Presenting research from a wide range of scientists working on intertidal

sediments - both mud flats and saltmarshes - this book is of importance to all
environmental scientists.
The individual chapters explore the underlying biogeochemical processes
controlling the behaviour of carbon, the nutrients nitrogen and phosphorus, and
contaminants such as toxic organics, trace metals and artificial radionuclides in
intertidal environments. The biogeochemistry of these environments is critical to
understanding their ecology and management. All of the chapters include both a
comprehensive review and the results of recent research. The authors are active
researchers and the book brings together their different perspectives on this
diverse and ecologically important environment.
This book is designed for researchers and managers working on intertidal
environments, but it will also serve as a valuable senior undergraduate and
graduate reference text in environmental chemistry, environmental science,
earth science and oceanography.
BIOGEOCHEMISTRY OF INTERTIDAL SEDIMENTS
CAMBRIDGE ENVIRONMENTAL CHEMISTRY SERIES

Series Editors:
P.G.C. Campbell, Institut National de la Recherche Scientifique,
Universite du Quebec a Quebec, Canada
R.M. Harrison, School of Chemistry, University of Birmingham, UK
SJ. De Mora, Department d'oceanographie, Universite du
Quebec a Rimouski, Canada

Other books in the series:

P. Brimblecombe Air Composition and Chemistry Second Edition
A.C. Chamberlain Radioactive Aerosols
M. Cresser and A. Edwards Acidification of Freshwaters
M. Cresser, K. Killham and A. Edwards Soil Chemistry and its Applications
R.M. Harrison and S.J. de Mora Introductory Chemistry for the
Environmental Sciences Second Edition
S.J. de Mora Tributyltin: Case Study of an Environmental Contaminant
Biogeochemistry of
intertidal sediments

Edited by
T . D . J I C K E L L S AND J . E . RAE

CAMBRIDGE
UNIVERSITY PRESS
CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sao Paulo

Cambridge University Press

The Edinburgh Building, Cambridge CB2 2RU, UK

Published in the United States of America by Cambridge University Press, New York

www.cambridge.org

Information on this title: www.cambridge.org/9780521483063

© Cambridge University Press 1997

This book is in copyright. Subject to statutory exception

and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without
the written permission of Cambridge University Press.
First published 1997

This digitally printed first paperback version 2005

A catalogue recordfor this publication is available from the British Library

Library of Congress Cataloguing in Publication data

Biogeochemistry of intertidal sediments / [edited by] T.D. Jickells and
J.E. Rae.
p. cm. — (Cambridge environmental chemistry series; 9)
Includes index.
ISBN 0 521 48306 9
1. Marine sediments. 2. Biogeochemistry. I. Jickells, T.D.
(Tim D.) II. Rae, J.E. (Joy E.) III. Series.
GC380.15.B55 1997
551.46'083-dc20 96-38998 CIP
ISBN-13 978-0-521-48306-3 hardback
ISBN-10 0-521-48306-9 hardback
ISBN-13 978-0-521-01742-8 paperback
ISBN-10 0-521-01742-4 paperback
 Contents
o o o o o o o o o o o o o o o o o o o

List of contributors ix
Preface xi

1 Biogeochemistry of intertidal sediments 1

T.D. Jickells and J.E. Rae

2 Trace metals in deposited intertidal sediments 16

J.E. Rae
3 Modelling adsorption and desorption processes
in estuaries 42
A. Turner and A.O. Tyler

4 A critical appraisal of the methodology used in studies of

material flux between saltmarshes and coastal waters 59
K. Carpenter

5 Nutrient recycling in intertidal sediments 84

SJ. Malcolm and D.B. Sivyer

6 An overview of carbon and sulphur cycling in

marine sediments 99
G. Ruddy
7 Microbial activity and diagenesis in saltmarsh sediments.
North Norfolk, England 119
K. Pye, M.L. Coleman and W.M. Duan

8 The behaviour of radionuclides in the coastal and

estuarine environments of the Irish Sea 152
P. McDonald and S.R. Jones

9 The sorption of hydrophobic pyrethroid insecticides to

estuarine particles: a compilation of recent research 173
J.L. Zhou and SJ. Rowland

Index 189
 Contributors
ooooooooooooooooooo

K. Carpenter, Hydraulics Research, Wallingford, Howbery Park,

Wallingford, Oxfordshire, OX10 8BA, UK
M.L. Coleman, Postgraduate Research Institute for Sedimentology,
University of Reading, PO Box 227, Whiteknights Park, Reading,
RG6 2AB, UK
W.M. Duan, Postgraduate Research Institute for Sedimentology,
University of Reading, PO Box 227, Whiteknights Park, Reading,
RG6 2AB, UK
T.D. Jickells, School of Environmental Studies, University of East
Anglia, Norwich, NR4 7TJ, UK
S.R. Jones, The Princess Royal Building, Westlakes Research Institute,
Westlakes Science and Technology Park, Moor Row, Cumbria,
CA24 3LN, UK
S.J. Malcolm, Ministry of Agriculture, Fisheries and Food, Directorate
of Fisheries Research, Pakefield Road, Lowestoft, Suffolk, NR33 OHT,
UK
P. McDonald, The Princess Royal Building, Westlakes Research
Institute, Westlakes Science and Technology Park, Moor Row,
Cumbria, CA24 3LN, UK
K. Pye, Postgraduate Research Institute for Sedimentology, University
of Reading, PO Box 227, Whiteknights Park, Reading, RG6 2AB, UK
J.E. Rae, Postgraduate Research Institute for Sedimentology, University
of Reading, PO Box 227, Whiteknights Park, Reading, RG6 2AB, UK
S.J. Rowland, Petroleum and Environmental Chemistry Group,
Department of Environmental Sciences, University of Plymouth, Drake
Circus, Plymouth, PL4 8AA, UK
Contributors

G. Ruddy, School of Environmental Studies, University of East Anglia,

Norwich, NR4 7TJ, UK. (Present address Plymouth Marine
Laboratory, Citadel Hill, Plymouth, PL1 2PB, UK)
D.B. Sivyer, Ministry of Agriculture, Fisheries and Food, Directorate of
Fisheries Research, Pakefield Road, Lowestoft, Suffolk, NR33 OHT, UK
A. Turner, Department of Environmental Sciences, University of
Plymouth, Drake Circus, Plymouth, PL4 8AA, UK
A.O. Tyler, BMT Port & Coastal Limited, Grove, House, 5 Ocean
Way, Ocean Village, Southampton, SOI 1TJ, UK
J.L. Zhou, Petroleum and Environmental Chemistry Group,
Department of Environmental Sciences, University of Plymouth, Drake
Circus, Plymouth, PL4 8AA, UK
 Preface
o o o o o o o o o o o o o o o o o o o

This book arises from a meeting of the same name held at the Postgraduate
Research Institute for Sedimentology, Reading, UK, in April 1994. The
meeting was sponsored by the Challenger Society and the Geochemistry
Group of the Mineralogical Society: we are grateful to both these
organisations for their generous support. Eight scientists active in
research on the biogeochemistry of intertidal systems were invited to
present papers on their own particular interests within the field. We
deliberately selected a mixture of established and young UK scientists and
invited them to both briefly review thefieldand to present some new data
from their own research. These papers have now been formalised into this
book together with an overview chapter written by ourselves. The various
authors have taken different approaches to the task we set them: some
have achieved a balance between review and current research, whereas
others have chosen to concentrate mainly on either review or recent
research. As editors we have encouraged their diversity, allowing the
authors the freedom to deal with the subject as they see best. We wish to
thank all the authors for their good humour and hard work. All the
chapters have been peer reviewed so we are additionally grateful to D.
Cooper, A. Grant, J. Hamilton-Taylor, D. Hydes, S. Malcolm, R.
Mortimer, J. Scudlark and S. Wakefield. (Chapter one was reviewed by
the authors of the other book chapters.)
The work presented here demonstrates that despite the diversity of
intertidal habitats there are common biogeochemical principles in operation.
It is our hope that the book might enhance the appreciation and
understanding of these complex and beautiful environments and that in so
doing it might contribute to their preservation.
This book is University of Reading PRIS Contribution No 441.
 1
o o o o o o o o o o o o o o o o o o o
Biogeochemistry of intertidal
sediments
T.D. Jickells and J.E. Rae

In some places the boundary between land and sea is in the form of abrupt
and often spectacular cliffs but, elsewhere, the boundary can take the form
of a complex environment of intertidal sediments. These environments
include shingle banks, sandy beaches, mud flats, saltmarsh and mangrove
(or mangal) communities. In some cases one or other of these environments
will occur, in others they will be associated with one another. For
example, on many North Sea and North American East Coast shorelines,
mud flats grade into saltmarshes behind the shelter of shingle spits and
sand dunes. In general, saltmarshes and mangroves occupy similar
ecological niches with mangroves at lower latitudes (winter temperatures
greater than 10°C) and saltmarshes at high latitudes, though in some
locations both communities coexist (Chapman, 1977).
The considerable global significance of these intertidal systems is
evident from Figure 1.1. Around the coast of Britain alone there are 44 370
hectares (ha) of saltmarsh (Allen & Pye, 1992) and 589 429 ha on the US
East Coast (Reimold, 1977) with a total global area of 3.8 x 10 7 ha
(Steudler & Peterson, 1984 and references therein). There are 365 500 ha of
mangal forest around the Indian subcontinent and another 250 000 ha in
the Mekong delta (Blasco, 1977). The total global area of mangal is about
2.4 x 10 7 ha (Twilley, Chen & Hargis, 1992). The total area of intertidal
sediments is likely to be similar to that of adjacent saltmarshes and mangals.
These intertidal environments afford very effective coastal protection
(Brampton, 1992). The importance of such coastal protection is increasing
as a result of two processes. Firstly, the growth in world population is
increasing the pressure on land in general. Moreover, the increase in
populations in coastal areas is disproportionately large, compounding
this pressure (Hinrichsen, 1994). Secondly, the process of sea level rise is
threatening low lying land with increased risk of flooding. Sea level rise is
 JANUARY

Mangal
Saltmarsh

Figure 1.1 Map showing global distribution of saltmarshes and mangroves (mangal) in relation to average temperature, redrawn
from Chapman, 1977.
 Biogeochemistry of intertidal sediments 3

in part a natural process, as the coastal system slowly responds to the end
of the last glaciation, but is now being exacerbated by anthropogenically
induced climate change - the greenhouse effect (e.g. Wigley & Raper,
1992). The threat to coastal areas from climate change arises not just from
sea level rise associated with global warming, but also potentially from
changes in weather patterns and hence wave and current patterns (Tooley,
1992).
In addition to providing a valuable sea defence, intertidal sediments
offer an important habitat for wildlife, food and recreation. The very
nature of intertidal areas has left them relatively undisturbed by human
activity compared to inland areas. This coupled with the rich food supply
in the muddy sediments means that many intertidal areas are now very
important wildlife sanctuaries and nursery grounds for fish and inverte-
brates (Adam, 1990). Thus, for example, up to 12 million birds of 50
different species live for at least part of the year on the vast shallow water
or intertidal muds of the Wadden Sea off the north Dutch and German
coast (North Sea Task Force, 1993). Beside these quantifiable environmental
roles as wildlife habitats and coastal defences, these intertidal areas have
an intrinsic beauty that has always attracted people.
For all these reasons, intertidal sedimentary environments are important
areas for scientists to study, and there is a long history of such work. This
has often focused on the geomorphology of these areas and how this can be
used to interpret the geological record (e.g. Nummedal, Pilkey & Howard,
1987; Allen and Pye, 1992) and the ecology of these areas (e.g. Adam, 1990;
Chapman, 1977; Mathieson and Nienhuis, 1991). In addition many studies
have concentrated on one type of ecosystem (e.g. saltmarsh or mangal).
This book aims to take a different and complementary approach. Firstly, it
focuses on the chemistry of these systems as modulated by and interacting
with the geological and biological environment - hence the term
biogeochemistry. Secondly, we have drawn no distinction between the
different types of intertidal ecosystem since we believe the fundamental
biogeochemical processes are the same in all the systems, though of course
the final chemical system observed by our measurement varies as a function
of the ecosystem. Thus the biogeochemical environment of a mangal
swamp and an arctic mud flat may be very different, but the fundamental
principles regulating these environments will be similar. Some of the
authors in this book have even drifted out into more open waters to
illustrate their points; something that only serves to emphasise both the
generality of the biogeochemical principles and also the limited research
effort expended into some aspects of the biogeochemistry of intertidal areas.
4 T.D. Jickells and J.E. Rae

The detrital mineral phases of the intertidal sediments (e.g. clays and
quartz) are of limited geochemical interest in the context of this book.
Rather it is the organic matter and the chemicals adsorbed to the mineral
phases of the sediment and to one another that drive most of the
geochemical processes. Adsorption to sediments is a function of sediment
surface area and hence particle size, with the finest sediments adsorbing
the most material, as illustrated in the chapters by McDonald & Jones,
Zhou & Rowland, and Rae. This means that in a geochemical (and also an
ecological) sense the coarse sediments of high energy environments like
shingle and sand beaches are of limited interest. The fine-grained organic
matter which feeds the flora and fauna and fuels the geochemical reactions
is washed out of these areas, to accumulate in the low energy environments
such as the mud flats, mangals and saltmarshes that are the main focus of
this book. Hence, the sunbathers can keep the sandy beaches and leave the
geochemists to wallow in the mud!
These areas of fine sediment accumulate in both estuarine and open
coast environments, though in the latter case they usually form in areas of
reduced tidal and wave energy, such as behind barrier island complexes or
in embayments (Allen & Pye, 1992). As noted earlier, in such environments
there may be a continuum from sand dune through saltmarsh to mud
flats, and the relative sizes of these can vary greatly. The marshes of Sapelo
Island Georgia (USA) occupy 75% of the available coastal lagoon
(Wiegert, Pomeroy & Wiebe, 1981) and in the 'big swamp' phase (6000 yr
BP) many Australian estuaries were almost completely filled with mangal
forest, to be replaced more recently by a mixture of mangal and mud flats
(Woodroffe, 1990). By contrast in the sheltered embayment of the Wash
coast in England (Malcolm & Sivyer, this volume) about 4000 ha of
saltmarsh (Doodey, 1992) are fronted by about 30 000 ha of intertidal mud
flats (S.J. Malcolm, pers. comm.).
These areas of sediment accumulation are transitory features on a
whole range of time scales. They have migrated dramatically over the
Holocene period (last 10000 years) as relative sea levels have risen
globally due to deglaciation, though this pattern of sea level change has
varied locally (Allen & Pye, 1992). This is illustrated for the North Sea in
Figure 1.2, and similar changes occurred in other coastal areas. As noted
earlier, increases in sea level over the next 100 years may accelerate as a
result of the greenhouse effect. Intertidal systems have responded to these
changes by migrating where this is practical or accreting sediment
sufficiently quickly to retain their position relative to sea level (e.g.
Funnell & Pearson, 1989; Tooley, 1992). Maximum marsh vertical
 Biogeochemistry of intertidal sediments 5

accretion rates appear to be 10-20 mm y" 1 , similar to maximum pre-

dicted rates of sea level rise, so, provided sediment supply is maintained
(which may depend on policies of coastal defence in sediment supply
regions), marshes should be able to keep pace with sea level rise
(Boorman, Goss-Custard & McGrorty, 1989; Reed, 1990). Because of the
longer life span and slower growth rates of mangrove trees, mangals may
not be able to respond so readily (Woodroffe, 1990).
Predicting the exact response of the whole intertidal sedimentary
ecosystem to sea level rise is extremely difficult, though progress is being
made, often with surprising results. Thus in the Wadden Sea, a rise in sea

Figure 1.2 Map showing estimated locations of shore lines in the North
Sea from 18000 years ago to present, based on Jelgersma, 1979.
18 000 yr BP, sea level 130 m below present day, 10300
yr BP, sea level 65 m below present day, 8700 yr BP,
sea level 36 m below present day, 7800 yr BP, sea level 20 m below
present day.
6 T.D. Jickells and J.E. Rae

level may result in increased exposure of intertidal areas and hence

increased primary production (Peerbolte, Eysink & Ruardij, 1991). In
areas where sediment supply is not maintained, loss of the intertidal
ecosystems will occur. This is evident in areas such as the Nile and
Mississippi deltas where damming has dramatically reduced sediment
supply. This, combined with local subsidence, has meant that sea level rise
exceeds sedimentation rates and contributes to marsh deterioration, at
least in Louisiana (Reed, 1990), but probably also in the Nile delta, and
indirectly in the Indus delta (Halim, 1991).
Other human impacts on intertidal systems are less subtle with a long
history of drainage and protection from inundation to yield agricultural
land. The data of Doodey (1992) imply that the UK has lost over 50% of
its saltmarshes to reclamations that have gone on since Roman times and
continue to the present day. In Holland, 40% of the present country is
reclaimed land (Walker, 1990). In recent years in Europe, agricultural
surpluses and rising sea levels have begun to encourage policies to reverse
this trend and to allow recreation of marsh systems.
On shorter time scales intertidal systems are of course defined by the
dramatic diurnal tidal cycles of inundation and drying out, and the
saltmarsh ecological zonation by the spring/neap tidal cycles (Adam,
1990). On slightly longer time scales, storm events may cause dramatic
short term erosion, though the systems can recover from this, often by
exchange of sediments between the marshes and the adjacent intertidal
mud flats (Pethick, 1992).
These short term changes present an enormous challenge to scientists
studying intertidal systems as illustrated here in the chapters by Carpenter,
Malcolm & Sivyer and Ruddy. Sampling at low tide in these systems is a
relatively straightforward matter, though the terrain is often difficult and
dangerous with the added requirement of returning to dry land before the
tide returns. Sampling at high tide is more difficult but marine scientists
have a wide range of tools for such tasks, and indeed for such expeditions
the problem is one of retreating to sea before the tide falls and strands
researchers. Mangal forests and swamps are difficult to penetrate at any
state of the tide though some nutrient flux data from these systems are
now becoming available (Boto & Robertson, 1990; Lugo, Brown &
Bronson, 1988; Rivera-Monroy et a/., 1995). However, the real problem
for scientists working in intertidal systems is coping with the dramatic and
rapid changes from a system under water to one above water, and one
where water flows reverse and change on very short time scales. The
contrasting approaches of Carpenter, Ruddy and Malcolm & Sivyer in
 Biogeochemistry of intertidal sediments 7

this volume illustrate two of the available strategies. These approaches are
designed to address changes associated with diurnal tidal rhythms and
also spring/neap tidal cycles. Pethick (1992), French & Spencer (1993) and
Nyman, Crozier & de Laune (1995) present evidence that storm events
may dominate sedimentation on saltmarshes. Such events are extremely
difficult to study by direct observations because of their unpredictability,
as well as the dangers of working in these environments under such
conditions. We therefore have no direct measurements of fluxes during
storm events, though they can be indirectly estimated (e.g. Nyman,
Crozier & de Laune, 1995). Therefore we must be extremely cautious in
extrapolating data from relatively low energy situations to such high
energy conditions.
The primary productivity of many of these intertidal communities is
high relative to offshore communities - 200-500 gC m~ 2 yr" 1 for
saltmarshes and mangroves (Mann, 1982) and about 100 gC m~ 2 yr" 1
for benthic algae on mud flats (Cadee & Hegeman, 1974). This high
productivity is supported in part by very high rates of N 2 fixation.
Capone and Carpenter (1982), for example, estimate that almost half of
all marine N 2 fixation takes place in saltmarshes and mangroves, though
more recent data (e.g. Howarth et a/., 1988; Boto & Robertson, 1990)
suggests substantially lower rates. Very little of the productivity of
marshes and mangroves appears to be grazed while alive, but is rather
decomposed as detritus by micro-organisms (Mann, 1982; Adam, 1990)
partly in or on the sediments, together with the products of benthic algal
production, and partly after export to adjacent coastal waters. Based on
studies of carbon isotopes in particulate organic matter in saltmarsh
creeks, this export appears to be minor (Adam, 1990) though the subject
of considerable debate (Carpenter, this volume). For nitrogen, even if net
exchanges are minor, transformations may occur such as from dissolved
to particulate forms (e.g. Rivera-Monroy et a/., 1995) or nitrate to
ammonium (Carpenter, this volume). Thus much of this plant detritus,
together with decomposing below-ground plant material and organic
carbon associated with imported fine sediment, provides the sediments
with a rich supply of organic matter. This organic matter ultimately feeds
the invertebrates and birds of these areas and also drives the geochemical
reactions in the sediments.
It is the decomposition of this organic matter which provides one of the
unifying biogeochemical concepts in all these intertidal environments. In
all areas subject tofine-grainedsedimentation, organic matter content of
sediments is relatively high. This organic matter is subsequently oxidised
8 T.D. Jickells and J.E. Rae

by bacteria using a variety of oxidising agents or terminal electron

acceptors (TEA) which are themselves reduced, hence the term redox
reactions. These different TEAs yield different amounts of energy from
their reactions with organic carbon (Table 1.1), so the microbial community
using the highest energy yielding system always dominates, assuming
there is a significant amount of that TEA present. Thus, as long as there is
oxygen present, this will always be the oxidising agent used, but, once this
is exhausted, alternative oxidants are used in a fixed sequence beginning
with nitrate (Coleman, 1985; Ruddy, this volume). It is for this reason that
these intertidal sediments can consume large amounts of nitrate very
quickly (Malcolm & Sivyer, this volume). In many systems nitrate levels
are naturally low, but increasing concentrations of nitrate in rivers arising
from human activity can mean that nitrate is an important electron
acceptor in some coastal systems today. Once nitrate is exhausted in the
sediments the readily available oxidised iron and manganese are reduced
to their more soluble reduced form. Since phosphate in sediments is
usually strongly associated with iron, this iron mobilisation can also

Table 1.1 Redox equilibria (based on Andrews et ai, 1995)

