauged at
4.6.1 Catchment geology Mislingford, the Meon exhibits the characteristics of a typical
Chalk river, with a high baseflow index (Table 4.9). Where
The geology of East Hampshire is dominated by the the river flows over the Palaeogene deposits, its response to
asymmetrical syncline of the Hampshire Basin (see rainfall is more flashy. Further south, at Funtley, the Meon
Figures 1.11 and 2.2). Mainland East Hampshire lies on the again flows over the Chalk where minor (previously major)
northern limb of the syncline, which in general dips at a springs occur (Environment Agency, 1999c) after which it
relatively low angle to the south. This limb carries two major again crosses Palaeogene deposits.
subsidiary anticlinal structures, the Warnford Dome in the
north, which brings Lower Chalk to crop in the head waters of
the River Meon, and the Portsdown Anticline, which brings 4.6.3 Hydrogeology
a west–east-trending ridge of Upper Chalk (Portsdown The Chalk is the main aquifer in East Hampshire. Test
Chalk Formation and the Spetisbury Chalk Member of pumping shows that the underlying Upper Greensand is often
the Culver Chalk Formation) to surface within the area of not considered to be in hydraulic continuity with the Chalk,
Palaeogene cover north of Portsmouth. Well-defined, strike- and its depth over the most of the area precludes its use as
orientated, near-vertical, narrow fracture zones are a feature an aquifer, although usable water supplies can be obtained
of the chalk members exposed on the Portsdown structure. where it occurs at shallower depths. Water supplies can
also be obtained from the arenaceous units of the overlying
Palaeogene strata, although these vary considerably in both
Table 4.8 Water balance calculation for the Itchen quantity and quality.
catchment 1991 to 1995 (data from Entec, 2005). In the east, there is rapid and substantial movement of
groundwater from the north under the Palaeogene deposits
Itchen to Allbrook of the Chichester–Bere Forest Syncline to major springs at
and Highbridge Havant and Bedhampton. To the west, the Chalk under the
Bere Forest Syncline appears to have poor aquifer properties
Area (km2) 541
and limited resource potential, presumably as a result of the
Total available water (Ml d–1) 635 thickness of overburden.
The hydrogeology of the Portsdown Anticline (for
Recharge (Ml d )–1
602 (406mm a–1)
example, its effect on the Bedhampton springs) is becoming
Discharges (Ml d–1) 12 better understood. Groundwater appears to flow off the
Abstractions (Ml d ) –1
125
anticline in a generally radial pattern, supplying springs
along the north coast of Portsmouth Harbour and springs
River flow (Ml d )
–1
503 feeding the River Wallington.
Baseflow (Ml d )
–1
474 The Palaeogene cover has an important bearing on the
Chalk aquifer properties in this area because it inhibits direct
Imbalance (Ml d–1) 18
recharge to the underlying Chalk aquifer; in particular, the

67
Table 4.9 Gauging station data from selected sites on rivers in east Hampshire
(source: Marsh and Lees, 2003).

Station name Grid Catchment Period Mean annual Mean flow Q10 Q95 Base
(river name) reference area (km2) of rainfall (m3 s–1) (m3 s–1) (m3 s–1) flow
record (mm) index
Frogmill (Hamble) SU 52 14 56.6 1972–2000 882 0.42 0.8 0.10 0.67
Mislingford SU 58 14 72.8 1958–2000 930 0.97 2.0 0.20 0.93
(Meon)
North Fareham SU 58 07 111.0 1951–2000 855 0.62 1.6 0.04 0.41
(Wallington)

lowest formations are usually poorly permeable and inhibit Groundwater flow pathways are generally from north
recharge. Run-off from the Palaeogene deposits is directed to south. Between catchments, flow pathways can be in
into surface watercourses that flow onto the Chalk at other directions: for example, between the Meon and upper
discrete points, thus providing a mechanism for enhanced Itchen catchment to the east, groundwater flow is initially
local dissolution of the Chalk. The effect can be exacerbated to the north (Institute of Geological Sciences, 1979a and
if the water has picked up an acidic chemical signature from 1979b). As with most Chalk catchments, the depth to the
the Palaeogene strata. Enhanced dissolution features (e.g. water table will vary considerably between interfluve
dolines) are, for example, observed on the northern edge of areas, where the unsaturated zone is thickest (sometimes
the outcrop of the Palaeogene deposits of the Bere Forest exceeding 40 m), and valley areas, where the unsaturated
Syncline, where the topography is undulating, locally zone is thinnest. Valleys tend to provide the best sites for
permitting run-off to the north onto the Chalk outcrop. boreholes because it is here that there has been maximum
The existence of these karstic features suggests that some fracture development.
groundwater can move rapidly along highly developed but
relatively thin, solution-enhanced fractures at the Chalk–
Palaeogene boundary, passing below the Bere Forest 4.7 ISLE OF WIGHT
Syncline to re-emerge near sea level through the Chalk of the
Portsdown Anticline. That such rapid transit systems exist in 4.7.1 Catchment geology
certain topographically/structurally controlled situations is
well shown by the major spring supplies of Bedhampton and The Isle of Wight lies on the southern, steeply dipping
Havant. These form the largest public supply spring source limb of the synclinal Hampshire Basin. Here, the Chalk
from the Chalk in the UK and are reputed to be the largest forms a spine running west–east across the island and
group of springs used for public supply in Europe. They bedding has been rotated to near-vertical along much of the
produce between 53 and 170 million litres of water every outcrop in a structure known as the Purbeck–Isle of Wight
day (Portsmouth Water website: www.portsmouthwater. Monocline. The Chalk is some 300 m in thickness on the
co.uk). Rapid karstic groundwater flow from the north under Isle of Wight and extends as high as the Upper Campanian
the Bere Forest Syncline to the springs has been shown by Portsdown Chalk Formation beneath the Palaeogene
a tracer study (see Chapter 3 for description). Similar rapid unconformity. Mortimore et al. (2001) provide descriptions
flows under the syncline are considered to occur further of the complete Chalk succession from the classic coastal
east, along the syncline axis, at Fishbourne and possibly as exposures at Compton Bay and between Sandown Bay and
far west as the Wallington valley. Whitecliff Bay. The spine of Chalk crossing the island is
Further westwards, the flexuring becomes much less generally narrow with a dip approaching vertical, but in
marked, so that the Chalk outcrop becomes narrower and the area south-west of Newport the Chalk crop expands
then disappears beneath the Palaeogene cover west of the to form the broad Idlecombe Down where bedding in the
Meon valley. The increasing thicknesses of low-permeability, Chalk does not dip more than 10 degrees. This broad area
Palaeogene, clay-dominant strata in this direction removes of low dips creates an offset in the narrow steeply dipping
outlet opportunities away from the incised axes of the main Chalk spine and is thought to reflect an offset in the
rivers crossing the structural grain. This reduces the scope underlying faults whose reactivation controlled the uplift
for highly productive flow systems to have developed and and rotation of the Chalk (Underhill and Patterson, 1998).
it is therefore less likely that well-developed, connected, The northern part of the island is underlain by Palaeogene
solution-enhanced flow horizons persist far beneath the strata contained within the Hampshire Basin whose dip is
feather edge of the Palaeogene cover west of the Meon valley. steep close to the Purbeck–Isle of Wight Monocline but
which flattens out rapidly to the north. To the south of the
4.6.3.1 Water level variations and groundwaterflow Purbeck–Isle of Wight Monocline, near horizontal Upper
There is an extensive north–south, dry-valley network across Greensand, Lower and Middle Chalk outcrop within the
the outcrop of Chalk. The water table here is well below the Southern Downs of the Isle of Wight.
valley bottoms and only in exceptional circumstances does
water flow in these valleys. An example of this was during 4.7.2 Surface water flows
spring of 1994, when surface flows were noted in certain
valleys and the potentiometric level was measured some The River Medina and Eastern Yar on the Isle of Wight have
20 m higher than average (Hopson, 2000). This resulted their origins on the chalklands of the Southern Downs on the
in flooding in the villages of Hambledon, Finchdean and island. They flow northwards, initially across an extensive
Stoughton. crop of gently inclined Lower Greensand, before cutting

68
Table 4.10 Gauging station data from selected sites on rivers in the Isle of Wight
(source: Centre for Ecology & Hydrology, 2003).

Station name Grid Catchment Period Mean annual Mean flow Q10 Q95 Base
(river name) reference area (km2) of rainfall (m3 s–1) (m3 s–1) (m3 s–1) flow
record (mm) index
Upper Shide SZ 50 87 29.8 1965–2000 857 0.27 0.5 0.08 0.63
(Medina)
Burnt House SZ 58 85 59.6 1982–1999 831 0.41 0.8 0.04 0.48
(Eastern Yar)

through the central chalk ridge formed by the Isle of Wight Afton (Freshwater) and at Brading, but there are ‘overflow’
Monocline. Surface water flow data are given in Table 4.10. springs at Ashey, Knighton (historically), Shalcome and
Brighstone (Buddlehole) and, at high water levels, at points
along the northern Chalk–Palaeogene contact. It is likely
4.7.3 Hydrogeology
that the Chalk is highly fractured in these areas. There is
The Chalk and Upper Greensand aquifer system in the Isle a flow of groundwater out of the high elevation ‘Plateau
of Wight falls into three distinct parts: the Southern Downs, gravels’ at St George’s Down, just east of Newport onto the
the steeply dipping central ‘spine’ and the central Chalk– Chalk and down a sinkhole.
Upper Greensand area. In the Central Chalk–Upper Greensand area, there are
The Southern Downs are composed of a gently south- numerous dry valleys and the only significant stream
ward dipping, almost planar, surface. Groundwater occurs is the Lukely Brook, which discharges to the north-
mostly within the Upper Greensand, with the overlying east through Carisbrooke. The Bowcombe valley is
Chalk being generally unsaturated. Groundwater discharge underlain by a remarkable historic water gathering
is to the headwaters of the Eastern Yar and Medina and structure — the Idlecombe main, which is part adit
to the (unstable) landslip area on the southern coast and part pipe. The other discharge areas are to the Caul
of the island. Flow within the Upper Greensand is likely to Bourne, the Sheat Stream (Chillerton), the Shorwell
be strongly influenced by the presence of an upper fractured stream and Gatcombe stream. There is an internal spring
sandstone (Malm Rock), intervening fractured chert beds in the Central Downs area at Froglands Farm (just
and by underlying silty sands, the Passage Beds. south of Carisbrooke Castle) at the Upper Greensand–Chalk
Steeply dipping Chalk forms a ‘spine’ across the Island contact, so the hydraulic connectivity between Chalk and
and groundwater flow must be strongly influenced by Upper Greensand must be low. Some sources (Bowcombe)
the dip. There are spring outflows on the down-dip side at abstract from the Upper Greensand as well as from the Chalk.

69
5 Hydrogeochemistry

5.1 INTRODUCTION
BOX 3 MINERALOGY OF THE CHALK
GROUP
The chemical characteristics of groundwater in the Chalk are
determined by a wide variety of processes, but the dominant Chalk is a microporous limestone comprising mainly cocco-
features (such as hardness) are acquired through water–rock lithic fragments with lesser amounts of other fossils (e.g. fo-
interactions with the chalk matrix. The fine-grained nature raminifera) and shelly debris. The coccoliths are composed of
and composition of the Chalk (see Box 3) make it highly a relatively pure calcite (CaCO3). However, small amounts of
reactive to incoming solutions. The initial solute inputs to other elements (Mg, Sr, Mn) are present in the calcite structure,
the aquifer are derived from rainfall, which contains mainly which helped to stabilise the calcite in the marine environment.
marine-derived salts. Most rainfall is slightly acidic due to the This low-Mg calcite is relatively stable at low temperature
presence of dissolved carbon dioxide from the atmosphere. and pressure and little recrystallisation has occurred within
the matrix, except at depths greater than 1000 m (Downing et
The chemistry of rainfall has, in recent decades, been al., 1993). The Chalk in the south Dorset area has undergone
affected by anthropogenic inputs, which may have increased intense deformation and parts of the Chalk in this area have
its acidity due to the formation of low concentrations of undergone significant recrystallisation. Original biogenic silica
sulphuric and nitric acid. As rainfall percolates through the has undergone extensive diagenetic processes to form the typi-
soil, it picks up additional CO2 from soil respiration and cal ‘flint’ bands of the Chalk.
organic matter decomposition and concentrations are often The non-carbonate fraction of the Chalk generally comprises
10 to 100 times that of atmospheric CO2. clays but minor amounts of zeolite, quartz, collophane, dolo-
The soil zone provides an important control on the mite, feldspar and barite have been noted (see Hancock, 1993 for
chemistry of shallow waters in the Chalk due to the high CO2 summary). Significant amounts of clay and terrigenous material
concentrations developed as a result of biological activity. are present in the Lower Chalk, dominantly smectite and quartz
(Bath and Edmunds, 1981). Montmorillonite forms the dominant
The CO2 and the extent to which the system remains open clay in most formations but is particularly abundant in the Upper
with respect to CO2 is a major control on determining the and Middle Chalk (Morgan-Jones, 1977) where it often forms
evolution of waters in carbonate terrains. The CO2 dissolves distinct marl bands which may contain as much as 30 per cent
to produce carbonic acid (H2CO3), which reacts with the clay (Hancock, 1993). The origin of the montmorillonite has
carbonate minerals. The kinetics of this reaction are very been the subject of debate with both neoformational origin and
rapid; the acidity is quickly neutralised by reaction with volcanic origin being suggested. Kaolinite is less abundant and
the carbonate-rich chalk soils and saturation with calcite is often considered to be detrital in origin (Morgan-Jones, 1977)
generally reached within the top few metres of the Chalk. The although it may in part be derived from alteration of glauconite.
pH of water in the soil zone overlying chalk is typically 7.5 to The presence of muscovite and illite is also considered to be de-
8.3 (Price et al., 1993). The chemistry of most groundwaters trital in origin.
Knowledge is limited concerning the complex structures and
is dominated by reactions of the carbonate system. mineralogy developed on fracture surfaces and in the weath-
Other reactions which are important in controlling base- ered mantle immediately below the soil horizon. The develop-
line water quality include redox reactions and ion exchange ment of secondary mineral phases on shallow chalk fracture
reactions, although the strongest influence of these is within surfaces is generally very extensive and is typified by clays and
the confined parts of the aquifer. Fe and Mn sesquioxides (Shand and Bloomfield, 1995).
This chapter aims to characterise the groundwater chem-
istry in the Chalk aquifer of the Wessex Basin and to deter-
mine the dominant geochemical processes that control the Chalk groundwaters in the Lulworth–Poole area from both
spatial variations in hydrochemistry. The source of the data public abstraction sources and BGS cored boreholes. The
used is a combination of Environment Agency and water results showed that groundwaters from the outcrop and
company data and detailed studies carried out by the BGS near outcrop areas had very similar compositions with low
(Buckley et al., 1998; Edmunds et al., 2002). salinity and generally contained high nitrate and low iron
concentrations. The limited data from the confined aquifer
showed that these waters were more reducing, contained
5.2 PREVIOUS HYDROCHEMICAL STUDIES high Fe and nitrate was below the detection limit. Interstitial
water at depth in the aquifer near the coast (in a borehole at
Alexander (1981) presented a summary of unconfined and Lulworth) was fresh, suggesting that fresh water had flowed
confined Chalk groundwaters of south Dorset and discussed to greater depths in the aquifer during late Pleistocene times
the different chemical facies in the area. No clear patterns in when the sea level was lower. In general, more saline water
major ion groundwater chemistry appeared from this study is found in the pore space whilst fresher water flows through
and the groundwaters tended to be of a similar type and clus- the fissures to depths of at least 250 m (Edmunds, 1996).
tered together, dominantly of a Ca-HCO3 type. Alexander Buckley et al. (1998) undertook geophysical logging and
defined two groups of groundwaters: Group I waters from depth sampling of groundwaters in the confined aquifer
the South Winterbourne catchment and the western edge of in the region around Wareham. They showed that there
the outcrop areas, and Group II waters from around the Lul- is significant hydrochemical stratification in the aquifer
worth area, with Group I waters containing slightly higher with waters evolving from Ca-HCO3 to Na‑HCO3-Cl-
Na levels than those of Group II. Edmunds (1996) analysed dominated waters deeper in the aquifer. The chemistry of

70
the groundwaters of Dorset was studied by Edmunds et al. Basin. However, pore-water samples were collected
(2002) who discussed the spatial variations and geochemical from a borehole at Gussage, near Blandford Forum, for
controls on groundwaters, mainly in the unconfined Chalk of tritium analyses by Geake and Foster (1989). Tritium was
Dorset. This Dorset report is in a series describing baseline produced in the 1960s during thermonuclear testing and
groundwater quality and is complemented by a study of the acts as a tracer for the water molecule with the maximum
Palaeogene of the Wessex Basin (Neuman et al., 2004). concentration occurring in 1963. Cores obtained from the
Gussage Borehole in 1970 were sampled subsequently
to provide information on the rate of movement of water
5.3 HYDROCHEMICAL CHARACTERISTICS OF through the unsaturated zone. The three sets of samples
THE UNSATURATED ZONE showed how the tritium peak has moved down the profile.
The preservation of the peak, although decreasing in size
Several processes occur in the unsaturated zone which affect by radioactive decay over time (modified by dispersion),
the pore-water chemistry including: showed that piston flow is an important process in water
transfer through the matrix of the Chalk. The rate of flow
• evapotranspiration, which may significantly concentrate was estimated to be around 1 m per year. Elsewhere in
solutes southern England, the slow transfer of nitrate and sulphate
• uptake of solutes by biomass in the unsaturated zone beneath agricultural land has been
• dissolution of aquifer matrix minerals shown (Foster et al., 1982), which poses a future risk to
• precipitation of minerals (e.g. calcite, clays, groundwater quality.
sesquioxides)
• ion-exchange reactions
5.4 HYDROCHEMICAL CHARACTERISTICS OF
The last three processes may significantly modify both the GROUNDWATERS
character and reactivity of fracture surfaces. It has been
found that enhanced porosity may occur to a depth of A total of around 190 chemical analyses from groundwaters
1 to 1.5 mm away from fractures in the Chalk, related to covering much of the Wessex Basin (Figure 5.1) have
carbonate dissolution (Bloomfield, 1997). Fracture surfaces been used to characterise the groundwater chemistry and a
are likely to be important loci for both dissolution and summary of the data is shown in Tables 5.1 and 5.2. The
precipitation. Shand and Bloomfield (1995) found extensive tables show minimum and maximum concentrations as well
Fe-rich clays and oxide minerals coating fracture surfaces in as statistical averages (mean and median).
the shallow, unconfined Chalk and concluded that this may
exert an important control on flow through the unsaturated 5.4.1 Physicochemical characteristics
zone. Fracture minerals may also be indicators of flowpaths
through the Chalk, e.g. the presence of abundant manganese The groundwaters show a significant range of temperatures
spots on the surfaces of marl horizons implies that these (from 8 to 17ºC) although the majority lie in the range 10 to
may be important for the lateral movement of water. 12ºC. The pH is typically between 7.0 and 7.6 (median pH
There are no major or trace-element data for pore is 7.4) but some of the confined groundwaters are as high
waters from the unsaturated zone of the Chalk in the Wessex as 8.7. Specific electrical conductance (SEC) is typically

Figure 5.1 Location !

of samples used in the Chalk outcrop ! Newbury
!
present study. 1
60
! Chemistry samples
! !
contains OS data © Crown Rivers !
! !
( Basingstoke
copyright and database ! !
!( !
! !
( !
rights 2017 Ordnance Survey !
(( !
! Andover ! Aldershot
Wells ( !
! !
(
! !!!( ! !
[100021290 EUL]. Use of this !
(
1
40 !
data is subject to terms and ! !
( ! ! !
!
conditions. !
! ! ! !
!
( !
(
!
(
Winchester
!! !
(
Salisbury ! !
! !
! (
! !
(
! !
!
(
(
!!
1
20
Yeovil !
! !
( ! Southampton
! ! ! !
!
!
! !
! !! !
1
00 ! !
( ! !
! ! !
!! ! ! ! !
(( Portsmouth
!! ! !!!!
( !
( !
(
! ( ! !
( ! !
(
!! ! ! !
Dorchester Bournemouth !
! Newport N
!(!
! !! ( !
! ! ! !
(!! ! !
!
!
0
80 !
! Weymouth !! !!
(

0 50 km

0
60
3
60 3
80 4
00 4
20 40
4 4
60 80
4

71
Table 5.1 Summary of chemical analyses of Chalk groundwaters from the Wessex Basin.

Parameter Units Minimum Maximum Mean Median N

T °C 7.6 17.2 11.3 11.1 111
pH 6.66 8.65 7.38 7.30 163
Eh mV –16 884 455 503 35
DO mg l–1 <0.1 13.6 7.8 8.4 101
SEC µS cm –1
343 1409 545 520 151
2
H ‰ –47 –36 –42 –41 37
18
O ‰ –7.2 –5.8 –6.6 –6.5 37
13
C ‰ –16.2 –3.8 –13.0 –14.2 38
Ca mg l–1 50 176 101 103 173
Mg mg l –1
0.76 22 3.0 2.2 172
Na mg l–1 5.0 155 14 9.9 143
K mg l –1
0.07 7 1.5 1.2 117
Cl mg l–1 4.5 223 23 19 188
SO4 mg l –1
2.5 69 17 14 143
HCO3 mg l –1
107 400 274 280 152
NO3-N mg l–1 <0.03 18 5.6 5.6 188
NO2-N mg l –1
<0.001 0.057 0.005 0.005 125
NH4-N mg l–1 <0.003 0.31 0.019 0.010 184
P mg l –1
<0.02 0.163 0.04 0.031 62
TOC mg l–1 0.49 5.8 1.6 1.2 30
DOC mg l –1
0.25 4.2 1.1 0.748 56
F mg l –1
0.03 4.3 0.237 0.090 122
Br mg l–1 0.010 0.289 0.076 0.069 66
I mg l –1
0.002 0.047 0.007 0.004 39
Si mg l–1 3.3 13.5 5.5 5.0 147

between 400 and 600 µS cm–1 in the unconfined part of the in the groundwaters of the Chalk. It is apparent that the
aquifer, increasing to a maximum of 1409 µS cm–1 in the data are not generally log normal but often show bimodal
confined part. The redox status of the groundwater largely or multimodal distributions. Calcium and HCO3 are the
reflects the presence or absence of confining conditions dominant solutes and display a relatively linear distribution
with oxidising conditions (dissolved oxygen present, high limited by the solubility of calcite. Nitrate varies over three
Eh) in the unconfined aquifer and reducing conditions in the orders of magnitude with low concentrations present in the
confined part of the aquifer. confined groundwaters.