Aerobic respiration

AG = - 29.9 kcal/mol
Denitrification
| N O 3 - + CH2O = | H + ->|N 2 4- CO2 + lH 2 O
AG= - 28.4 kcal/mol
Manganese reduction
2MnO2 + CH2O + 2HCCV + 2H+ ->2MnCO3 + 3H2O + CO2
AG = - 23.3 kcal/mol
Nitrate reduction
i N ( V + CH2O + H + - ^ N H 4 + + CO2 + ^H2O
AG = - 19.6 kcal/mol
Iron reduction
4FeOOH + 4HCO 3 " + 4H + + CH 2 O^4FeCO 3 + 7H2O + CO2
AG = - 12.3 kcal/mol
Sulphate reduction
iSO 4 2 " + CH2O + ^H + - ^ H S " + CO2 + H2O
AG = - 5.9 kcal/mol
Methane formation
CH 2 O-^iCH 4 + iCO 2
AG= - 5 . 6 kcal/mol

CH2O is a simplified representation of organic matter

 Biogeochemistry of intertidal sediments 9

result in phosphorus mobilisation (Sundby, Gobeil, Silverberg & Mucci,

1992). These reactions involving the important nutrients, nitrate and
phosphorus, illustrate the ways in which intertidal sediment processes can
influence the productivity of the intertidal ecosystems and potentially the
adjacent coastal waters. These potential effects provide a rationale for flux
studies such as those of Carpenter (this volume) and similar studies in
mangals (e.g. Rivera-Monroy et a/., 1995). Once the reservoirs of available
oxidised iron and manganese are depleted, sulphate reduction begins as
discussed by Ruddy (this volume).
In deep ocean waters, organic carbon inputs, microbial activity,
sedimentation and bioturbation rates are relatively low and the various
TEA reaction zones occur as distinct vertical layers. As the chapter by
Ruddy emphasises, this is not the case in these intertidal sediments where
organic inputs are high and heterogenous, giving rise to high benthic
bioturbation rates. Heterogeneity may be particularly important in
mangal forests where there is evidence that the different physiology and
biochemistry of tree species may result in markedly different patterns of
accumulation and degradation of organic matter in sediments (Lacerda,
Ittekkot & Patchineelam, 1995). Together these factors mean that the
different sediment reaction zones will be concentrated around micro-
environments, yielding complex sediment profiles of chemicals involved
in these redox cycles and a variety of resultant diagnetic mineral
assemblages (Coleman, 1985). Pye et al. in this book consider the
mineralogical evidence for the mechanisms of carbon oxidation and
further emphasise that the classic theories of clear transitions from one
TEA zone to another with depth in the core are inappropriate. This
chapter also emphasises that microbiological investigations can provide a
useful insight into the controls on these processes.
Estimations offluxesout of these intertidal sediment systems cannot be
based solely on passive and diffusional flux measurement. Rather such
flux estimates must take account of the active exchange arising from
burrow irrigation, the effect of the tidal 'pumping' (arising from the
drainage of sediment pore waters at low tide and their recharge at high
tide which can affect both water and gas exchange, Nuttle & Hemond,
1988) and the effect of uptake by algal and nutrient communities at the
surface of the sediments. Many of these biologically mediated processes
(bioturbation, plant growth, benthic algae growth) will have a marked
seasonality in temperate regions. In addition, redox reactions in sediments
can show a seasonality both directly via the seasonal input of organic
matter from the growth cycle of marsh plants (at least in temperate
10 T.D. Jickells and J.E. Rae

regions) and also via the seasonal production of organic ligands in pore
waters by higher plants which can regulate the sedimentary iron cycle
(Luther et a/., 1992). Redox and pH conditions can also change on short
time scales (hours) as a result of tidal pumping and intrusion of air during
the tidal drying of tidal flats (Kerner & Wallmann, 1992). All these
complications mean that studies of fluxes from these intertidal systems
tend to take one of two forms; either involving studies of individual
sediment cores to assess rates of individual processes or alternatively
fluxes from the whole system. Both are difficult and uncertain as the
chapters by Carpenter and Malcolm illustrate. This means that we are a
long way from currently being able to effectively quantifyfluxesin and out
of these intertidal sediment systems. The chapter by Malcom & Sivyer in
particular does, however, illustrate the potential importance of these
fluxes to coastal ecosystems in general.
A further consequence of the oxidation of organic carbon by a sequence
of TEAs is the production of various gaseous products including reduced
sulphur species such as dimethyl sulphide (Steudler & Peterson, 1984) and
carbonyl sulphide (Chin & Davis, 1993), methane and N 2 O. These gases
all have potential climate modifying roles (Houghton, Callander and
Vaney, 1992). For the sulphur gases, saltmarshes (and possibly all
intertidal systems) appear to produce relatively highfluxesper unit area,
though their contributions to globalfluxes,in comparison to ocean and
terrestrial soil sources which occur over much larger areas, is necessarily
modest (Charlson et al, 1987). Methane emissions from wetlands are also
potentially important (Dacey, Drake & Klug, 1994) though saltmarshes
are likely to be small sources compared to terrestrial wetlands because of
the greater role in marine systems of sulphate reduction which precedes
methanogenesis (Table 1.1; Harriss et a/., 1988). However, Barber, Burke
& Sackett (1988) have reported very high methane fluxes from Florida
mangrove environments which they suggest may reflect rapid consumption
of sulphate in these organic rich environments. Malcolm & Sivyer (this
volume) indicate the potential for large scale N 2 O emissions from
intertidal sediments as a biproduct of nitrate reduction.
Since all these gas emissions are byproducts of organic matter
decomposition they are natural sources, but the magnitude of these fluxes
may be altered by human intervention both directly, via destruction of
intertidal sediments, and also indirectly, via increases in atmospheric CO 2
levels (Dacey, Drake & Klug, 1994) or changes in nitratefluxes(Malcolm,
this volume). Thus while intertidal sediments are very sensitive to climate
change, they are not wholly passive since they can themselves influence
 Biogeochemistry of intertidal sediments 11

climate change via atmospheric trace gas cycling. However, the global
significance of this is inevitably limited by their areal extent at the land/sea
interface.
On a global scale, shelf sediments appear to be major sinks for organic
carbon storage (Wollast, 1991) and hence important regulators of
atmospheric CO 2. Within the shelf system, wetlands may contain a third
of this carbon storage (Twilley, Chen & Hargis, 1992). The bulk of this
wetland storage is in mangal systems, largely because of the slower decay
of the woody material (Twilley, Chen & Hargis, 1992). The relatively slow
degradation of the woody mangrove material, compared to the detritus of
saltmarshes, also means that the residence time of carbon in the mangal
system will be longer. A further consideration in some areas is the long
term fate of these organic rich sediments during sea level rise. Some may
be buried but others eroded and perhaps oxidised.
The other major unifying geochemical theme in intertidal systems is the
adsorption of a wide range of chemicals to the sediments. As noted earlier
this is a strong function of sediment particle size, but it also depends on the
redox state of the sediments, the chemical behaviour of the particular
component of interest, the physico-chemical nature of the sediment, and
interstitial water salinity as discussed in the chapters by McDonald &
Jones, Zhu & Rowland, Turner & Tyler and Rae. This adsorption allows
intertidal sediments to provide an historical record of environmental
contamination, provided this record is interpreted with care to allow for
the complications noted above, as shown in the chapter by Rae. Intertidal
sediments can represent important sinks for contaminants, carbon and
nutrients, and, as with nitrate reduction (Malcolm & Sivyer, this volume),
can represent a mechanism for mitigating coastal waters contamination.
However, this storage can also leave these systems potentially vulnerable
to pollution, though apart from the direct effects of oil pollution and
enhanced algal growth due to inputs, documented pollution of intertidal
systems is limited (Adam, 1990).
The other critical issue with storage of contaminants in intertidal
systems is their potential for re-release - or 'chemical bombs' in the
terminology of Liss et al. (1991). Such release could occur via erosion or
changes in the chemical environment in terms of, for example, redox
(Kerner & Wallmann, 1992) or salinity. An interesting natural example of
such release in an undisturbed system is provided by barium and 226 Ra in
the Ganges/Brahmaputra system. Here barium and 226Ra rich sedimentary
material is deposited under low salinity conditions in mangal swamps
during winter high riverflowconditions. In summer under low river flows,
12 T.D. Jickells and J.E. Rae

saline waters intrude further into the mangal system leading to barium
and 226 Ra desorption and flux to sea (Carroll et al, 1993). This process
may be of widespread significance for barium cycling (Coffey et al, in
press). In the case of long lived radionuclides, the potential hazard arising
from re-release from sediments persists for hundreds, or even thousands of
years, but for other contaminants, cessation of inputs, dilution by
uncontaminated sediments and slow degradation of 'persistent' organic
molecules probably reduce the time scales over which re-release represents
a major concern to tens or hundreds of years. Given human use of
intertidal sedimentary environments for recreation and food supplies,
adjacent waste discharge is clearly unsatisfactory.
However, the bigger threats to these intertidal systems almost certainly
come from our apparently endless appetite for destroying them to
reclaim agricultural land or for coastal 'development'. Faced with the
inevitability of sea level rise it is to be hoped that this trend will now
reverse and allow us to celebrate and encourage these beautiful and
complex environments.

Acknowledgement
We would like to thank all the authors of other chapters in the book
whose useful comment improved the manuscript. TDJ wishes to thank
Kate Carpenter, Tom Church and Joe Scudlark for all they have taught
him about saltmarshes. This chapter represents a contribution of the
JONUS programme funded by MAFF, DoE and NRA. It is University of
Reading PRIS Contribution No. 442.

References
Adam, P. (1990) Saltmarsh Ecology, Cambridge University Press, Cambridge, 461 pp.
Allen, J.R.L. & Pye, K. (1992) Coastal saltmarshes: their nature and importance. In
Saltmarshes, 1-18, eds. J.R.L. Allen and K. Pye, Cambridge University Press,
Cambridge, 184 pp.
Andrews, J.E., Brimblecombe, P., Jickells, T.D. & Liss, P.S. (1995) An Introduction to
Environmental Chemistry, Blackwell Science, Oxford.
Barber, T.R., Burke, R.A. & Sackett, W.M. (1988) Diffuse flux of methane from warm
wetlands. Global Biogeochem. Cycl. 2, 411-25.
Blasco, F. (1977) Outlines of ecology, botany and forestry of the mangals of the Indian
subcontinent. In Wet Coastal Ecosystems, 241-60, ed. V.J. Chapman, Elsevier,
Amsterdam, 428 pp.
Boorman, L.A., Goss-Custard, J.D. & S. McGrorty (1989) Climate change, rising sea level
and the British Coast, ITE Research Publication no. 1, NERC, HMSO, London,
24 pp.
 Biogeochemistry of intertidal sediments 13

Boto, K.G. and Robertson, A.I. (1990) The relationship between nitrogen fixation and
tidal exports of nitrogen in a tropical mangrove system. Est. Coastal Shelf Sci. 31,
531-40.
Brampton, A.H. (1992) Engineering significance of British saltmarshes. In Saltmarshes,
115-22, eds. J.R.L. Allen and K. Pye, Cambridge University Press, Cambridge,
184 pp.
Cadee, G.C. & Hegeman, J. (1974) Primary production of the benthic microflora living on
tidal flats in the Dutch Wadden Sea. Neth. J. Sea Res. 8, 260-91.
Capone, D.G. & Carpenter, E.J. (1982) Nitrogen fixation in the marine environment.
Science 217, 1140-2.
Carroll, J , Falkner, K.K., Brown, E.T. & Moore, W.S. (1993) The role of the
Ganges-Brahmaputra mixing zone in supplying Ba and 2 2 6 Ra to the Bay of
Bengal. Geochim. Cosmochim. Acta 57, 2981-990.
Chapman, V.J. ed. (1977) Wet Coastal Ecosystems. Elsevier, Amsterdam, 428 pp.
Charlson, R J., Lovelock, J.E., Andreae, M.O. & Warren, S.G. (1987) Oceanic phytoplankton,
atmospheric sulphur, cloud albedo and climate. Nature 326, 655-61.
Chin, M. & Davis D.D. (1993) Global sources and sinks for OCS and CS 2 and their
distributions. Global Biogeochem. Cycl. 7, 321-37.
Coffey, M., Dehairs, F., Collette, O., Luther, G., Church, T. & Jickells, T.D., The
behaviour of dissolved barium in estuaries. Est. Coastal Shelf Sci. (in press).
Coleman, M.L. (1985) Geochemistry of diagenetic non-silicate minerals: kinetic consider-
ations. Phil. Trans. R. Soc. London A, 315, 39-56.
Dacey, J.W.H., Drake, B.G. & Klug, MJ. (1994) Stimulation of methane emission by
carbon dioxide enrichment of marsh vegetation. Nature 370, 47-9.
Doodey, J.P. (1992) The conservation of British saltmarshes. In Saltmarshes, 80-114, eds.
J.R.L. Allen and K. Pye, Cambridge University Press, Cambridge, 184 pp.
French, J.R. & Spencer, T. (1993) Dynamics of sedimentation in a tide-dominated back
barrier salt marsh, Norfolk, UK. Mar. Geol. 110, 315-31.
Funnell, B.M. & Pearson, I. (1989) Holocene sedimentation on the North Norfolk barrier
coast in relation to relative sea-level change. J. Quat. Sci. 4, 25-36.
Halim, V. (1991) The impact of human alterations on the hydrological cycle on ocean
margins. In Ocean Margin Processes in Global Change, 301-27, eds. R.F.C.
Mantoura, J-M. Martin and R. Wollast, Wiley, Chichester, 469 pp.
Harriss, R.C., Sebacher, D.I., Bartlett, K.B., Bartlett, D.S. & Gill, P.M. (1988) Sources of
atmospheric methane in the south Florida environment. Global Biogeochem.
Cycl. 2, 231-43.
Hinrichsen, D. (1994) Coasts under pressure. People and the Planet 4, 6-9.
Houghton, J.T., Callander, B.A. & Vaney, S.K. (1992) Climate Change 1992. The
Supplementary Report to the IPCC Scientific Assessment. Cambridge University
Press, Cambridge, 200 pp.
Howarth, R.W., Marino, R., Lane, J. & Cole, J.J. (1988) Nitrogen fixation in freshwater,
estuarine and marine ecosystems. 1. Rates and importance. Limnol. Oceanogr. 33,
669-87.
Jelgersma, S. (1979) Sea-level changes in the North Sea basin in The Quaternary History of
the North Sea. Acta Univ. Ups. Symp. Univ. Dps. Annum. Quingentesium
Celebrantis: 2, Uppsala, 233-48.
Kerner, M. & Wallmann, K. (1992) Remobilisation events involving Cd and Zn from
intertidal flat sediments in the Elbe estuary during the tidal cycle. Est. Coastal
Shelf Sci. 35, 371-93.
Lacerda, L.D., Ittekkot, V. & Patchineelam, S.R. (1995) Biogeochemistry of mangrove soil
14 T.D. Jickells and J.E. Rae

organic matter: a comparison between Rhizophora and Avicennia in south-eastern

Brazil. Est Coastal Shelf Sci. 40, 713-20.
Liss, P.S., Billen, G., Duce, R.A., Gordeev, V.V., Martin, J-M., McCave, I.N., Meincke, J.,
Milliman, J.D., Sicre, M-A., Spitzy, A. & Windom, H.L. (1991) What regulates
boundary fluxes at ocean margins. In Ocean Margin Processes in Global Change,
111-26, eds. R.F.C. Mantoura, J-M. Martin and R. Wollast, Wiley, Chichester,
469 pp.
Lugo, A.E., Brown, S. & Bronson, M.M. (1988) Forested wetlands in freshwater and
salt-water environments. Limnol. Oceanogr. 33, 894-909.
Luther, G.W., Kostka, J.E., Church, T.M., Sulzberger, B. & Stumm, W. (1992) Seasonal
iron cycling in the salt-marsh sedimentary environment: the importance of
ligand complexes with Fe(II) and Fe(III) in the dissolution of Fe(III) minerals
and pyrite respectively. Mar. Chem. 40, 81-103.
Mann, K.H. (1982) Ecology of Coastal Waters, Blackwell, Oxford, 322 pp.
Mathieson, A.C. & Nienhuis, P.H. (1991) Intertidal and littoral ecosystems, Elsevier,
Amsterdam, 564 pp.
North Sea Task Force (1993) North Sea Quality Status Report. Oslo and Paris
Commissions, London. Olsen and Olsen, Fredensborg, Denmark, 132 pp.
Nummedal, D., Pilkey, O.H. & Howard, J.D. eds. (1987) Sea-level fluctuation and coastal
evolution, Spec. Pub. Soc. Economic Palaeontologists and Mineralogists 41,
Tulsa, USA, 267 pp.
Nuttle, W.K. & Hemond, H.F. (1988) Salt marsh hydrology: implications for biogeochemical
fluxes to the atmosphere and estuaries. Global Biogeochem. Cycl. 2, 91-114.
Nyman, J.A., Crozier, C.R. & de Laune, R.D. (1995) Roles and patterns of hurricane
sedimentation in an estuarine marsh landscape. Est. Coastal Shelf Sci. 40,665-79.
Peerbolte, E.B., Eysink, W.D. & Ruardij, P. (1991) Morphological and ecological effects of
sea level rise: an evaluation for the Western Wadden Sea. In Ocean Margin
Processes in Global Change, 329—47, eds. R.F.C. Mantoura, J-M. Martin and R.
Wollast, Wiley, Chichester, 469 pp.
Pethick, J.S. (1992) Saltmarsh geomorphology. In Saltmarshes, 41-62, eds. J.R.L. Allen and
K. Pye, Cambridge University Press, Cambridge, 184 pp.
Reed, D J . (1990) The impact of sea-level rise on coastal salt marshes. Prog. Phys. Geog. 14,
465-81.
Reimold, R.J. (1977) Mangals and saltmarshes of eastern United States. In Wet Coastal
Ecosystems, 1-18, ed. V.J. Chapman, Elsevier, Amsterdam, 428 pp.
Rivera-Monroy, V.H., Day, J.W., Twilley, R.R., Vera-Henera, F. & Coronado-Molina, C.
(1995) Flux of nitrogen and sediment in a fringe mangrove forest in teminos
lagoon Mexico. Est. Coastal Shelf Sci. 40, 137-60.
Steudler, P. A. & Peterson, B.J. (1984) Contribution of gaseous sulphur from salt marshes
to the global sulphur cycle. Nature 311, 455-7.
Sundby, B., Gobeil, C , Silverberg, N. & Mucci, A. (1992) The phosphorus cycle in coastal
marine sediments. Limnol. Oceanogr. 37, 1129-45.
Tooley, M.J. (1992) Recent sea-level changes. In Saltmarshes, 19-40, eds. J.R.L. Allen and
K. Pye, Cambridge University Press, Cambridge, 184 pp.
Twilley, R.R., Chen, R.H. & Hargis, T. (1992). Carbon sinks in mangroves and their
implications to carbon budget of tropical coastal ecosystems. Water Air and Soil
Pollut. 64, 265-88.
Walker, H.J. (1990) The Coastal Zone. In The Earth as Transformed by Human Action,
271-9>4, eds. B.L. Turner III, W.C. Clark, R.W. Kates, J.F. Richards, J.T.
Matthews and W.B. Meyer, Cambridge University Press, Cambridge, 713 pp.
 Biogeochemistry of intertidal sediments 15

Wiegert, R.G., Pomeroy, L.R. & Wiebe, W J . (1981) The Ecology of a Salt Marsh,
Springer-Verlag, New York, 271 pp.
Wigley, T.M.L., & Raper, S.C.B. (1992) Implications for climate and sea level of revised
IPCC emissions scenarios. Nature 357, 293-300.
Wollast, R. (1991) The coastal organic carbon cycle: fluxes, sources and sinks. In Ocean
Margin Processes in Global Change, 365-81, eds. R.F.C. Mantoura, J-M. Martin
and R. Wollast, Wiley, Chichester, 469 pp.
Woodroffe, C D . (1990) The impact of sea-level rise on mangrove shorelines. Prog. Phys.
Geog. 14, 483-520.
o o o o o o o o o o o o o o o o o o o
Trace metals in deposited intertidal
sediments
J.E. Rae

Introduction
Trace metals, for example elements from the first row of transition metals
including cobalt, nickel, copper and zinc (Co, Ni, Cu and Zn), occur
naturally in intertidal sediments. The natural (or 'background' inputs)
reflect a combination of the composition of drainage-basin rocks and
marine-derived sediment. The global anthropogenic input of many trace
metals to the environment currently equals or exceeds the amount
released by weathering (de Groot, Salomons & Allersma, 1976), and the
extent of anthropogenic influence in intertidal sediments is particularly
high, since intertidal areas are often considered as convenient dumping
grounds for industrial and other waste.
Anthropogenic inputs to intertidal environments are often direct,
through point-source waste disposal, but they are also indirect, from
riverine, marine and/or atmospheric sources. Trace metals are partitioned
between each component of the intertidal sediment-water system: they
are found in solution ('bulk' water or interstitial water) and associated
with suspended and deposited sediments. This chapter is concerned with
the biogeochemistry of trace metals in deposited intertidal sediments.
Two main sections follow: in the first, an overview of surface sediments
and sediment depth profiles is presented, and in the second, a case study is
given of the historic record of Zn from saltmarsh sediments in the Severn
Estuary, UK.
 Trace metals in deposited intertidal sediments 17

2.1 Overview of trace metals in deposited

intertidal sediments
2.1.1 Surface sediments
Concentrations and partitioning of trace metals and relation
to bioavailability
Concentrations of trace metals (particularly the pollutants, Cr, Mn, Fe,
Co, Cu, Ni, Cd, Hg and Pb) are commonly reported in surficial intertidal
surface sediments, for example in the St Lawrence Estuary, Canada
(Loring, 1978), San Francisco Bay, USA (Thompson-Becker & Luoma,
1985), the Kelang Estuary, Malaysia (Law & Singh, 1988), the Brisbane
River Estuary, South East Queensland, Australia (Mackey, Hodgkinson
& Nardella, 1992), the Severn Estuary, UK (French, 1993a, b), and the
Sado Estuary, Portugal (Cortesao & Vale, 1995). Natural trace metal
concentrations are less commonly reported, although some useful
compilations of data exist, e.g. Windom et al. (1989) have considered
background data for estuarine and coastal sediments of the southeastern
United States.
Estuarine environments in particular are often sites of intense human
and animal activity (e.g. sites of leisure pursuits and breeding grounds for
many species of birds) so the level of contamination of intertidal sediments
is of particular interest in relation to environmental health. Total
concentrations of individual trace elements in UK estuaries, for example,
vary widely (Table 2.1), reflecting the natural sediment characteristics (e.g.
organic content and surface area), and the level of anthropogenic
contamination of individual systems (Bryan & Langston, 1992). The order
of variability of concentrations of individual elements in Table 2.1 is
Sn > As > Cu > Pb > Hg > Ag > Zn > Cd > Se > Cr > Mn > Co >
Ni > Fe, which to some extent can be considered as the order of
anthropogenic influence.
Trace metals are held in deposited sediments in a number of different
'sites'. First of all there are metals in the mineral lattices (often referred to
as the 'residual' phase metals). In uncontaminated sediments, this is
normally the most important site quantitatively. Secondly, trace metals,
particularly anthropogenically introduced metals, occur associated with
particle surfaces, hence there is often a significant correlation between
trace metal content and grain-size or surface area, which is in turn related
to the sediment mineralogy (Figure 2.1). There has been a lot of detailed
work in recent years on the mode of occurrence of trace metals in
sediments and one of three different approaches is normally taken.
Table 2.1 Average concentrations of metals (ugg~1 dry wt) in sediments from 19 UK estuaries3 (placed in order of decreasing Cu
concentration) (after Bryan & Langston, 1992)

Site Ag As Cd Co Cr Cu Fe Hg Mn Ni Pb Se Sn* Zn
Restronguet Creek 3.76 1740 1.53 21 32 2398 49071 0.46 485 58 341 — 55.9 2821
Fal 1.37 56 0.78 9 28 648 28063 0.20 272 23 150 — 39.5 750
Tamar 1.22 93 0.96 21 47 330 35124 0.83 590 44 235 1.40 8.3 452
Gannel 4.13 174 1.35 26 24 150 25420 0.08 649 38 2753 — 8.5 940
Tyne 1.55 24.8 2.17 11 46 92 28206 0.92 395 34 187 1.23 5.4 421
Mersey 0.70 41.6 1.15 13 84 84 27326 3.01 1169 29 124 0.30 8.3 379
Humber 0.42 49.8 0.48 16 77 54 35203 0.55 1015 39 113 0.84 5.05 252
Medway 1.45 18.4 1.08 11 53 55 32216 1.00 418 26 86 0.48 3.4 220
Poole 0.82 14.1 1.85 11 49 50 29290 0.81 185 26 96 1.51 7.4 165
Severn 0.42 8.9 0.63 15 55 38 28348 0.51 686 33 89 0.23 8.0 259
Hamble 0.16 18.4 0.34 10 37 31 28132 0.43 241 19 56 0.41 3.9 105
Loughor 0.18 17.5 0.47 10 207 27 19337 0.16 597 21 48 — 161 146
Dyfi 0.19 9.7 0.62 17 32 24 45683 0.12 1127 33 166 0.20 0.98 212
Wyre 0.21 6.4 0.35 8 37 20 16970 1.52 590 17 44 — 3.15 122
Avon 0.06 13.0 0.08 10 28 18 18361 0.12 326 23 68 — 3.9 82
Teifi 0.09 10.6 0.17 10 29 13 30280 0.04 684 23 25 0.16 1.14 87
Axe 0.13 4.8 0.17 7 27 12 14004 0.20 248 14 26 — 1.39 76
Rother 0.17 12.4 0.13 6 29 11 15648 0.09 259 15 20 0.18 0.62 46
Solway 0.07 6.4 0.23 6 30 7 14816 0.03 577 17 25 0.11 0.40 59

"Data are for the < 100urn fraction of surface sediment following digestion in HNO 3 .
h
A large proportion of Sn in some samples occurs as cassiterite and is not dissolved in HNO 3 (fusion technique necessary).
Figures in bold are the highest concentrations.
 Trace metals in deposited intertidal sediments 19

i oo

75

o 50
"O

25

organic matter

25 50 75
Grain size (%<l6jjm)
(a)

Mn x IO 2 = ppm
32 Cu x 2 = ppm
Co = ppm
Fe x '/2 = %
Hg x '/4 = ppm
28

24

20 40 60 80 100
Grain size (%<l6y>jm)

Figure 2.1 (a) Relationship between mineralogical composition and

grain-size distribution (% < 16|im) for samples from the same locality (after
de Groot, 1976). {b) Relationship between trace element concentrations and
grain-size distribution (% < 16jim) for sediment samples (after de Groot,
1976).
20 J.E. Rae

An early way of tackling the problem involved the use of theoretical

Eh-pH diagrams to examine the stability fields of trace metal compounds
(e.g. Hem, 1972). This is theoretically a rather elegant and even satisfying
approach, but probably bears little resemblance to the actual field
distribution of trace metals, since among other factors the important role
of organic matter is not included. (See review on metal-organic matter
interactions by Saxby (1969), and more recent detailed work, e.g. Fengler
et al, 1994.)
A second and well-used approach to determining the mode of occurrence
of trace metals in sediments involves the application of chemical extraction
techniques such as those of Chester & Hughes (1967) and Tessier,
Campbell & Bisson (1979). (See review of extraction techniques by
Towner, 1984.) Chemical extraction techniques have, in fact, been used
and developed by soil scientists for several decades (Pickering, 1981) and
many of these extraction schemes can be equally well applied to
sediments. A series of investigations and collaborative studies on current
extraction techniques was recently initiated by the (European) Community
Bureau of Reference (BCR) and the results have been reported, together
with agreed procedural protocols (Ure et a/., 1993). Whatever the detail of
any one technique, however, the principle is the same, i.e. the sediment is
sequentially leached with progressively more aggressive extractants in
order to define a number of trace-element-bearing phases, e.g. exchangeable,
carbonate, Fe and Mn oxide and hydroxide, organic, and residual. It must
be noted that all such extraction techniques are largely empirical, and the
limitations of the approach must always be recognised in data interpretation
(Martin, Nirel & Thomas, 1987; Kheboian & Bauer, 1987; Nirel & Morel,
1990; Whalley & Grant, 1994).
The third and most recent technique for investigating the binding sites
for trace metals in sediments involves the use of computer modelling. The
model GEOCHEM (Mattigod & Sposito, 1979) goes some way to
establishing speciation in waters, sediments and soils but it does have
shortcomings, not least because it ignores the macroionic nature of fulvic
and humic acids. More recently a 'competitive Gaussian' model for fulvic
acid was introduced into the MINTEQ code (Allison & Perdue, 1994), but
the available database for metal interactions is limited (Susetyo et al,
1991). Another recent model, the Windermere Humic Aqueous Model
(WHAM) (Tipping, 1994) is particularly useful since it can be applied to
equilibrium speciation problems involving waters, sediments and soils
with humic substances present in dissolved and particulate form. WHAM
is a combination of several submodels: the Humic Ion-Binding Model V
 Trace metals in deposited intertidal sediments 21

(Tipping & Hurley, 1992; Tipping, 1993a, b), and models of inorganic
solution chemistry, precipitation of aluminium and iron oxyhydroxides,
cation exchange on a representative clay and the adsorption-desorption
reactions of fulvic acids (Tipping & Woof, 1990, 1991). Other recent
models include SCAMP (Surface Chemistry Assemblage Model for
Particles) which is now being developed as part of a large project in the
UK (Land-Ocean Interaction Study (LOIS)). Computer models, although
an important advance, should not be treated as a panacea. They are, for
example, currently limited in the number of variables which can be
considered. Perhaps we should view these models as an adumbration of
detailed computer models of the future, which may more closely resemble
the real situation.
An important reason for studying the concentrations and partitioning
of trace metals in intertidal sediments is to enhance our understanding of
bioaccumulation and biological effects of trace metals in intertidal
environments. Deposited sediments are important for at least three
reasons: firstly, because uptake of metals by organisms from solution is
very significant and concentrations of metals in interstitial and overlying
water can be controlled by equilibria between dissolved metals and those
adsorbed by sediment particles; secondly, concentrations of most trace
metals in sediments are orders of magnitude higher than concentrations in
solution; lastly, some trace metals, e.g. As, Sn, Hg and Pb, can be
converted to more toxic organic (methyl) compounds during their
association with sediment (Bryan & Langston, 1992). Filter feeders and
burrowing organisms are particularly at risk from sediment-bound trace
metals, but biomagnification with increasing trophic levels along food
chains can also occur (e.g. Goede, 1985). An important question must,
however, be addressed: that is, how accurately can we determine the
biological availability (bioavailability) of sediment-bound trace metals?
This question has been usefully discussed by Luoma (1989), who draws
attention to pitfalls awaiting the uncritical researcher. One approach to
investigating metal bioavailability in sediments is through the use of
partial chemical extractions such as those previously discussed (e.g.
Tessier & Campbell, 1987; Tessier et a/., 1984), but Luoma (1989) lists a
number of inherent limitations:

(1) since extraction techniques rarely (if ever) remove metals from
entirely specific components of the sediment, correlations of
biological availability with specific metals components is problematic;
(2) reactions in solution (such as complexation), and competition
22 J E. Rae

among cations for transport into the animal are not accounted for;
(3) biological uptake processes are complex; for example, digestion is
a 'flexible, adaptive, multi-faceted living process that can change
in response to environmental conditions or with life history'.

Despite such limitations, significant relationships have been observed

between concentrations of metals in deposit-feeding organisms and in the
sediments (Bryan & Langston, 1992), Table 2.2. Some of the better
relationships that have been found depend on the use of metal/Fe or
metal/% organic matter ratios in the sediments. Luoma & Bryan (1978),
for example, in a study of English estuaries, showed that 80% of the
variance in Pb concentrations in Scrobicularia plana could be explained
by the ratio of Pb/Fe, (the weak acid (IN HC1) extractable fractions) in the
sediments, and Langston (1982) demonstrated decreased bioavailability
of total Hg where sediments are rich in organic matter.
A number of future research needs in this area are suggested by Bryan &
Langston (1992), particularly:

(1) the mechanisms by which metal-binding sediment components

(e.g. oxides of Fe and Mn and organics) control bioavailability in
the field must be investigated (It is uncertain whether the effects of
such components are important owing to their control over the
levels of dissolved metals, or because they control the uptake of
Table 2.2 Examples of significant relationships between metal concentrations
in benthic species and surface sediments from a range of estuaries (after Bryan &
Langston, 1992)

Sediment extract best

Organism Metal predicting bioavailability Reference

Macoma As As/Fe in IN HC1 (r = 0.94) Langston (1986)

balthica
(bivalve) Ag Ag in IN HC1 (r - 0.83) Bryan (1985)
Hg Hg in H N O 3 extract/% Langston (1982)
organics (r = 0.83)
Pb Pb/Fe in IN HC1 (r = 0.76) Bryan (1985)
Nereis As As/Fe in IN HC1 (r = 0.74) Langston (1980)
diver sicolor
(polychaete) Ag Ag in IN HC1 (r = 0.79) Bryan (1985)
Hg Hg in H N O 3 extract/% Langston (1986)
organics (r = 0.61)
Pb Pb in HNO3 extracts (r = 0.73) Luoma & Bryan
(1982)
 Trace metals in deposited intertidal sediments 23

metals from ingested sediment in the gut.);

(2) more work is required on the composition of interstitial waters in
relation to bioavailability and toxicity;
(3) experimental work on pH and metal/metal interactions in the
digestive tract is required for a range of sediment-ingesting species.

Methods used for reporting trace metal concentrations

of sediments
It is well known that the trace metal content of a sediment is often to a
large extent a function of its chemical and mineralogical characteristics. It
is, therefore, very important to use a reliable 'normalising' technique for
reporting trace metal concentrations. The key sediment characteristic is
surface area or particle size since many of the trace metal binding
components (e.g. organic matter, Fe and Mn oxides and hydroxides) are
very well correlated with both characteristics.
Methods for correcting for grain-size effects in studies on heavy metal
concentrations in estuarine and coastal sediments have been discussed by
Ackermann (1980). There is, unfortunately, no one standard method for
particle-size normalisation and a wide range of techniques are in use
(Table 2.3). The method which often involves the least effort is the
correction which uses comparison with rubidium (Rb) as a conservative
element (Ackermann, 1980). This technique relies on the fact that Rb has a
similar ionic radius to potassium (K) and so substitution of Rb for K will
take place in clay minerals. Furthermore, Rb is present in the sand
fraction in very much smaller concentrations than in the clay or silt
fraction and concentrations of the element in sediments are rarely
influenced by anthropogenic activity. Another advantage of the use of Rb
is that it is often routinely analysed by X-ray fluorescence along with a
suite of pollutant trace metals.

Variability of trace-metal concentrations of surface sediments with

space and time and implications for field sampling strategies
Any consideration of environmental quality issues based on concentrations
of trace metals in intertidal sediments must also address the question of
variability of these concentrations in space and time. Spatial variation of
trace metal concentrations in an estuarine system clearly depends on the
number and magnitude of point-source inputs and on the degree of
sediment mixing. In a physically well-mixed system, the degree of spatial
variation might be expected to be slight. Allen (1987c) and French
(1993a) have investigated the spatial distribution of trace metals (and
Table 2.3 Methods for the reduction of grain-size effects in sediment samples (references listed are examples of the applications
and/or discussion of the use of the techniques)

Method Example Reference

1. Trace metal analysis of a < 64 um (sieving) O'Reilly Wiese, Bubb & Lester (1995)
specific grain size fraction of < 20 um (sieving) Ackermann, Bergmann & Schleichert (1983)
sediment < 2 jam (settling tube) Banat, Forstner & Muller (1972)

2. Correction for inert mineral Quartz correction i.e. trace metal (qtz
constituent corrected)
trace metal (observed) x 100
Thomas & Jaquet (1976)
(100 - quartz %)
3. Ratios of trace metals to trace metal Cortesao & Vale (1995)
'conservative' elements Al Din (1992)

- Grain size Salomons & Mook (1977)

e.g. trace metal concentrations at
50% < 16 urn
4. Use of regression
curve based on ' Conservative elements Ackermann (1980)
e.g. Cs Allen & Rae (1986)
Rb Loring (1990)
Li Rowlatt & Lovell (1994)
Al Oliver (1973)
" Surface area
 Trace metals in deposited intertidal sediments 25

other components) in contemporary muddy sediments from the Severn

Estuary, UK, a large and relatively well-mixed system. A streamwise
variation in grain-size corrected trace metal values was observed (Figure
2.2a) and was interpreted as reflecting a combination of inputs from
land-based sources and re-working of pollutant-rich saltmarsh sediments.
A similar study for the Rhine-Meuse Estuary System (de Groot et a/.,

I150
•

^ 900 rv; •.*• •

.• . ^&
_ • *• •.,
a. 800 _• ••. a> . ..• •
••..: • CL
- ../:';/• .s.:.. : \ . ; a IOO
o 700
- \»: •.:•;;.• '. :.:y - Q •

• •
* 600 -

500 .. ". * " * 50 -

400 •* • -

~l i i i i 1 . i . | 1 i i . i 1 i i "i i i i i i . • i i i i i i . i i •
0 50 100 150 0 50 100 150
(a) Distance downstream (km) Distance downstream (km)

20

8 40
£
O
o
c
S 60

80

Seaward direction •
(b)

Figure 2.2 (a) Spatial (downstream) variation in grain-size corrected

concentrations of zinc and copper, in surface sediments from the Seven
Estuary, UK (after French, 1993a). [Note: data are presented in smoothed
form calculated using a three point moving average.] (b) Spatial
(downstream) variation in trace metal concentrations of sediments from the
Rhine Estuary, expressed as percentages of the contents in fluvial
sediments (after de Groot, 1976).
26 J.E. Rae

1976) (Fig. 2.2b) also demonstrated a general decrease in sediment trace

metal contents in a seaward direction, although grain-size variation does
not appear to have been taken into account. De Groot et al. (1976)
suggest that the downstream trend in the Rhine-Meuse system is a
function of a combination of two very different processes: chemical
desorption of trace metals as the ionic strength of estuarine waters
increases downstream, and physical mixing of polluted fluvatile sediments
with uncontaminated estuarine sediments. (Recent work by Turner &
Mill ward (1993) has explored the former process in more detail.) A
contrasting picture is provided by Mackey et al (1992), working in a
mangrove estuarine environment in Eastern Australia. These authors
report a general increase in the trace metal content of intertidal sediments,
seaward, suggesting a tidal influence on trace metal accumulation.
Temporal fluctuations in the physicochemical character of surface
intertidal sediments must also be considered. For example Thomson-Becker
& Luoma (1985) and Luoma & Phillips (1988) investigated such fluctuations
in intertidal surface sediments of San Francisco Bay, and French (1993b)
has explored the temporal variability of contemporary intertidal muds at
Aust Cliff in the Severn Estuary. In each case samples of surface sediment
were collected at monthly intervals from fixed sites over a period of at least
two years. Temporal fluctuations (seasonal and annual) in grain-size of
surface sediments were found by both sets of workers and were in each
case related to meteorological conditions. In the Severn Estuary, periods
of meteorological stability over one twelve-month period correlated with
relatively little variation in particle size, whereas sediments deposited
during the preceding twelve months were coarser in the winter and finer in
the summer, corresponding to a significantly stormy winter and a
relatively storm-free summer. In San Francisco Bay, sediment tended to
be finer and richer in total organic carbon in autumn and early winter
when winds were minimal (fine particles deposited from run-off), and
coarsest in summer after most terrigenous fine-particle input had ceased,
and seasonal winds had accelerated. (Immediately following run-off
events, sediment tended to be poorer in total organic carbon than
expected from the grain-size distribution.)
Since it is well known that sediment grain-size and organic content
influence trace metal concentrations, it would be expected that temporal
variability in trace metal concentrations would also be found. This is
indeed the case (French, 1993a), but additional controls appear to also
exert an influence: zinc exhibits a temporal variability even after grain-size
correction in Severn Estuary sediments (French, 1993b), probably reflecting
 Trace metals in deposited intertidal sediments 27

a number of processes including variability of anthropogenic inputs and

of riverine discharge rates. Araujo, Bernard & Van Erieken (1988),
working in the Scheldt Estuary, sampled sediment at two stations
periodically during two years. The standardised values of the composition
of the bulk sediments and of the fraction < 63 \im showed large variations
in trace metal content, with the highest values found in samples collected
in June and/or September. It is suggested that these high values may be
more closely related to the increase in bacterial activity during the
summer, after the spring phytoplankton bloom, than they are to an
increase in anthropogenic input.
The design of a field sampling strategy for investigating temporal
variability in heavy metal concentrations of surface sediments clearly
needs careful thought applied to it. There are two potential problems:
firstly, the problem of confounding of temporal differences by spatial
variation, and secondly, the problem of variations at shorter time scales
than the scale of interest. For example, to demonstrate differences from
one season to another, it must be shown that there is more variation
between seasons than is found from time to time within a season. These
problems are discussed by Morrisey et al. (1994), in a study of the
distribution of Cu, Pb and Zn in the sediments of Botany Bay. The
authors suggest that nesting of spatial and temporal scales should be done
in a pilot study, and scales at which significant variation is not detected
can then be omitted from the subsequent sampling.

2.1.2 Sediment depth profiles as records of intertidal pollution history

Trace metals in intertidal sediment depth profiles have been extensively
studied since the late 1970s. There were two key factors which resulted in
the proliferation of studies at this time: firstly, a rise in environmental
awareness led to interest in using sediment depth profiles as historic
records of pollution, and secondly, radiometric techniques for sediment
dating became well established. In particular, 210 Pb dating of modern
sediments has been widely used since it was first outlined by Goldberg
(1963) and several workers have used the technique for dating intertidal
sediments (e.g. McCaffrey, 1977; Goldberg, et al.9 1979; McCaffrey &
Thompson, 1980; Sharma et al, 1987; Bricker, 1990 and Buckley, Smith &
Winters, 1995), all of whom have investigated trace metal concentrations
in sediment depth profiles.