5.4.2 Major element characteristics 5.4.3 Minor and trace element characteristics
The Wessex Basin Chalk groundwaters have been plotted The ranges in concentrations of minor and trace elements
on a Piper diagram (Figure 5.2), which illustrates the are shown in Table 5.2 and on Figures 5.5 and 5.6. Iron and
relative proportions of major elements and the groundwater Mn concentrations are generally low but reach relatively
composition. The vast majority of groundwaters are from high concentrations in the confined groundwaters. The
the unconfined aquifer and these are of Ca-HCO3 type concentrations of most metal species are low because of
with low Mg/Ca ratios similar to the solid chalk matrix. their low solubility in circumneutral, oxidising groundwater.
Groundwaters in the confined aquifer trend towards higher
relative proportions of Mg and Na. A similar trend towards 5.4.4 Depth variations
higher relative Cl also occurs but is less pronounced than
that towards Na. Sulphate concentrations are generally Samples obtained from pumped boreholes often represent
low in these groundwaters and only two groundwaters plot mixtures of water, possibly of different ages and chemistry.
towards sulphate dominance. There appear to be two trends They are also derived largely from fractures, which often
on the central Piper plot: a few groundwaters trend towards have different chemistry from water contained within
the upper part due to relatively high Ca and Cl and several the pores of the matrix. There are limited data on depth
sites trend towards the lower right due to relatively high Na, variations in individual boreholes in the region, however,
HCO3 and Cl. Edmunds et al. (2002) reported porewater chemistry from a
The range in concentrations is highlighted using box plots borehole at Lulworth in unconfined Chalk and Buckley et al.
(Figure 5.3) and cumulative probability plots (Figure 5.4). (1998) analysed depth samples collected from boreholes in
These highlight the considerable range of concentrations the confined aquifer in the region of Wareham.

72
Table 5.2 Summary of trace element analyses of Chalk groundwaters from the Wessex Basin.

Parameter Units Minimum Maximum Mean Median N

Ag µg l–1 <0.05 6.0 0.32 0.03 68
Al µg l –1
<1 927 14 2.5 150
As µg l–1 <0.2 75 1.1 <0.2 122
Au µg l–1 <0.05 <0.05 <0.05 <0.05 34
B µg l–1 <20 399 32 20 93
Ba µg l –1
2.50 135 21 13 158
Be µg l–1 <0.05 0.50 0.22 <0.05 66
Bi µg l–1 <0.05 0.07 0.03 <0.05 39
Cd µg l –1
<0.05 0.50 0.11 0.10 101
Ce µg l–1 <0.01 0.04 <0.01 <0.01 39
Co µg l–1 <0.01 15 0.77 0.46 66
Cr µg l –1
<0.05 4.0 0.89 <0.5 96
Cs µg l–1 <0.01 0.04 <0.01 0.01 39
Cu µg l–1 0.20 370 9.5 2.0 101
Dy µg l –1
<0.01 <0.01 <0.01 <0.01 39
Er µg l–1 <0.01 <0.01 <0.01 <0.01 39
Eu µg l–1 <0.01 0.01 <0.01 <0.01 39
Fe µg l –1
0.20 36600 401 10 172
Ga µg l–1 <0.05 <0.05 <0.05 <0.05 39
Gd µg l–1 <0.01 0.02 <0.01 <0.01 39
Ge µg l–1 <0.05 0.32 0.05 <0.05 39
Hf µg l –1
<0.02 <0.02 <0.02 <0.02 34
Hg µg l–1 <0.1 0.50 <0.01 <0.1 60
Ho µg l–1 <0.01 0.01 <0.01 <0.01 39
In µg l –1
<0.01 <0.01 <0.01 <0.01 34
Ir µg l–1 <0.05 <0.05 <0.05 <0.05 34
La µg l–1 <0.01 0.02 <0.01 <0.01 39
Li µg l –1
0.20 179 12 1.4 43
Lu µg l–1 <0.01 <0.01 <0.01 <0.01 39
Mn µg l–1 <2 726 13 4.2 173
Mo µg l –1
<0.1 2.8 0.30 0.05 39
Nb µg l–1 <0.01 0.01 <0.01 <0.01 34
Nd µg l–1 <0.01 0.02 <0.01 <0.01 39
Ni µg l –1
<0.2 20 2.2 0.80 39
Os µg l–1 <0.05 <0.05 <0.05 <0.05 34
Pb µg l–1 <2 22 1.3 <2 102
Pd µg l –1
<0.2 <0.02 <0.2 <0.2 34
Pr µg l–1 <0.01 0.01 <0.01 <0.01 39
Pt µg l–1 <0.01 <0.01 <0.01 <0.01 34
Rb µg l–1 0.71 6.1 1.7 1.5 39
Re µg l –1
<0.01 0.02 <0.01 <0.01 34
Rh µg l–1 <0.01 <0.01 <0.01 <0.01 34
Ru µg l–1 <0.05 <0.05 <0.05 <0.05 34
Sb µg l –1
<0.05 1.4 0.30 0.20 94
Sc µg l–1 <0.4 3.0 1.2 1.4 56
Se µg l–1 <1 2.9 <1 <1 120
Sm µg l –1
<0.5 <0.05 <0.05 <0.05 39
Sn µg l–1 <0.5 0.21 0.08 0.07 34
Sr µg l–1 142 3570 467 237 78
Ta µg l –1
<0.05 <0.05 <0.05 <0.05 36
Continued overleaf

73
Table 5.2 (cont) Summary of trace element analyses of Chalk groundwaters from the
Wessex Basin.

Parameter Units Minimum Maximum Mean Median N

Tb µg l –1
<0.01 <0.01 <0.01 <0.01 39
Te µg l–1 <0.05 0.10 <0.05 <0.05 34
Th µg l –1
<0.05 <0.05 <0.05 <0.05 39
Ti µg l–1 <10 <10 <10 <10 34
Tl µg l –1
<0.01 0.08 0.02 0.01 39
Tm µg l –1
<0.02 <0.02 <0.02 <0.02 39
U µg l–1 <0.05 1.4 0.31 0.26 39
V µg l –1
<1 3.0 <1 <1 83
W µg l–1 <0.1 <0.1 <0.1 <0.1 34
Y µg l –1
<0.01 0.15 0.02 <0.01 39
Yb µg l–1 <0.01 0.01 <0.01 <0.01 39
Zn µg l –1
0.80 1460 56 15 112
Zr µg l–1 <0.5 <0.5 <0.5 <0.05 34

5.4.4.1 Interstitial waters of the Lulworth Borehole temperature increases linearly with the geothermal gradient,
Interstitial waters were extracted from a research borehole at whilst at shallower depths the temperature is disturbed by
Lulworth drilled during the late 1970s to a depth of 170 m active groundwater flow. The SEC below this depth decreases
below OD (Edmunds et al., 2002). Geophysical logging slightly showing that fresh groundwater exists to a depth of
carried out in 1997 was used to determine the groundwater around 130 m along the Dorset coast.
flow at this site using the change of gradient of the temperature The presence of relatively high NO3-N in the top 65 m
log (Figure 5.7). The base of the present-day flow system of the borehole indicates the penetration of modern, pol-
was interpreted to be around 65 m; below this depth, the luted groundwater. Selected data from the deeper part of the

Figure 5.2 Piper 100

diagram showing the 100
90
relative proportions 90
of major cations and 80
80
anions in Wessex Basin 70
groundwaters. 70
60
60
50
50
40
40
30
30
20
20
10
10
0
0
Mg SO4
0 0
100 100
10 10
90 90
20 20
80 80
30 30
70 70
40 40
60 60
50 50
50 50
60 60
40 40
70 70
30 30
80 80
20 20
90 90
10 10
100 100
0 0
100 90 80 70 60 50 40 30 20 10 0 0 10 20 30 40 50 60 70 80 90 100
Ca Na HCO 3 Cl

74
 1000 5.10 for one spring and one borehole. The Litton Cheney
spring [SY 55 90]) lies at the western part of the outcrop
and has been monitored (by the Environment Agency) since
1992. Although there are no long-term trends in SEC, these
100 do vary seasonally, most likely indicating that at least one
source of the spring is of short residence time (probably
weeks or months). This seasonal trend is seen to a lesser
degree for SO4 and Cl. The seasonal changes may reflect
10 greater, less dilute recharge in the late winter months when
evapotranspiration is lower. Nitrate also shows seasonality
Concentration (mg l–1)

with higher concentrations during the times of lower SEC.

In contrast to the other elements, NO3-N shows an increasing
1
trend with time as well as a seasonal trend (Figure 5.9).
Data from a shallow borehole (West Houghton
[ST 82 04]) are shown on Figure 5.10. The Cl concentra-
tion is slightly lower than for the Litton Cheney spring
0.1
and may reflect the decrease in maritime influence (Ed-
munds et al., 2002) but seasonal trends or long-term trends
are not apparent. As with Litton Cheney, NO3-N shows an
increase in concentration with time, reflecting increased
N-loading or higher N en route from the recharge area.
0.01
Such increases are often found in the Chalk aquifer where
it is unconfined.
The natural baseline for species such as nitrate are diffi-
cult to determine from modern data. Some data (unconfined
0.001
CI Na SO4 Mg Ca K HCO3 NO3–N aquifer) from the BGS archives and from Whitaker and Ed-
wards (1926) are shown on Table 5.4 and indicate that NO3-
N concentrations were much lower in the early part of the
Figure 5.3 Box plot showing ranges in concentration of twentieth century.
major elements in groundwaters of the Wessex Chalk.

5.5 AGE OF THE GROUNDWATER

borehole are shown in Table 5.3 where it can be seen that
nitrate is present at much lower (probably baseline) concen- For the deeper, confined groundwaters, direct determination
trations. There is additionally an increase in some elements of groundwater age using 14C is difficult in aquifers such as
(e.g. Sr) below a depth of around 140 m, possibly indicating the Chalk because of problems with correction for carbon
that a more mature groundwater is present at depth. derived from the Chalk matrix. However, the indications are
that most of the groundwaters beneath Wareham, to a depth
5.4.4.2 Depth samples from the confined aquifer of 300 m below OD, are of Holocene age (around 11 000
years to present) (Edmunds et al., 2002) but are likely to be
Groundwater samples were collected from three boreholes several thousand years old since they only contain traces of
drilled into the confined aquifer west of Poole Harbour: 14
C.
Bulbury, Lytchett Minster and Holton Heath. These were Several indicators improve understanding of the
analysed during a preliminary study of the potential for residence time of groundwater in the unconfined aquifer,
aquifer storage and recovery (ASR) in the confined Chalk of for example the seasonality of springs, tritium data and
Wessex (Buckley et al., 1998; Gaus et al., 2002). The depths
of the boreholes increase from Bulbury through Lytchett
Minster to the deepest at Holton Heath and selected data
have been plotted on Figure 5.8. There is a general increase 99
98
in Na and Cl with depth, both in individual boreholes and Ca
Mg
between boreholes. This is also seen to a lesser degree 95
Na
with HCO3, K and several trace elements (F, B, Ba, Li) but 90 K
HCO3
Ca displays a decrease with depth (Figure 5.8). The high
Cumulative frequency

80
SO4
concentrations of F in these boreholes, increasing with depth, 70 CI
60 NO3–N
is of particular note and is discussed further in Chapter 6. An 50 DOC
increase in 13C and the ratios Sr/Ca and Mg/Ca indicate that 40
30
the extent of water–rock interaction is greater in the deeper
20
groundwaters. The increase in Na and decrease in Ca results
from ion-exchange reactions. The stable isotopes 2H and 18O 10
5
are consistent with a dominantly Holocene age but a slight
decrease may imply that an older component of Pleistocene 2
1
water is present. 0.01 0.1 1 10 100 1000
Concentration (mg l–1)

5.4.5 Temporal variations

Figure 5.4 Cumulative probability plot showing the
There is a lack of long-term data available for the study area. ranges in concentration of major elements in groundwaters
Examples of available data are shown on Figures 5.9 and of the Wessex Chalk.

75
Figure 5.5 100000

Box plot showing

the ranges in 10000
concentration of minor
and trace elements in
1000
groundwaters of the
Wessex Chalk.
100

Concentration (µg l–1)

10

0.1

0.01

0.001
Br Sr B Si F Li I Ba Mo Zn Ni As Cu Fe Mn V Al Co Cd Cr Pb Sc La Y Be

the presence of a wide range of introduced contaminants. the Chalk reaches saturation rapidly with calcite, controlled
The abstraction of groundwater is likely to have modified by the congruent dissolution of calcite:
the natural groundwater flow, distorting the original age
structure and promoting mixing of waters of different ages. Cax(Mg,Sr)1–xCO3 + CO2 + H2O = xCa + 1–x(Mg,Sr) +
The presence of nitrate and other introduced contaminants 2HCO3
shows the degree to which the shallow groundwater has
been affected by modern (post 1940s) recharge. However, Calcium, Mg and HCO3 show little change in the uncon-
detailed studies (e.g. using CFCs and other dating tools) fined aquifer reflecting the above reaction, but the confined
are required to assess the age stratification with depth. groundwaters show an increase in Mg and Sr that reflects
Seasonal oscillations in the hydrochemistry of springs (e.g. the change from congruent to incongruent dissolution of
Figure 5.9) and boreholes also show that at least some of the calcite.
groundwater is of recent origin. This highlights the fact that calcite continues to be
The available evidence therefore suggests that the dissolved, but in order to maintain chemical equilibrium
unconfined groundwaters are relatively young (or contain a similar amount is reprecipitated. Magnesium, Sr and
a substantially younger component) and that the confined other ‘impurities’ are effectively lost to solution as a purer
groundwaters are much older, of the order of thousands of calcite is precipitated. This change is most evident in
years. This is also substantiated by the chemical changes groundwaters of the confined aquifer and implies a distinct
that have occurred along the regional flow direction. age difference between the unconfined and confined
groundwaters. The change is also shown by an increase in
13
C: the groundwaters of the unconfined aquifer are around
5.6 DOWN-GRADIENT EVOLUTION OF CHALK -16 to -12 representing a mixture of C of biogenic origin
GROUNDWATER from the soil (13C of -25) and from the chalk matrix C
(ca. +2.4). There is a trend deeper in the confined aquifer
Groundwater in the Chalk aquifer evolves with time
as a result of time-dependent reactions and changes in
geochemical environment. Such changes, often seen down 99
Al
the hydraulic gradient, highlight the geochemical processes 98
As
occurring spatially and with time in the aquifer. 95 B
Ba
Hydrochemical changes are due both to mixing with 90
Fe
Cumulative frequency

continued recharge (especially in the unconfined aquifer), 80 Mn

Sr
70
mixing with older formation water and reaction with the 60 U
Zn
chalk matrix. These changes along the hydraulic gradient 50
40
are illustrated using a section from upland areas of Chalk 30
in the west to the coast close to Poole (Figures 5.11 and 20
5.12). However, it should be borne in mind that changes 10
also occur with depth and some of the differences may be 5
due to different borehole depths and borehole completions. 2
1
In addition, recharge occurs over the entire outcrop area.
Nevertheless, the general trends are considered to be due 0.0001 0.001 0.01 0.1 1 10 100 1000 10000
Concentration (µg l–1)
to time-dependant reactions, age and mixing as the waters
move along the hydraulic gradient.
The groundwaters within the unconfined aquifer are of Figure 5.6 Cumulative probability plot showing the
Ca-HCO3 type with low Mg/Ca (and Sr/Ca) ratios similar to ranges in concentration of major elements in groundwaters
the chalk matrix (Figures 5.2 and 5.12). The groundwater in of the Wessex Chalk.