A number of useful reviews exists on the use of sediment cores to

reconstruct historical pollution trends (Alderton, 1985; Farmer, 1991;
28 J.E. Rae

Valette-Silver, 1992, 1993) and there are perhaps two key issues which
emerge. Firstly, sediment depth profiles (including those from intertidal
environments, e.g. Buckley et a/., 1995) often reflect the globally averaged
profiles of historic sediment contamination (Figure 2.3). The global
pattern is one of significant heavy metal contamination of the environment,
beginning during the last decades of the nineteenth century and reaching
maximum levels around 1960-70 (Valette-Silver, 1993). Concentrations of
heavy metals infine-grainedintertidal sediments commonly range from
'background' levels of a few |igg - 1 or tens of ^igg"1 up to (commonly)
several hundred |igg~l for polluted sediments (Valette-Silver, 1993). The
second key issue to emerge from the literature is also a caution: that is, the
character of the sediments (e.g. grain-size, mineralogy, organic carbon

2OOO - i

1960-

1920-

1880-

1840-

1800 I
0 100 0 100
Relative Concentration of Contaminants (%)

Heavy metals ( Pb, Zn, Cu, etc )

Ni,Co

Figure 2.3 Globally averaged profiles of historic sediment contamination

by trace metals, in industrialised countries, providing a general idea of
the evolution of contamination versus time (years). 100% represents the
maximum value reached in cored sediments and 0% the lowest value
measured. These values are rough estimates, extrapolated from results
reported in the literature to 1990. Section (a) represents a background or
low contamination period, section (b) corresponds to the increase in
sediment contamination around World War I. Section (c) visualises the
rapid increases occurring after World War II that reach a maximum in
the mid-1970s. Section (d) represents the recent decrease observed in
concentrations of many trace metals (after Valette-Silver, 1993).
 Trace metals in deposited intertidal sediments 29

content), together with the diagenetic history can be important in

influencing trace metal concentrations and bioavailability. In other
words, it is necessary to establish, and where appropriate correct for, the
sediment characteristics before confirming historic trace metal records, in
addition to considering the extent of influence of diagenetic processes
(Farmer, 1991; Rae & Allen, 1993; Zwolsman, Berger & Van Eck, 1993).
As discussed in the previous section, pollutant trace metals are predomi-
nantly held in association with the surfaces of sedimentary materials, in a
number of different 'phases' which can be experimentally defined (e.g.
exchangeable, carbonate, co-precipitated with Fe and Mn oxides, organic-
bound (Tessier et a/., 1979)). It is not surprising, therefore, to find that
sediment characteristics such as grain-size and organic carbon content,
and diagenetic processes such as reduction of Fe and Mn oxides, are
important in influencing trace metal concentrations in sediment depth
profiles (Allen, Rae & Zanin, 1990; Farmer, 1991). An illustration of the
influence of sediment characteristics is given in Rae (1980), (Figure 2.4), in
which an apparent historic record of mercury (Hg) pollution from an
estuarine sediment core (Wyre Estuary, UK) is illustrated. The increase in
sediment Hg content above 10 cm may appear to reflect a simple historic
increase in environmental Hg pollution, but closer inspection reveals a

Sediment Grain Size

% Organic Carbon (%<63>jm)
0 I 2 50 60 70 80 90 100
"1— T—r T T"

Figure 2.4 Results of the analysis (Hg, organic carbon and grain size
(% < 63 jam)) for a sediment core from the Wyre Estuary, UK. Values of Hg
below 10 cm represent 'background' levels with an apparent increase in
contamination from 10cm depth to the sediment surface (after Rae, 1980).
30 J.E. Rae

corresponding increase in organic carbon content and percentage of fine

sediment (% < 63 |im). It is likely, therefore, that the influence of sediment
characteristics is here masking the historic pollution record. (The increase
in fines below 10 cm has little effect on the very low 'background' Hg
concentrations.) Intertidal sediments are particularly prone to variability
in sediment characteristics with depth, when compared to other sedimentary
environments such as lakes, in which historic pollution trends are also
frequently established. The greater variability in sediment characteristics
with depth in intertidal systems results primarily from tidal and wave
action which can have profound effects in influencing particle size and
sorting. Diagenetic history in intertidal sediment profiles may also be
more complex than in lakes, since intertidal environments may be more
subject to reworking and/or more rapidly fluctuating pore-water compo-
sitions. (It should be said, however, in their favour, that intertidal
environments are not subject to acidification which often complicates the
use of lake sediments as historical records of pollution (Valette-Silver, 1993).)
Finally, a recent approach to the study of trace metal distribution in
sediment depth profiles deserves mention. This is a factor-analysis
technique which is used to determine the main environmental condition
prevailing at the place and time when the sediment was deposited, or the
main process responsible for modification of the sediment after deposition
(Buckley et a/., 1995). The study in Halifax Harbour, Nova Scotia,
(Buckley et al, 1995), established the following groups:

(1) primary contaminants directly associated with waste discharge

(including total and organic-bound forms of Cu, Zn and Pb);
(2) secondary contamination attributed to leaching and modification
of primary contaminants (including acid labile forms of Zn, Ni
and Cu);
(3) diagenetic modification of buried sediments (total and labile Mn);
(4) dispersion of contaminants from land surface drainage (includes
total Cu, Hg, Pb, Zn).

Trace metal concentrations in intertidal sediment depth profiles will no

doubt continue to provide useful historic records of pollution in the
future. In particular, dated profiles that take into account sediment
characteristics and diagenetic processes (where appropriate) are of value.
Future studies will perhaps be more process-oriented, e.g. by combining
pore-water analysis with solid phase data and through a detailed
investigation of the nature and significance of the organic matter present.
 Trace metals in deposited intertidal sediments 31

Lastly, there is an urgent need for more biogeochemical work on

organometallic forms of trace metals owing to their presence in intertidal
sediments and their toxicity at relatively low concentrations; this work
may well profit from a detailed appreciation of microbiological processes.

2.2 The record of zinc pollution in a sediment depth

profile at Tites Point, Severn Estuary, UK: a
case study
2.2.1 Introduction
The Severn Estuary (Figure 2.5) lies on the west coast of Britain, facing the
prevailing winds and Atlantic swell. It is a large (c. 100 km long),
macrotidal system (12.3 m mean spring tidal range at Avonmouth), swept
by vigorous currents (Crickmore, 1982; Shuttler, 1982; Uncles, 1984).
Urban and industrial effluent from three main areas - southeast Wales,
the Bristol region, and the West Midlands - is currently being brought
into the estuary (Radford, 1982; Morris, 1984; Owens, 1984).
There has been much work reported in recent years on the sedimentology
and stratigraphy of the tidally influenced marshes and high tidal mud-flats
(Allen, 1985a, b, 1986,1987a, b, c, d, 1988,1990a, b, c, d, e, f, 1992a, b; Allen
& Rae, 1986, 1987, 1988).
The formation of mud-flats, which develop into saltmarshes, is occurring
in response to an apparently climate-driven, estuary-wide cycle of erosion

~ *\ v ^u/wv,u morsh
0 -
0-2-
0-4- layered
So-6- laminated
£ 0-8- silts
•o 1-0-
GLOUCESTER 1-2 -
1-4-
Tites Point
1-6- ^ t ^ ^ f z : bedrock
Avonmouth
> * * t ,y
^5^1^" ""111: c samples}^ |:Z£

^>-^b e d roclT^T^i^r^ 7

Tites Point Section

Figure 2.5 The Severn Estuary and Tites Point; setting in southwest
England; a schematic vertical profile of the saltmarsh cliff (after Rae & Allen,
1993).
32 J.E. Rae

and accretion (Allen & Rae, 1987), with a decadal to century time scale.
Mud-flat and saltmarsh deposits build up in the intertidal zone, but later
become eroded vertically and horizontally (toward land) to form a
wave-cut platform and bold cliff, on and against which a new mud-flat and
then a marsh grows up (Allen, 1992a, b). At least three repetitions of this
cycle have occurred throughout the Severn Estuary, resulting in present-day
saltmarshes that are visibly terraced, descending stair-like toward the
estuary, each terrace overlying an erosively based mud deposit that
offlaps the preceding unit (Allen & Rae, 1987).
Early work on trace metals in sediment depth profiles in the Severn
Estuary provided an overview of trace metal concentrations in marsh
sediments, in order to establish a 'chemostratigraphy' (Allen, 1987c; Allen
& Rae, 1987; Allen, 1988). Later work (Allen et al, 1990) provided a
preliminary investigation of the post-depositional behaviour of Cu, Zn
and Pb at one location (Tites Point). Subsequent intensive sampling of an
adjacent profile has allowed a more detailed appreciation of trace metal
concentrations and behaviour.

2.2.2 Site location and sampling methods

The location of Tites Point and the sampling site are given in Figure 2.5,
together with a schematic vertical profile of the saltmarsh cliff. A detailed
description of the field context is given in Allen et al. (1990). The deposit
sampled is attributed to the Awre Formation, initiated during the second
depositional cycle, which began late in the nineteenth century. At Tites
Point, the Awre Formation lies in the uppermost 3 m of the intertidal zone
and rests on and against an uneven bedrock surface and gravel-capped
cliff. The deposits at the sampling site (British National Grid SO 688046)
are 1.5-1.6 m thick on the mud cliff, and consist of interlaminated sandy
silts and clayey silts. The section sampled may be considered as an oxic,
open system, since root channels and desiccation cracks allow active
exchange between sediment and pore-water, meteoric and estuarine
water, and atmospheric gases.
The saltmarsh was sampled at 10 mm intervals to a depth of 1500 mm
using the device of Allen & French (1989). Bedding-parallel slices are
obtained using this device in which a cutting plate is positioned between
two slotted uprights pinned onto the marsh cliff.

2.2.3 Laboratory methods

Sediments were air-dried in the laboratory and lightly disaggregated.
Organic carbon determinations were carried out by the method of
 Trace metals in deposited intertidal sediments 33

Gaudette et al (1974), in which exothermic heating and oxidation with

K 2 CrO 7 and concentrated H 2 SO 4 are followed by titration of the excess
dichromate with 0.5N Fe(NH 4 ) 2 (SO 4 ) 2 6H 2 O. (This method excludes
elementary carbon which in intertidal sediments is commonly present as
coal, coke and fly-ash particles (Allen, 1987a, b).)
The sediment samples were sequentially leached in accordance with the
method of Tessier et al (1979) in which five fractions are experimentally
defined. The extracts were analysed by atomic absorption (Perkin Elmer
3030), (Pb and Rb), and inductively coupled plasma spectrometry (ARL
35000), (Cu, Zn, Cd, Ni, Fe and Mn). Instrumental precision was better
than 1% and accuracy was ensured by analysing a range of international
reference sediments.
The sediment profile was dated using 2 1 0 Pb and a number of other
techniques (Allen et al, 1993).

2.2.4 Results and discussion

Results of dating the sediments at Tites Point are presented in Allen et al
(1993) and a preliminary assessment of the significance of organic matter
concentrations in relation to trace metal mobility (Cu, Zn and Pb) is given
in Rae & Allen (1993). A complete report of trace metal concentrations
and behaviour at this site will be given in Allen & Rae (in preparation).
There follows here a summary of the results for Zn, in order to
demonstrate in broad terms (i) the approach to consideration of the
influence of sediment characteristics and early diagenetic processes on
trace metal profiles, and (ii) that the record can be, in some instances,
straightforward.
The variation in total sediment Zn with depth is illustrated in Figure
2.6. There is a clear trend of essentially constant values of total Zn (around
150ppm) from 1500 mm to approximately 1000 mm, with a steady
increase in concentrations from 1000 mm to 225 ppm at 400 mm.
Concentrations of Zn remain at approximately this latter value until they
fall to 200 ppm in the uppermost 50 mm. In contrast to this clear trend,
concentrations of zinc in the 'residual' phase (non-pollutant zinc) remain
approximately constant with depth (~ 75 ppm) (Figure 2.7). We are now
faced with the problem of interpretation and the key question is: does the
well-defined depth trend of total Zn represent a historic pollution record
or is it partly (or entirely) a function of sediment characteristics and/or
early diagenetic processes?
The most important sediment characteristic is usually the sediment
grain-size. The abundance of Rb in sediments is an established and
34 J.E. Rae

reliable grain-size proxy (Allen & Rae, 1987; Allen, 1987c), increasing with
clay-mineral content. There appears to be a slight but systematic increase
infineswith depth below 220 mm and a decrease infineswith depth from
the marsh surface to 220 mm. It is unlikely, however, that these grain-size
trends are of a magnitude that would interfere with the broad picture of
historic pollution at this site.
Having established that grain-size variability is not significant in this
case, the next step is to investigate any perturbations in the historic
pollution record owing to early diagenetic processes. One approach to
this investigation is through an examination of depth trends of Zn in
'carbonate', Te and Mn oxide' and 'organic' phases (Figure 2.7). (The
amount of Zn at Tites Point in the exchangeable phase (generally
< 1 ppm) is too low to influence the interpretation of the total zinc
profile.) The question now is as follows: to what extent, if at all, are the
profile of Zn in carbonate, Fe and Mn oxide and organic phases a function
of early diagenetic processes? Calculation of the Zn content at each depth,
in each phase, as a percentage of the 'pollutant' (i.e. non-residual) zinc at
corresponding depth intervals provides a clue, although the result of
sequential extraction techniques should always be used with caution and
with supporting data where possible (see Section 2.1.1). In this case,

Total Zn(ppm ) Figure 2.6 Variation in the saltmarsh sediment

0 200 400 concentration of total zinc with depth in a dated
i i (210Pb and archaeological evidence) profile from
10 -
-1970 Tites Point, Severn Estuary, UK.

f
-1960

L*f
20 - -1955
30 - -1950
-1945 H
40 -
/ -1940 §'

1
1
CD
50 - -1935^
60 - -1930 p
70 -
-1925^-
£
?
80 -

/
-1920
90 -

<
100 -
-1915
110 - -POTTERY
-1910 B E D
120 -
130 -
140 -
c
1
150 -
160 -
Trace metals in deposited intertidal sediments 35

ZnF2(ppm) ZnF3(ppm)
0 25 50 75

ZnF4(ppm) ZnF5(ppm)
0 25 50 50 100 150
I I
-1980
-1970
-1960
-1955
-1950
-1945 H
-1940 §'
-1935 ^

-1925 *

-1920

-1915
-POTTERY -POTTERY
-H_1910BED -1910 BED

Figure 2.7 Variation in the saltmarsh sediment concentration of zinc with

depth in four fractions, defined according to the sequential extraction
scheme of Tessier, Campbell & Bisson (1979) as: F2, carbonate; F3, Fe and
Mn oxide and hydroxide; F4, organic; F5, residual. The location of the
profile is Tites Point, Severn Estuary, UK.
36 J.E. Rae

although there is a little variation with depth, percentage values remain

similar (carbonate bound zinc ~ 15%, Fe/Mn bound zinc ~ 60%, organic
bound zinc -20%). Early diagenetic mobility of Zn at this site, therefore,
appears slight in comparison with the historic record, so the total Zn
profile can now be used as a historic record with a good degree of confidence.
Sources of Zn entering the deposits of the Severn Estuary are undoubtedly
numerous and historically complex, but the slightly mineralised Carbon-
iferous coals in the catchment, and the smelting of ores in the vicinity of
the estuary will no doubt have contributed Zn to the system. Perhaps the
most striking feature of the total Zn profile is the similarity to the globally
averaged profile of historic sediment contamination discussed in Section
2.1.2; that is, increasing Zn pollution is documented from around 1900,
reaching maximum levels around 1960-70.
Finally, it is important not to assume that the historic record for other
trace metals in the same sediment core will be so straightforward. Detailed
examination of depth-related phase associations for each metal is always
essential and additional data, e.g. on pore-water chemistry and the
amount and nature of organic matter present can be advantageous.
This paper represents University of Reading PRIS Contribution No. 443.

References
Ackermann, F. (1980) A procedure for correcting the grain size effect in heavy metal
analyses of estuarine and coastal sediments. Environmental Technology Letters 1,
518-27.
Ackermann, F., Bergmann, H. & Schleichert, V. (1983) Monitoring of heavy metals in
coastal and estuarine sediments - a question of grain size: < 20 um versus
< 60 um. Environmental Technology Letters 4, 317-28.
Alderton, D.H.M. (1985) Sediments. In Historical Monitoring, Report No. 31, 1-95,
MARC, London.
Allen, J.R.L. (1985a) Principles of Physical Sedimentology, Allen & Unwin, London.
Allen, J.R.L. (1985b) Intertidal drainage and mass-movement processes in the Severn
Estuary: rills and creeks (pills). Journal of the Geological Society, London 142,
849-61.
Allen, J.R.L. (1986) A short history of saltmarsh reclamation at Slimbridge Warth and
neighbouring areas, Gloucestershire. Transactions of the Bristol and Gloucestershire
Archaeologial Society 104, 139-55.
Allen, J.R.L. (1987a) Dimlington Stadial (late Devensian) ice-wedge casts and involutions
in the Severn Estuary, southwest Britain. Geological Journal 22, 109-18.
Allen J.R.L. (1987b) Late Flandrian shoreline oscillations in the Severn Estuary: the
Rumney Formation at its typesite (Cardiff area). Philosophical Transactions of
the Royal Society, B 315, 157-84.
Allen, J.R.L. (1987c) Towards a quantitative chemostratigraphic model for the sediments
of late Flandrian age in the Severn Estuary, U.K. Sedimentary Geology 53,73-100.
 Trace metals in deposited intertidai sediments 37

Allen, J.R.L. (1987d) Reworking of muddy intertidai sediments in the Severn Estuary,
Southwestern U.K. - a preliminary study. Sedimentary Geology 50, 1-23.
Allen, J.R.L. (1988) Modern-period muddy sediments in the Severn Estuary (southwestern
U.K.): a pollutant-based model for dating and correlation. Sedimentary Geology
58, 1-21.
Allen, J.R.L. (1990a) The Severn Estuary in southwest Britain: its retreat under marine
transgression, and fine-sediment regime. Sedimentary Geology 96, 13-28.
Allen, J.R.L. (1990b) Reclamation and sea defence in Rumney Parish (Monmouthshire).
Archaeologia Cambrensis 97, 135-40.
Allen, J.R.L. (1990c) Late Flandrian shoreline oscillations in the Severn Estuary: change
and reclamation at Arlingham, Gloucestershire. Philosophical Transactions of
the Royal Society, A 330, 315-34.
Allen, J.R.L. (1990d) Saltmarsh growth and stratification: a numerical model with special
reference to the Severn Estuary, southwest Britain. Marine Geology 95,
77-96.
Allen, J.R.L. (1990e) Constraints on measurement of sea-level movements from saltmarsh
accretion rates. Journal of the Geological Society, London 147, 5-7.
Allen, J.R.L. (1990f) The formation of coastal peat marshes under an upward tendency of
relative sea-level. Journal of the Geological Society, London 147, 743-5.
Allen, J.R.L. (1991a) Saltmarsh accretion and sea-level movement in the inner Severn
Estuary: the archaeological and historical contribution. Journal of the Geological
Society, London 148, 485-94.
Allen, J.R.L. (1991b) Fine sediment and its sources, Severn Estuary and inner Bristol
Channel, southwest Britain. Sedimentary Geology 75, 57-65.
Allen, J.R.L. (1992a) Large scale textural patterns and sediment processes on tidal
saltmarshes in the Severn Estuary, southwest Britain. Sedimentary Geology 81,
299-318.
Allen, J.R.L. (1992b) Tidally influenced marshes in the Severn Estuary, southwest Britain.
In J.R.L. Allen & K. Pye (eds.), Saltmarshes: Morphodynamics, Conservation and
Engineering Significance, Cambridge University Press, Cambridge, 123-47.
Allen, J.R.L. & French, P.W. (1989) An apparatus for sequentially sampling unconsolidated
cohesive sediments exposed on saltmarsh and river cliffs. Sedimentary Geology
61, 151-4.
Allen, J.R.L. & Rae, J.E. (1986) Time sequence of metal pollution, Severn Estuary,
southwestern U.K. Marine Pollution Bulletin 17, 427-31.
Allen, J.R.L. & Rae, J.E. (1987) Late Flandrian shoreline oscillations in the Severn Estuary:
a geomorphological and stratigraphical reconnaissance. Philosophical Transactions
of the Royal Society B 315, 185-230.
Allen, J.R.L. & Rae, J.E. (1988) Vertical saltmarsh accretion since the Roman period in the
Severn Estuary, southwestern Britain. Marine Geology 83, 225-35.
Allen, J.R.L., Rae, J.E. & Zanin, P.E. (1990) Metal speciation (Cu, Zn, Pb) and organic
matter in an oxic salt marsh. Severn Estuary, southwest Britain. Marine
Pollution Bulletin 21, 574-80.
Allen, J.R.L., Rae, J.E., Longworth, G., Hasler, S.E. & Ivanovich, M. (1993) A comparison
of the 2 1 0 Pb dating technique with three other independent dating methods in an
oxic estuarine saltmarsh sequence. Estuaries 16, 670-7.
Allison, J.A. & Perdue, E.M. (1994) Modelling metal-humic interactions with MINTEQA2.
Proceedings ofthe 6th International Meeting of the International Humic Substances
Society, Bari, Italy.
Araujo, M.F., Bernard, P.C. & Van Erieken, R.E. (1988) Heavy metal contamination in
38 J.E. Rae

sediments from the Belgian Coast and Scheldt Estuary. Marine Pollution Bulletin
19, 269-73.
Banat, K., Forstner, U. & Muller, G. (1972) Schwermetalle in den Sedimentendes Rheinso
Unschau 72, 192-3.
Bricker, S.B. (1990) The history of metals pollution in Narragansett Bay as recorded by
saltmarsh sediments. PhD Dissertation. University of Rhode Island, Kingston,
Rhode Island, USA.
Bryan, G.W. (1985) Bioavailability and effects of heavy metals in marine deposits. In B.H.
Ketchum, J.M. Capuzzo, W.V. Burt, I.W. Duedall, P.K. Park & D.R. Kestner
(eds.), Wastes in the Ocean Vol 6 Nearshore Waste Disposal, John Wiley & Sons,
New York, 42-79.
Bryan, G.W. & Langston, W.J. (1992) Bioavailability, accumulation and effects of heavy
metals in sediments with special reference to United Kingdom estuaries: a
review. Environmental Pollution 76, 89-131.
Buckley, D.E. & Winters, G.V. (1992) Geochemical characteristics of contaminated
surficial sediments in Halifax Harbour: impact of waste discharge. Canadian
Journal of Earth Sciences 29, 2617-39.
Buckley, D.E., Smith, J.N. & Winters, G.V. (1995) Accumulation of contaminant metals in
marine sediments of Halifax Harbour, Nova Scotia: environmental factors and
historical trends. Applied Geochemistry 10, 175-95.
Chester, R. & Hughes, M.J. (1967) A chemical technique for the separation of ferro-manganese
minerals, carbonate minerals and adsorbed trace elements from pelagic sediments.
Chemical Geology 2, 249-62.
Cortesao, C. & Vale, C. (1995) Metals in sediments of the Sado Estuary, Portugal. Marine
Pollution Bulletin 30, 34-7.
Crickmore, M.J. (1982) Data collection - tides, tidal currents, suspended sediment. In
Institution of Civil Engineers (eds.), Severn Barrage, Thomas Telford, London,
19-26.
de Groot, A.J., Salomons, W. & Allersma, E. (1976) Processes affecting heavy metals in
estuarine sediments. In J.D. Burton and P.S. Liss (eds.), Estuarine Chemistry,
Academic Press, London, 131-57.
Din, Z. (1992) Use of aluminium to normalize heavy metals data from the estuarine and
coastal sediments of the straits of Melaka. Marine Pollution Bulletin 24,484-91.
Farmer, J.G. (1991) The pertubation of historical pollution records in aquatic sediments.
Environmental Geochemistry and Health 13, 76-83.
Fengler, G., Grossman, D., Kersten, M. & Liebezeit, G. (1994) Trace metals in humic acids
from recent Skagerrah sediments. Marine Pollution Bulletin 28, 143-7.
French, P.W. (1993a) Areal distribution of selected pollutants in contemporary intertidal
sediments of the Severn Estuary and Bristol Channel, U.K. Marine Pollution
Bulletin 26, 692-7.
French, P.W. (1993b) Seasonal and inter-annual variation of selected pollutants in modern
intertidal sediments, Aust Cliff, Severn Estuary. Estuarine, Coastal and Shelf
Science 37, 213-19.
Gaudette, H.E., Flight, W.R., Toner, L. & Folger, D.W. (1974) An inexpensive titration
method for the determination of organic carbon in recent sediments. Journal of
Sedimentary Petrology 44, 249-53.
Goede, A.A. (1985) Mercury, selenium, arsenic and zinc in waters from the Dutch Wadden
Sea. Environmental Pollution 37A, 287-309.
Goldberg, E.D. (1963) Geochronology with 2 1 0 Pb. In Radioactive Dating, International
Atomic Energy Agency, Vienna.
 Trace metals in deposited intertidal sediments 39

Goldberg, E.D., Griffin, J.J., Hodge, V., Koide, M. & Windom, H. (1979) Pollution
history of the Savannah River Estuary. Environmental Science and Technology
3, 588-94.
Hem, J.D. (1972) Chemistry and occurrence of cadmium and zinc in surface water and
groundwater. Water Resources Research 8, 661-79.
Kheboian, C. & Bauer, C.F. (1987) Accuracy of selective extraction procedures for metal
speciation in model aquatic sediments. Analytic Chemistry 59, 1417-23.
Langston, W.J. (1980) Arsenic in UK estuarine sediments and its availability to benthic
organisms. Journal of the Marine Biological Association 60, 869-81.
Langston, W.J. (1982) The distribution of mercury in British estuarine sediments and its
availability to deposit feeding bivalves. Journal of the Marine Biological
Association 62, 667-84.
Langston, W.J. (1986) Metals in sediments and benthic organisms in the Mersey estuary.
Estuarine Coastal & Shelf Science 23, 239-61.
Law, A.T. & Singh, A. (1988) Heavy metals in the Kelang estuary, Malaysia. Malayan N.
Journal 41, 505-13.
Loring, D.H. (1978) Geochemistry of zinc, copper and lead in the sediments of the estuary
and Gulf of St. Lawrence. Canadian Journal of Earth Science 15, 757-72.
Loring, D.H. (1990) Lithium - a new approach for the granulometric normalization of
trace metal data. Marine Chemistry 29, 155-68.
Luoma, S.N. (1989) Can we determine the biological availability of sediment-bound trace
metals? Hydrobiologia, 176/177, 379-96.
Luoma, S.N. & Bryan, G.W. (1978) Factors controlling the availability of sediment-bound
lead to the estuarine bivalve Scrobicularia plana. Journal of the Marine Biological
Association 58, 793-802.
Luoma, S.N. & Bryan, G.W. (1982) A statistical study of environmental factors controlling
concentrations of heavy metals in the burrowing bivalve Scrobicularia plana and
the polychaete Nereis diversicolor. Estuarine Coastal & Shelf Science 15,95-108.
Luoma, S.N. & Phillips, D.J.H. (1988) Distribution, variability and impact of trace
elements in San Francisco Bay. Marine Pollution Bulletin 19, 413-25.
Mackey, A.P., Hodgkinson, M. & Nardella, R. (1992) Nutrient levels and heavy metals in
mangrove sediments from the Brisbane River, Australia. Marine Pollution
Bulletin 24, 418-20.
McCaffrey, R.J. (1977) A record of the accumulation of sediment and trace metals in a
Connecticut, U.S.A. saltmarsh. PhD Thesis, Yale University, New Haven,
Connecticut, USA.
McCaffrey, R.J. & Thomson, J. (1980) A record of the accumulation of sediment and trace
metals in a Connecticut saltmarsh. Advances in Geophysics 22, 165-236.
Martin, J.M., Nirel, P. & Thomas, AJ. (1987) Sequential extraction techniques: promises
and problems. Marine Chemistry 22, 313-41.
Mattigod, S.V. & Sposito, G. (1979) Chemical modelling of trace metal equilibria in
contaminated soil solutions using the computer programme GEOCHEM. In
E.A. Jenne (ed.), Chemical Modelling in Aqueous Systems, American Chemical
Society Symposium, Washington DC, 837-56.
Morris, A.W. (1984) The chemistry of the Severn Estuary and the Bristol Channel. Marine
Pollution Bulletin 15, 57-61.
Morrisey, D.J., Underwood, A.J., Stark, J.S. & Hewitt, I. (1994) Temporal variation in
concentrations of heavy metals in marine sediments. Estuarine Coastal and Shelf
Science 38, 271-82.
Nirel, P.M. & Morel, F.M. (1990) Pitfalls of sequential extractions. Water Research 24,1055-6.
40 J.E. Rae

Oliver, B.G. (1973) Heavy metal levels of Ottowa and Rideau river sediments. Environmental
Science and Technology 7, 135-7.
O'Reilly Wiese, S.B., Bubb, J.M. & Lester, J.N. (1995) The significance of sediment metal
concentrations in two eroding Essex salt marshes. Marine Pollution Bulletin 30,
190-9.
Owens, M. (1984) Severn Estuary- an appraisal of water quality. Marine Pollution Bulletin
15, 41-7.
Pickering, W.F. (1981) Selective chemical extraction of soil components and bound metal
species. CRC Critical Reviews. Analytical Chemistry 12, 233-66.
Radford, P J . (1982) The effect of a barrage on water quality. In Institution of Civil
Engineers (eds.), Severn Barrage, Thomas Telford, London, 203-8.
Rae, J.E. (1980) The geochemistry of mercury in estuarine mixing and sedimentation. PhD
Thesis, University of Lancaster, Lancaster, UK.
Rae, J.E. & Allen, J.R.L. (1993) The significance of organic matter degradation in the
interpretation of historical pollution trends in depth profiles of estuarine
sediment. Estuaries 16, 678-82.
Rowlatt, S.M. & Lovell, D.R. (1994) Lead, zinc and chromium in sediments around
England and Wales. Marine Pollution Bulletin 28, 324-9.
Salomons, W. & Mook, W.G. (1977) Trace metal concentrations in estuarine sediments:
mobilization, mixing or precipitation. Netherlands Journal of Sea Research, 11,
119-29.
Saxby, J.D. (1969) Metal-organic chemistry of the geochemical cycle. Reviews of Pure and
Applied Chemistry 19, 131-49.
Sharma, P.T., Church, M., Murray, S. & Biggs, R.B. (1987) Geochronology and trace metal
records in a Delaware saltmarsh sediment. Eos 68, 331.
Shuttler, R.M. (1982) The wave climate in the Severn Estuary. In Institution of Civil
Engineers (eds.), Severn Barrage, Thomas Telford, London, 27-34.
Susetyo, W., Carreira, L.A., Azarraga, L.V. & Grimm, D.M. (1991) Fluorescence
techniques for metal-humic interactions. Fres. Zeit Analytical Chemistry 339,
624-35.
Tessier, A. & Campbell, P.G.C. (1987) Partitioning of trace metals in sediments:
relationships with bioavailability. Hydrobiologia 149, 43-52.
Tessier, A., Campbell, P.G.C. & Bisson, M. (1979) Sequential extraction procedure for the
speciation of particulate trace metals. Analytical Chemistry 51, 844-51.
Tessier, A., Campbell, P.G.C, Auclair, J.C. & Bisson, M. (1984) Relationships between the
partitioning of trace metals in sediments and their accumulation in the tissues of
the freshwater mollusc Elliptio complanata in a mining area. Canadian Journal of
Fisheries and Aquatic Science 41, 1463-72.
Thomas, R.L. & Jaquet, J.M. (1976) Mercury in surficial sediments of Lake Erie. Journal of
the Fisheries Research Board of Canada 33, 404-12.
Thomson-Becker, E.A. & Luoma, S.N. (1985) Temporal fluctuations in grain size, organic
materials and iron concentrations in intertidal surface sediment of San Francisco
Bay. Hydrobiologia 129, 91-107.
Tipping, E. (1993a) Modelling the competition between alkaline earth cations and trace
metal species for binding by humic substances. Environmental Science and
Technology 27, 520-9.
Tipping, E. (1993b) Modelling ion-binding by humic acids. Colloid Surfaces 73, 117-31.
Tipping, E. (1994) WHAM - A chemical equilibrium model and computer code for waters,
sediments and soils incorporating a discrete site/electrostatic model of ion-binding
by humic substances. Computers and Geosciences 20, 973-1023.
 Trace metals in deposited intertidal sediments 41

Tipping, E. & Hurley, M.A. (1992) A unifying model of cation binding by humic
substances. Geochimica et Cosmochimica Acta 56, 3627-41.
Tipping, E. & Woof, C. (1990) Humic substances in acid organic soils: modelling their
release to the soil solution in terms of humic charge. Journal of Soil Science 41,
573-83.
Tipping, E. & Woof, C. (1991) The distribution of humic substances between the solid and
aqueous phases of acid organic soils; a description based on humic heterogeneity
and charge-dependent sorption equilibria. Journal of Soil Science 42, 437-48.
Towner, J.V. (1984) Studies of chemical extraction techniques used for elucidating the
partitioning of trace metals in sediments. PhD Thesis, University of Liverpool,
Liverpool, UK.
Turner, A. & Millward, G.E. (1993) Application of the KD concept to the study of trace
metal removal and adsorption during estuarine mixing. Estuarine, Coastal and
Shelf Science 36, 1-13.
Uncles, R J. (1984) Hydrodynamics of the Bristol Channel. Marine Pollution Bulletin 15,47-53.
Ure, A.M., Quevauviller, Ph., Muntau, H. & Griepink, B. (1993) Speciation of heavy
metals in soils and sediments. An account of the improvement in harmonisation
of extraction techniques under the auspices of the BCR of the Commission of the
European Union. International Journal of Environmental Analytical Chemistry
51, 135-51.
Valette-Silver, N.J. (1992) Historical reconstruction of contamination using sediment
cores: A review. NOAA Technical Memorandum NOS.ORCA 65, Rockville,
Maryland.
Valette-Silver, N.J. (1993) The use of sediment cores to reconstruct historical trends in
contamination of estuarine and coastal sediments. Estuaries 16, 577-88.
Whalley, C. & Grant, A. (1994) Assessment of the phase selectivity of the European
Community Bureau of Reference (BCR) sequential extraction procedure for
metals in sediment. Analytical Chimica Acta 291, 287-95.
Windom, H.L., Schropp, S.J., Calder, F.D., Ryan, J.D., Smith, R.G. Jr, Burney, L.C., Lewis,
F.G. & Rawlinson, C.H. (1989) Natural trace metal concentrations in estuarine
and coastal sediments of the southeastern United States. Environmental Science
and Technology 23, 314-20.
Zwolsman, J.J.G., Berger, G.W. & Van Eck, G.T.M. (1993) Sediment accumulation rates,
historical input, postdepositioned mobility and retention of major elements and
trace metals in salt marsh sediments of the Scheldt estuary, SW Netherlands.
Marine Chemistry 44, 73-94.
o o o o o o o o o o o o o o o o o o o
Modelling adsorption and desorption
processes in estuaries
A. Turner and A.O. Tyler

3.1 Introduction
Many inorganic and organic pollutants are of great concern to water
quality managers owing to their persistence, toxicity and liability to
bioaccumulate (Tanabe, 1988; Robards & Worsfold, 1991; Bryan &
Langston, 1992; McCain et al, 1992; Forstner, 1993). The major temporary
or ultimate sink for such pollutants in estuaries is the sedimentary
reservoir, including inter tidal deposits (Hanson, Evans & Colby, 1993;
van Zoest & van Eck, 1993; Kennicutt et a/., 1994), and definition of the
biogeochemical mechanisms by which they absorb onto, desorb from and
repartition amongst natural, heterogeneous particle populations is essential
in order to assess their environmental fate. In estuaries, prediction of the
distribution of pollutants is further compounded by intense temporal and
spatial gradients of the reaction controlling variables such as salinity,
dissolved oxygen concentration and particle composition, occurring both
within the sediment and in the water column during particle suspension.
This chapter focuses on an empirical technique to study the sorptive
behaviour of trace metals and trace organic compounds in estuaries.
Thus, the partitioning of constituents between particles and solution is
determined experimentally, without identifying the inherent reaction
mechanisms or reactant speciation, under controlled laboratory conditions
using natural samples spiked with radio tracer analogues of the constituent
of interest (Turner et a/., 1993). Adsorption and desorption processes may
be modelled as a function of particle concentration, and the controlled
variables of salinity and dissolved oxygen concentration, by incorporating
empirically derived results into simple mass balance equations. This
approach is exemplified herein using site-specific results from two
contrasting estuarine environments, namely the Clyde and Humber, and
the calculated results are discussed in the context of their agreement with
 Modelling adsorption and desorption processes 43

field measurements of trace constituents in these estuaries. Although the

arguments and discussion are presented explicitly for sorption processes
occurring in the water column they may be extrapolated to the intertidal
environment where chemical reactivity is also promoted by changes in
dissolved oxygen concentration (Kerner & Wallmann, 1992) and salinity
(Carroll et al.9 1993).

3.1.1 The Clyde and Humber Estuaries

The physical characteristics of the Clyde and Humber Estuaries are listed
in Table 3.1, and typical axial distributions of salinity, concentration of
suspended particulate matter (SPM) and dissolved oxygen concentration
are shown in Figure 3.1. The Clyde is a partially stratified, mesotidal
estuary. Irregularities in its axial salinity distribution arise from a
multiplicity of fresh water sources along the tidal estuary. The SPM
concentrations are low and relatively invariant and an occasional, weak
turbidity maximum results from trapping of fine sediment by convergent
subsurface flows (Curran, 1986). The high biochemical oxygen demand of
sewage discharges combined with the poor exchange between subsurface
and surface waters results in an extensive area of oxygen depletion in the
upper estuary (Curran, 1986). In contrast, the Humber is a well-mixed,
macrotidal estuary, and a turbidity maximum generated by tidal
resuspension of bed sediment is a regular feature of the low salinity zone
(Turner, Millward & Morris, 1991).

3.2 Determination of distribution coefficients

The distribution coefficient, or partition coefficient, X D (v/w), parameterises
the ratio of adsorbed particulate concentration, P (w/w), to dissolved
concentration, C (w/v), of a constituent, and may be determined empirically

Table 3.1 Physical characteristics of the Clyde and Humber Estuaries

Clyde0 Humber*

Axial length (km) 40 62

Mean river flow (m3 s" 1 ) 44 ( + 63C) 246
Tidal range at mouth, neaps/spring (m) 1.9/3.0 3.5/6.2
Residence time (days) 10-20 -40
Riverine sediment discharge (t a" 1 ) 83 000 170000

a
Mackay & Leatherland (1976); Curran (1986)
* Turner et al (1991)
c
contribution from tributaries to the tidal estuary
44 A. Turner and A.O. Tyler

CLYDE HUMBER

10 20 30 20 30 40

Distance (km) Distance (km)

10 20 30 20 30 40
Distance (km) Distance (km)

10 20 30 20 30 40

Distance (km) Distance (km)

Figure 3.1 Distributions of salinity, suspended particulate matter (SPM)

and dissolved oxygen determined during axial samplings of the Clyde (8/92;
ebb tide, river flow including tributary inputs= 152 m3 s~1) and Humber (1/88;
flood tide, river flow = 234m3 s~1). Solid lines denote near surface
measurements and broken lines denote near bottom measurements.
Distance is measured from the weir at Glasgow on the Clyde and from
Trent Falls on the Humber.
 Modelling adsorption and desorption processes 45

by doping natural samples with radiotracer analogues and analysing the

adsorbed and dissolved reactant concentrations by a suitable radiochemical
method.
Water samples used for the experimental studies of trace metals and
chlorinated organic compounds were collected from the shore near to the
river and marine end-members of the Humber and Clyde Estuaries and
stored cool and in the dark before being used. Trace metal studies using
Humber samples were undertaken onboard Royal Research Ship Challenger
within 48 hours of collection and are described in detail elsewhere (Turner
et a/., 1993); the remaining experiments were undertaken in the laboratory
in Plymouth within one week of sample collection. The simulation of
estuarine mixing was achieved by mixing aliquots of unfiltered end-member
samples in different proportions. For the trace metal studies mixed
samples of 50 or 100 ml were contained within acid-cleaned 150 ml
polyethylene vessels and spiked with 50 or 100 |il of a cocktail consisting
of sub-nanomolar concentrations of 109 Cd and 137 Cs (Amersham
International) in dilute HC1. This resulted in a small reduction in pH in
the Humber experiments which was neutralised by the addition of a spike
of ammonia solution diluted accordingly. The samples were equilibrated
for 120 hours and filtered through pre-weighed 0.45 |im pore size
Millipore cellulose acetate filters mounted in a Millipore filtration unit.
The filtrates and filters were stored in unused 150 ml polyethylene vessels
and 50 mm diameter petri dishes, respectively, and their activities were
determined using a high purity Ge detector connected to a Canberra
Series 80 Multichannel Analyser. Disintegrations were counted for 1000
seconds and this gave statistical counting errors ( ± 0.5 a) of < 5%. The
distribution coefficient, KD (ml g" 1 ), is then given as follows:

K = — = — E'— (3 I)
C Acmf

where Ap and Ac are the activities on the particles and in solution,

respectively, m is the mass of particles on the filter (g), V is the volume of
filtrate (ml), a n d / i s a geometry factor which corrects for the differential
sensitivity of the detector towards filter and filtrate samples. Replicate
experiments using estuarine samples have indicated that KDs determined
by this technique are generally reproducible within 10% (Turner &
Mill ward, 1994).
The effect of oxygen depletion on the partitioning was investigated by
bubbling nitrogen through a Clyde river end-member sample for 12 hours
46 A. Turner and A.O. Tyler

prior to spiking and thereafter throughout the incubation. This achieved a

dissolved oxygen concentration of 1.5 mg 1"* compared with 8.0-8.5 mg
1 ~* under ambient, aerated laboratory conditions, and thus approximates
the observed range in concentrations of the upper reaches of the estuary
(Figure 3.1).
Two 14C-labelled chlorinated organic compounds were selected for
investigation; 2,3,7,8-TCDD (Chemsyn), the most toxic of the dioxin
congeners; and 22/55'-TCB, IUPAC 52 (Sigma Chemicals), an abundant
PCB congener. A 25 JLXI spike of 0.02 |aCi compound in hexane was placed
on the side of a hexane-washed 50 ml glass centrifuge tube using a glass
microsyringe; the hexane was then evaporated under a laminar flow hood
and a sample of 20 ml was added (Zhou, Rowland & Mantoura, 1995;
Zhou & Rowland, 1997). After shaking the stoppered tube for 12 hours,
the contents were centrifuged at 3000 rpm for 30 minutes to separate the
particulate and dissolved phases. A 2 ml aliquot of supernatant was
pipetted into a glass vial containing 10 ml Ultima Gold Liquid Scintillation
Cocktail, taking care not to disturb particles at the bottom of the
centrifuge tube. The particles were then discarded along with the
remaining water and the residual compound adsorbed to the centrifuge
walls was extracted in 4 ml hexane for 12 hours on a shaker. A 2 ml aliquot
of the extract was then pipetted into a glass vial containing 10 ml
scintillation cocktail. All sample vials were counted on a Philips 4700
scintillation counter and calibrated against matrix-matched standards
quenched using CC14. The KD (ml g"~ *) was derived from mass balance as
follows:

C Ac m

where As9 Ac and Aw are the respective activities of the original spike, the
supernatant and that adsorbed onto the walls of the centrifuge tube, Kis
the volume of supernatant (ml) and m is the mass of particles (g) in the
original suspension derived from subsample filtration through pre-weighed
Whatman GF/F filters. Experiments were performed in triplicate or
quadruplicate and the reproducibility of the resulting KDs was generally
better than 10% for 22'55'-TCB, and up to about 40% for 2,3,7,8-TCDD
due to its poor solubility and consequent low activity in solution.
 Modelling adsorption and desorption processes 47

3.3 Results and model calculations

3.3.1 Fraction of constituent in solution
The KDs for the trace metals and chlorinated organic compounds are
shown against salinity in Figure 3.2. End-member KDs have been
combined with representative end-member SPM concentrations (SPM)
(Figure 3.1) in order to calculate the fraction of constituent in solution, £,
as follows:

K D (mlg- 1 )

12,3,7,8
f -TCDD

10 15 20 25 30
Salinity

K D (mlg- 1 )
10 7
HUMBER
e
10
2,3,7,8
-x -TCDD
10£ - * 22*55*
-TCB

10*-

10'
10 15 20 25 30

Salinity
Figure 3.2 Distribution coefficients, /CDs, against salinity for the Clyde and
Humber mixed samples. Boxed data points denote KDs determined using
the sample incubated under de-oxygenated conditions.
48 A. Turner and A.O. Tyler

* C + PSPM (3.3)
1
" 1 + KDSPM

The magnitude of this fraction (Table 3.2), which has important implications
concerning the bioavailability, transport and fate of these constituents
(Balls, 1989; Pankow & MacKenzie, 1991; Webster & Ridgway, 1994)
varies from more than 90% for Cd and Cs in sea water and oxygen
depleted waters of the Clyde, to less than 5% for 2,3,7,8-TCDD in the
Humber Estuary and the low salinity zone of the Clyde.