76
Figure 5.7 0
Interstitial water
20
chemistry, temperature
and SEC profiles from 40
Na Cl Sr
the research borehole
at West Lulworth (from 60
Edmunds et al., 2002).
80

Depth (m)
100

120

140

160

180 NO3-N T SEC

200
50 0 50 100 0.0 0.5 1.0 5 10 15 10 12 14 510 515 520 525
Na (mg l–1) Cl (mg l–1) Sr (mg l–1) NO3-N (mg l–1) Temperature (°C) SEC µS cm–1

towards the Chalk reaching -2 in the deep Holton Heath present. These minerals may explain the excess F noted
Borehole close to Wareham. Groundwaters with relatively during aquifer storage and recovery trials in the confined
high Mg/Ca and Sr/Ca ratios are due to a much greater Chalk (Gaus et al., 2002).
amount of reaction and are, therefore, indicative of a Strontium and Ba also increase along the hydraulic
longer residence time. gradient. The Sr is derived dominantly from the calcite
Chloride concentrations are generally low but variable matrix, but unlike Ca it is not limited in these groundwaters
in the unconfined aquifer. Many are slightly higher than by saturation with respect to any mineral phase. Barium
expected for rainfall that has undergone evapotranspiration is likely to be derived either from the chalk matrix or
and these probably reflect some degree of anthropogenic the mineral baryte (Shand and Bloomfield, 1995). Most
input e.g. agricultural, domestic pollution or road salt metals are present at low concentrations, their solubility
application. The deeper confined groundwaters show an being limited at the circum-neutral pH typical of the chalk
increase in Na and Cl (Figure 5.11) indicating that these groundwaters.
older waters contain a component of chalk formation water Dissolved oxygen and redox potential are uniformly high
which has not been flushed from the aquifer. Nevertheless, in the unconfined groundwaters (Figure 5.11). There is a
Cl concentrations remain relatively low and are typically distinct change in these parameters close to the unconfined–
less than 100 mg l–1 Cl, which indicates that the aquifer is confined boundary where a redox boundary has been
well flushed of any original connate water. established. The groundwaters in the confined aquifer are
The increase in Cl in the confined groundwater is reflected reducing and denitrification has led to loss of nitrate from
in a similar increase in Na, but Na/Cl ratios are slightly the groundwater. The reducing nature of the groundwater
higher than expected for a marine-derived source indicating has also led to high concentrations of dissolved Fe and an
an additional source of Na (Figure 5.12). The confined increase in NH4 (Figure 5.11). Sulphate concentrations
groundwaters (pumped and depth samples) from around are generally low but show no distinct pattern along the
Wareham show a decrease in Ca concomitant with the hydraulic gradient and it appears that sulphate reduction is
increase in Na. The changes in Na and Ca (one mole of Ca not a dominant control on sulphate concentrations in these
for every two moles of Na) are consistent with ion exchange groundwaters.
of Ca in solution for Na on exchange sites (probably clay
minerals). This explains the trends observed on the Piper
diagram, which are intermediate between that expected for Table 5.3 Chemistry of deep interstitial waters from the
mixing and ion exchange of a freshening aquifer (Figure 5.2). Lulworth Borehole (after Edmunds et al., 2002).
Chalk porewaters present in isolated parts of the aquifer,
which are probably very old, may show extreme ion exchange
Depth (m bgl)
with concentrations of Ca as low as 3 mg l–1 (Shand, 1999).
Fluoride concentrations are generally low in the Parameter 69.6 104.9 125.1
unconfined aquifer but increase in the confined groundwaters pH 7.40 7.84 7.58
(Figure 5.12). The source of fluoride is most likely to be
Ca (mg l ) –1
67 70 68
derived from apatite or fluorapatite (present mainly as fossils
in the Chalk and particularly associated with marl horizons), Mg (mg l–1) 2.1 1.8 2.0
which have slow dissolution kinetics. This implies that Na (mg l ) –1
23 22 21
there is a significant difference in residence time between K (mg l–1) 2.7 1.0 1.3
the groundwaters of the unconfined and confined aquifer.
Saturation with respect to the mineral fluorite is reached HCO3 (mg l ) –1
179
in some of the deeper groundwaters. Detailed studies have Cl (mg l–1) 27 34 35
shown that in some areas where very old groundwaters NO3-N (mg l ) –1
1.2 7.1 3.6
are present (Shand, 1999) fluorite and amorphous F-rich
Sr (mg l ) –1
0.30 0.24 0.25
phosphate minerals (with rapid dissolution kinetics) are

77
 Lytchett AB Bulbury Holton Heath

50 50 50 50 50 50 50

75 75 75 75 75

100 100 100 100 100 100 100

125 125 125 125 125

Depth (m)

150 150 150 150 150 150 150

175 175 175 175 175

200 200 200 200 200 200 200

225 225 225 225 225

250 250 250 250 250 250 250

275 275 275 275 275

300 300 300 300 300 300 300

0 40 80 120 160 200 1 2 3 4 5 6 20 30 40 50 60 70 80 0 5 10 15 20 25 180 220 260 300 340 10 15 20 25 30 35 40 0 40 80 120 160 200
Na (mg l−1) K (mg l−1) Ca (mg l−1) Mg (mg l−1) HNO3 (mg l−1) SO4 (mg l−1) CI (mg l−1)
50 50 50 50 50 50 50

75 75 75 75 75

100 100 100 100 100 100 100

125 125 125 125 125

150 150 150 150 150 150 150

Depth (m)

175 175 175 175 175

200 200 200 200 200 200 200

225 225 225 225 225

250 250 250 250 250 250 250

275 275 275 275 275

300 300 300 300 300 300 300

.05 .10 .15 .20 .25 .30 1.0 2.0 3.0 4.0 0.00 0.10 0.20 0.30 0.40 0.01 0.02 0.03 0.04 0.05 0.00 0.10 0.20 0.30 7.2 7.4 7.6 7.8 0.01 0.03 0.05 0.07 0.09
NH4 (mg l ) −1
Sr (mg l ) −1
Ba (mg l ) −1
Li (mg l ) −1
Fe total (mg l ) −1
pH Sr/Ca
50 50 50 50 50 50 50

75 75 75 75 75

100 100 100 100 100 100 100

125 125 125 125 125

150 150 150 150 150 150 150

Depth (m)

175 175 175 175 175

200 200 200 200 200 200 200

225 225 225 225 225

250 250 250 250 250 250 250

275 275 275 275 275

300 300 300 300 300 300 300

0.0 0.1 0.2 0.3 0.4 0.5 0 1 2 3 4 5 6 -46 -44 -42 -40 -38 -36 -10 -8 -6 -4 -2 0 0.8 1.0 1.2 1.4 1.6 1.8 -0.4 -0.2 0.0 0.2 0.4 -0.8 -0.4 0.0 0.4 0.8
Mg/Ca F (mg l−1) δ 2H δ13C molar Na/CI SI calcite SI Fluorite

Figure 5.8 Depth variations in groundwaters of the confined aquifer in the Wareham area of the
Wessex Basin.

5.7 SPATIAL VARIATIONS IN CHALK contains low Na, but high Cl. This also had the highest
GROUNDWATERS IN THE WESSEX BASIN specific electrical conductance (SEC). The Cl is balanced
by high Ca and it is possible that this unusual composition
The geochemical evolution along the detailed section is related to ion-exchange of Na for Ca.
described above can also be seen regionally across the aquifer. Distributions for Ca and Mg are shown on Figure 5.15.
Selected parameters have been plotted on Figures 5.13 The highest Ca is generally in areas close to the junction
to 5.17. The unconfined groundwaters are oxidising and with the Greensand but there is otherwise little consistent
have low to moderate electrical conductance (Figure 5.13). variation across the outcrop area. Lower concentrations of
Although data for the confined aquifer are limited, the Ca are present in groundwaters in the deeper part of the
groundwaters typically have much higher electrical confined aquifer around Wareham (where ion exchange
conductance and low dissolved oxygen concentrations. has occurred). Magnesium, in contrast, has the highest
Sodium and Cl distributions (Figure 5.14) are similar, with concentrations in the confined groundwaters as incongruent
concentrations being higher in the confined groundwaters, dissolution increases with residence time. The minor
reflecting mixing with formation water, although there is elements Sr and F (Figure 5.16) both reflect residence time
tendency for higher Na and Cl in unconfined groundwaters and typically display a rapid increase in concentration where
on the Isle of Wight and towards the Dorset coast, which the groundwaters are confined.
may reflect a higher marine (sea salt) input. Higher Na/Cl The effects of diffuse pollution are reflected in high NO3
ratios in the confined groundwaters are due to ion exchange concentrations in the outcrop areas (Figure 5.17), which are
(Na for Ca) as discussed previously. In contrast, one of stable under the oxidising conditions prevalent where the
the samples to the east of Southampton, sampled from aquifer is unconfined. Nitrate concentrations decrease in the
the confined aquifer close to the edge of the Palaeogene, confined aquifer due to denitrification and are typically below

78
 Litton Cheney West Houghton
600 5.5

550 5.0

500 4.5
SEC (µS cm –1)

NO3 -N (mg l–1)

450 4.0

3.5
400

3.0
350

2.5
300
1992 1993 1994 1995 1996 1997 1998 1999 2000
2.0
50

1.5
40 1980 1985 1990 1995 2000
CI (mg l–1)

30
20

20

15
10
O2 (mg l–1)

0
1992 1993 1994 1995 1996 1997 1998 1999 2000 10

8
5
7
NO3-N (mg l–1)

6
0
5
1980 1985 1990 1995 2000

3 30

2 25
1992 1993 1994 1995 1996 1997 1998 1999 2000
50
20
CI (mg l–1)

40
15
SO4 (mg l–1)

30 10

20 5

0
10

0
1980 1985 1990 1995 2000
1992 1993 1994 1995 1996 1997 1998 1999 2000

Figure 5.9 Temporal variations in hydrochemical Figure 5.10 Temporal variations in hydrochemical
parameters in the Litton Cheney spring [SY 55 90] (from parameters from the West Houghton Borehole [ST 82 04]
Edmunds et al., 2002). (from Edmunds et al., 2002).

detection limit in the deep groundwaters sampled around in Figure 5.18. The major element chemistry of the Chalk
Wareham. Some locally high concentrations of metals do groundwaters is controlled by the carbonate system and the
occur (e.g. As, Ni) but may simply be a consequence of not groundwaters in the outcrop areas are of Ca-HCO3 type.
filtering during sampling. Mixing with remnant formation water (probably within the
A summary of the dominant processes affecting chalk matrix) has led to an increase in salinity in the deepest,
groundwater quality in the Wessex Basin is illustrated on oldest waters. Ion exchange of Na for Ca in these waters has
the schematic cross-section through the Chalk of Dorset led to a trend intermediate between that of Na-HCO3 and

79
Table 5.4 Selected historical nitrate and chloride data 250

from BGS borehole archives.

200
Unconfined Confined

CI (mg l–1)
150
Site Date (year) NO3-N C1 (mg l–1)
(mg 1–1)
100
Corfe Mullen 1908 1.0 30
50
Durweston 1911 1.3 17
0
Upwey 1910 0.88 23 0 10 20 30 40 50

Sutton Poyntz 1913 0.73 19

12
Redox boundary
Alton Pancras 1946 1.5 – 10

DO (mg l–1)
8

6
Na-Cl types. The largest changes in water chemistry occur
where the aquifer becomes confined beneath the overlying 4

Palaeogene sediments. A redox boundary has been 2

established close to this boundary and beyond this zone the 0
groundwaters are reducing and contain low concentrations 0 10 20 30 40 50
of nitrate. The reducing nature of the system has, however, 14
led to an increase in Fe and Mn. There are limited data
12
on trace metals over much of the aquifer but the available
10
information indicates that concentrations are generally low,
NO3 (mg l–1)
a notable exception being F, which generally exceeds the 8

EU-MAC for drinking water in confined parts of the aquifer. 6

The large chemical differences between the unconfined and 4
confined groundwaters in the Wessex Basin are consistent 2
with differences in residence time and indicate that they
0
may be controlled by different flow systems. 0 10 20 30 40 50
0.35
0.30
5.8 HYDROCHEMISTRY AND SURFACE – 0.25
GROUNDWATER INTERACTIONS, BOURNE CASE
NH4 (mg l–1)

0.20
HISTORY 0.15
0.10
5.8.1 Introduction
0.05

A preliminary study of the hydrochemistry of the Bourne 0.00

catchment was undertaken by the BGS as a contribution to
0 10 20 30 40 50
the Environment Agency’s hydrogeological investigation.
50
The purpose of the hydrochemical study was to assess
spatial variations and provide some constraints on flow 40
pathways in the Bourne catchment.
SO4 (mg l–1)

Samples were collected both from the River Bourne 30

and from boreholes along most of the catchment area

20
(Figure 5.19) between April 1999 and February 2000. A
number of parameters were measured on site including 10
temperature, SEC, total alkalinity (as bicarbonate) by
titration, pH, Eh and dissolved oxygen (DO). Filtered 0
0 10 20 30 40 50
samples were then taken for subsequent laboratory analysis.
Distance (km)

5.8.2 Hydrochemical characteristics Figure 5.11 Major and minor element characteristics
The surface and groundwaters of the Bourne catchment are of Chalk groundwaters along a line of section between
all of Ca-HCO3 type and show remarkably little variation in Powerstock and Wareham (from Edmunds et al., 2002).
the relative proportions of major elements on a Piper dia-
gram (Figure 5.20). The Mg/Ca ratio increases with water–
rock interaction (and hence residence time) due to incon- The groundwaters of the Bourne catchment may be ex-
gruent dissolution and the low ratios in the Bourne waters pected to show variations in baseline chemistry as the wa-
indicates that the waters are relatively immature. The pH of ters evolve with increased residence time in the aquifer. A
the groundwaters is neutral to slightly alkaline (7.03 to 7.54) comparison with surface waters will also indicate the role of
and typical of those found in the Chalk. Dissolved oxygen groundwater discharge to the river. The groundwater chemi-
and Eh are both moderate to high and indicate that the waters cal data have been plotted against northing in Figures 5.21
are all oxidising. The groundwaters are moderately mineral- and 5.22 because this is the general direction of surface wa-
ised with SEC showing some variation: 552 to 718 µS cm–1 ter flow (north to south), however, the groundwater catch-
in the river and 417 to 640 µS cm–1 from groundwaters. ment and flow may be very different from that of the surface.

80
 2.5 5.8.3 Comparison between groundwaters and River
Bourne
2.0
The pH of the groundwaters shows no significant trend
1.5
along the Bourne valley (Figure 5.21). The higher pH values
Na/CI

1.0
in the river waters occur as a consequence of the degassing
of CO2 from this groundwater-dominated river. The pH of
0.5 the river rises between Collingbourne Kingston (northing
5500) and Tidworth (5000) as CO2 degasses with time,
0.0
0 10 20 30 40 50
leading to oversaturation with respect to calcite. However,
0.005
the river pH is lower downstream of the central part of
the catchment where the river often dries (i.e. between
0.004 northings 5000 and 4000), implying more recent discharge
from the groundwater. There is significant overlap in several
0.003
major (Ca, HCO3, Na) and minor (F, Sr, Ba) elements
Br/CI

0.002
between the groundwaters and the river (Figures 5.21 and
Sea water
5.22), which indicates that, at the time of sampling, the
0.001 rivers were dominated by groundwater inputs. In addition,
the same spatial trends present in the rivers are a good
0.000
0 10 20 30 40 50
indication that the river is dominated by local inputs from
0.7
the groundwater system. If the river water represented a
continuous accreting discharge, the concentrations would
0.6
be different from the local groundwater because they would
0.5 consist of a mixture of all inputs upstream of the sampling
point. The concentrations of some dissolved components
Mg/Ca

0.4

0.3 in the river do show significant differences to the local

0.2
groundwater: e.g. higher concentrations of Cl, SO4, P and
0.1
NO3 were present in the river (Figures 5.21 and 5.22) and
indicate some anthropogenic input to the River Bourne but
0.0
0 10 20 30 40 50
it is difficult to quantify volumes due to lack of knowledge
of the input chemistry. However, the concentrations of these
0.030
elements varied significantly at sites where sampling was
0.025 repeated (e.g. Newton Tony where NO3 varied between 15
0.020 and 41 mg l–1), in contrast to most major and trace elements,
probably indicating a shallow source of short residence time.
Sr/Ca

0.015

0.010
5.8.4 Regional variations and chemical evolution
0.005
In general, groundwaters normally show an evolution of
0.000
increasing total dissolved solids (TDS) along flowlines and
0 10 20 30 40 50
5000
with depth, related to increased water–rock interaction and
residence time. The chemical trends present in the Bourne
4000 catchment are anomalous in that the highest TDS waters
are present near the top of the catchment, with the freshest
F (ug l–1)

3000
waters close to the confluence with the River Avon. The
2000 following general observations may be made:

1000 • There is a general decrease in SEC, Cl, Ca, HCO3,

SO4, Si and Sr between Leckford Bridge in the north
0
0 10 20 30 40 50
and Winterbourne Gunner in the south. There are no
significant trends for elements such as K, Mg, F, NO3
Distance (km)
and Ba.
• The river waters and groundwaters show similar
Figure 5.12 Hydrochemical characteristics of Chalk
trends.
groundwaters along a line of section between Powerstock
and Wareham (after Edmunds et al., 2002). It is difficult to envisage a process whereby the groundwaters
in the south of the catchment can have directly evolved
from those in the north. Although the precipitation of
The chemistry of the river water and the Idmiston Spring mineral phases (e.g. calcite, which would lower Ca and
have also been plotted for comparison. Note that the Upavon HCO3) is feasible, there is no geochemical reason for this
boreholes are within the Avon surface water catchment. to occur. In addition, Sr (which is derived initially from
Several discharges are present within the Bourne catch- calcite) does not precipitate with calcite and would be
ment and are likely to affect both surface water and ground- expected to increase with progressive reaction and time.
water quality. However, it is not possible to quantify these The lower Sr southward cannot be due to differences in
because of the lack of data on discharge chemistry. Anthro- the host Chalk as Sr has been found to increase towards
pogenic inputs are clearly seen for the elements P, NO3, Cl the top of the Chalk7 (this may explain the slightly higher
and SO4. Sr/Ca south of Cholderton). Similar arguments can be

81
Figure 5.13
Regional plot for 1
60 N
electrical conductance SEC (µS cm−1)
in the Wessex Basin ! 343–462
Chalk.
contains OS data © Crown ! 462–504
copyright and database ! 504–520
rights 2017 Ordnance Survey
1
40

[100021290 EUL]. Use of ! 520–554

this data is subject to terms
and conditions.
! 554–614

! 614–1409
1
20

1
00

0
80

0 50 km

3
60 3
80 4
00 4
20 40
4
60
4

Palaeogene Lower Chalk Lower and Middle Chalk Middle and Upper Chalk Pre-Chalk

applied to Cl, which behaves as a conservative element, trends are closely linked to groundwater chemistry also
and other elements such as SO4. Therefore, it is not implies that discharge to the river system is dominated by
possible to produce a direct evolutionary sequence along local inputs of groundwater.
the postulated flowlines and the changes must be due to The chemical trends of decreasing concentration
different sources of groundwater. The source of higher SEC of many elements down the catchment shows that the
waters in the upper reaches of the catchment may well be a groundwaters in the lower reaches of the catchment cannot
consequence of contributions from the Upper Greensand. be directly derived from those of the upper catchment. The
This is consistent with hydrochemical data collected from hydrochemical trends from north to south may imply mixing
a borehole at Aughton [SU 23 57] where discharge was between two (or more) end members that have evolved
partly from the Upper Greensand (Figures 5.21 and 5.22). independently. The simplest explanation for this is that
higher SEC waters originating from the Upper Greensand
5.8.5 Summary are affecting the composition of the Chalk groundwater in
the upper part of the catchment, but that these groundwaters
The groundwaters of the Bourne catchment show significant do not generally flow to the lower part of the catchment,
changes in hydrochemistry along a north–south traverse, where groundwaters are fresher and of relatively local
these changes being mirrored to a large degree by the origin. This explanation would concur with the conceptual
water chemistry of the River Bourne. This implies that model of the Bourne summarised in Chapter 4, in which
the river waters, at the time of sampling, are dominated groundwaters in the upper parts of the catchment tend to
by groundwater inputs. This is consistent with physical diverge to the adjacent Avon and Test catchments at lower
measurements, which show that the unsaturated zone is close elevations. Such an interpretation is also supported by the
to zero in the valley areas and reaches depths of greater than fact that when the central section of the Bourne is dry there
60 m in the interfluves. The evidence that river chemistry are some differences between the River Bourne chemistry
upstream and downstream of the dry river bed. This implies
that after recharging the aquifer, the upstream river flow
7
For example Sr increases above the Chalk Rock at Banterwick Barn in does not re-emerge further downstream, but rather tends to
the Berkshire Downs (Murphy, 1998). join groundwater flowing to the Avon and Test catchments.