3.3.2 Salinity and particle character dependence of partitioning

All constituents in the Clyde exhibit a reduction in KD with increasing
salinity; in the Humber, KDs for Cd and Cs decrease with increasing
salinity whereas KDs for 22'55-TCB and 2,3,7,8-TCDD are relatively
invariant. Regarding trace metals, although inter-estuarine variability of
end-member KDs arises through differences in end-member particle and
water composition, the distribution of KDs induced by mixing of end-
members is controlled principally by salinity and not changes in net
particle character (Turner et al, 1993). Thus, a reduction in KD with
increasing salinity, S, which is characteristic of many metals (Li, Burkhardt
& Teraoka, 1984), results from an increase in the proportion of the
sorbable species being complexed by sea water anions (Cl~ and SO 4 2 ~)
and an increasing occupation of particle sorption sites by sea water

Table 3.2 The percentage of trace metals and chlorinated organic compounds
in solution, calculated according to Equation (3.3) using end-member KDs and
representative concentrations of suspended particulate matter (SPM). REM and
MEM denote river and marine end-members, respectively

SPM(mg I" 1 ) Cd Cs 22'55'-TCB 2,3,7,8-TCDD

Humber
REM 100 75.1 57.0 15.2 4.2
MEM 100 93.0 94.2 8.5 3.0
Clyde
REM 10 71.5 95.7 35.8 3.1
REM* 10 98.8 98.7 — —
MEM 10 99.4 99.9 75.0 11.2

*De-oxygenated river end-member sample

 Modelling adsorption and desorption processes 49

cations (Na + , K + , Ca 2 + , Mg 2 + ), and may be defined by the following

general relationship (Bale, 1987; Turner & Millward, 1994):

lnKD = b-ln(S + 1) + lnK° (3.4)

where K£, the partitioning in fresh water, and b are constants. Values of
these constants for Cd and Cs in the Clyde and Humber have been derived
from regressions of In KD versus In (S + 1) and are given in Table 3.3. The
parameterisation of the sorptive behaviour of constituents in the form of
such generic relationships is extremely useful for the refinement of
chemical codings in contaminant transport models (Plymsolve, 1991; Ng
et al9 1996).
Regarding chlorinated organic compounds, the information presented
herein and elsewhere (Tyler et al, 1994) suggests that particle character
rather than salinity exerts the key influence on their estuarine distributions.
The chemical characteristics of typical Clyde and Humber end-member
particles are given in Table 3.4. Thus, Humber river and marine
end-member particles have similar chemical characteristics engendering a
relatively uniform distribution of KD, whereas in the Clyde, riverine
particles have a higher content of Fe-Mn hydroxides and carbon for the
sorption of compounds and a reduction in KD is observed as the
proportion of riverine particles in the admixture decreases. More specifically,
recent studies suggest that adsorption of trace organic compounds onto
natural particles is controlled dominantly by the apolar lipid content
(Preston & Raymundo, 1993; Tyler et a/., 1994). Higher absolute K Ds in
the Clyde than in the Humber are, therefore, reflected by inter-estuarine
differences in particulate lipid content (Clyde sediment = 6.5 + 5.0 mg
g " 1 cf. Humber sediment 1.6 + 0.7mg g" 1 ; Tyler et a/., 1994).

Table 3.3 Regression analyses of In/CD versus ln(S+1) (n = 5)forCd andCs (see
Equation (3.4))

Intercept, X D ° r2
1
Estuary Slope, b (mlg" ) (%) p

Clyde Cd - 1.42 80800 96.5 0.002

Cs - 1.25 11000 95.8 0.002
Humber Cd - 0.653 6000 92.8 0.005
Cs - 1.05 21000 99.0 < 0.001
50 A. Turner and A.O. Tyler

3.3.3 Desorption modelling

The foregoing discussion suggests that the estuarine distributions of trace
organic compounds are determined largely by natural particle mixing
processes and the rates at which sorbed compounds redistribute amongst
different particle types. In contrast, the partitioning of trace metals in
estuaries is controlled principally by salinity (Equation (3.4)) and their
distributions will be determined to some extent by desorption from
seaward fluxing particles. Assuming no loss or gain of seaward fluxing
particles, and neglecting possible additional sources of trace metals such
as pore-water infusion, this effect may be quantified from mass balance of
constituent as follows (Li et a/., 1984):

CS1 + pSl.SpMf = CS2 + pS2. S p M f ( 3 5)

where SPMr is the concentration of seaward fluxing particles traversing

an axial or vertical salinity gradient, S1-S2. Assuming that the sorption
reaction is fully reversible, the relative changes in adsorbed particulate
and dissolved concentrations along this salinity gradient are given as follows:

PS2
[ j
PS1 " SPM{

and:
r-S2 pS
- (3 7)
CS1
PslKl2 (
'
Table 3.4 The composition of typical end-member particles from the Clyde and
Humber. Fe and Mn were determined following extraction by hydroxylamine
hydrochloride-acetic acid. Carbon and nitrogen were determined using an
elemental analyser. Specific surface area (SSA) was determined using a BET
nitrogen adsorption technique (nd = not determined). REM and MEM denote
river and marine end-members, respectively.

Fe Mn Carbon Carbon/ SSA

(mgg" 1 ) (mg g"') (%) Nitrogen (m2 g"')
Clyde
REM 8.40 1.02 9.9 9.2 7.1
MEM 3.52 0.29 9.3 10.3 nd
Humber*
REM 10.1 1.32 4.9 15.8 31.4
MEM 11.2 1.05 5.2 13.9 20.6

a
Turner et al. (1993)
 Modelling adsorption and desorption processes 51

Note that these equations are also valid for trace organic compounds if
the seaward fluxing particle population is chemically modified in situ
through, for example, precipitation or dissolution of Fe-Mn oxides
and/or organic coatings.
The relative change in dissolved or adsorbed concentration of trace
metals due to the flux of particles along a dissolved oxygen gradient,
O2-O1, may be calculated likewise using KDs determined under different
oxygen conditions. Thus:

P°2 SPM< + I/Kg 1

P ~ SPMC + I/Kg 2
O1 (
'

and:
/^O2 pO2 fOl

Figure 3.3 shows dissolved and adsorbed particulate concentration ratios

for Cd as a function of seaward fluxing particle concentration, SPMf,
calculated according to Equations (3.6) and (3.7) using end-member KDs
for the Humber and Clyde, and according to Equations (3.8) and (3.9)
using X D s determined under extreme dissolved oxygen concentrations in
the Clyde. Although for a given concentration, SPM f , the relative change
in dissolved concentration in the Clyde effected by variations in both
salinity and dissolved oxygen is greater than the relative change in
dissolved concentration in the Humber effected by variations in salinity,
the magnitude of the advecting particle population is considerably
smaller in the Clyde. Thus, for a representative concentration of SPMr in
the Clyde of ~ lOmg 1" x (Figure 3.1), CS2/CS1 = CO2/COi ~ 1.4, pS2/psl
~ 0.02 and pO2/pO1 ~ 0.04. Particle fluxes in dynamic macro tidal estuaries
are spatially and temporally variable (Morris et al, 1986), but a hypothetical
particle population of concentration 250mg I " 1 traversing the full
salinity range of the Humber Estuary yields the following concentration
ratios: C S2 /C S1 = 1.5,andP S2 /P sl = 0.35. According to these calculations
the largest relative change in Cd concentration occurs in Clyde suspended
particles and results in a 25- to 50-fold reduction of the adsorbed
particulate fraction.

3.3.4 Adsorption modelling

Sorptive removal of dissolved constituents is engendered by the addition
of sorption sites. In estuaries this is achieved through tidally induced
52 A. Turner and A.O. Tyler

resuspension of bed particles that are depleted of the adsorbed constituent

relative to equilibrium. This effect may be quantified by the following
equation derived from mass balance (Morris, 1986):

C_ KD-SPM°
(3.10)
c° 1 + KDSPM° + KD-SPM%1 - a)
where C/C° is the ratio of dissolved constituent concentration in
turbidised water to that in ambient water (i.e. fractional removal), SPMa is
the concentration of added or resuspended particles, SPM° is the
concentration of ambient suspended particles, and a defines the depletion

20 40 60 80 100
1
SPM (mg I" )

200 400 600 800 1000

f 1
SP/W (mg I )

Figure 3.3 Dissolved and adsorbed particulate concentration ratios as a

function of fluxing particle concentration (SPMf) for Cd in the Clyde and
Humber, calculated according to Equations (3.6M3.9).
 Modelling adsorption and desorption processes 53

of adsorbed constituent on the added particles relative to equilibrium in

the turbidised water (i.e. Pa/P). Such constituent-depleted particles are
derived from the marine reaches of an estuary where anthropogenic
inputs are generally less significant and particle sorption sites are
occupied by sea water cations, and are transported up-estuary by
asymmetric tidal currents. The calculated sorptive removal of constituents
is shown for the Humber Estuary in Figure 3.4 within an added particle
concentration range of 0-1000 mg 1 ~ *, using river end-member KDs and
an ambient particle concentration of 5mg I" 1 ; the value of a of 0.50 is
based on empirical estimates given by Morris (1986). Under these
conditions 50% removal of dissolved constituent is achieved at the
following approximate added particle concentrations: Cd, 700mg I" 1 ;
Cs, 250mg I" 1 ; 22'55'-TCB, 50mg I" 1 ; 2,3,7,8-TCDD, 20mg I" 1 .
Application of the sorption model to the Clyde Estuary is not appropriate
for the following reasons. Firstly, although localised disturbance of the
bed occurs during dredging operations, there is no evidence of extensive
resuspension from axial and vertical SPM distributions (Figure 3.1).
Moreover, bottom current speeds are lower than the critical erosion
velocity for representative grain diameters (Curran, 1986). Secondly, there
is little potential for up-estuary transport of marine particles due to slow
subsurface currents and deep basins in the outer estuary which act as

O
O

2,3,7,8
-TCDD
0.01
200 400 600 800 1000
SPM* (mg I1)
Figure 3.4 The removal of dissolved trace metals and chlorinated organic
compounds as a function of added particle concentration {SPMa) in the
river end-member of the Humber Estuary, calculated according to Equation
(3.10) for SPM°= 5mg M and a = 0.50.
54 A. Turner and A.O. Tyler

sediment traps (Mackay & Leatherland, 1976). It is conceivable, however,

that removal of certain trace metals (including Cd) may be achieved via
the precipitation of insoluble sulphide species (Zwolsman & van Eck,
1993). A continual source of sulphide (namely, the reducing bottom
sediments) is not accounted for in the laboratory simulations and this
serves to exemplify potential limitations with our attempts to replicate
multiple, interactive environmental processes in laboratory enclosures.

3.4 Discussion
Although chemical parameters such as distribution coefficients are
fundamental to understanding the geochemical reactivity of trace
CLYDE

b<0

_/T_S_hajqcline r \ -C-*\ <

REM
MEM

| low:oxygen kprie
_±J_ ; t i : : ; ; ; N; ;

HUMBER
b<0

SPM' turbidity SPM1

maximum

MEM REM
SPMC

bed load

oc<1

Figure 3.5 Schematic illustration of particle fluxes in the Clyde and

Humber Estuaries. Arrow size denotes relative magnitude of the flux. The
broken line represents the boundary of the saline intrusion and the river.
REM and MEM denote river and marine end-members, respectively. All
terms are defined in the text.
 Modelling adsorption and desorption processes 55

constituents in estuaries, the foregoing calculations have emphasised the

important role of sediment dynamics on chemical distributions. In the
Humber Estuary sorption reactions are promoted by substantial resus-
pending and advecting particlefluxes,as conceptualised in Figure 3.5, and
the suspended particle populations are in a dynamic state of exchange
with subtidal and intertidal material. In highly energetic, macrotidal
estuaries a proportion of riverine suspended particles are transported into
the turbidity maximum, where the magnitude of the seaward advecting
population is augmented by particles resuspended by tidal currents
(Morris et a/., 1986). Trace metals are desorbed from these particles
downestuary in accordance with the magnitude of b, the rate of change of
KD with salinity, until they are deposited as the current speeds are
reduced. An admixture of this sediment and marine derived material is
then transported up-estuary as bedload by asymmetric tidal currents to
the null zone where it is resuspended into the turbidity maximum
engendering sorptive removal of dissolved constituent in accordance with
the value of oc, the extent of depletion (relative to equilibrium) of adsorbed
constituent on the particles. Field investigations of various dissolved trace
metals (diagnosed as constituent-salinity plots) in the Humber Estuary
have demonstrated conservativeness (Balls, 1985; Kitts et al, 1994) as well
as a variety of reactive distributions (Gardiner, 1982; Edwards, Freestone
& Urquhart, 1987; Coffey & Jickells, 1995). Differential behaviour
amongst constituents reflects differences in chemical parameters such as
end-member KDs, ot, and the reversibility of sorption, whereas intra-
constituent variability arises from temporal variability of component
sediment fluxes.
Chemical reactivity in the Clyde is governed by the flux of particles
through lateral and vertical gradients of salinity or dissolved oxygen and
there is little exchange of particles between the water column and the bed
(Figure 3.5). Because of the low and invariant concentrations of suspended
particles, calculations predict that sorptive reactivity in the water column
will result in a significant modification of the trace metal composition of
suspended particles. Seasonal measurements of trace metals on suspended
particles available to 1M NH4OAc (an operational measure of an
exchangeable particulate fraction) indicate a persistent reduction from the
upper estuary to the lower estuary (Muller, Tranter & Balls, 1994); for Cd,
an exchangeable particulate concentration ratio, XS2/XS1, is about 0.20.
Although an order of magnitude greater than the adsorbed particulate
ratio calculated (Equation (3.6)), reflecting analytical difficulties associated
with the determination of an exchangeable fraction and the assumption in
56 A. Turner and A.O. Tyler

the calculations that the adsorbed metal is completely exchangeable

(Equations (3.6)-(3.9)), these field measurements demonstrate a distinct
seaward modification of the Cd content of the suspended particle
population. Calculations also predict that accompanying changes in the
dissolved phase are likely to be difficult to detect. Thus, conservative
distributions of dissolved trace metals are generally observed (Mackay &
Leatherland, 1976; Muller, Tranter & Balls, 1994), deviations from the
theoretical dilution line being accounted for by multiple straight line
segments resulting from a variety of tributary inputs of different compositions
(Mackay & Leatherland, 1976), and direct anthropogenic inputs to the
tidal estuary (Muller Tranter & Balls, 1994).
Although the arguments and calculations presented herein are qualitat-
ively compatible with field observations from two contrasting estuaries,
accurate quantitative predictions will rely on a better definition of
chemical parameters such as a and sorption time constants, and a greater
understanding of estuarine sediment dynamics including the interaction
between suspended populations and subtidal/intertidal deposits. Never-
theless, the partitioning data and sorption models are particularly useful
per se in providing a general understanding of the likely environmental
transport and fate of constituents about which relatively little is known,
such as PCBs and dioxins.

References
Bale, A.J. (1987) The characteristics, behaviour and heterogeneous reactivity of estuarine
particles. Ph.D. Thesis, Polytechnic South West, Plymouth, UK, 216 pp.
Balls, P.W. (1985) Copper, lead and cadmium in coastal waters of the western North Sea.
Marine Chemistry 15, 363-78.
Balls, P.W. (1989) Trend monitoring of dissolved trace metals in coastal sea water - a waste
of effort? Marine Pollution Bulletin 20, 546-8.
Bryan, G.W. & Langston, W.J. (1992) Bioavailability, accumulation and effects of heavy
metals in sediments with special reference to United Kingdom estuaries: a
review. Environmental Pollution 76, 89-131.
Carroll, J., Falkner, K.K., Brown, E.T. & Moore, W.S. (1993) The role of the
Ganges-Brahmaputra mixing zone in supplying barium and 2 2 6 Ra to the Bay of
Bengal. Geochimica et Cosmochimica Ada 57, 2981-90.
Coffey, M.J. & Jickells, T.D. (1995) Ion chromatography-inductively coupled plasma-
atomic emission spectrometry (IC-ICP-AES) as a method for determining trace
metals in estuarine water. Estuarine, Coastal and Shelf Science 40, 379-86.
Curran, J.C. (1986) Effluent disposal and the physical environment. Proceedings of the
Royal Society of Edinburgh 90B, 97-115.
Edwards, A., Freestone, R. & Urquhart, C. (1987) The Water Quality of the Humber
Estuary, 1986. A Report of the Humber Estuary Committee, 27 pp.
 Modelling adsorption and desorption processes 57

Forstner, U. (1993) Metal speciation - general concepts and applications. International

Journal of Environmental and Analytical Chemistry 51, 5-23.
Gardiner, J. (1982) Nutrients and persistent contaminants. In A.L.H. Gameson (ed.) The
Quality of the Humber Estuary. Yorkshire Water Authority, Leeds, 27-33.
Hanson, P.J., Evans, D.W. & Colby, D.R. (1993) Assessment of elemental contamination
in estuarine and coastal environments based on geochemical and statistical
modelling of sediments. Marine Environmental Research 36, 237-66.
Kennicutt, M.C. II, Wade, T.L., Presley, B.J., Requejo, A.G., Brooks, J.M. & Denoux, G.J.
(1994) Sediment contaminants in Casco Bay, Maine: inventories, sources, and
potential for biological impact. Environmental Science and Technology 28, 1-15.
Kerner, M. & Wallmann, K. (1992) Remobilisation events involving Cd and Zn from
intertidal flat sediments in the Elbe Estuary during the tidal cycle. Estuarine,
Coastal and Shelf Science 35, 371-93.
Kitts, H.J., Millward, G.E., Morris, A.W. & Ebdon, L. (1994) Arsenic biogeochemistry in
the Humber Estuary, U.K. Estuarine, Coastal and Shelf Science 39, 157-72.
Li, Y.-H., Burkhardt, L. & Teraoka, H. (1984) Desorption and coagulation of trace
elements during estuarine mixing. Geochimica et Cosmochimica Acta 48,1659-64.
McCain, B.B., Chan, S-L, Krahn, M.M., Brown, D.W, Myers, M.S., Landahl, J.T, Pierce,
S, Clark, R.C. & Varanasi, U. (1992) Chemical contamination and associated
fish diseases in San Diego Bay. Environmental Science and Technology 26,
725-33.
Mackay, D.W. & Leatherland, T.M. (1976) Chemical processes in an estuary receiving
major inputs of industrial and domestic wastes. In J.D. Burton & P.S. Liss (eds.),
Estuarine Chemistry. Academic, London, 185-218.
Morris, A.W. (1986) Removal of trace metals in the very low salinity zone of the Tamar
Estuary, England. The Science of the Total Environment 49, 297-304.
Morris, A.W., Bale, A J , Howland, R.J.M, Millward, G.E, Ackroyd, D.R, Loring, D.H. &
Rantala, R.T.T. (1986) Sediment mobility and its contribution to trace metal
cycling and retention in a macrotidal estuary. Water Science and Technology 18,
111-19.
Muller, F.L.L, Tranter, M. & Balls, P.W. (1994) Distribution and transport of chemical
constituents in the Clyde Estuary. Estuarine, Coastal and Shelf Science 39,105-26.
Ng, B , Turner, A, Tyler, A.O, Falconer, R.A. & Millward, G.E. (1996) Modelling
contaminant geochemistry in estuaries. Water Research 30, 63-74.
Pankow, J.F. & MacKenzie, S.W. (1991) Parameterising the equilibrium distribution of
chemicals between the dissolved, solid particulate matter, and colloidal matter
components in aqueous systems. Environmental Science and Technology 25,
2046-53.
Plymsolve (1991) An Estuarine Contaminant Simulator. User Manual. Natural Environment
Research Council, Swindon, 91 pp.
Preston, M.R. & Raymundo, C.C. (1993) The associations of linear alkyl benzenes with the
bulk properties of sediments from the River Mersey Estuary. Environmental
Pollution 81, 7-13.
Robarbs, K. & Worsfold, P. (1991) Cadmium: toxicology and analysis. A review. Analyst
116, 549-68.
Tanabe, S. (1988) PCB problems in the future: foresight from current knowledge.
Environmental Pollution 50, 5-28.
Turner, A. & Millward, G.E. (1994) The partitioning of trace metals in a macrotidal
estuary. Implications for contaminant transport models. Estuarine, Coastal and
Shelf Science 39, 45-58.
58 A. Turner and A.O. Tyler

Turner, A., Millward, G.E. & Morris, A.W. (1991) Particulate metals from five major
North Sea estuaries. Estuarine, Coastal and Shelf Science 32, 325-46.
Turner, A., Millward, G.E., Bale, A.J. & Morris, A.W. (1993) Application of the KD concept
to the study of trace metal removal and adsorption during estuarine mixing.
Estuarine, Coastal and Shelf Science 36, 1-13.
Tyler, A.O., Millward, G.E., Jones, P.H. & Turner, A. (1994) Polychlorinated dibenzo-para-
dioxins and polychlorinated dibenzofurans in sediments from U.K. estuaries.
Estuarine, Coastal and Shelf Science 39, 1-13.
van Zoest, R. & van Eck, G.T.M. (1993) Historical input and behaviour of hexachlorobenzene,
polychlorinated biphenyls and polycyclic aromatic hydrocarbons in two dated
sediment cores from the Scheldt Estuary, SW Netherlands. Marine Chemistry 44,
95-103.
Webster, J. & Ridgway, I. (1994) The application of the equilibrium partitioning approach
for establishing quality criteria in two UK sea disposal and outfall sites. Marine
Pollution Bulletin 28, 653-61.
Zhou, J.L. & Rowland, S.J. (1997) The sorption of hydrophobic pyrethroid insecticides to
estuarine particles: a compilation of recent research (this volume).
Zhou, J.L., Rowland, S.J. & Mantoura, R.F.C. (1995) Partition of synthetic pyrethroid
insecticides between dissolved and particulate phases. Water Research 29,1023-31.
Zwolsman, J.J.G. & van Eck, G.T.M. (1993) Dissolved and particulate trace metal
geochemistry in the Scheldt Estuary, S.W. Netherlands (water column and
sediments). Netherlands Journal of Aquatic Ecology 27, 287-300.
o o o o o o o o o o o o o o o o o o o
A critical appraisal of the methodology
used in studies of material flux between
saltmarshes and coastal waters
K. Carpenter

4.1 Introduction
In the context of saltmarsh/coastal flux studies, afluxis the net exchange
of a substance between these two ecosystems. Thefluxesof several types of
material, including nutrients, trace metals, herbicides and radionuclides,
have been studied by researchers. Although this paper primarily appraises
the methodology of nutrient flux estimations, many of the problems are
also shared by researchers working on other types of material fluxes.
Since thefirstdetritalfluxstudies conducted on Sapelo Island, Georgia,
USA (Teal, 1962; Odum & de la Cruz, 1967), there has been a research
interest in nutrient fluxes between saltmarshes and the adjacent coastal
water. The major drive behind this work was to ascertain whether
saltmarshes affect the concentration of dissolved inorganic nutrients and
detritus (and hence the ecology) of coastal water. The link between
saltmarsh processes and coastal productivity may be of relevance to
coastal managers who wish to predict the impact of future saltmarsh loss
(due to reclamation, erosion or drowning due to sea level rise) or possibly
saltmarsh gain (if managed retreat becomes an extensively used strategy
for coastal defence in rural areas).
The published flux results indicate that some saltmarshes appear to
process nutrients in a very different way to others (Nixon, 1980; Carpenter,
1993). These discrepancies could be due to actual differences in nutrient
budgeting caused by variations in saltmarsh characteristics. However, the
disparity may be artificial, caused by differences infieldmethodology and
the large uncertainties involved in the calculation of tidal and annual
fluxes. It is difficult to differentiate between these two possibilities as error
margins have never been reported for the published flux values. The total
error on afluxvalue includes both methodology errors (e.g. uncertainties
60 K. Carpenter

in the measurement of water discharge and nutrient concentration) and

errors due to extrapolating instantaneous fluxes in space and time to
produce tidal and annualfluxes.The resulting potential error margin may
be larger than the calculatedflux.Perhaps confidence limits are omitted by
some researchers due to a natural reluctance to draw attention to this issue.
In this paper, I aim to draw attention to the large uncertainties which
are involved in the estimation of annual nutrient flux values, using the
various methodologies in current use. In particular, the paper highlights
the differences in methodology (both field methods and during flux
calculations) of flux studies to date, and how these would lead to major
differences in the apparent flux even if none actually existed. It is hoped
that by doing so, these errors may be reduced in future studies and that the
data collected will be comparable. Only then may it be possible to
establish how variability in saltmarsh characteristics such as plant species,
temperature, and sediment type affect nutrient flux. This would bring
researchers closer to the goal of developing a quantitative model that
could apply to unsurveyed marshes to estimate their nutrient budgets and
assess their influence on coastal ecology.

4.2 Differences in methodology which would

cause variation in the calculated flux
A review paper published by Nixon in 1980 painstakingly brought
together the research carried out in the previous 20 years on the role of
saltmarshes in estuarine productivity and water chemistry. After considering
the errors and assumptions of estimating annual flux using the direct
method, he concluded that no research project had produced an accurate
flux value. The major sampling problems included poor discharge
estimates, incomplete knowledge of nutrient transport mechanisms and
the unknown effects of storm tides on the total flux of materials (Wolaver
et a/., 1983). Twelve years later this conclusion still stands as there has not
been a significant improvement in the methods.
There is no standard method for determining the flux between
saltmarshes and the coastal water. Differences in methodology arise
during both the field work and the method of flux calculation. Table 4.1
lists the stages influxmeasurement which have been tackled differently by
saltmarsh researchers. Each of these stages will be discussed individually
and I aim to point out that for a particular saltmarsh, the apparentfluxof
a substance would vary substantially depending on the set of methods
chosen.
 Methodology used in studies of material flux 61

(a) Field technique chosen for flux measurement

There are four main field techniques which have been used to quantify
nutrient fluxes. These are:

(a) community budgeting (e.g. Teal, 1962; Day et a/., 1973)

(b) direct tidal creek measurements (e.g. Axelrad, 1974; Valiela et a/.,
1978; Roman, 1984; Dankers et al9 1984; Whiting et al, 1985;
Azni, Abd. Aziz & Nedwell, 1986)
(c) flume studies (e.g. Wolaver et al, 1983; Wolaver & Spurrier, 1988;
Whiting et al.9 1989)
(d) diffusion chamber studies (e.g. Lee, 1979; Scudlark & Church,
1989; Chambers, 1992)

These methods are illustrated in Figure 4.1.

Methods (b), (c) and (d) usually involve sampling over a flood/ebb tidal
cycle; the calculated tidal fluxes are subsequently extrapolated to give an
estimate of the annual flux. Historically, method (a) was the first used. It is
an indirect approach, whereby nutrient flux is derived from the difference
between the annual production and aerobic decomposition. Teal used this
method for his study published in 1962, which first cited saltmarshes as
exporters of organic carbon. This method has the advantage of being long
term, unlike tidal studies, which can only take snapshots of the flux during
the sampled tides and may not be representative of the seasonal average.
However, due to its indirectness, and the assumptions made (e.g. there was
no anaerobic decomposition, nitrification or denitrification), this type of
flux measurement was superseded by method (b).

Table 4.1 Stages in the methodology in which differences would change the
calculated flux value

Field techniques
(a) Technique chosen for flux measurement
(b) Position of sampling station in marsh system
(c) Water sampling strategy
(d) Accuracy of discharge measurement
(e) Frequency of discharge and concentration measurements
(f) Frequency of tidal monitoring
Flux calculations
(g) Tidal flux calculations:
(i) Combination of discharge and nutrient concentration to obtain flux
(ii) Treatment of tidal inequality
(h) Annual flux calculations:
Method of extrapolating results from the sampled tides to give annual flux
62 K. Carpenter

Method (b) is a direct method that measures the amount of nutrient

flowing into and out of the marsh through a closed tidal creek system
during a flood/ebb cycle. To calculate the total number of moles of a
nutrient species entering the marsh system during thefloodingportion of
the tide, the concentration of this nutrient and the rate of water discharge
throughout the flood tide are multiplied together. The number of moles
leaving the marsh with the ebbing tide is similarly calculated and the flux
is the difference between the estimated flood and ebb totals. The value of
the results from this method are greatly related to the accuracy to which
water discharge can be measured (see section (e)). This method integrates
the results of all the nutrient transforming processes occurring in the
various subsystems of the saltmarsh (e.g. creek water column, sediment,
vegetated surface and creek bank) and thus gives the netfluxvalue for the
marsh as a whole, unlike methods (c) and (d).
The flume technique (c) has been used by Wolaver and his colleagues
to measure marsh surface processes on the marshes of North Inlet,
South Carolina, USA. It can only be used for tides that are large enough

Chamber
Studies (d)

Tidal Creek Measurements (b)

Figure 4.1 Schematic representation of four main techniques used to

quantify nutrient fluxes between saltmarshes and the coastal water.
 Methodology used in studies of material flux 63

to spill over the banks and onto the marsh surface. Theflumeconsists of
two walls of plexiglass fencing pushed partly into the sediment, parallel
to the flow of overtopping water, stretching from the creek or mudflat
into the interior of the marsh. To estimate the nutrient flux from this
subsystem, water samples are taken from the end of theflumenearest the
creek, throughout the portion of the tidal cycle when water is flowing
over the marsh surface. The flux is again estimated by combining
nutrient and discharge data, but the flow is calculated from the change
in volume of the water within the flume per unit time instead of using
velocity meters. Water volume estimates are made using water level
elevation and the topography of the marsh surface within the flume
(Wolaver et al, 1983).
It is important to realise that the flume method should not be used to
estimate the overall flux from the saltmarsh. It is only suitable for
assessing the importance of surface processes, such as diffusion of
substances between pore-water and inundating tidal water, uptake by
vegetation and wash off processes. The results can not be extrapolated to
determine the total flux between the saltmarsh and the coastal water, as
the nutrient transforming processes associated with the marsh surface
only play a small role in determining the overall nutrientflux(see Figure 4.2).
The fourth method (d) which has been used to measure saltmarsh
nutrient fluxes is the diffusion chamber technique. The basic apparatus
consists of a perspex cylinder which is partly pushed into the surface of the
marsh andfilledwith water. Periodically water samples are removed from
the chamber and analysed for nutrient species. It would seem logical to
use creek water taken at high tide, so that an increase or decrease in the
nutrient concentration with time would indicate whether the marsh
surface was a sink or source under natural concentration gradients.
Chambers (1992) resolved a major problem associated with the earlier
chamber studies. These had an artificial flooding regime, where the depth
of the water above the marsh surface remained constant throughout the
tidal cycle. He managed to produce a natural flooding regime in the
chamber using an ingeniously simple design, involving a collapsible
reservoir which floods and drains the chamber with water. Water
movement between the reservoir and the chamber is controlled by the
pressure of the water on the marsh surface. As the water that floods the
chamber is isolated and separated from the tidal waters, this experimental
design is ideal to investigate how chemical and biological variables affect
nutrient cycling. Another advantage of the chamber method is that the
flux from small areas can be determined with confidence; this would be
64 K. Carpenter

useful for studying the effect of different plant species, for example.
Furthermore, by comparing the predicted flux values (calculated from
pore-water concentration gradients, using a diffusion model) with the
actualfluxrates (derived fromfluxchamber work), the relative importance
of surface and subsurface processes in regulating sedimentary nutrient
fluxes can be evaluated (Scudlark & Church, 1989).
Like the flume method, fluxes derived from chamber experiments do
not integrate all marsh processes, and only measure changes in nutrient
concentration caused by diffusion or uptake during over bank tides; the
technique may also perturb the system. It is also inappropriate to use this
technique to estimate annual nurient fluxes, via the marsh surface, if the
calculated fluxes are based on constant inundation of the marsh surface
(as were those of Scudlark & Church, 1989) and not the periodic
submergence which actually occurs. It would be relatively easy to correct
for this error, if surface water level had been continually monitored at the
site, by calculating the periods of inundation from the tidal records and
details of the marsh topography. However, it should be noted this
extrapolation is another source of uncertainty.
To summarise,flumeand chamber studies are helpful for investigating
the relative importance of surface diffusion, wash off and plant uptake/release
from aerial shoots, when used in conjunction with direct tidal creek flux
measurements. However, the results from these experiments should not be
extrapolated to give a flux value for the marsh as a whole as they do not
account for sediment drainage, water column and creek bank processes.
Figure 4.2 illustrates why the direct tidal creek method is best for
determining whether a saltmarsh is a source or a sink of nutrients to the
coastal water. The saltmarsh nitrogen cycle has been used as an example
to indicate which of the four main techniques, outlined earlier, accounts
for each of the sources and sinks of nitrogen. Although the direct creek
method treats the saltmarsh as a black box (so that the sites at which the
nutrient transforming processes occur cannot be identified without
further work), it is the only method which integrates the nutrient
transforming processes in all the different subsystems. The other methods
only account for some of these processes.
The difference the field methodology makes to the estimation of the
nutrient flux between a saltmarsh and coastal water can be illustrated
using studies conducted on North Inlet marsh, South Carolina. Wolaver
& Spurrier (1988), used the flume technique to estimate the budget for a
whole marsh. They concluded that North Inlet marsh imported both
phosphate and particulate phosphorus. However, an earlier direct tidal
 Methodology used in studies of material flux 65

study in the same marsh by Whiting et al (1985) reported an export of

these two nutrient species. Although Whiting's study was confined to
eight tidal cycles in a single month (May), and thus cannot be considered
as an annual budget, it still shows that the integrated approach was not in
agreement with the flume study as the latter found the marsh to be an
importer in all seasons. Furthermore, all the other American flux studies
apart from those using chambers orflumes,found saltmarshes to be net
sources of phosphate. Thus, from the field studies, it does appear that
flume or chamber studies do not reflect the total flux from the marsh, as
predicted earlier from the fundamental flaws in the approach.
It is not surprising that flume chamber techniques indicate a different
flux direction for phosphate. It is well known from fertilisation experiments
(e.g. Tyler, 1967; Marshall, 1970; Patrick & Delaune, 1976; Valiela & Teal,
1976) that primary production on saltmarshes is nutrient limited, so
nutrients will be taken out of the overlying water and converted into
biomass, the majority of which will be below ground. After death, some of
(a)-Community Budgeting

(b)-Tidal Creek Studies

(c)-Flume Studies

(d)-Diffusion

Other water column

processes(b)

Figure 4.2 Diagram to illustrate the sources and sinks of fixed nitrogen in
a saltmarsh and to indicate which of these processes are taken into acount
by each of the alternative techniques for the estimation of 'saltmarsh/coastal
water' flux.
66 K. Carpenter

this biomass, plus organic matter from marsh sedimentation, will be

remineralised in the sediment to release the inorganic and organic
nutrients to the pore-waters. Drainage of saltmarsh groundwater into the
creek, followed by tidal flushing, may export these nutrients from the
marsh. Therefore, a marsh could be a source of dissolved inorganic
nutrients even if surface processes suggest it to be a sink.

(b) Position of sampling station in marsh system

The next methodology difference which could cause disparity of flux
results, even if researchers were working on the same marsh, is the
position of the sampling station. Younger, lower marshes seem to function
differently to the older, upper marshes; there is evidence that the former
are net importers while the latter are net exporters (Boorman et al, 1994;
Dame, 1991). The variation in the way nutrients are cycled within these
two areas of the marsh probably relates primarily to hydrological
differences (i.e. the degree of drainage and flushing of the pore-water and
the position of the water table) which in turn relate to differences in plant
community, redox potential and hence biogeochemical processes. Therefore
within the marsh system the annual flux estimate would differ depending
on the position of the sampling station. Figure 4.3 indicates how the siting
of the sampling station may result in different flux estimates. A sampling
station positioned at site C on the upper saltmarsh could indicate an
export of inorganic nutrients, whereas the measured flux at site B (where
the creek drainage basin is in the lower marsh) could show an import.
Station A, at the seaward end of the creek, would probably be the best site
for determining the flux to the coastal water, as the whole saltmarsh
ecosystem would be integrated. Unfortunately, it is not practical to site a
station here as the errors on the measurement of water discharge would be
very large, and so the uncertainties on the estimated flux would be
unacceptable.

(c) Water sampling (recording interval and spatial sampling)

To obtain good estimates of nutrientfluxes,it is important to determine
accurately the variation in its concentration over a tide. Therefore a
sufficient number of samples should be taken to reflect the temporal
variability. It should be noted that the concentration of particulate
material is prone to greater short-term fluctuations than dissolved
constituents and so the former requires more intensive sampling than the
latter. In addition to temporal variability, spatial variability needs also to
 Methodology used in studies of material flux 67

be considered. The water in large creeks is not necessarily well mixed. To

determine an adequate sampling regime an initial experiment should be
conducted to determine the degree of spatial variation in nutrient
concentration across the creek cross-section. Again, the spatial variation
is greater for particulate than dissolved materials as suspended solid
concentration varies strongly with depth, with more sediment moving
close to the bed than at the surface.
To ensure that the water samples maintain the same nutrient
concentration until they are analysed, the samples should be treated (e.g.
by filtering, addition of biocide, freezing) as soon as they are collected.

Figure 4.3 Illustration of how the positioning of sampling stations may

influence the estimated flux, due to the different nature of the drainage
basins.
68 K. Carpenter

Failure to adequately preserve water samples results will compromise the

estimated flux values.

(d) Accuracy of water discharge estimates

The main difficulty with direct tidal creek measurements is obtaining
reliable water velocity/discharge data. The hydrology of tidal creeks is
different fromfluvialchannels due to tidalflowwhich causes rapid change
in water level and the rate of discharge (making accurate gauging by
velocity meters extremely difficult). Furthermore, the bidirectional nature
of water flow in creeks renders gauging by weirs inappropriate.
The uncertainty in discharge measurement is the greatest factor
contributing to the error on the calculated tidal nutrient flux. Velocity
meters have been universally used to gaugeflows;most tide studies have
taken velocity readings from only one or two points across the cross
section of the creek and used this value to estimate instantaneous
discharge across the whole creek. The hydrodynamics of tidal creeks are
complex (Kjerfve & Proehl, 1979) and are not well suited to gauging by
this method, hence the error on an instantaneous discharge estimate is
large. Roman (1984) calculated hourly instantaneous discharge values (i.e.
the rate of discharge through the entire creek cross section) for three tidal
cycles using 18 sampling positions. The average cross-sectional velocity
was found for each hourly interval and these data were regressed with the
velocity values from the 18 sampling positions. The station that best
predicted the average cross-sectional velocity was identified and this
position alone was subsequently used to gauge discharge. Roman
estimated that the average error between the calculated instantaneous
discharges using the single sampling position and the dense array of
meters was 11%. However, the actual error would be much larger than
this as even the densely metered value would still only approximate the
real instantaneous discharge. Furthermore, with only 12 discharge
measurements for the whole tidal cycle, irregular surges in discharge
could be missed, thus the random error on the total discharge volume for
the flood and ebb positions of the tide could be substantial. This error
would have an effect on the value and even the direction of the flux; as
illustrated below. Assume, for example,

Actual tidal volume of flood tide = 4000 m3

Actual tidal volume of ebb tide = 4500 m3
 Methodology used in studies of material flux 69

Actual total N H 4 + in flood tide = 30 moles

Actual total N H 4 + in ebb tide = 40 moles
Therefore, actual flux = — 10 moles

If the measured flood discharge was 30% higher and the ebb 10% lower
than the actual, and the variation in concentration through the tide was
monitored accurately, the results would be:

Estimated flood discharge = 5200 m 3

Estimated tidal volume in ebb = 4050 m 3
Calculated N H 4 + in flood tide = 39 moles
Calculated N H 4 + in ebb tide = 36 moles
T h e r e f o r e , e s t i m a t e d flux = 4 - 3 m o l e s

The hypothetical calculation above illustrates how crucial accurate

discharge data is for accurate flux estimates and the necessity for an
accurate gauging method.
An improvement in the methodology to measure flows through tidal
creeks would be the single most efficient way of reducing the uncertainty
on the calculated flux value. In my research on Stiffkey marsh, North
Norfolk, UK, I used an alternative method of gauging creek flow that
was developed by David Sargent (Sargent, 1981, 1982a, b; Carpenter,
1993). This method, known as the integrating rising air float technique,
is ideally suited to overcoming the difficulties in gauging tidal creek
flows, which velocity meters fail to overcome. The technique works on
the principle that horizontal displacement of a buoyant float rising from
the bed to the surface, is proportional to the unit discharge. Air is
pushed through a flexible pipe, which is pulled straight and lies along
the banks and bed of the creek, using the air pressure from an aqualung
type cylinder (see Figure 4.4). The air exits the pipe from nozzles which
occur at 0.5 m intervals along the pipe. As the bubbles rise they are
displaced laterally by the current. The major advantage of this technique
is that the bubbles integrate the varying flow velocities through the
water column; velocity meter readings, at various depths down a
vertical, only sample this velocity profile. Furthermore, instantaneous
discharge across the whole creek cross section can be gauged by this
method as frequently as desired by recording the positions of the
surfacing bubbles across the creek with photography and deducing the
horizontal area between the air pipe and the bubble envelope (area A in
70 K. Carpenter

Figure 4.5) using photogrammetry. Discharge through the creek is

calculated by multiplying this area with the rise velocity of the bubble.
Photographs could be taken every few minutes during periods of the
tidal cycle experiencing the greatest changes in discharge; hence this
method could pick up discharge pulses which velocity meters may miss
if the frequency at which instantaneous discharge can be calculated is
too low.
The instantaneous rising air float technique is not only inexpensive to
make and involves little field labour, it is also measures water discharge
through creeks more accurately than velocity meters. Sargent (1981)
calibrated the rising air float technique against a gauging weir and found
the average discrepancy was 1.3% (although the range was — 6 to
+ 10%). This is considerably lower than the error on the creek water
discharge rates using the velocity meter method. Hence the integrating
rising air float technique seems superior to velocity meters and should be
considered as a preferable gauging option in future saltmarsh material
flux studies.

Figure 4.4 The equipment used in the integrating rising air float
technique.
Methodology used in studies of material flux 71

Figure 4.5 Photograph of the emerging air bubbles, from which the rate
of water discharge can be calculated. Discharge = area Ax bubble rise
velocity. Area A is calculated using photogrammetry. PF is the principal
focus.
72 K. Carpenter

(e) Frequency of discharge and concentration measurements

The last major difference in the way researchers have conducted the field
work of tidal flux studies is in the frequency of recording water discharge
and nutrient concentration: ranging from 10 minutes to 1 hour, depending
on the position in the tidal cycle (Carpenter, 1993); every 30 minutes (e.g.
Boorman et a/., 1994) to every 1.5 hours (Whiting et a/., 1985). The error of
extrapolating the data to estimate the total flow during the tidal cycle
increases with time between measurements. To establish the optimal
sampling frequency, it would be a good idea to take water samples and
discharge measurements every five minutes and use this data set to
investigate the difference that various sampling frequencies have on the
calculated flux value.