82
Figure 5.14 a)
a) Regional plot for Cl 1
60 N
in the Wessex Basin CI (mg l−1)
Chalk. b) Regional plot ! 8.56–14
for Na in the Wessex
Basin Chalk. ! 14.0–16.7
contains OS data © Crown ! 16.7–18.9
1
40
copyright and database
rights 2017 Ordnance Survey ! 18.9–22.0
[100021290 EUL]. Use of
this data is subject to terms
! 22.0–32.0

and conditions. ! 32.0–254.0

1
20

1
00

0
80

0 50 km

3
60 3
80 4
00 4
20 4
40 60
4

Palaeogene Lower Chalk Lower and Middle Chalk Middle and Upper Chalk Pre-Chalk

b)
1
60 N
Na (mg l−1)
! 5.03–7.0
! 7.0–9.0

1
40
! 9.0–11.0

! 11.0–13.3

! 13.3–18.1

! 18.1–155.0
1
20

1
00

0
80

0 50 km

3
60 3
80 4
00 4
20 4
40 60
4

Palaeogene Lower Chalk Lower and Middle Chalk Middle and Upper Chalk Pre-Chalk

83
Figure 5.15 a)
a) Regional plot for Ca 1
60 N
in the Wessex Basin Ca (mg l−1)
Chalk. b) Regional plot ! 37.0–86.6
for Mg in the Wessex
Basin Chalk. ! 86.6–98.0
contains OS data © Crown ! 98.0–104.0
1
40
copyright and database
rights 2017 Ordnance Survey ! 104.0–110.0
[100021290 EUL]. Use of
this data is subject to terms
! 110.0–117.0

and conditions. ! 117.0–265.0

1
20

1
00

0
80

0 50 km

3
60 3
80 4
00 4
20 4
40 60
4

Palaeogene Lower Chalk Lower and Middle Chalk Middle and Upper Chalk Pre-Chalk

b)
1
60 N
Mg (mg l−1)
! 0.76–1.59
! 1.59–1.90

1
40
! 1.90–2.30

! 2.30–2.73

! 2.73–3.90

! 3.90–21.50
1
20

1
00

0
80

0 50 km

3
60 3
80 4
00 4
20 4
40 4
60

Palaeogene Lower Chalk Lower and Middle Chalk Middle and Upper Chalk Pre-Chalk

84
Figure 5.16 a)
a) Regional plot for F 1
60 N
in the Wessex Basin F (mg l−1)
Chalk. b) Regional plot ! <0.05–0.07
for Sr in the Wessex
Basin Chalk. ! 0.07–0.079
contains OS data © Crown ! 0.079–0.09
1
40
copyright and database
rights 2017 Ordnance Survey ! 0.09–0.11
[100021290 EUL]. Use of
this data is subject to terms
! 0.11–0.20

and conditions. ! 0.20–4.30

1
20

1
00

0
80

0 50 km

3
60 80
3 4
00 4
20 4
40 60
4

Palaeogene Lower Chalk Lower and Middle Chalk Middle and Upper Chalk Pre-Chalk

b)
1
60 N
Sr (mg l−1)
! 142.37–178.19
! 178.19–190.00

1
40
! 190.00–236.24

! 236.24–286.51
! 286.51–528.18
! 528.18–3570.00
1
20

1
00

0
80

0 50 km

3
60 3
80 4
00 4
20 4
40 4
60

Palaeogene Lower Chalk Lower and Middle Chalk Middle and Upper Chalk Pre-Chalk

85
Figure 5.17
Regional plot for 1
60 N
NO3-N in the Wessex NO3 −N (mg l−1)
Basin Chalk. ! <0.2–3.07
contains OS data © Crown
copyright and database ! 3.07–4.70
rights 2017 Ordnance Survey
! 4.70–5.60
[100021290 EUL]. Use of 1
40
this data is subject to terms ! 5.60–6.75
and conditions.
! 6.75–8.10

! 8.10–18.37
1
20

1
00

0
80

0 50 km

60
3 3
80 00
4 4
20 4
40 4
60

Palaeogene Lower Chalk Lower and Middle Chalk Middle and Upper Chalk Pre-Chalk

Figure 5.18 North South

Possible
Conceptual diagram Rain water solutes Aquifer highly recharge
of the Chalk aquifer an important vulnerable to through
diffuse Marine aerosol
source of some solution influence decreasing
of Dorset highlighting elements pollution e.g. hollows inland from coast
agrochemicals/ Coastal
the main geochemical organic Major springs
processes controlling fertilisers spring
discharges
water quality (after
Edmunds et al., 2002).
Pa l a e o g
e n e c o n f i n in g b e d s
R E D OX
NDA R Y

B OU

Upper Greensand
OU
XB

in hydraulic
ND

DO CHA LK
RE
AR

continuity with
Y

the Chalk
Oxidising fresh- Fresh water to
Up p e r G r e e n s a n d total depth of
waters of good Reducing waters G aul t Clay a q ui t ar d
quality, low in iron Chalk at
but containing with some iron but Traces of CI coastline
nitrate absent. Some remaining in
some nitrate waters with high F pore spaces

86
Figure 5.19 Bourne " Borehole

hydrochemistry # River sampling site

sampling sites. ! Spring

Avo
n
1
60 tern
contains OS data © Crown Rivers Eas
We
copyright and database Surface water catchment boundary ste
rights 2017 Ordnance Survey
rn
Av
on ""
#Collingbourne
Kingston

[100021290 EUL]. Use of

Bagshot, Barton and Bracklesham groups
Thames Group
Upavon #Collingbourne
Ducis
this data is subject to terms Lambeth Group
#" Leckford Bridge

k
and conditions. "#

Broo
Portsdown Formation Tidworth "
"

e
Culver Chalk Formation

lern
Newhaven Chalk Formation
""

Chit

e
" Cholderton

urn
Seaford Chalk Formation

Bo
Till
Lewes Nodular, Seaford, Newhaven, Culver and 1
40 Rive #Newton Tony

River
Portsdown Chalk formations (undifferentiated) r Wy
lye
"
Lewes Nodular Chalk Formation #
! Idmiston
New Pit Chalk Formation Winterbourne
Holywell Nodular Chalk Formation "
" Gunner
Formations (Undifferentiated)
"
N
Holywell Nodular Chalk Formation
""
Zig Zag Chalk Formation Salisbury !
West Melbury and Zig Zag Chalk
e
formations (undifferentiated) River Ebbl

Nine Mile R.
West Melbury Marly Chalk Formation 0 10 km

von
Stockbridge Rock Member

R. A
4
00 4
20
Upper Greensand Formation

Figure 5.20 Piper 100

100
diagram showing 90
90
samples of the 80 Chalk groundwater
River Bourne and 80
River Bourne
groundwaters collected 70
70 Upper Greensand
in the study. 60
60
50
50
40
40
30
30
20
20
10
10
0
0
Mg SO4
0 0
100 100
10 10
90 90
20 20
80 80
30 30
70 70
40 40
60 60
50 50
50 50
60 60
40 40
70 70
30 30
80 80
20 20
90 90
10 10
10

0
10

0 0
0

100 90 80 70 60 50 40 30 20 10 0 0 10 20 30 40 50 60 70 80 90 100
Ca Na + K HCO 3 Cl

87
 River Bourne Boreholes Spring Aughton

9.0 900

800
8.5

700

SEC (µS l−1)

8.0
pH

600

7.5
500 Upavon
boreholes
7.0
400

6.5 300
2500 3000 3500 4000 4500 5000 5500 6000 2500 3000 3500 4000 4500 5000 5500 6000

Northing Northing

30 360

28 340
26 320
24
300
22
HCO3 (mg l−1) 280
CI (mg l−1)

20
260
18
240
16
220
14

12 200

10 180

8 160
2500 3000 3500 4000 4500 5000 5500 6000 2500 3000 3500 4000 4500 5000 5500 6000

Northing Northing

3 200

180

160
Mg (mg l−1)

Ca (mg l−1)

140
2
120

100

80

1 60
2500 3000 3500 4000 4500 5000 5500 6000 2500 3000 3500 4000 4500 5000 5500 6000

Northing Northing

Figure 5.21 Hydrochemical measurements and major element concentrations for surface waters and
groundwaters in the Bourne catchment.

88
 River Bourne Boreholes Spring Aughton

80 0.8

70

60 0.6

50
SO4 (mg l−1)

P (mg l−1)
40 0.4

30

20 0.2

10

0 0.0
2500 3000 3500 4000 4500 5000 5500 6000 2500 3000 3500 4000 4500 5000 5500 6000

Northing Northing

50 0.16

0.14
40

0.12
NO3 (mg l−1)

30
F (mg l−1)

0.10

20
0.08

10
0.06

0 0.04
2500 3000 3500 4000 4500 5000 5500 6000 2500 3000 3500 4000 4500 5000 5500 6000

Northing Northing

1.5 16
1.4
1.3 14
1.2
1.1 12
1.0
10
0.9
Sr (mg l−1)

Si (mg l−1)

0.8
8
0.7
0.6
6
0.5
0.4 4
0.3
0.2 2
0.1
0.0 0
2500 3000 3500 4000 4500 5000 5500 6000 2500 3000 3500 4000 4500 5000 5500 6000

Northing Northing

Figure 5.22 Selected major, minor and trace element concentrations for surface waters and groundwaters
in the Bourne catchment.

89
6 Groundwater management

6.1 INTRODUCTION and Edwards, 1926). The gradual increase in use of boreholes
has led to the decline in usage of natural spring sources.
Groundwater management is essential in order to maintain a Historically, water use in the Avon and Dorset area of the
balance between meeting water demands and safeguarding Wessex Basin was chiefly agricultural with light industry
the environment and the resource. Where groundwater is in the towns and cities. Water-meadow construction was a
readily available, it is normally less expensive to use than common practice in order to produce an early crop of grass.
surface water as it usually needs less treatment. However, The water was returned to the watercourse downstream of the
over-abstraction of groundwater can lead to problems such intake point. This type of industry has now virtually ceased
as reduced baseflow to rivers, with consequent impacts because of a change in agricultural practice to larger-scale crop
on ecology and amenity value; reduced well yields, and production. Similar types of agriculture also predominated
movement of saline water into the aquifer. In addition to in the Hampshire countryside in the Wessex Basin. Cress
managing abstraction, it is also vital to protect the quality cultivation was undertaken on the rivers close to the major
of groundwaters by, for example, preventing or limiting springs, dairy and arable farming on the upland areas of Chalk
activities that are likely to pollute the aquifer. and intensive horticulture on the heathlands, typically found
Groundwater management is therefore of great on Palaeogene cover (Hampshire River Authority, 1970).
importance in a region such as the Wessex Basin, where Several water companies were established in the
groundwater from the Chalk aquifer forms a vital (and nineteenth century that exploited groundwater resources
commonly the principal) source of water supply. In addition, in the Wessex Basin. For example in 1809, the Farlington
much of the water abstracted from surface sources has also Waterworks Co. was established and constructed a pumping
passed through the groundwater system (as shown by the station at Farlington Marshes. Also in this year, the Portsea
high baseflow indices of many of the rivers) and this further Island Water Company was started. The two companies
increases the importance of groundwater management. merged as the Portsmouth and Farlington Waterworks Co.
in 1840. In 1857, this company became the Borough of
Portsmouth Waterworks Co.
6.1.1 Historical perspective
Bournemouth Water (established 1863) and West Hamp-
In Hampshire, water supply has evolved from the use of shire Water (established 1893) were two former statutory
spring and surface water to a mixture of springs, boreholes/ water companies established by an Act of Parliament. The
wells and river takes (Whittaker, 1910). For example, in the companies merged in 1994 through Schemes of Arrangement
latter part of the 19th century, Southampton obtained its under the Water Industry Act 1991, to become Bournemouth
water from the River Itchen and Bournemouth took its water and West Hampshire Water Plc.
from the River Stour; nowadays groundwater abstractions Following the 1963 Water Resources Act (see Box 4) the
are important. In many parts of the Wessex Basin, Wessex Basin lay within the remit of the Avon and Dorset,
abstractions from the Chalk aquifer have grown to become and Hampshire river authorities. In 1970 there were nine
an essential component of water supply. For example, water statutory water undertakings in the Avon and Dorset River
in the Andover region is entirely sourced from groundwater. Authority area, together with the licence-exempt Ministry
The significant springs near Portsmouth at Havant [SU 71 of Defence supplies. Of these, the Dorset Water Board was
06] and Bedhampton [SU 70 06] were first used in 1741. As the largest and supplied the area including Poole in the
many as 25 individual springs can be used providing what south-east, Bridport in the west and Blandford Forum in the
is believed to be the largest spring water supply used for north. Other undertakings were the West Wiltshire Water
public consumption in the world, with a combined yield of Board, South Wiltshire Water Board, North Wiltshire Water
between 53 and 170 Ml d–1 depending on Chalk groundwater Board, West Hampshire Water Company, Bournemouth and
levels (http://www.portsmouthwater. co.uk/). District Water Company, East Devon Water Board, Wessex
In addition to rivers, surface water sources include the Water Board and the Lulworth Castle Estate (Avon and
Blashford Lakes, situated near Ringwood in Hampshire, Dorset River Authority, 1970).
which are a series of water-storage reservoirs created from The Hampshire River Authority in 1970 contained whole
worked-out gravel pits. The lakes store water for use during or part of a number of water undertakings: Southampton
the high-demand summer period and act as a backup to Undertaking (covering the largest area), Portsmouth Water
other sources in case of an emergency. They are filled from Company, West Hampshire Water Company, South Wiltshire
the nearby River Avon when river conditions are suitable, Water Board, Swindon Undertaking, Thames Valley Water
usually during the winter months. Board, Mid Wessex Water Company, Wey Valley Water Com-
In Dorset, spring sources and shallow wells were pany, Lymington Undertaking and Winchester Undertaking.
commonly used, the former especially on the more elevated At the present time, the Chalk aquifer of the Wessex Basin
areas of the Chalk catchments where springs form the (to the south of the Thames Region groundwater divide) lies
headwaters of the main rivers. Along certain rivers, it was within the areas covered by the Southern Water and Wessex
necessary to deepen boreholes because the river flows Water Service Companies. The main water-supply companies
decreased during the summer periods. For example, in parts operating in the Wessex Basin are: Bournemouth and West
of the Winterbourne, the water levels were noted to be up to Hampshire Water, Portsmouth Water and Cholderton and
30 m deep in summer but overflowing in winter (Whitaker District Water Company (Figure 6.1).

90
 the River Meon in Hampshire to the River Arun in West
BOX 4 MAIN WATER LEGISLATION Sussex. The average demand is around 180 Ml d–1, around
50 per cent of which is supplied from boreholes and wells
In 1945 the Water Act defined national water policy for in the Chalk, 35 per cent from the major springs at Havant
the first time, including some abstraction control (in and Bedhampton and the remainder from an abstraction on
areas recognised as being over-pumped or where this was the River Itchen (Portsmouth Water, 2006). Further west, the
considered likely to occur) and data collection. Test and Itchen catchments are mainly served by Southern
The 1963 Water Resources Act created 29 catchment- Water, with the area split into northern (Hampshire Andover)
based River Authorities to provide regional, integrated and southern (Hampshire South) water resource zones. The
management of water resources and introduced abstraction southern zone, extending from the Solent as far as Winchester,
licensing. For the Chalk of the Wessex Basin, the western serves a population of around 600 000 (260 000 customers)
section of the aquifer fell into the Avon and Dorset River
with an average daily demand of nearly 160 Ml. Eight
Authority area and the eastern part lay within the Hampshire
groundwater abstractions from the Chalk are used, with two
River Authority’s region.
The 1973 Water Act formed ten Regional Water surface water abstractions. Water is also transferred from this
Authorities to provide complete regional management of the water resource zone to the Isle of Wight via the cross-Solent
water cycle. These were given responsibility for, amongst main (up to 12 Ml d–1). To the north, the Hampshire Andover
other things, water supply, sewage disposal, prevention of water resource zone serves a much smaller population of
pollution and land drainage. For the Chalk aquifer of the 65 000 (27 500 customers). Average daily demand typically
Wessex Basin the western half of the aquifer lay within the averages 17 Ml and is supplied entirely from six groundwater
Wessex Water Authority’s region and the eastern side fell stations (Southern Water, 2006).
within that of the Southern Water Authority. The Isle of Wight water resource zone serves a popula-
The 1989 Water Act disbanded the Regional Water tion of 135 000 (63 000 customers), with an average daily
Authorities. Their utility functions were transferred to water demand of 37 Ml. Water is supplied from eight groundwater
service companies; for the Wessex Basin these comprised abstractions from the Chalk and Lower Greensand aquifers
Wessex Water and Southern Water. Under the Act, the (which are generally low yielding) and a surface-water ab-
National Rivers Authority was created as an independent straction. Supplementary water is supplied by the cross-So-
regulatory body, later (1995) to be subsumed into the lent main (Southern Water, 2006).
Environment Agency. Currently under the Water Resources Bournemouth and West Hampshire Water supplies a popu-
Act 1991, the Environment Act 1995, and The Water lation of around 420 000 (nearly 184 000 customers) with an
Environment (Water Framework Directive) (England and average demand of around 160 Ml d–1. Water is supplied from
Wales) Regulations 2003, the Environment Agency has a four groundwater sources (three of which are in the Chalk
duty to conserve, redistribute or otherwise augment water aquifer), three surface water sources and one joint ground-
resources and secure their proper use. The Wessex Basin
water/ surface water source; surface water provides most of
Chalk aquifer falls into the Environment Agency’s South
the supply (Bournemouth and West Hampshire Water, 2006).
West and Southern regions.
The main European legislation that intended to protect Within the total area covered by Wessex Water (which
groundwater from contamination was the Groundwater extends to the west beyond Taunton and north to include
Directive 80/68/EEC, until it was repealed by the Water Malmesbury), the company supplies on average 365 Ml d–1
Framework Directive in 2013. In December 2000 the of which 80 per cent is derived from groundwater sources.
European Water Framework Directive (2000/60/EC) came In the Wessex Basin, the area supplied by Wessex Water
into force. This is the most significant piece of European includes most of the upper Avon catchment and the Stour
legislation relating to water management for at least two and Frome/Piddle catchments. The Wessex Water area
decades. The Directive was a response to the fragmented is divided into four zones, of which two — the east and
nature of existing legislation relating to water and aims to south zones — broadly encompass the Chalk outcrop in the
provide a framework to integrate this and to expand the Wessex Basin. The south resource zone (mainly the Avon
scope of water protection to all waters, and aimed to achieve catchment) is supplied by groundwater sources and by a
‘good status’ (involving both quantitative and qualitative river intake and reservoirs near Ringwood. Water can be
aspects) by 2015. A significant feature of the Directive is the transferred from Poole across to Dorchester and Weymouth.
emphasis on ecological objectives, to be used as measures The east resource zone (mainly the Dorset part of the Wessex
of the success of water management strategies. Basin) comprises six discrete areas of demand with limited
Under the Water Framework Directive, the Groundwa- transfers between them. The sources in these zones are
ter Daughter Directive (2006/118/EC) became effective. exclusively groundwater, with typically only two sources in
This Directive addresses the discharge of polluting matter each of these subzones (www.wessexwater.co.uk).
to groundwater, particularly with regards to certain harmful
substances. These pollutants are classified under hazardous
and non-hazardous substances. Hazardous pollutants must
6.2 GROUNDWATER RESOURCE ISSUES AND
be prevented from entering groundwater, while discharge of
non-hazardous substances must be controlled to ensure that PRESSURES
pollution of groundwater is avoided. In England and Wales
the Water Framework Directive is implemented by Environ- There are a number of issues that are important to the
ment Agency through The Water Environment (Water Frame- management of groundwater resources in the Wessex
work Directive) (England and Wales) Regulations 2003. Basin. Quantitative issues are discussed below, while those
relating to quality are addressed in the subsequent section
on groundwater pollution and protection.
6.1.2 Current water supply in the Wessex Basin
6.2.1 Increasing demand
The eastern part of the basin is supplied mainly by
Portsmouth Water. The company supplies water to around The demand for water for public supply increased from the
290 000 domestic and commercial customers in an area from 1950s as industrial, agricultural and per capita demand rose,

91
 as the number of wetlands and wetland species of plants
Andover Basingstoke
and animals is diminishing globally. Important sites are
40
1 protected by the RAMSAR designation (RAMSAR sites are
wetlands of international importance designated under the
Salisbury
Winchester
RAMSAR Convention). There are a number of RAMSAR
sites in Wessex (Figure 6.2a) including the Arun and Avon
valleys.
Southampton
Other designations such as Sites of Special Scientific Interest
(SSSI) and Special Protection Areas (SPA, designated under
the EC Birds Directive) are relevant to groundwater. There
Portsmouth
are many groundwater-dependant wetlands that are classified
0
90 Newport
Dorchester
Bournemouth as SSSIs in Wessex, in recognition of their importance and
the fact that they support many different communities of
Weymouth
0 50 km plants and aquatic life forms, including the Frome, the Itchen,
the Moors and the River Avon (Figure 6.2b). The locations of
3
80 30
4 4
80
SPAs in the Wessex Basin are shown in Figure 6.2c.
Southern Water Cholderton and District Portsmouth Water An important element of environmental legislation is the
Wessex Water Bournemouth and 1992 Habitats Directive, whose main aim is to help to ensure
West Hampshire Water South East Water
biodiversity by requiring member states to take measures
Figure 6.1 Water supply map of the Wessex Basin to maintain or restore natural habitats and wild species at
(boundaries approximate). a favourable conservation status. The Directive introduced
contains OS data © Crown copyright and database rights 2017 Ordnance
significant protection for those species or habitats of European
Survey [100021290 EUL]. Use of this data is subject to terms and importance. Special Areas of Conservation (SAC) are specified
under the Directive and are shown in Figure 6.2d. The Directive
has been an important reason for groundwater investigations
and probably peaked in Wessex in the 1980s. In the Ports- in the Wessex Basin in recent years (for example, studies in
mouth Water area, for example, demand for public supply the Hampshire Avon and Itchen). In addition to providing
water has declined significantly since 1988 (Environment habitats for plants and wildlife, rivers support fisheries and
Agency, 1999c). This is thought to be largely due to both provide an important resource for recreational activities; there
a reduction in leakage and reduced demand. However, it is are therefore economic, ecological and amenity interests to
likely that demand for water over much of the Wessex Basin consider. The Environment Agency has the responsibility of
will increase in the future. This is likely to result partly from setting targets for river flows, but given the fact that the majority
an increase in population, which in Hampshire for example of the rivers across the basin have been significantly altered by
is forecast to rise by 11 per cent between 2001 and 2021 humans, setting targets raises many economic, technical and
(Hampshire Water Partnership, 2003). environmental issues, such as whether the aim is the protection
The additional water consumption from such a popula- of the current condition of the river or the improvement of the
tion increase is likely to be enhanced by the current trend river environment and its associated benefits.
of reducing numbers of occupants per household and the Judging the worth of environmental improvement
increasing use of household appliances such as power show- schemes is a complex problem. In order for cost-benefit
ers. As a result, the domestic daily use of water per person analyses of schemes to be performed, it is necessary to
in Hampshire is predicted to rise from the current value of understand how flows affect the parameters of interest and
160 litres to between 178 and 228 litres by 2025 (Hampshire how to measure the environmental quality of a river. One
Water Partnership, 2003). approach, for example, is a Lotic-invertebrate Index for Flow
The Isle of Wight has historically suffered water supply Evaluation (LIFE) technique, which links the population of
problems due to its limited surface water and groundwater invertebrates in a river (considered to be useful indicators of
sources and high summer population. During droughts and the ‘health’ of a river) to the river flow regime (Extence et
in particular in 1976, the island suffered unreliability in al., 1999). A strong correlation has been observed between
public water supplies. This led to a significant investment in the LIFE index and mean summer flows for Chalk streams.
mains reinforcement, leakage reduction and development of
new sources of supply, which included bringing water under 6.2.2.1 Habitats Directive
the Solent from the Testwood Reservoir in Hampshire. The The 1992 EC Habitats Directive requires the Environment
Environment Agency predicts that licensed water resources, Agency, as a ‘competent authority’, to maintain a
together with support from the cross-Solent main, are suf- ‘favourable conservation status’ of those habitats that are
ficient to meet the demand for water on the island for the afforded international protection, and which may be affected
foreseeable future (Environment Agency, 1999b). by Agency activities or authorisations. Special Areas of
Conservation (SAC) are specified under the Directive and
are shown in Figure 6.2d. Decisions must be based on a
6.2.2 Environmental issues
sound understanding of the key species, including the flow
Many rivers across the Wessex Basin are supported by Chalk requirements of aquatic plants, salmonids and other fauna,
groundwaters and these tend to have clear waters with a by working closely with water companies, other companies
stable temperature regime, characteristic plant communities and Natural England. Water companies also act as a
and a rich diversity of invertebrates. The UK Biodiversity competent authority and as such are required to identify the
Action Plan (www.ukbap.org.uk) lists Chalk streams as a impacts of their comments and authorisations. An example
‘high priority’ habitat. is the Test and Itchen (Environment Agency, 1999a), where
In addition to providing baseflow to rivers, groundwater the Agency must contribute to maintaining the favourable
feeds certain wetlands that are important habitats for conservation status of the River Itchen SAC. This applies to
wildlife. Protection of wetlands is of particular importance, all responsibilities as an operator, regulator and influencer,

92
a) b) 1
60
1
60
Basingstoke Basingstoke

Andover N Andover N

1
40
1
40

Salisbury Salisbury Winchester

Winchester
1
20
1
20
Southampton Southampton

1
00
1
00
Portsmouth Portsmouth
Newport Bournemouth
Dorchester Bournemouth Dorchester Newport
0
80
0
80
Weymouth 0 50 km Weymouth 0 50 km
3
60 3
80 4
00 4
20 4
40 4
60 4
80 3
60 3
80 4
00 4
20 4
40 4
60 4
80

RAMSAR sites (Natural England data) Rivers SSSIs (Natural England data) Rivers

c) 1
60
d) 1
60
Basingstoke Basingstoke
Andover N Andover N

1
40 1
40

Salisbury Winchester Salisbury Winchester

1
20 1
20
Southampton Southampton

1
00 1
00
Portsmouth Portsmouth
Bournemouth Newport Bournemouth Newport
Dorchester Dorchester
0
80 0
80
Weymouth 0 50km Weymouth 0 50km
3
60 3
80 4
00 4
20 4
40 4
60 4
80 3
60 3
80 4
00 4
20 4
40 4
60 4
80

SPAs (Natural England data) Rivers SACs (Natural England data) Rivers

Figure 6.2 Examples of protected sites in Wessex. a) RAMSAR designated sites. b) Sites of special scientific interest
(SSSIs). c) Special protection areas (SPAs). d) Special areas of conservation (SACs).
contains OS data © Crown copyright and database rights 2017 Ordnance Survey [100021290 EUL]. Use of this data is subject to terms and conditions.

and the Agency is required to review all existing consents lower streamflows and higher demand in summer. In addition
and authorisations that may have a significant impact upon it is likely that for an aquifer with a low specific yield and
the site. The Agency must also properly assess the likely rapid response, such as the Chalk, a small reduction in the
effect of new applications and plans. length of the recharge season could significantly reduce the
groundwater stored above the outflow base level.
6.2.3 Climate change
6.2.4 Drought
While the precise nature of climate change in southern
England is uncertain, there is growing consensus that A drought may cause both environmental effects and
winters will be wetter, summers will be drier and that annual deficiency in water supplies and the balance between supply
temperatures will increase. The likelihood of extreme events and demand may be affected in different ways by different
such as droughts and storms will also increase. Predictions droughts. For example, in the Southern Water area in 1976,
from the UK Climate Impacts Programme suggest that low groundwater levels and river flows were experienced
if high fossil-fuel use continues, then by 2080 summer because of the preceding very dry winter and this, coupled
precipitation in the south-east of England could reduce by with high summer demand, resulted in a shortage of
up to 60 per cent with winter precipitation increasing by up resources (Southern Water, 2006). In the same region during
to 30 per cent. The net effects of these changes on total water 1989 to 1992, a succession of dry winters and somewhat
resources are not known — for example the extent to which elevated summer demands put groundwater sources under
enhanced infiltration caused by increased winter rainfall stress, whereas in 1995, while groundwater levels were high
will be offset by higher evapotranspiration caused by higher after a wet winter, unprecedented summer demand caused
temperatures and the higher soil moisture deficit resulting problems. Between 2004 and 2006, prolonged dry periods
from lower summer rainfall — but it is likely that seasonal and below-average winter rainfall resulted in both reduced
stresses on water resources will increase, for example by river flows and low groundwater levels.

93
 Groundwater and surface water sources tend to react cycle. Thus recent legislation, such as the Water Framework
differently to winter drought. While the lack of recharge Directive, requires that the condition of the resource is
resulting from low winter rainfall will cause groundwater assessed by taking into account recharge, abstraction and the
sources to suffer, the large storage capacity of aquifers effects of groundwater discharge on surface-water ecologies
gives sources some resilience against a single winter with and dependent terrestrial ecosystems.
low rainfall, although two or more dry winters will result
in reduced yield. Surface sources, however, can be severely 6.3.1 Licensing and sustainable development
affected by one dry winter, where this results in significantly
reduced baseflow in the following year. Summer drought While abstractors must be protected from drought, it is
tends to affect both groundwater and surface water sources, also important to minimise the impact of abstraction on the
as demand is increased. environment. The term ‘sustainable development’ has been
used to mean a balanced approach to abstraction, allowing
groundwater to be exploited for purposes such as public
6.2.5 Groundwater flooding
water supply, industrial and agricultural uses, while avoiding
Groundwater flooding is a common natural phenomenon unacceptable derogation of surface water sources, groundwater
that occurs when the natural storage capacity of an aquifer quality and the environment, so that future generations continue
is exceeded. It can arise in two main ways as discussed in to have access to these resources. Abstractions of greater
Chapter 3. The first is associated with river floodplain that 20 m3 d–1 are subject to licensing by the Environment
deposits and results in water flowing onto the floodplains Agency and this forms an important mechanism for managing
when the flow in a watercourse exceeds its capacity and river groundwater resources in the Wessex Basin.
alluvium becomes fully saturated. These natural floodplains
provide extra capacity for storage and allow floodwater 6.3.2 Sustainable abstraction — river augmentation
to pass downstream; they are also important for species and alleviation of low flows
dependent on seasonal flooding. The combination of building
in the floodplain and an expansion of impermeable surfaces 6.3.2.1 River regulation
has increased the flood risk from this type of flooding in some Following the Water Resources Act of 1963 the newly
areas. created Water Resources Board, tasked with the planning
The second type of groundwater flooding occurs when the of water resource development on a national scale,
water level in consolidated, unconfined, rock aquifers rises instigated several major regional studies of England and
to emerge at the surface as springs and seepages. The low Wales. A number of subsidiary studies, including specific
specific yield of the Chalk means that if the transmissivity groundwater resource studies, were also undertaken as part
of the aquifer and imposed head gradient are insufficient to of the programme. The most important of the groundwater
cope with increased recharge, then water levels may rise resource studies were schemes in the Thames, Great Ouse,
substantially. In the winter of 2000/2001 over one hundred Severn, Waveney and Itchen valleys. The purpose of these
villages in Hampshire, in addition to roads and fields, were schemes was to assess the feasibility of regulating rivers
flooded by groundwater after a long period of heavy rain by pumping groundwater into them for abstraction in
caused groundwater levels to rise substantially (Hampshire their lower reaches while maintaining river flows for other
Water Partnership, 2003). The north-west of the county environmental requirements (Downing, 1993).
suffered again in the winter of 2002–2003. In 2012, between The Candover scheme was the initial phase of a proposal
April and September, unusually high rainfall affected the area to regulate the Itchen. The purpose of the scheme was to
of the Wessex Basin leading to groundwater levels close to or use seasonal groundwater abstraction to augment low river
above long-term maximum values in most areas of Dorset, flows and allow direct river abstraction for public water
south Wiltshire and Hampshire, and even groundwater supply near the tidal limit of the river. It was considered that
flooding in west Dorset and Hampshire. These areas were Hampshire was ideally suited for such a scheme because
again affected by groundwater floods during the winter of the northern part of the county is underlain by Chalk and is
2012/2013 due to continuous rainy conditions, low soil- drained by rivers flowing south towards population centres.
moisture deficit values and exceptionally high groundwater The scheme involved the construction of six abstraction
levels, which followed from the unexpected recharge of the boreholes (in three pairs) sited away from the perennial
previous wet summer. In January 2014, groundwater floods head of the Candover stream, but linked to it by a pipeline.
affected areas susceptible to groundwater flooding, where After six months, pumping during the dry summer of 1976,
monitored groundwater levels showed maximum recorded the cumulative net gain was an encouraging 80 per cent
values for January in several boreholes on the Chalk outcrop. of the total pumped volume of 5.1 x 106 m3, representing
This flooding event was a result of persistent rainfall that 60 mm of rainfall over the 65 km2 catchment (Headworth et
continued from December 2013 for several weeks, showing al.,1982). The second stage of the scheme, in the Alre valley,
the heaviest rainfall ever recorded since 1766 for England gave a lower net gain because of a low storage coefficient of
and Wales. Such flooding can carry on for long periods until the aquifer (Downing, 1993).
groundwater levels fall sufficiently, and can be a particular
danger to public health where the water is polluted from 6.3.2.2 Low flows in rivers
overflowing cesspits and septic tanks. Across the Wessex region, several rivers have been identified
as requiring measures to be implemented to alleviate low
flows. Examples in Hampshire include the Rivers Hamble,
6.3 GROUNDWATER MANAGEMENT Meon and Itchen, and the Candover stream. Modelling of
STRATEGIES these catchments has shown that abstraction in this area is
aggravating the problem of low flows.
Groundwater resource management has, in recent years, Wessex Water’s abstractions for public water supply were
increasingly moved towards recognition of the need to found to be causing unacceptably low river flows in the
manage groundwater in the context of the whole water rivers Wylye, Piddle and Upper Bristol Avon, which affects

94
the ecology, fisheries and amenity value of these rivers. It developed into local Environment Agency plans (LEAPs)
also has detrimental effects on the chalk river species for that fulfilled the same role, by highlighting specific issues
which the Wylye is designated an SSSI and a candidate for particular catchment systems through consultation
SAC. To resolve these issues, the Environment Agency documents and summarised in action plans.
proposed a staged approach with the result that abstraction The LEAP reports describe the main environmental
from the Chitterne and Cowbridge boreholes stopped and issues in the plan areas, and the state of the environment at
abstractions were reduced from the Alton Pancras Borehole, the time of publication. The Wessex Chalk aquifer system
to restore flows in the Wylye, Upper Bristol Avon and Piddle was covered by several catchment-based LEAP reports,
respectively. which set out the work that the Environment Agency and
The River Allen has been identified as a low-flow river others planned to undertake to protect and enhance the
and is one of the top 20 low-flow sites in England and Wales environment in the plan area over the following five years.
(Environment Agency, 1997). Investigations conducted by The LEAP reports also set out water resources management
the Environment Agency concluded that the principal reason actions for the future, such as demand management.
for derogation of the flow of the River Allen was from the Two main issues were highlighted in the Wessex LEAP
Bournemouth and West Hampshire Water abstraction borehole reports that are relevant to groundwater: low summer flows
at Stanbridge. In 1993, after a lengthy investigation, the in rivers fed by baseflow and groundwater quality.
National Rivers Authority (NRA) proposed an action plan that
identified the need for a reduction for the licence at Stanbridge, 6.3.4 Catchment abstraction management strategies
in association with the setting of revised flow targets for the
management of stream-flow support from existing boreholes. Catchment abstraction management strategies (CAMS) are
The River Tarrant has historically experienced parts of a more recent approach to managing water resources in
the river drying up during the summer months. In 1995, the England and Wales, and are developed on an area-by-area
river dried along all of its length and in 1996, it dried up basis (Environment Agency, 2002b). The CAMS areas have
downstream of Tarrant Keyneston. Data gathered by local been based mainly on surface-water catchments.
concern groups indicates that in nine of the 22 years from The CAMS operate on a six-year review cycle, during
1974 to 1996, the river dried up below Tarrant Monkton, which time the Environment Agency undertakes a detailed
and also that the lower reaches have dried up. In 1995, a assessment of the catchment, including total available
major fish rescue operation was mounted as the river resource, environmental requirements, licensed quantities
progressively dried out both from the source and from its and actual abstracted quantities. The balance between the
confluence with the Stour, leaving little flow in the vicinity committed and available resources determines the ‘resource
of Tarrant Rushton (Environment Agency, 1997). There has availability status’ for each water-resource management
been concern about the effects of Wessex Water abstractions unit within the area i.e. whether further abstraction licences
in the Tarrant catchment and the effects that these may can be granted without derogating the environment or other
have had on the river flow. These abstractions are located users (Table 6.1). The following ‘sustainability appraisal’
at Stubhampton, at the head of the river, and at Shapwick process considers what the resource availability status for
in the Stour valley. Investigations have concluded that the each unit should be at the end of the six-year cycle. For
Stubhampton source has little influence on the natural
occurrence of events, but there may be a connection with
the Shapwick source (Environment Agency, 1997). Table 6.1 Definition of the CAMS ‘resource availability
On the Isle of Wight, the Chalk aquifer is heavily status’ classifications.
exploited for public water supply. This has an impact on
the flows of Chalk streams, particularly in summer months,
and results in minimum flow conditions, which restrict Indicative resource Definition (relating to the
abstractions from Bowcombe, Chillerton and Calbourne availability status availability of water for
pumping stations when flows in the adjacent rivers fall abstraction licences)
below a certain level. The Medina–Yar Transfer Scheme Water available Water likely to be available at
and the Lower Greensand Scheme are used to augment river all flows including low flows.
flows for subsequent abstraction at Sandown. Restrictions may apply.

6.3.3 Catchment management plans and LEAPs No water available No water available for further
licensing at low flows although
The NRA produced catchment management plans (CMPs) water may be available at higher
for individual river catchments in the 1990s. The purpose flows with appropriate restrictions.
of CMPs was to provide a focus for the development and
implementation of NRA policies and a decision framework Overlicensed Current actual abstraction is
with guidelines on the sustainable use of catchments. resulting in no water available
The plans described the physical nature of each of the at low flows. If existing licenses
catchments, together with catchment statistics and details were used to their full allocation,
of the catchment usage. They then highlighted key issues they would have the potential to
and management proposals to be addressed by the NRA. cause unacceptable environmental
Finally, the management plans incorporated an ‘action plan’ impact at low flows.
with suggested timings for completion together with a broad
indication of the costs involved in implementing the actions. Overabstracted Existing abstraction is causing
Several of these reports were produced for river catchments unacceptable environmental
across the Wessex Basin. impact at low flows. Water may
With the formation of the Environment Agency in 1995 still be available at high flows
and its inception in 1996, the work of the NRA’s CMPs was with appropriate restrictions.

95
example, in a catchment that is ‘overabstracted’, the Agency deterioration of the status of all bodies of groundwater
may attempt to recover some licences. • protect, enhance and restore all bodies of groundwater
An important aspect of the CAMS process is that it is and ensure a balance between abstraction and recharge
designed to enable interested parties such as abstractors and of groundwater, with the aim of achieving good
environmental organisations to become involved in managing groundwater status 15 years at the latest after the date of
the water resources of a catchment. CAMS documents are entry into force of the Directive
also be open to the public, providing more open access to • implement necessary measures to reverse any significant
information than was available in the past. and sustained upward trend in the concentration of any
There are 129 CAMS areas covering England and Wales. pollutant resulting from the impact of human activity in
For the Wessex Basin Chalk aquifer south of the Thames order to progressively reduce pollution of groundwater.
Region catchment divide, the relevant CAMS areas are
shown in Figure 6.3 and Table 6.2. For groundwater, the term ‘status’ in the Directive is a
Within the CAMS framework, new licences are of finite measure of the condition of the groundwater body and
duration. Existing licences will gradually be brought under has both quantitative and qualitative components. Good
this new system and converted to time-limited status. ‘quantitative status’ for a groundwater body broadly means
that the rate of groundwater abstraction is not sufficient to
adversely affect the environmental (essentially ecological
6.3.5 The Water Framework Directive and chemical) objectives of associated surface waters or
An important feature of the Water Framework Directive to damage terrestrial ecosystems or cause saline intrusion.
is that the management of waters will be based upon Good ‘chemical status’ broadly means that pollutants in the
the concept of integrated river basin management. All groundwater body do not exceed specified standards and
waters will be managed within ‘river basin districts’, with also are not sufficient to adversely affect the environmental
groundwater assigned to ‘groundwater bodies’ within these. (as above) objectives of associated surface waters or to
Groundwaters in the Chalk aquifer of the Wessex Basin fall damage terrestrial ecosystems.
within one of two river basin districts: a western district The river-basin management plan to achieve the
including the Avon and catchments to the west, and an environmental objectives must be based on an analysis of
eastern district including the Test and catchments to the east. the pressures (both quantitative and qualitative) on the water
Central to the Directive is the requirement to produce bodies within the river basin and an assessment of their
a strategic management plan for each river basin district likely impact. For water bodies at risk of failing to meet the
setting out how the Directive’s environmental objectives environmental objectives, a more detailed analysis is required.
are to be achieved. Environmental objectives are specified This allows a comprehensive programme of measures to be
for surface waters, groundwaters and protected areas. For drawn up, tailored to the specific circumstances in each river
groundwaters these are essentially to: basin district, and in particular to target those water bodies
most at risk of failing to meet their environmental objectives.
• implement necessary measures to prevent or limit the The initial assessment of whether a groundwater body is at
input of pollutants into groundwater and to prevent the risk of failing to meet its environmental objectives (‘initial

Figure 6.3
CAMS boundaries in 1
60 N
Wessex. Basingstoke
contains OS data © Crown
copyright and database Andover
rights 2017 Ordnance Survey
[100021290 EUL]. Use of this
data is subject to terms and 1
40
conditions.

Salisbury Winchester

1
20

Southampton

1
00
Portsmouth

Dorchester Bournemouth Newport

0
80
Weymouth
0 50 km

3
60 3
80 4
00 4
20 4
40 4
60 80
4

CAMS (Environment Agency) Rivers

96
Table 6.2 CAMS areas for the Wessex Basin Chalk. groundwater bodies in the Chalk of the Wessex Basin
from the Avon eastwards, including the Isle of Wight, are
considered to be at risk, principally as a result of diffuse-
CAMS area number CAMS area name pollution pressures. Those from the Stour to the Dorset
65 East Hampshire coast were classified as probably at risk.
The subsequent river basin management plans covering
66 Test and Itchen
the Wessex Basin (Environment Agency 2009a, 2009b)
67 New Forest showed overall groundwater status (combined chemical and
68 Isle of Wight quantiative status) to be classified as poor in essentially all of
the unconfined Chalk groundwater bodies across the basin.
69 Hampshire Avon
70 Dorset Stour 6.3.6 Aquifer storage recovery
71 Frome, Piddle, Poole Harbour The principle of aquifer storage recovery (ASR) is to use
and Purbeck aquifers, including those containing poor-quality water,
72 West Dorset streams to store good-quality water during periods when there is
excess and then to recover the water when it is required.
The technique works by injecting the water into the aquifer
characterisation’) was carried out by December 2004. via boreholes, creating a volume of clean water around
Groundwater monitoring programmes were established and the injection borehole from which the native water has
river basin management plans, including programmes of been displaced. Later, when the water is required, it is
measures designed to enable the environmental objectives to pumped from the aquifer, using the injection borehole as an
be achieved, were finalised in late 2009. By December 2015 abstraction source. Given that some mixing between injected
good status must have been achieved for all groundwater water and native groundwater will occur, it is not possible
bodies, except where derogations have been agreed. to recover the same volume of water as that injected before
water quality deteriorates; however, the ratio of useable
6.3.5.1 Results of initial characterisation abstracted water to injected water can be high, particularly
The results of initial characterisation for groundwater bodies after several injection–abstraction cycles, and a number
in the Chalk of the Wessex Basin were reported in the Article of successful schemes have been undertaken in the USA.
5 summary characterisation reports covering the south-west Given that ASR can provide storage in parts of aquifers that
and south-east river basin districts (Defra 2005a, 2005b). In are otherwise unused, it potentially offers a useful way of
terms of the Chalk aquifer, the groundwater bodies identified enhancing water resources, provided that acceptable levels
during the characterisation study were in general delineated of environmental impact and water quality can be achieved.
by the surface outcrop of the Chalk and then by CAMS A major ASR trial was undertaken in the Wessex Basin in
boundaries. Each of the resulting groundwater bodies was the confined Chalk aquifer at Lytchett Minster, near Poole
assessed as to whether it was at risk of failing to meet its in Dorset, in the late 1990s by Wessex Water in conjunction
environmental objectives. with CH2M Hill and the BGS. An existing production
The results indicated that, in terms of abstraction borehole was modified to form an ASR borehole and several
pressures, much of the groundwater in the Chalk outcrop fell initial cycle tests and full-scale tests were undertaken. After
into the category of being ‘probably at risk’of failing to meet analysis of the results of the tests, two main conclusions
good status by 2015. Only one Chalk groundwater body were reached (Eastwood and Stanfield, 2001).
was classified as ‘probably not at risk’, and no groundwater
bodies fell into the ‘not at risk’ category. To the north-east 1. Hydraulically, the injection and recovery cycles caused
of Portsmouth, the groundwater bodies extending to the very little environmental impact. Injection of water
east, away from the Wessex Basin, towards Brighton, were into the aquifer tended to increase the net volume of
assessed as being ‘at risk’ as a result of abstraction pressures. storage rather than displacing water at the margins of
Pressures from diffuse pollution8 were considered to the Palaeogene, and recovery of water from the aquifer
render almost all of the eastern side of the basin as ‘at risk’ tended to utilise aquifer storage. Thus an ASR scheme
as well as the Chalk on the Isle of Wight. To the west this would be likely to have little impact on rivers either on
designation includes the Avon catchment; beyond this, the the Palaeogene deposits or at the margins where the
bodies in the south-west of the basin (i.e the Dorset Stour Chalk is unconfined.
catchment and beyond) were classified as ‘probably at risk’. 2. The quality of abstracted water, however, was poor,
No chalk groundwater bodies in the basin were classified compared with required standards, as a result of high
as ‘not at risk’ from diffuse pollution pressures. Point- fluoride levels. The groundwater in the Chalk in the
source pollution pressures are not considered sufficient to Lytchett Minster area has a fluoride concentration of
render any of the groundwater bodies at risk and all were 4 mg l–1 while the current drinking water standard is
designated as probably not at risk. 1.5 mg l–1, and a maximum safe threshold of 0.7 mg l–1
In summary therefore, given the range of pressures had been proposed for seasonal ASR operation. During
considered during the characterisation process, all of the initial cycle testing the fluoride concentration in recovered
water tended to stabilise at around 2 mg l–1 and did not
improve during the main full-scale cycle tests. The results
of water quality modelling suggested that most (80 per
8
The diffuse-pollution pressures considered by the characterisation cent) of the fluoride response was caused by mixing with
process were nutrients, pesticides and sheep dip, sediment, urban land native groundwater rather than by water–rock interaction.
use, acidification and mines and minewaters. Although not specified in the
Defra characterisation reports, it is assumed that nutrients and pesticides
and sheep dip would provide the causes of pollution pressure in the Thus, while the ASR trial indicated that, as a resource
Wessex Basin Chalk aquifer. management technique, ASR in the confined part of the

97
Wessex Basin would have little detrimental environmental Wales. Any sources considered to be at a significant risk of
effect, the recovered water would not be of sufficient quality Cryptosporidium contamination have a more frequent raw-
to be used for public supply, although the water quality water monitoring regime for the Cryptosporidium oocysts
would be significantly better than that of the local aquifer. in accordance with the regulations. Sophisticated treatment
plants that are capable of removing the infective oocysts
may then be required.
6.4 GROUNDWATER POLLUTION AND
6.4.1.2 Urban areas
PROTECTION
Most urban areas tend to be associated with a wide range of
6.4.1 Groundwater pollution human activities that are potentially polluting to groundwater.
Sources of pollution can include leaking sewers (where
One of the main advantages of groundwater for water supply bacterial contamination, e.g. Escherichia. coli, is of concern
is its generally favourable and consistent quality, which in the highly vulnerable Chalk aquifer), run-off from roads,
reduces the need for water treatment. However, urbanisation, sanitary waste disposal, industrial activities and weed
industrial growth and widespread, intensive agricultural control. Even essentially residential districts may contain
activity can have a serious impact on groundwater quality. dispersed small-scale service industries.
Groundwater is particularly vulnerable to long-term
pollution as water movement through the Chalk can be Solid waste disposal
slow and therefore current groundwater quality may reflect Solid waste disposal can be an important source of the
pollution events from many years or even decades ago. subsurface contaminant load. Under EC water quality
6.4.1.1 Karstic chalk and pollution directives on groundwater protection and landfill, there
has been a move away from ‘dilute and disperse’ sites,
The Chalk can possess high transmissivities and flow is which rely on natural processes to attenuate pollutants
predominantly via fractures, so there is the potential for to acceptable levels, towards artificial containment with
rapid flow between sources of contamination and abstraction natural or artificial barriers. Although modern landfills
points, particularly where the Chalk has a karstic nature. are engineered to minimise the risk of leachate reaching
For example in Hampshire, the springs at Bedhampton and the water table, many existing facilities remain a potential
Havant are heavily used for public water supply and represent source of pollution, particularly where sites are located in
a valuable and irreplaceable public asset. However, swallow areas with shallow water tables. The greatest risk occurs
holes in the Chalk at Lovedean, Cowplain, Horndean and where disposal sites are located directly on Chalk, whereas
Rowlands Castle are in direct connection with the springs Palaeogene or recent cover may reduce the opportunity for
and need to be protected against contamination. Flow leachate to access the water table. The development of new
measurements using tracers (Atkinson and Smith, 1974) landfills is now steered away from the most vulnerable areas.
have shown that the speed of groundwater movement to the In the Chalk the carbonate rocks buffer the leachate
springs is very fast and capable of travelling more than 2 km and thus reduce heavy-metal mobility. The high buffering
in 24 hrs. capacity of the Chalk has also been found to be conducive
Where conditions on the Chalk are suitable, weathering to microbial metabolism (Blakey and Towler, 1988), which
occurs from the surface down to the water table. This forms may further attenuate leachate components.
solution features that allow surface water to migrate rapidly
both vertically and horizontally into the aquifer. Some work Road run-off
has been done on identifying areas vulnerable to pollution due The quality of first-flush urban storm run-off may contain a
to these solution features around Otterbourne and Tangley, but wide range of compounds such as hydrocarbon derivatives
further work is required to identify all the areas at risk. from exhaust gases and oil spills, heavy metals from engines
In the clay-with-flint domains, the high clay-mineral con- and sulphate from tyre wear (Christensen et al., 1978).
tent (usually between 50 and 70 per cent) and iron-oxide Road drainage is therefore potentially a pollution risk
content has the ability to attenuate contaminants by sorbtion wherever major roads cross the catchments of public supply
and cation-exchange reactions. However, solution features sources. The risk is minimised by trying to route roads
beneath, and sandy horizons within, the deposits focus sur- outside the public supply protection zones. Where this is not
face run-off adjacent to clay-with-flints deposits, and may possible, full drainage for new roads is typically provided
actively promote recharge to the groundwater system. Su- rather than soakaways. Elsewhere, soakaways are kept as
perficial coverings of clay-with-flints are found over much shallow as possible, to make use of the maximum thickness
of the south Dorset area, covering about 10 per cent of the of the unsaturated zone, and are provided with interceptors
Chalk at the surface. There is a tendency both for lower angle to prevent the ingress of petroleum compounds.
slopes (commonly the Chalk dip slopes) and, to the north of
the Frome valley, for south-facing slopes to be preferentially Non-agricultural pesticides
covered with these deposits. Larger areas of clay-with-flints Weeds are widely controlled by the application of a variety
occur on the interfluve areas (Klinck et al., 1998). of herbicides, the majority of which are applied to paved
One implication of rapid groundwater flow is that travel areas, industrial sites and railways, and may be integrated
times between sources of pathogens and abstraction points within rapid run-off events. The possibility of infiltration to
may not be long enough for all pathogens to be removed. groundwater is greatly enhanced where surface drainage is
Abstracted water is therefore routinely disinfected at all public directly to soakaways, bypassing the biologically active soil
water supply sources to remove any potential pathogens. zone.
Some pathogens (notably Cryptosporidium parvum)
are resistant to the normal methods of disinfection and Industrial impact
legislation (the Water Supply (Water Quality) (Amendment) Inorganic pollution occurs in many of the major urban areas
Regulations 1999) requires a risk assessment to be carried of the UK where groundwater is unconfined. The dominant
out for each public water supply source in England and contaminants are chloride and nitrate. Many organic

98
compounds have been identified in UK urban groundwaters. solution will vary with affinity for organic matter and/
Subsurface transport and attenuation of these chemicals is more or clay minerals. Pesticides that are strongly sorbed onto
complex than for inorganic compounds and many are soluble organic matter or clay particles are likely to remain in the
and stable in groundwater. Of main concern are industrial soil zone rather than leaching to groundwater, with the
solvents and petroleum hydrocarbons that have limited possible exception of transport through fractures. Pesticide
aqueous solubility and may accumulate in the subsurface as compounds may also degrade in the soil zone to produce
a separate phase, acting as a long-term subsurface source. (ultimately) simple compounds such as ammonia and carbon
Chlorinated solvents are a particular problem: they are widely dioxide. Soil half-lives for commonly used pesticides range
used and their low viscosity and solubility coupled with their from 10 days to years, although the most mobile pesticides
high relative density means that rapid and deep penetration of are normally less than 100 days. It is unlikely that matrix
an aquifer can occur. transport rates in the unsaturated zone exceed 1 m a–1 and
Heavy industry is largely absent from the Wessex Basin only the most persistent pesticides are able to reach the water
region (apart from petrochemical works and other waterside table. However, under certain conditions, rapid bypass flow
plants). Point-source pollution incidents are now commonly may occur. This could permit low to moderately persistent
associated with the storage of industrial chemicals, heating components to reach the water table relatively rapidly.
oil and with the construction industry. Spillage or illegal
disposal of some toxic substances, such as industrial solvents, Nitrate
can cause serious pollution of aquifers and may result in the Dissolved nitrate in groundwater is of concern because
closure of groundwater abstractions — as has happened in of potential human-health issues caused by the uptake of
Hampshire (Hampshire Water Partnership, 2003). Isolated nitrogen in the blood after the ingestion of water containing
historical problems also exist, as for example in the Andover high concentrations of nitrate. These concerns have
area due to the effects of historical solvent contamination resulted in a nitrate standard of 50 mg l–1 NO3 (equivalent
(Hampshire County Council, 2000). to 11.3 mg l–1 N) being set in the UK by the Drinking
Water Inspectorate. The source of anthropogenic nitrate in
6.4.1.3 Agriculture groundwater is principally artificial fertilisers and animal
The quality of groundwater in the Wessex Chalk aquifer manure; it is also produced by the dissolution of nitrous
system is generally good, although the intensification oxide released from vehicles and industry. Dissolved nitrate
of agriculture in the second half of the 20th century has also results from the decay of humic acids in soils and is
produced changes in quality as a result of diffuse pollution. present as a natural constituent in most groundwaters (see
Nitrate and pesticide concentrations are of particular concern Chapter 5).
and have required costly treatment at some UK public water Most nitrate leaching in the UK occurs during the autumn
supply sources. The intensive production of watercress and winter when the soil reaches field capacity and recharge
is common in the region, particularly in Hampshire, and occurs. Furthermore, nitrate accumulated in the soil prior to
requires the use of pesticides and fertilisers, which may the onset of recharge is transported through the unsaturated
result in the pollution of surface and groundwaters. zone by the infiltrating water. Careful management of the
Farm-waste storage can lead to point-source pollution, land at this time is important to reduce the amount of nitrate
however, over recent years there have been significant that is available to be leached.
improvements by farmers in farm-waste storage facilities. The Environment Agency (and its predecessor
For example, these have resulted in a significant reduction organisations) have monitored nitrate levels in groundwater
in the numbers of point-source pollution incidents from the Dorset and Hampshire Chalk aquifer for over
attributed to dairy-farming practices in the Stour catchment 30 years as part of its remit to prevention pollution of
(Environment Agency, 1997). groundwater under UK and European legislation. In a recent
study9 (Roy et al., 2007) an initial assessment was made of
Pesticides spatial and temporal variations in the groundwater nitrate
Crop rotation avoids continuous heavy application of single concentration dataset (of over 8000 values) for the Dorset
compounds to individual fields. This lessens the danger of and Hampshire Chalk aquifer.
accumulation of pesticide residues in the soil and subsurface The data indicated a clear seasonal variability (which
environment. The most rapid growth in pesticide use has was seen to be most pronounced in the east of the aquifer)
been in certain herbicides and fungicides. The greatest with the highest nitrate values in late winter and spring
threat to groundwater is generally the herbicides as these are and lowest values between summer and early winter. The
relatively soluble. As weed growth is less intense on light, data suggested that seasonal cycles were superimposed on
thin soils typical of the Chalk outcrop than it is on clayey a long-term trend of rising nitrate concentration, and the
soils, the increased vulnerability of the chalk soils may be rate of increase of nitrate in groundwater in the dataset as
countered by lower rates of herbicide application. a whole was determined to be 0.12 mg N l–1 a–1. Both the
The natural processes that control the fate and transport seasonal variation in nitrate and the increasing trend with
of pesticides include leaching, sorption, volatilisation, time are well illustrated by the Litton Cheney spring data
degradation and plant uptake. The mode of application and discussed in Chapter 5 (Figure 5.9).
action of the pesticide are important with regard to leaching Contour plots of five-yearly mean nitrate data from the
potential as those targeted at plant roots and soil insects Environment Agency study (Figure 6.4) showed significant
are usually significantly more mobile than those acting on increases with time, with an average increase of 30 per
the leaves. Many herbicides are applied to the soil before cent in the period 1976 to 2006, and at one site the increase
the weeds emerge and some insecticides are used for soil
treatment. Given the timing of these applications, they are
sufficiently persistent to remain in the soil for significant 9
The study also noted that phase lags between effective precipitation,
periods, so that leaching may occur. groundwater level and nitrate peaks of about two to three months
Most pesticide compounds have water solubilities in suggested that piston flow within the Chalk matrix is the main
excess of 10 mg l–1, and the mobility of pesticides in soil transport mechanism for nitrate to the saturated zone.

99
 20
3 3
40 60
3
80
3
00
4 4
20 4
40 60
4 reached 115 per cent. By comparing the data with land-use
information, it was seen that higher groundwater nitrate
40
1
levels were more likely to be associated with arable and
urban land use than with managed grassland.
20
1
In conclusion, both basin-wide contouring and time-
series data imply that the nitrate problem in the Chalk of the
Wessex Basin is likely to worsen with time, with breaches of
00
1
the quality limits likely to occur with increasing frequency,
1976 – 1980 especially during the winter.

20
3 3
40 60
3
80
3
00
4 4
20 4
40 60
4

6.4.2 Groundwater protection

40
1

If groundwater does become polluted, low flow rates and

limited microbiological activity restrict self-purification,
20
1
and remediation is often expensive and difficult and, in
some cases, impossible. It is therefore very important that
00
1
groundwater resources are protected both with regard to
aquifer vulnerability and source protection.
1981 – 1985
6.4.2.1 Groundwater protection policy
20
3 3
40 60
3
80
3
00
4 4
20 4
40 60
4

The powers and duties of the Environment Agency for

40
1 the protection of groundwater were set out in the Water
Resources Act 1991. For groundwater quality these were to:
20
1
• achieve statutory quality objectives for groundwater
• control discharges to groundwater (discharge consents
00
1
process)
1986 – 1990 • prevent pollution (through regulations)
• enforce against pollution events
3
20 3
40 3
60 3
80 4
00 4
20 4
40 4
60 • take remedial action once pollution has occurred.

40
1 In addition, the Environment Agency has indirect powers
enforced through other bodies under the Environmental
Protection Act 1990 and the Control of Pollution Act 1974
20
1

to control certain discharges to natural waters and to control

waste disposal to land where pollution of water resources
00
1
might result.
1991 – 1995 In 1992, the National Rivers Authority (NRA) adopted
a policy framework for protecting groundwater, which
3
20 3
40 3
60 3
80 4
00 4
20 4
40 4
60 was intended to cover all types of threat to groundwater of
whatever size, from point or diffuse sources, and by both
40
1
conservative and degradable pollutants (National Rivers
Authority, 1992b). This policy was updated in 1998 by the
20
1 Environment Agency’s framework document ‘Policy and
Practice for the Protection of Groundwater’, (Environment
Agency, 1998b), in which two approaches to groundwater
00
protection were set out. Given that both European and
1

1996 – 2000 national legislation require that all groundwater should

be protected, one approach was to assess the vulnerability
20
3 3
40 60
3
80
3
00
4 4
20 4
40 60
4
of aquifers to pollution. In addition, it was recognised
that protection may be required for specific sources
40
1
and therefore groundwater source protection zones
were introduced involving consideration of each source
20
1 catchment area.
Aquifer vulnerability
00
1

In its framework document (Environment Agency, 1998b),

2001 – 2006
the Environment Agency discussed the vulnerability of
groundwater resources to pollution in the context of ground-
water-resource protection. Factors defining groundwater
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 vulnerability were given as:

Figure 6.4 Mean nitrate (as mg nitrogen from nitrate • presence and nature of overlying soil
per litre) in Chalk groundwater in the Wessex Basin for • presence and nature of superficial deposits
consecutive five-year periods since 1976 (after Roy et al., • nature of strata
2007). • thickness of unsaturated zone.

100
While acknowledging that a full assessment of groundwater following Environment Agency Groundwater Vulnerability
resource vulnerability can only be achieved by local maps: Southern Cotswolds (Sheet 37); Somerset Coast
studies using a range of investigations and data types, (Sheet 42); East Somerset and South West Wiltshire (Sheet
the Environment Agency considered it valuable to use 43); North West Hampshire (Sheet 44); West Sussex and
existing soil and geological data to produce groundwater Surrey (Sheet 45); East Devon and South Somerset (Sheet
vulnerability maps for England and Wales. To that end in the 50); Dorset (Sheet 51), and Southern Hampshire (Sheet 52).
1990s, the NRA and its successor the Environment Agency Figure 6.5 shows the groundwater vulnerability over
produced vulnerability maps covering the whole of England the Wessex catchment. As would be expected, the Chalk
and Wales. The methods used are described in the document outcrop across most of Wessex is highly vulnerable where it
‘Policy and Practice for the Protection of Groundwater’ is overlain by less than five metres of overburden. However,
(Environment Agency, 1998b). these classifications can be misleading, as they do not take
Geological strata are classified as major, minor or non- into account important factors such as the thickness of low-
aquifers depending on their permeability, the nature of the permeability superficial deposits and the thickness of the
aquifer (e.g. whether it is fractured) and its productivity. unsaturated zone. Such factors are important and need to be
The vulnerability of aquifers to contamination is then considered when dealing with site-specific issues. In 2010,
classified by assessing the type and thickness of overlying the Environment Agency groundwater protection policy
soils. Superficial deposits with low permeability (e.g. till, started to use aquifer designations consistent with the Water
soliflucted material, peat, lacustrine deposits, clay-with- Framework Directive. Aquifers are designated as ‘principal’
flints and brick earth) are identified by a distinctive stipple or ‘secondary’ and subdivided into ‘superficial’ or ‘bedrock’
overlay. The maps for Wessex include all or part of the materials.

Figure 6.5
N
Groundwater 1
60
vulnerability map for
Basingstoke
the Wessex Basin.
contains OS data © Crown Andover
copyright and database
rights 2017 Ordnance Survey
[100021290 EUL]. Use of 1
40
this data is subject to terms
and conditions.
Salisbury
Winchester

1
20

Southampton

1
00
Portsmouth
Dorchester
Newport
Bournemouth

0
80
Weymouth

0 50 km

0
60
3
60 3
80 4
00 4
20 40
4
60
4 4
80

MAJOR AQUIFER MINOR AQUIFER

From 1: 100 000 scale
Soil leaching capacity Soil leaching capacity Groundwater Vulnerability
Map
Drift High High (Sheets 43, 44, 45, 51 and 52)
© Environment Agency
Rivers Intermediate Intermediate
© NERC (British Geological
Survey)
Low Low
© Soil Survey and Land
Research Centre (National Soil
Non-aquifer Resources Institute).

101
Source protection zones shape and size of the SPZ. The zones are delineated by
In addition to the groundwater vulnerability approach to numerical modelling or semi-analytical techniques that
protecting groundwater resources, the Environment Agency assume uniform permeability and transmissivity to calculate
developed a methodology for protecting groundwater travel times. Further information about this process is given
sources using source protection zones. The approach is by the NRA (1995b) and Keating and Packman (1996).
based on the concept that the proximity of a potentially However, in reality, the Chalk aquifer has a spectrum
polluting activity to a groundwater abstraction source is of velocities associated with the various apertures and
likely to be one of the most important factors in assessing distribution of the fractures, but this is poorly understood.
the risk of source pollution. Therefore, principle zones The manner in which the SPZs have been delineated in the
based on groundwater travel time drawn around a source Chalk aquifer is therefore simplified as a consequence of the
can be used to assess pollution risk and to guide pollution lack of knowledge regarding the range of velocities. This
prevention measures. means that recharge may be able to reach a borehole more
Source protection zones (SPZ) have been defined by the quickly than predicted by the SPZ. The Agency undertook
Environment Agency for nearly 2000 groundwater sources a research and development project to assess existing
(wells, boreholes and springs) used for public drinking methods to develop a rigorous and defensible methodology
water supply in England and Wales. The SPZs provide an for deriving SPZs in fractured/fissured aquifers (Robinson
indication of the risk to groundwater supplies, for which and Barker, 2000).
SPZs have been defined, that may result from potentially
polluting activities and accidental release of pollutants.
Three zones (an inner, outer and total catchment) are usually 6.5 NITRATE VULNERABLE ZONES
defined although a fourth zone (zone of special interest) is
occasionally used. The zones are defined based on travel In the late 1990s, two schemes were developed to protect
times, i.e. how long it would take recharge reaching the groundwater through changes in farming practice to reduce
water table to arrive at the source. Zone 1 is delineated the amount of excess nitrate leached from the soil, the
by a travel time of 50 days, the time over which faecal- nitrate vulnerable zone (NVZ) and nitrate sensitive area
indicator bacteria are reduced to very low or non-detectable (NSA) schemes. Farmers volunteered to participate in
concentrations by die-off and dilution. Any recharge the NSA scheme, for which compensation was offered.
contaminated with pathogenic microbes occurring within The groundwater NVZ designation is compulsory where
this zone may therefore cause an unacceptable concentration groundwater quality is significantly threatened by nitrate
of viable microbes at the groundwater source. It is therefore from agricultural sources. They are defined as areas where
wise to minimise potentially polluting activities within groundwater contains or could contain, if preventative
this zone and thus activities like sewage sludge spreading measures are not taken, nitrate concentrations greater than
or landfilling of waste are not permitted in Zone 1. Zone 2 50 mg l–1. Farmers who are farming land within NVZs are
is the 400-day travel time zone, while Zone 3 contains the required to comply with measures to control and reduce
entire catchment for the source. leaching of nitrate to rivers and groundwater.
The direction of groundwater flow and the properties In the Wessex Basin, essentially all of the Chalk outcrop
of the surrounding strata determine the orientation, is currently classified as NVZ.

102
7 References

Most of the references listed below are held in the Library migration and attenuation of landfill leachate.   Water
of the British Geological Survey at Keyworth, Nottingham. Science and Technology, Vol 20, 119–128.
Copies of the references may be purchased from the Library Bloomfield, J P. 1997.   The role of diagenesis in the
subject to the current copyright legislation. hydrogeological stratification of carbonate aquifers:
an example from the Chalk at Fair Cross, Berkshire,
Abraham, D, and Ritchie, J D. 1991.  The Darwin UK.  Hydrology and Earth Systems Sciences, Vol. 1,
Complex, a Tertiary igneous centre in the Northern Rockall 19–33.
Trough.  Scottish Journal of Geology, Vol. 27, 113–126.
Bloomfield, J P, Brewerton, L J, and Allen, D J. 1995.  
Alexander, L S. 1981.   The hydrogeology of the Chalk Regional trends in matrix porosity and bulk density of the
of south Dorset.   Unpublished PhD thesis, University of Chalk of England.   Quarterly Journal of Engineering
Bristol. Geology, Vol. 28, S131–S142.
Allen, D J, Brewerton, L J, Coleby, L M, Gibbs, B R, Bournemouth and West Hampshire Water. 2006.  
Lewis, M A, MacDonald, A M, Wagstaff, S J, and Drought plan.   Version 1.   Draft for consultation.
Williams, A T. 1997.   The physical properties of major
aquifers in England and Wales.   British Geological Survey Bristow, C R, Barton, C M, Freshney, E C, Wood,
Technical Report, WD/97/34.   Environment Agency R & C J, Evans, D J, Cox, B M, Ivimey-Cook, H C, and
D Publication 8. Taylor, R T. 1995.   Geology of the country around
Shaftesbury.  Memoir of the British Geological Survey,
Aspinwall and Company. 1997a.  Groundwater
Sheet 313 (England and Wales).
Protection Zone Project Phase III MODFLOW Modelling:
Frome and Piddle MODFLOW Model Report.  Completed Bristow, C R, Mortimore, R, and Wood, C J. 1997.  
for Environment Agency, South West Region. Lithostratigraphy for mapping the Chalk of southern
Aspinwall and Company. 1997b.  Groundwater England.  Proceedings of the Geologists’ Association,
Protection Zone Project Phase III MODFLOW Modelling: Vol. 108, 293–315.
Allen and Crane MODFLOW Model Report.  Completed British Geological Survey. 2001.  Dorchester.
for Environment Agency, South West Region. England and Wales Sheet 328.   Solid and drift.  
Aston, J J. 1999.   The River Frome catchment, 1:50 000.   (Keyworth, Nottingham: British Geological
Dorset — development of a conceptual framework to Survey.)
support environmental and modelling research.   Un- Bromley, R G, and Gale, A S. 1982.  The litho-
published MSc thesis, Imperial College of Science, stratigraphy of the English Chalk Rock.   Cretaceous
Technology and Medicine. Research, Vol. 3, 273–306.
Atkinson, T C, and Smith, D I. 1974.  Rapid groundwater Buckley, D K. 1996   A review and interpretation
flow in fissures in the Chalk: an example from south of geophysical logging performed for the Wareham
Hampshire.  Quarterly Journal of Engineering Geology, groundwater project.   British Geological Survey Technical
Vol. 7, 197–205. Report, WD/96/45C.
Avon and Dorset River Authority. 1970.  Periodic Buckley, D K, Shand, P, and Gale, I N. 1998.  Geo-
survey of the water resources and future demand within physical and geochemical investigations related to the
the Avon and Dorset River Authority area.   Internal ASR trial at Lytchett Minster, Dorset.   British Geological
Report.   Avon and Dorset River Authority, 61. Survey Technical Report, WD/98/27C.
Avon and Dorset River Authority, Water Resources Centre of Ecology & Hydrology, British Geologial
Board and West Wiltshire Water Board. 1973.  The Survey, 2003.   Hydrological data, United Kingdom
Upper Wylye investigation.   Avon and Dorset River hydrometric register and statistics 1996–2000. A catalogue
Authority, Poole. of river-flow gauging stations and observation boreholes
Bath, A H, and Edmunds, W M. 1981.  Identification of with summary hydrometric statistics. (Wallingford:
connate water in interstitial solution of chalk sediments.   NERC Centre for Ecology & Hydrology.) http://nora.necr.
Geochimica Cosmochimica Acta, Vol. 34, 1449–161. ac.uk/6950/
Bevan, T G, and Hancock, P L. 1986.   A late Cenozoic Chadwick, R A. 1986.   Extension tectonics in the Wessex
regional mesofracture system in southern England and Basin, southern England.   Journal of the Geological
Northern France.   Journal of the Geological Society of Society of London, Vol. 143, 465–488.
London, Vol. 143, 355–362. Chadwick, R A. 1993.   Aspects of basin inversion in
Birmingham University School of Civil Engineering. southern Britain.   Journal of the Geological Society of
1997.   Wareham groundwater scheme — phase III London, Vol 150, 311–322.
pumping test analysis (unpublished). Christensen, E R, Scherfig, J, and Kolde, M. 1978.  
Blakey, N C, and Towler, P A. 1988.  The effect Metals from urban run-off in dated sediments of a very
of unsaturated/saturated zone properties upon the shallow estuary.   Environmental Science and Technology,
hydrochemical and microbial processes involved in the Vol. 12, 1168–1173.

103
Culshaw, M G, and Waltham, A C. 1987.  Natural and Environment Agency. 1997.  Local Environment Agency
artificial cavities as ground engineering hazards.   Quarterly action plan — Dorset Stour consultation report.
Journal of Engineering Geology, Vol. 20, 139–150. Environment Agency. 1998a.  Local Environment
Darling, W G, and Bath, A H. 1988.   A stable isotope Agency action plan — Hampshire Avon consultation draft
study of recharge processes in the English Chalk.   Journal (December 1998).
of Hydrology, Vol. 101, 31–46. Environment Agency. 1998b.  Policy and practice for the
Department for Environment, Food and Rural Affairs Protection of Groundwater (second edition).   (London:
(Defra). 2005a.  Water Framework Directive.  Summary Stationery Office.)
report of the characterisation, impacts and economics Environment Agency. 1999a.  Local environment action
analyses required by Article 5.   South West River Basin plan — Test and Itchen environmental overview.
District.
Environment Agency. 1999b.  Local environment action
Department for Environment, Food and Rural Affairs plan — Isle of Wight environmental overview.
(Defra). 2005b.  Water Framework Directive.  Summary
report of the characterisation, impacts and economics Environment Agency. 1999c.  Local environment action
analyses required by Article 5.   South East River Basin plan — East Hampshire environmental overview.
District. Environment Agency. 2001.   Bourne and Nine Mile
Dewar, G A B. 1898.  South country trout stream.   Rivers low flow investigation.   Conceptual Modelling
London: Lawrence and Bullen Ltd. Photocopies of relevant Report.
sections obtained from Dorset County Council, Corporate Environment Agency. 2002a.   The east Hampshire
Services, County Library Services, Dorchester.   (Agency catchment abstraction management strategy consultation
File: WR/05/30/03 — No. 4) document.
Downing, R A. 1993.  Groundwater resources, their Environment Agency. 2002b.  Managing Water
development and management in the UK: an historical abstraction: the catchment abstraction management
perspective.  Quarterly Journal of Engineering Geology strategy process.
and Hydrogeology, Vol. 26, 335–358. Environment Agency. 2003.   The Isle of Wight
Downing, R A, Price, M, and Jones, G P. 1993.  The catchment abstraction management strategy consultation
making of an aquifer.   14–34 in The hydrogeology of the document.
Chalk of north-west Europe, Downing, R A, Price, M, and Environment Agency. 2004.   A comparison of Chalk
Jones, G P (editors).   (Oxford: Clarendon Press.) groundwater models in and around the River Test
Drummond, P V O. 1970.   The mid-Dorset swell.   catchment.
Evidence of Albian–Cenomanian movements in Wessex.   Environment Agency. 2009a.   Water for life and
Proceedings of the Geologists’ Association, Vol. 81, 679–714. livelihoods. River Basin Management Plan, South West
Eastwood, J C, and Stanfield, P J. 2001.  Key success River Basin District.
factors in an ASR scheme.   Quarterly Journal of Environment Agency. 2009b.   Water for life and
Engineering Geology and Hydrogeology, Vol. 34, 399–409. livelihoods. River Basin Management Plan. South East
Edmonds, C N. 1983.   Towards the prediction of River Basin District.
subsidence risk upon the Chalk outcrop.   Quarterly Environment Agency and Water Management Consultants.
Journal of Engineering Geology, Vol. 16, 261–266. 2004.   Bourne and Nine Mile Rivers RSA
Edmunds W M. 1996.   Groundwater quality in the Chalk; project — phase 2 investigations; numerical modelling
Lulworth to Poole Harbour.   British Geological Survey final report to the Environment Agency.
Technical Report, WD/96/49C. Extence, C A, Balbi, D M, and Chadd, R P. 1999.  
Edmunds, W M, Darling, W G, Kinniburgh, D G, Dever, River flow indexing using British benthic macro-
L, and Vachier, P. 1992.   Chalk groundwater in England invertebrates: a framework for setting hydroecological
and France: hydrogeochemistry and water quality.   British objectives.  Regulated Rivers: Research and Management,
Geological Survey Research Report, SD/92/2. Vol. 15, 543–574.
Edmunds, W M, Doherty, P, Griffiths, K J, Shand, P, Farrant, A R, Hopson, P M, Booth, K A, and Aldiss, D T.
and Peach, D. 2002.   Baseline report series: 4. The Chalk 2001.   Geology of the Bourne River catchment.   Final
of Dorset.   British Geological Survey Commissioned report of the geology of the Bourne and Nine Mile River
Report, CR/02/268N. catchments for the Environment Agency.   Internal Report
Edwards, R A, and Freshney, E C. 1987.  Geology of of the British Geological Survey, IR/01/157.
the country around Southampton.   Memoir of the British Food and Agriculture Organisation of the United Nations.
Geological Survey, Sheet 315 (England and Wales). 1998.   Crop evapotranspiration — guidelines for
Entec. 2002.   River Itchen catchment groundwater computing crop water requirements — Food and
modelling study: phase 1 data synthesis and draft Agriculture Organisation irrigation and drainage, Paper
conceptual model report.   Report for Environment Agency 56.  Rome.
Southern Region. Ford, D C, and Williams, P. 1989.  Karst hydrology and
Entec. 2005.   Test and Itchen groundwater geomorphology. (Chapman and Hall.)
modelling study, phase 2: model development and Foster, S S D, Cripps, A C, and Smith-Carrington, A K.
requirement.   Environment Agency Southern Region. 1982.   Nitrate leaching to groundwater.   Philosophical
Environment Agency. 1996.  Lead data request Transactions of the Royal Society, Vol. 196, 477–489.
proforma: determination of groundwater protection zones, Gardner, C M K, Cooper, J D, Wellings, S R, Bell, J P,
Piddle and Frome catchments, October 1996. Hodnett, M G, Boyle, S A, and Howard, M J. 1990.  

104
Hydrology of the unsaturated zone of the Chalk of south- Hopson, P M. 2005.   A stratigraphical framework for the
east England.  in Chalk.  Burland, J B, Mortimore, Upper Cretaceous Chalk of England and Scotland with
R N, Roberts, T S, Jones, D L, and Corbett, B O statements on the chalk of Northern Ireland and the UK
(editors).   (London: Thomas Telford.) offshore sector. British Geological Survey Research Report,
Gaus, I, Shand, P, Gale, I N, Williams, A, and Eastwood, J. RR/05/01.
2002.   Geochemical modelling of fluoride concentration Hopson, P M, Farrant, A R, Newell, A J, Marks, R J
changes during aquifer storage and recovery (ASR) in the Booth, K A, Bateson, L B, Woods, M A, Wilkinson, I P,
Chalk aquifer in Wessex, England.   Quarterly Journal of Brayson, J and Evans, D J. 2008.  Geology of the
Engineering Geology and Hydrogeology, Vol. 35, 203–208. Salisbury district.   Sheet description of the British
Geake, A K, and Foster, S S D. 1989.  Sequential Geological Survey.   Sheet 298 (England and Wales).
isotope and solute profiling in the unsaturated zone of Houston, J F T, Eastwood, J C, and Cosgrove, T K P.
British Chalk.   Hydrological Sciences, Journal des 1986.   Locating potential borehole sites in a discordant
Sciences Hydrologiques, Vol. 34, 79–95. flow regime in the Chalk aquifer at Lulworth using
Giles, D M, and Lowings, V A. 1990.   Variation in the integrated geophysical surveys.   Quarterly Journal of
character of the chalk aquifer in east Hampshire.   619– Engineering Geology, Vol. 19, 271–282.
626 in Chalk.  Burland, J B, Mortimore, R N, Roberts, Howden, N J K, Wheater, H S, Peach, D W, and Butler,
T S, Jones, D L, and Corbett, B O (editors).  (London: A P. 2004.   Hydrogeological controls on surface/
Thomas Telford.) groundwater interactions in a lowland permeable chalk
Greenaway, P W. 1995.   Computer modelling of a catchment.  113–122 in Hydrology: science and practice
small Chalk catchment with reference to the river Allen, for the 21st century.  Volume II.  Webb, B, Acreman, M,
Dorset.   Unpublished MSc thesis, University of Reading. Maksimovic, C, Smithers, H and Kirby, C (editors).  (The
Netherlands: British Hydrological Society.)
Grindley, J. 1967.   The estimation of soil moisture
deficits.  Meteorological Magazine, Vol. 96, 97–108. Howden, N J K. 2004.  Hydrogeological controls on
surface/groundwater interactions in a lowland permeable
Groundwater Development Consultants (GDC). 1992.   Chalk catchment: implications for water quality and
Groundwater study: River Allen catchment, National numerical modelling.   Unpublished PhD thesis,
Rivers Authority, Wessex Region. University of London.
Halcrow. 1992.  Upper Hampshire Avon groundwater Institute of Geological Sciences. 1979a.  Hydrogeological
study: report on the phase 1 studies.   National Rivers map of the Chalk and associated minor aquifers of
Authority, Wessex Region. Wessex.   (Institute of Geological Sciences and Wessex
Halcrow. 1995.   River Piddle water resources manage- Water Authority.)
ment study volume 1 — main report, report to the National Institute of Geological Sciences. 1979b.  Hydro-
Rivers Authority, South Western Region. geological map of Hampshire and the Isle of Wight.  
Halcrow. 1996.  Upper Hampshire Avon groundwater (Institute of Geological Sciences and Southern Water
study.   phase 2: the Wylye catchment. National Rivers Authority.)
Authority, South Western Region. Irving, A A K. 1993.   The Alre augmentation scheme —
Hampshire County Council. 2000.   Water in Hampshire a model to calculate its effect on groundwater levels and
— a comprehensive review. river flows.   Unpublished MSc Thesis, University College
Hampshire River Authority. 1970.  First periodical survey. London.
Hampshire Water Partnership. 2003.  Hampshire water Johnson, I W, Elliott, C R N, and Gustard, A. 1995.  
strategy. Modelling the effect of groundwater abstraction on
salmonid habitat availability in the River Allen, Dorset,
Hancock, J M. 1993.   The formation and diagenesis England.  Regulated Rivers: Research and Management,
of Chalk.   14–34 in The Hydrogeology of the Chalk of Vol. 10 (2–4), 229–238.
north-west Europe. Downing, R A, Price M, and Jones, G
P (editors).   (Oxford: Oxford University Press.) Jones, D K C. 1999.   Evolving models of the Tertiary
evolutionary geomorphology of southern England, with
Headworth, H G. 1972.   The analysis of natural ground- special reference to the Chalklands.   1–23 in Uplift,
water level fluctuations in the Chalk of Hampshire.   erosion and stability: perspectives on long-term landscape
Journal of the Institute of Water Engineers and Scientists, development, Smith, B J, Whalley, W B, and Warke, P A
Vol. 26, 107–124. (editors).  Special Publication of the Geological Society
Headworth, H G. 1978.  Hydrogeological characteristics of London, No. 162.
of artesian boreholes in the Chalk of Hampshire.   Quarterly Jones, H K, Morris, B L, Cheney, C S, Brewerton, L J,
Journal of Engineering Geology, Vol. 11, 139–144. Merrin, P D, Lewis, M A, McDonald, A M, Coleby, L M,
Headworth, H G, Keating, T, and Packman, M J. 1982.   Talbot, J C, McKenzie, A A, Bird, M J, Cunningham, J E,
Evidence for a shallow highly permeable zone in the Chalk and Robinson, V K. 2000.   The physical properties of
of Hampshire, UK.   Journal of Hydrology, Vol. 55, 93–112. minor aquifers in England and Wales.   British Geological
Higginbottom, I, and Fookes, P. 1971.  Engineering Survey Technical Report, WD/00/4.
aspects of periglacial features in Britain.   Journal of Jones, H K, and Robins, N S (editors). 1999.  The Chalk
Engineering Geology, Vol. 3, 85–117. aquifer of the South Downs.  (Keyworth, Nottingham:
Hopson, P M. 2000.   Geology of the Fareham and Ports- British Geological Survey.)
mouth district.   Sheet explanation of the British Geological Jukes-Browne, A J, and Hill, W. 1903.  The Cretaceous
Survey.   Sheet 316 and part of Sheet 331 (England and rocks of Britain.   II.   The Lower and Middle Chalk of
Wales). England.  Geological Survey of Great Britain Memoir.

105
Jukes-Browne, A J, and Hill, W. 1904.  The Cretaceous Symposium on Spatial Data Handling, Vol. 1 July 23–27
rocks of Britain.   III.   The Upper Chalk of England.   Zurich, 250–262.
Geological Survey of the Great Britain Memoir. Mortimore, R N. 1983.   The stratigraphy and sediment-
Keating, T. 1978.   A method for the analysis of draw- ation of the Turonian–Campanian in the southern province
down from multiple-source test pumping.   Journal of of England.   Zitteliana, Vol. 10, 515–529.
Hydrology, Vol. 39, 185–191. Mortimore, R N. 1986.   Stratigraphy of the Upper
Keating, T. 1982.   A lumped parameter model of a chalk Cretaceous White Chalk of Sussex.   Proceedings of the
aquifer stream system in Hampshire, United Kingdom.   Geologists’ Association, Vol. 98, 97–139.
Groundwater, Vol. 20, 430–436. Mortimore, R N, and Pomerol, B. 1991.  Upper
Keating, T, and Packman, M J (editors). 1996.  Manual Cretaceous tectonic disruption in a placid chalk sequence
of standard zone delineation methodologies. Environment in the Anglo–Paris Basin.   Journal of the Geological
Agency. Society of London, Vol. 148, 391–404.
Klinck, B A, Hopson, B A, Lewis, P M, MacDonald, D Mortimore, R N, and Wood, C J. 1986.  The distribution
M J, Inglethorpe, S D J, Entwisle, D C, Harrington, J F, of flint in the English Chalk, with particular reference
and Williams, L. 1998.   The hydrogeological behaviour to the ‘Brandon Flint Series’ and the high Turonian
of the clay-with-flints in Southern England.   British flint maximum. 7–20 in The scientific study of flint and
Geological Survey Technical Report, WE/97/5. chert.  Sieveking, G de G, and Hart, M B (editors).
Lerner, D N. 1997.  Groundwater Recharge.  109–150 (Cambridge: Cambridge University Press.)
in Geochemical processes, weathering and groundwater Mortimore, R N, Argent, K, Caillard, P, Snopok, P G,
recharge in catchments.  Saether, O L, and De Caritat, P Smith, A J, Tracey, N, Holliday, J K, and Honeyman, W N.
(editors).  (Rotterdam: Balkema.) 1990.   Geophysical surveys over solution pipes and
Neolithic flint mines in the Chalk of the South Downs,
Lewis, M A, Jones, H K, MacDonald, D M J, Price, M,
Sussex, England.   95–102 in Cahiers de Quarternaire,
Barker, J A, Shearer, T R, Wesselink, A J, and, Evans
No. 17 –Le Silex De Sa Genese A L’outil.   Acts du vo,
D J. 1993.   Groundwater storage in British aquifers:
Colloque International sur le Silex, 95–102.
Chalk.  National Rivers Authority, Bristol.  R & D Note
169. Mortimore, R N, Wood, C J, and Gallois, R W. 2001.  
British Upper Cretaceous stratigraphy.   Geological
Maclean, R D. 1969.   The effects of tipped domestic Conservation Review Series, No. 23.   (Peterborough:
refuse on groundwater quality: a survey in north Kent.   Joint Nature Conservation Committee.)
Proceedings of the Society for Water Treatment and
Examination, Vol. 18, 18–34. Murphy, A M. 1998. Sediment and groundwater
geochemistry of the Chalk in southern England.
Mansell-Moullin, M. 1986.   The River Piddle’s water Unpublished PhD thesis, Kingston University.
resources.   The River Piddle Protection Association.
National Rivers Authority. 1992a.  River Piddle action
Mansell-Moullin, M. 1994.   Report on the River Frome: plan, Blandford Forum.
water resources, environmental characteristics and local
National Rivers Authority. 1992b.  Policy and practice
concerns.   Report to the Frome, Piddle and West Dorset
for the protection of groundwater.   (London: HMSO.)
Fishery Association, 76.
National Rivers Authority. 1994.  Hampshire Avon
Marcus Hodges Environment Ltd. 1999.  Upper River catchment management plan — final report.
Piddle — assessment of strategies to alleviate low flow.  
Final report No. 51060/R2 and 51060/R2. National Rivers Authority. 1995a.   The Frome and
Piddle catchment management plan consultation report,
Marsh, T J, and Lees, M L (editors). 2003.  Hydrological Blanford Forum.
Data United Kingdom, hydrometric register and
statistics.  1996–2000.  Centre for Ecology & National Rivers Authority. 1995b.  Guide to
Hydrogeology. groundwater protection zones in England and Wales.
(London: HMSO.)
Mimran, Y. 1976.   Strain determination using a density-
National Rivers Authority. 1996.   The Frome and
distribution technique and its application to deformed
Piddle catchment management plan — action plan.  
Upper Cretaceous Dorset chalks.   Tectonophysics, Vol. 31,
NRA South Western Region.
175–192.
Neumann, I, Cobbing, J E, Tooth, A, and Shand P. 2004.  
Moore, R J, and Bell, V A. 2002.  Incorporation of Baseline Report Series: 15. The Palaeogene of the Wessex
groundwater losses and well level data in rainfall-runoff Basin.  British Geological Survey Commissioned Report,
models illustrated using the PDM.   Hydrology and Earth CR/04/245N.
System Sciences, Vol. 6 (1) 25–38.
Newell, A J, Royse, K R, Barron, A J M, and Woods, M A.
Moore, R V, Morris, D G, and Flavin R W. 1994.   2002.   The geological framework of the Frome–Piddle
Subset of UK digital 1:50 000 scale river centre-line catchment.  British Geological Survey Commissioned
network.   NERC, Institute of Hydrology, Wallingford. Report, CR/02/197.
Moore, W G. 1973.   A dictionary of Geography A. Paolillo, S. 1969.   Hydrogeology of the River Frome
(Harmondsworth: Penguin Books.) catchment.   Unpublished MSc Thesis Vol XI, Istituto di
Morgan-Jones M. 1977.   Mineralogy of the non- Geologia Applicata, Napoli.
carbonate material from the Chalk of Berkshire and Pearce, M A, Jarvis, I, Swan, A R H, Murphy, A M,
Oxfordshire, England.   Clay Minerals, Vol. 12, 331–344. Tocher, B A, and Edmunds, W M. 2003.  Integrating
Morris, D G, and Flavin, R W. 1990.   A digital terrain palynological and geochemical data in a new approach
model for hydrology.   Proceedings 4th International to palaeoecological studies: Upper Cretaceous of the

106
Banterwick Barn Chalk borehole, Berkshire, UK.   Marine southern Wessex Basin.   British Geological Survey
Micropalaeontology, Vol. 47, 271–306. Technical Report, WD/99/55C.
Penman, H L. 1948.   Natural evaporation from open Shand, P, and Bloomfield, J P. 1995.  Mineralisation of
water, bare soil and grass.   Proceedings of the Royal shallow fracture surfaces in the Chalk and implications
Society of London, Series A, Vol. 193, 120–145. for contaminant attenuation.   British Geological Survey
Perkins, M A, and Robertson, A S. 1980.  Groundwater Technical Report, WD/95/15.
investigations and geophysical logging at Shapwick, Dorset.   Shore, T W. 1894.   Springs and streams of Hampshire.
British Geological Survey Technical Report, WD/ST/80/14. 33–58 in Minns, G W (editor).   Papers and proceedings of
Pitman, J I. 1979.   Carbonate chemistry of groundwater the Hampshire Field Club, Vol. II 1890 – 1893.
from chalk, Givendale, East Yorkshire.   Geochimica et Smith, D B, Wearn, P L, Richards, H J, and Rowe, P.
Cosmochimica Acta, Vol. 42 1885–1897. 1970.   Water movement in the unsaturated zone of
Portsmouth Water. 2006.  Draft statutory drought high and low permeability strata by measuring natural
plan — public summary. tritium.  Isotope Hydrology: Proceedings of Symposium,
International Atomic Energy Agency, Vienna.
Price, M. 1987.   Fluid flow in the Chalk of England.  
141–156 in Fluid flow in sedimentary basins and aquifers. Southern Science. 1991.   Report on the 1989 test
Goff, J C, and Williams, B P J (editors).   Special pumping of the Alre scheme, Report No. 91/R/160, Vol. 1,
Publication of the Geological Society of London.  No. 34. Southern Science, Worthing.
Price, M, Bird, M J, and Foster, S S D. 1976.  Chalk Southern Water. 2006.   Draft drought plan for
pore-size measurements and their significance.   Water consultation.
Services, October, 596–600. Southern Water Authority. 1979.  Itchen groundwater
Price, M, Robertson, A S, and Foster, S S D. 1977.   regulation scheme; final report on the Candover pilot
Chalk permeability — a study of vertical variation using scheme, Southern Water Authority, Hampshire.
water injection tests and borehole logging.   Water Southern Water Authority. 1984.  Further Itchen
Services, Vol. 81, 603–610. river augmentation scheme, 1984 test pumping analysis,
Price, M, Morris, B, and Robertson, A. 1982.  A Southern Water Authority, Hampshire.
study of intergranular and fissure permeability in chalk Sperling, C H B, Goudie, A S, Stoddart, D R, and Poole,
and Permian aquifers using double packer injection G G. 1977.   Dolines of the Dorset Chalklands and other
testing.  Journal of Hydrology, Vol. 54, 401–423. areas in southern Britain.   Transactions of the Institute
Price, M, Downing, R A, and Edmunds, W M. 1993.   British Geographers, NS2, 205–223.
The Chalk as an aquifer.   14–34 in The hydrogeology of Stuart, J D. 2000.   An analysis of surface and ground-
the Chalk of north-west Europe.  Downing, R A, Price, M, water interaction and the creation of an integrated
and Jones, G P (editors). (Oxford: Clarendon Press.) catchment groundwater model for the Frome/Piddle
Price, M, Low R G, and McCann, C. 2000.  Mechanisms catchment in Dorset.   Unpublished MSc Thesis, Imperial
of water storage and flow in the unsaturated zone of the College of Science, Technology and Medicine, University
Chalk aquifer.   Journal of Hydrology, Vol. 233, 54–71. of London.
Underhill, J R, and Stoneley, R. 1998.  Introduction
Rawson, P, Allen, P, and Gale, A S. 2001.  The Chalk
to the development, evolution and petroleum geology of
Group — a revised lithostratigraphy.   Geoscientist, Vol. 11,
the Wessex Basin.  1–18 in Development, Evolution and
21.
Petroleum Geology of the Wessex Basin. Underhill, J R
Rennie, A. 1994.   The hydrogeology of the Candover (editor).  Special Publication of the Geological Society of
catchment in Hampshire with special reference to the London, No. 133.
influence of fissures on the Candover Stream.   Un-
Underhill, J R, and Paterson, S. 1998.  Genesis of
published MSc thesis, University of Reading.   Post-
tectonic inversion structures: seismic evidence for the
graduate Research Institute for Sedimentology.
development of key structures along the Purbeck–Isle of
Robins, N S, and Lloyd, J W. 1975.   A surface resistivity Wight disturbance.   Journal of the Geological Society of
investigation over an area of the English Chalk aquifer.   London, Vol. 155, 975–992.
Journal of Hydrology, Vol. 27, 285–295. Waltham, A C, Simms, M J, Farrant, A R, and Goldie, H.
Robinson, J N, and Barker, J A. 2000.  A fractured/ 1997.   Karst and caves of Great Britain.   Geological
fissured rock approach to GPZs.   Environment Agency Conservation Review, No. 12.   (London: Chapman and
R & D Technical Report, W223. Hall.)
Roy, S, Speed, C, Bennie, J, Swift, R, and Wallace, P. Wellings, S R, Cooper, J D, and Bell, J P. 1982.  The
2007.   Identifying the significant factors influencing physics of solute movement in the unsaturated zone of
chronological and spatial trends in nitrate concentrations in the British Chalk.   Proceedings of the Symposium on the
the Dorset and Hampshire Basin Chalk aquifer of Southern Impact of Agricultural Activities on Groundwater, IAH,
England.  Quarterly Journal of Engineering Geology and Prague, p379–388.
Hydrogeology, Vol. 40, 377–392. Wellings, S R. 1984a.   Recharge of the Upper Chalk
Scholle, P A. 1977.   Chalk diagenesis and its relation aquifer at a site in Hampshire, England, 1.   Water balance
to petroleum exploration: oil from chalks, a modern and unsaturated flow.   Journal of Hydrology, Vol. 69,
miracle?  American Association of Petroleum Geologists, 259–273.
Bulletin, Vol. 61, 982–1009. Wellings, S R. 1984b.   Recharge of the Upper Chalk
Shand, P. 1999.   A geochemical investigation of chalk aquifer at a site in Hampshire, England, 2.   Solute
porewater samples from pilot ASR boreholes in the movement.  Journal of Hydrology, Vol. 69, 275–285.

107
Wessex Water Authority. 1988.   The River Piddle Whitaker, W, and Edwards, W. 1926.   Wells and springs
study. (London: HMSO.) of Dorset.   Memoir of the Geological Survey of Great
Wessex Water. 1995.   Wareham groundwater invest- Britain. (London: HMSO.)
igation — phase 3 preliminary report (unpublished). Williams, R B G. 1987.   Frost weathered mantles on the
Westhead, R K. 1992.   Geology of the Godmanstone– Chalk. 127–133 in Periglacial processes and landforms in
Dorchester (Dorset).   British Geological Survey Technical Britain and Ireland.  Boardman, J (editor).  (Cambridge:
Report, WA/92/36 Cambridge University Press.)
White, H. 1921.   A short account of the geology of the Wooldridge, S W, and Linton, D L. 1955.  Structure,
Isle of Wight.   Memoir of the Geological Survey of Great surface and drainage in south-east England.  
Britain. (London: George Philip.)
White, H. 1923.   Geology of the country south and west Younger, P. 1989.   Devensian periglacial influences on
of Shaftesbury.   Memoir of the Geological Survey of the development of spatially variable permeability in the
Great Britain, Sheet 313 (England and Wales). Chalk of south-east England.   Quarterly Journal of Engi-
neering Geology, Vol. 22, 343–354.
Whitaker, W. 1910.   The water supply of
Hampshire.  Memoir of the Geological Survey of Great
Britain.

108
Appendix

List of boreholes and shafts

Borehole name Grid Ref SOBI No.
Abbotstone SU 55 34 SU53SE1
Alton Pancras ST 70 01 ST70SW17
Aughton SU 23 57 SU25NW10
Axford 1A SU 61 42 SU64SW1
Axford 1B SU 61 42 SU64SW43
Blackhorse Farm SU 56 14 SU51SE152
Bradley 2A SU 62 41 SU64SW3
Bradley 2BS SU 62 41 SU64SW2
Brixton Deverill ST 85 38 ST83NE5
Chilgrove House SU 83 14 SU81SW7
Chitterne ST 99 45 ST94NE2
Compton House SU 77 14 SU71SE10
Corfe Mullen SY 97 98 SY99NE1C
Durweston ST 84 08 ST80NW6
East Holton Farm SY 95 91 SY99SE232
Figheldean SU 15 46 SU14NE17
Fonthill Bushes ST 95 35 ST93NE43
Itchen Down 2 SU 54 33 SU53SW4
Itchen Down Farm SU 54 33 SU53SW3
Leckford Bridge SU 23 51 SU25SW16
Lower Wield Farm SU 63 40 SU64SW21
Lulworth West SY 82 80 SY88SW3
Lymington SZ 30 96 SZ39NW22
Marchwood 1 SU 39 11 SU31SE227
Musseldean Copse ST 90 36 ST93NW7
Padnall Grange SU 70 11 SU71SW62
Raglington Farm SU 54 13 SU51SW114
Sandhills SZ 45 90 SZ49SE3
Sompting TQ 16 06 TQ10NE80
Stoborough SY 92 86 SY98NW298
Sutton Poyntz SY 70 84 SY78SW9
Totford SU 56 38 SU53NE1
Upwey SY 65 85 SY68NE97
Wareham Common SY 91 87 SY98NW297
West Houghton ST 82 04 ST80SW27
West Woodyates Manor SU 01 19 SU01NW92
Wield 3A SU 61 40 SU64SW4
Wield 3B SU 61 40 SU64SW5
Willoughby Hedge ST 86 33 ST83SE62
Wilmingham SZ 36 87 SZ398NE9
Woodside SU 33 56 SU35NW39

109
110
HYDROGEOLOGICAL MAPS 1:100 000
6 South Downs and part of the Weald, 1978
Hydrogeological maps have been published at various 7 South West Chilterns, 1978
scales. They are colour-printed maps, supplied as either 8 Chalk of Wessex, 1979
flat sheets or folded sheets in plastic sleeves, and are 9 Hampshire and Isle of Wight, 1979
available only from the BGS. 10 East Yorkshire, 1980
11 Northern East Midlands, 1981
12 Southern Yorkshire, 1982
13 Permo-Trias and other aquifers of SW England, 1982
INDEX TO AREAS COVERED BY
HYDROGEOLOGICAL MAPS 14 Between Cambridge and Maidenhead, 1984
16 Fife and Kinross, 1986
19 Clwyd and the Cheshire Basin, 1989
20 Eastern Dumfries and Galloway, 1990
1:63 360
15 Dartford (Kent) district, 1968 (out of print, available
as a colour photographic print. Also covered in ref. 3)
1:50 000
21 The Carnmenellis Granite: hydrogeological,
hydrogeochemical and geothermal characteristics,
1990
18 1:25 000
22 Jersey, 1992
16

OTHER REPORTS IN THIS SERIES

20
NORTHERN
IRELAND The Chalk aquifer of the South Downs. H K Jones and
N S Robins (editors), 1999.
10 Hydrogeochemical processes determining water quality in
12
upland Britain. P Shand, W M Edmunds, S Wagstaff, R Flavin
19 11 2 and H K Jones. 1999.
4 The Chalk aquifer system of Lincolnshire. E J Whitehead and
1 A R Lawrence. 2006.
5
14 The Dumfries Basin aquifer. N S Robins and D F Ball. 2006.
17 7 The Chalk aquifer of Yorkshire. I N Gale and H K Rutter. 2006.
15 3
13 9 6 The Chalk aquifer of the North Downs. B Adams (editor). 2008.
8
21
OTHER KEY BGS HYDROGEOLOGICAL
22 ­PUBLICATIONS

Hydrogeology of Scotland. N S Robins, 1990

1:625 000 (ISBN 0 11 884468 7)
1 England and Wales, 1977 Long-term hydrograph of groundwater levels in the Chilgrove
18 Scotland, 1988 House well in the Chalk of southern England, 1990. Poster
1:126 720 Long-term hydrograph of groundwater levels in the Dalton
2 North and east Lincolnshire, 1967   Holme estate well in the Chalk of Yorkshire, 1992. Poster
(out of print, available as a colour photographic print)
3 Chalk and Lower Greensand of Kent, 1970 Hydrogeology of Northern Ireland. N S Robins, 1996.
(two sheets) (ISBN 0 11 884524 1)
1:125 000 The physical properties of major aquifers in England and
4 Northern East Anglia, 1976 (two sheets). Flat only Wales. D J Allen et al., 1997. WD/97/34
5 Southern East Anglia, 1981 (two sheets) The hydrogeological behaviour of the clay-with-flints of
17 South Wales, 1986 southern England. B A Klink et al., 1997. WE/97/5
The physical properties of minor aquifers in England and
Wales. H K Jones et al., 2000. WD/00/04
The natural (baseline) quality of groundwater in England and
Wales. P Shand, W M Edmunds, A R Lawrence, P L Smedley
and S Burke. 2007. RR/07/06 and NC/99/74/24
 Chalk aquifers

Lower Greensand aquifers

Jurassic Limestone aquifers

Permo-Triassic sandstone aquifers

Magnesian Limestone aquifers

Post-Carboniferous
aquitards and minor aquifers

Carboniferous Limestone aquifers

Devonian and Carboniferous

aquitards and minor aquifers

Pre-Devonian strata

Igneous and metamorphic rocks

Dumfries
Basin

Yorkshire
Chalk

Upland Britain

Lincolnshire
Chalk

North
Downs

South Downs

Wessex Basin