(f) Frequency of tidal monitoring

Nutrientfluxhas been found to differ both seasonally and with tidal range
(spring-neap cycle). The seasonal variation is illustrated by data collected
during my PhD research of nutrient fluxes between Stiffkey upper
saltmarsh, on the North Norfolk coast, and the North Sea. Figure 4.6
illustrates the seasonal differences between nutrient concentrations in the
North Sea with those in the water seeping from the saltmarsh sediments.
The bars represent the concentration at the end of the ebb tide just before
the flooding water of the following tide reached the sampling station
(when the water flow is fed by groundwater seepage from the saturated
saltmarsh sediment). The points joined together by a line are the
concentrations at high water (which most closely resembles the character
of coastal sea water). By comparing the nutrient concentrations at these
two points in the tidal cycle, it is possible to establish whether the net effect
of the nutrient transforming processes in the sediment is uptake or release.
Thus when the bar is larger than the line, the sediment-creek bank system
is a source of the nutrient and conversely a higher line than the bar denotes
the system is a net sink. It is clear that the size of the flux varies
considerably, and in the case of phosphate and nitrate the direction of the
flux also differs seasonally.
In order to get a good estimate of annualflux,tidal monitoring should
be as frequent as possible to reduce the errors associated with interpolation
and extrapolation. Ideally, several tides, covering the range of tidal
heights should be sampled each month. This is particularly important for
sediment flux studies as suspended sediment concentration varies with
tidal range with the highest concentrations occurring at spring tides due
to the associated large tidal velocities. Furthermore, attempts should be
 Seasonal Patterns In Dissolved Nutrients Seasonal Patterns In Dissolved Nutrients Seasonal Patterns In Dissolved Nutrients
Nitrate Nitrila

Phoaphata oono. (wM) Nitrate oono. (uM) Nitrlta oono. (uM)

JanFabMarAprMayJun Jut AugSapOotNovDeoJanFebMarAprMayJun JanFabMarAprMayJun Jul AugSapOotNovDaoJanFabMarAprMayJun JanFabMarAprMayJun Jul AugSepOotNovDeoJanFebMarAprMayJun

8amplad montha (1090 and 1991) Samplad montha (1990 and 1991) 8amplad montha (1990 and 1991)

— - High watar E23 Low water —— High wafer E353 Low watar — High watar

Seasonal Patterns In Dissolved Nutrients Seasonal Patterns in Dissolved Nutrients Seasonal Patterns In Dissolved Nutrients
Ammonia Total diaaolved inorganio nitrogan Sifioate
Ammonia oono. (uM) 8llloata oono. (uM)

JanFabMarAprMayJun Jul Aug8epOotNovDeoJanFabMarAprMayJun J F M A M J J A 8 0 N D J F M A M J JanFabMarAprMayJun Jul AugSapOotNovOaoJanFabMarAprMayJun

8ampled montha (1990 and 1991) 8amplad montha (1990 and 1991) 8ampled montha (1990 and 1991)

— High watar ES3 Low water —— High watar E-S! Low water — High water £23 Low water

Figure 4.6 Seasonality in nutrient concentration in the creek water at high and low tide, Stiffkey saltmarsh, North Norfolk UK.
74 K. Carpenter

made to measure the flux during episodic events likely to cause large
fluxes. However, tight research budgets, manpower restraints and impossible
working conditions due to harsh weather normally prevent this level of
sampling intensity.

(g) Tidal flux calculations

Basically, flux is calculated as the difference between the amount of
nutrient in the flood and ebb tide, but there are several ways of doing this.
The amount of nutrient is estimated by combining discharge and water
sample concentration. Calculation details are normally glossed over in
published papers, and the absolute method is not defined, e.g. The
transport was calculated on the basis of the weights of a component
imported or exported on each tide' (Boorman et al, 1994) and 'By
multiplication with concentration values obtained from the sample, the
particulate and dissolved matter could be calculated for each period'
(Dankers et al, 1984). This does not help readers assess the confidence
they may place in the final flux values.
I assume there are two obvious ways of combining discharge and
concentration data to get theflux.These two methods are outlined in Box 4.1.
The way researchers deal with the tidal inequality affects the calculated
tidal flux. In all studies there has been substantial asymmetry between the
flood and ebb discharge volumes (e.g. Boon, 1975; Kjerfve & Proehl, 1979;
Valiela et al, 1978; Reed, 1987). This may partly be an artefact due to
either technique problems or real volume changes (e.g. rainfall, marsh
surface storage after an overbank tide). Discharge estimates using a
velocity meter in a fixed position may indicate a tidal inequality even if the
actual volumes are the same, due to spatial asymmetry in the flow pattern
between ebb and flood tides. Jay & Musiak (as reported at the joint
ECSA/Estuarine Research Federation conference, Plymouth, UK, 1992)
found that ebb flows were concentrated near the surface and flood flows
were either vertically uniform or bottom intensified. Thus the regression
between velocity at the meter's sampling position and the integrated
velocity for the whole creek cross section will not be constant for both
halves of the tidal cycle. As tidal discharge rates through the creek are
calculated by multiplying the velocity meter data by the wetted cross
section, differing regressions for the two tidal halves would mean that the
error between the actual and calculated discharge rates would differ for
the flood and the ebb.
The discrepancy between tidal volumes may also be due to real volume
changes due to atmospheric precipitation, evaporation or change in
 Methodology used in studies of material flux 75

Box 4.1 Two possible ways of combining discharge 1 and concentration 2

data to obtain flux values
(1) By interpolation of best fit lines through Q and C data

Total no. of moles in flood (A) = X fl (6'0' c

j __
Total no. of moles in ebb (B) = £e((>0*c
Tidal flux = A - B

(2) Via the flow rate of substance (mol s" *) from the multiplication of Q
and C data which were recorded simultaneously

moles
S1

Tidal flux = Area A — Area B

storage. Precipitation effects are usually insignificant due to the small

ratio between precipitation volume and tidal input. Land-derived
ground-water input could potentially cause a significantly inflated ebb
volume (e.g. in the Sapelo Island marshes), especially in sandy marshes
backed onto land of high elevation. However, it is unlikely that this is the
76 K. Carpenter

dominant cause of the tidal discrepancy as it does not explain the many
cases which exhibit significantly larger flood than ebb volumes. Furthermore,
large inputs of fresh water would easily have been identified by researchers
from changes in salinity over measured tidal cycles, so researchers would
have noticed if groundwater input explained the asymmetry. Most studies
have selected sites where land-derived groundwater input is negligible, to
simplify their calculations.
One of the most likely causes of the tidal inequality is the considerable
change in stage height between consecutive tides. If a tide flooded the
surface of the marsh, but the subsequent one was contained within the
banks of the creek (as may occur during the spring to neap tidal cycle), the
water table in the sediment could be higher than the water level in the
creek. Thus more water would drain into the creek during the ebb tide
than infiltrates through the creek banks during the flood tide. Hence the
ebb volume would be larger than the flood. During the neap to spring
cycle, the reverse situation exists and the flood could be bigger than the
ebb. The first 'overbank' tide of the cycle would theoretically show the
biggest discrepancy between flood and ebb volumes because flood water
would remain on the marsh surface due to the replenishment of pans and
infiltration into the unsaturated surface sediment. If the described events
were the major cause of observed tidal asymmetry, one would expect that
randomly sampled tides would be equally split between ebb dominated
and flood dominated tides. This does not seem to be the case in at least
some studies, the three marshes in the European comparative study
(Mont St Michel, France; De Slufter, the Netherlands; and Tollesbury,
UK) which measured creek flows, reported mean flood-ebb differences of
— 13%, — 11%, and — 16%, i.e. an apparent loss or storage of water
(Boorman et al, 1994).
Tidal asymmetry could also occur if the selected creek cross section was
not situated within a closed creek network, which would mean that a
parcel of water could exit from a different place to where it entered. This
could occur if two creek systems were connected, or could exchange water
during an overtopping tide. The proportion of water entering and leaving
the marsh from the two mouths would depend on wind direction and tidal
flow patterns.
Some researchers have ignored inequality in the flood and ebb
volumes completely (e.g. Dankers et a/., 1984), whilst others have
compensated for the tidal discrepancy by multiplying the total amount of
nutrient entering the marsh on the flooding tide by the ratio of the
flood/ebb volume (e.g. Valiela et a/., 1978). The way researchers treat the
 Methodology used in studies of material flux 77

asymmetry obviously has a great effect on the calculated flux value, so

which is the best approach to adopt? It would be incorrect to equalise an
ebb flow, which has been reduced in volume by evaporation, to the
preceding flood, as this could falsely show an export of material when
none existed. On the other hand it would be erroneous to ignore the
inequality of a large tide which had greater flood than ebb, due to
recharge and surface ponding. The calculated results most likely show a
marsh input of material even though the nutrients may still exist
unchanged in the ponded water and may later return to the creek via
advection. It would be even worse not to equalise tides that were
unbalanced due to transference of water between interconnected creek
systems. In terms of nutrient cycling and determining the direction of the
flux, it does not matter that some of the ebb water entered the marsh via
a different creek mouth, as that water package has experienced the same
marsh processes as water that entered from the main creek mouth.
However, it is possible that the magnitude of the flux per meter cubed of
tidal water is altered if the secondary package had spent a different
length of time on the marsh or covered less surface area. Thus if the tidal
inequality was due to the interconnection between two or more creek
systems it would be correct to equalise the volumes. Therefore, although
there are errors associated with both ignoring and compensating for the
flow imbalance, it seems better to equalise the flows unless major
precipitation or evaporation has taken place.

(h) Annual flux calculations

The confidence that can be placed in annualfluxvalues is even lower than
for individual tidal cycles. Previous research has concentrated on the
determination of a value for annual flux, almost all papers reporting this
value despite the uncertainty connected with it. Less attention seems to be
paid to seasonal variation in the flux value, which seems odd as the
direction of the flux may change throughout the year but the net annual
flux may be insignificant, thus masking the true impact the saltmarsh has
on the ecology of the adjoining coastal water. For example, a saltmarsh
may be a sink of phosphate in the winter and a source in the summer
(Asjes, presentation at INTECOL's IV International Wetlands conference,
1992, relating to a Dutch saltmarsh; Carpenter, 1993). The annualfluxof
phosphate from Asjes's study was negligible: if this value was presented
alone (as in the style of Nixon, 1980; Dankers et a/., 1984, for example) one
would conclude that phosphorus cycling in the marsh is insignificant.
However, the seasonal fluxes are potentially very important to coastal
78 K. Carpenter

production as phosphate is released at a time when the concentration in

coastal water is at its minimum.
Calculated nutrient fluxes between Stiffkey saltmarsh and the North
Sea, from my PhD research (Carpenter, 1993) are given in Tables 4.2 to
4.4. These three tables express the results in different ways which suit
different purposes. Table 4.2 expresses the semi-diurnal tidalfluxin moles.
This total value is useful for evaluating the impact the saltmarsh has on
nutrient concentration in the coastal water. Table 4.3 describes the fluxes
in volumetric terms, i.e. the flux per cubic metre of the flood tide. This is
calculated by dividing the semi-diurnal tidal nutrientfluxby the averaged
tidal volume. This form of expression is very useful for making comparisons
with other research. The total tidal flux can not be used to compare
saltmarshes as the value depends on both the drainage area and the tidal
volume. Table 4.4 expresses thefluxesas a percentage of the total number
of moles in the flood tide (after the quantity in the flood had been
corrected for tidal volume inequalities). This gives an idea of the size of the
flux in comparison to the amount flowing through the creek during tidal
flushing and also the degree of confidence which can be placed in the
direction and absolute value of the calculatedflux.The values depicted in
bold type indicate thefluxeswhich were larger than the uncertainty. The
flux results show that there is seasonal variation in all the measured
nutrient species, which is not appreciated ifjust the averaged annualfluxis
reported.

Table 4.2 Nutrient fluxes (moles) for measured semi-diurnal tidal cycles,
Stiffkey Saltmarsh, North Norfolk, UK. A negative sign denotesthesaltmarshasa
source of nutrients to the offshore waters

Flux (moles)
Date NO2" NO3- NH4 + PO 4 3 ~ SiO 4 3 ~

12 Aug. 1990 -2.8 1.9 - 1.94 - 1.99 -50.7

12 Oct. 1990 -0.88 -0.19 -7.09 0.1 3.17
8 Nov. 1990 0.21 - 1.9 1.5 - 0.027 -
22 Jan. 1991 0.24 -6.5 -1.79 0.035 -7.69
22 Mar. 1991 0.03 0.21 8.16 0.71 20.1
20 Apr. 1991 -0.42 -6.7 6.83 0.16 10.7
19 May 1991 - 10.7 2.41 0.13 4.76
14 Jun. 1991 Spring - 0.001 2.7 4.5 -0.01 - 12.5
15 Jun. 1991 to -0.60 - 10.9 -2.67 -4.45 -58.9
17 Jun. 1991 Neap - 7.2 2.5 -0.38 -24.7
19 Jun. 1991 - 5.3 5.04 -0.23 -4.3
21 Jun. 1991 . -0.36 2.2 - 1.02 -0.11 -2.8
 Methodology used in studies of material flux 79

To my knowledge, there are four major methods of calculating the

annual flux from tidal flux estimates. All four methods suffer from
assumptions and huge extrapolations of the data set, to the extent that I
did not attempt to calculate an annual flux for my study as I did not
think the resulting figure would be meaningful. Instead I compared the
tidal fluxes which were measured monthly during the same point of the
spring-neap tidal cycle to identify seasonal changes in nutrient fluxes.
It has not been possible to decipher the methods of annual flux
calculation in detail, as they are not usually reported in published
Table 4.3 Volumetric nutrient fluxes (flux per cubic metre of tidal volume),
Stiffkey Saltmarsh, North Norfolk, UK
3
Volumetric flux (umoles m of tidal volume)
Date NO2" NO3" NH 3 PO43" SiO 4 3 ~

12 Aug. 1990 -282 186 - 194 - 199 -5070

12 Oct. 1990 - 152 -33 - 1226 17 548
8 Nov. 1990 121 - 1095 846 - 15 -
22 Jan. 1991 82 -2182 -601 12 -2581
22 Mar. 1991 5.2 3.S1 1471 128 3677
20 Apr. 1991 -92 - 1472 1493 35 4461
19 May 1991 - 1803 406 27 801
14 Jun. 1991 -0.6 163 271 -0.6 -749
15 Jun. 1991 - 17 -307 -75 -125 - 1655
17 Jun. 1991 - 652 226 -34 -2229
19 Jun. 1991 - 936 896 41 -760
21 Jun. 1991 -210 1302 -595 -64 -1535

Table 4.4 Nutrient fluxes as a percentage of the total number of moles in the
flood tide, Stiffkey Saltmarsh, North Norfolk, UK

Percentage flux (%)

Date NO2" NO3- NH 3 PO 4 3 ~ SiO 4 3 "

12 Aug. 1990 - 140.0 12.4 -6.7 31.0 -61.8

12 Oct. 1990 -3.5 -3.8 -23.0 6.9 5.5
8 Nov. 1990 9.0 -9.1 - 21.7 -3.4 -
22 Jan. 1991 12.0 -5.3 - 1.8 18.9 -12.2
22 Mar. 1991 0.9 0.1 22.9 16.8 20.9
20 Apr. 1991 -6.5 -4.6 40.7 2.9 13.5
19 May 1991 - 13.9 11.4 9.7 16.0
14 Jun. 1991 -0.2 43.0 17.9 5.7 -16.8
15 Jun. 1991 -30.6 48.9 -3.5 42.0 -133
17 Jun. 1991 - 27.1 7.5 14.2 -46.1
19 Jun. 1991 -22.6 21.0 27.1 14.1 -10.3
80 K. Carpenter

literature, but the general approaches are given below. It is possible that
each method would give a different flux value given the same data set.
(i) The simplest method averaged the fluxes of the tides sampled
throughout the year and multiplied this value by 706, i.e. the total number
of tides per annum (Wolaver et a/., 1983; Whiting et a/., 1989). In these two
studies, tidal cycles were sampled once every 2-4 weeks, selecting tides to
cover all ranges and times of day. It would be very difficult to investigate
seasonal nutrient cycles using this method as the flux is affected by both
season and tidal range (Wolaver & Spurrier, 1988), i.e. if a spring tide was
sampled in summer and a neap tide in autumn, seasonal variation would
be masked by tidal height effects.
(ii) Dankers et a\. (1984) grouped their sampled tides into bands
governed by the elevation of high water. For each group the mean flux was
calculated and multiplied by the number of tides in the year within that
range. The resulting products were summed to give the annual flux. This
method assumed that tidal height, rather than seasonal factors, is the
dominant variable controlling the flux.
(iii) Asjes (pers. com., 1992) took storm tides into account. He managed
to measure the tidal flux during a storm tide (this has rarely been achieved)
and assumed that all storm tides throughout the year had the same flux.
These fluxes were included in the annual flux estimation.
(iv) Arguably the most accurate way of calculating annual flux would
be through the use of a regression model. By knowing how the key
variables behave throughout the year the fluxes for the unsampled tides
could be estimated. However, saltmarsh science has not yet reached the
stage at which this could be possible. Wolaver & Spurrier (1988) made a
contribution towards this aim when they correlated their flume derived
flux results, from a South Carolina saltmarsh, with pertinent chemical,
biological and physical parameters, to find the most significant factors
influencing phosphate flux. They concluded that the major controlling
variables, which individually explained up to 10% of the flux, included
tide height, water temperature, salinity, wind and rain. However, they did
not attempt to produce a multiple regression equation for these data so
that phosphate flux between an unstudied marsh and the coastal water
could be estimated.

4.3 Summary
(1) Nutrient fluxes have been obtained using a range of field techniques
and methods to calculate tidal and annualfluxes.The methodology
 Methodology used in studies of material flux 81

chosen affects the value of the estimated flux. In order to compare

how various saltmarshes process nutrients, research institutes
should coordinate studies using the same methodology.
(2) Four field techniques have been used to estimate the import/export
of nutrients between saltmarshes and the adjacent coastal water:
community budgeting, direct creek studies, flume studies and
diffusion chamber studies. The direct tidal creek method is the
only one which takes account of the processes in all of the
saltmarsh subsystems and hence appears to be the best technique,
provided the uncertainties in water discharge are minimised.
(3) In order to determine whether the differences in nutrient fluxes
from saltmarshes around the world are due to real differences in
nutrient budgeting as opposed to uncertainties in the calculated
value, confidence limits should be assigned to published flux values.
(4) The uncertainties in calculated flux values are large due to
methodology errors and errors associated with extrapolating the
collected data in space and time. A major problem with saltmarsh/
coastal water flux studies is that the flux is often smaller than the
error margin, hence one is not even sure of the direction of the flux.
The error associated with an annual flux estimate is much higher
than the tidal flux.
(5) Past researchers have concentrated on producing an annual flux
value, nearly all reporting this value (despite the huge uncertainties)
rather than drawing attention to seasonal variations in flux. These
will be masked by an annual average, but have a significant effect
on the ecology of the adjacent water body.
(6) The confidence that can be placed in material flux studies relies
heavily on the accuracy of the method for measuring water flow
through the creek. Sargent's integrating rising air float technique
has a significantly lower uncertainty on the calculated instantaneous
discharge through tidal creeks than velocity meters. The adoption
of this technique could make an important contribution to the
accuracy of future tidal creek material flux studies.
(7) The validity of nutrient flux studies as a method to establish
whether saltmarsh processes have a significant effect on the
productivity of coastal water should be questioned. Unless
uncertainties are reduced by improvements in methodology, and
an intensive sampling regime undertaken (sampling a semi-diurnal
tidal cycle every few days), creek nutrient flux studies do not seem
to be an appropriate method to quantify the impact of saltmarsh
82 K. Carpenter

processes on coastal water productivity. Perhaps future research

should concentrate on comparing nutrient concentrations, species
composition and productivity of coastal water adjacent to
saltmarshes with that remote from saltmarsh influence, in order to
ascertain the importance of saltmarshes to coastal productivity.

Acknowledgements
The research was conducted at the University of East Anglia with the
support of NERC funding

References
Axelrad, D.M. (1974) Nutrient flux through the salt marsh ecosystem. Ph.D Thesis,
College of William and Mary, Virginia, USA.
Azni, S., Abd. Aziz, S. & Nedwell, D.B. (1986) The nitrogen cycle of an East Coast, U.K.
saltmarsh: II. Nitrogen fixation, nitrification, denitrification, tidal exchange.
Estuarine, Coastal and Shelf Science 22, 689-704.
Boon, J. (Ill) (1975) Tidal discharge asymmetry in a saltmarsh drainage system. Limnology
and Oceanography 20, 71-80.
Boorman, L.A., Hazelden, J., Loveland, P.J., Wells, J.G. & Levasseur, J.E. (1994)
Comparative relationships between primary productivity and organic and
nutrient fluxes in four salt marshes. In Global Wetlands; Old World and New,
Mitch W.J. (ed.), Elsevier, Amsterdam.
Burt, T.N. & Stevenson, J.R. (1986) Monitoring cohesive sediment transport in estuaries.
In International Conference on Measuring Techniques of Hydraulic Phenomena in
Offshore, Coastal and Inland Waters, London, England 9-11 April 1986.
Carpenter, K.E. (1993) Nutrient fluvial and groundwater flux between a North Norfolk
saltmarsh and the North Sea. Ph.D Thesis, University of East Anglia, Norwich, UK.
Chambers, R.M. (1992) A fluctuating water level chamber for biogeochemical experiments
in tidal marshes. Estuaries 15 (1), 53-8.
Dame, R.F., Spurrier, J.D., Williams, T.M., Kjerfve, B., Zingmark, R.G., Wolaver, T.G.,
Chrzanowski, T.H., McKellar, H.N. and Vernberg, F.J. (1991) Annual material
processing by a salt marsh-estuarine basin in South Carolina, USA. Marine
Ecology Progress Series 72, 153-66.
Dankers, N., Binsbergen, M., Zegers, K., Laane, R. & van der Loeff, M.B. (1984)
Transportation of water, particulate and dissolved organic and inorganic matter
between a salt marsh and the Ems-Dollard Estuary, The Netherlands. Estuarine,
Coastal and Shelf Science 19, 143-65.
Day, J.W., Jr, Smith, W.G. & Wagner, P.R. (1973) Community structure and carbon
budget of a salt marsh and shallow bay estuarine system in Louisiana. Louisiana
State University, Center of Wetland Resources. Publ. LSU-SG-72-04. 74 pp.
Kjerfve, B. & Proehl, J.A. (1979) Velocity variability in a cross-section of a well mixed
estuary. Journal of Marine Research 37, 409-18.
Lee, V. (1979) Net nutrient flux between the emergent marsh and the tidal waters. MSc
Thesis, University of Rhode Island, USA, 67 pp.
Marshall, D.E. (1970) Characteristics of Spartina marsh receiving municipal sewage waste.
MSc thesis, University of Carolina, USA.
 Methodology used in studies of material flux 83

Nixon, S.W. (1980) Between coastal marshes and coastal water - a review of twenty years
of speculation and research on the role of saltmarshes in estuarine productivity
and water chemistry. In Estuarine and Wetland Processes, Hamilton, P. &
Macdonald, K.B. (eds.), Plenum.
Odum, E.P. & de la Cruz, A.A. (1967) Particulate organic detritus in a Georgia salt
marsh-estuarine ecosystem. In Estuaries, Lauff, G.H. (ed.), American Association
for the Advancement of Science, Publication No. 83.
Patrick, W.H., Jr. & Delaune, R.O. (1976) Nitrogen and phosphorus utilization by
Spartina alterniflora in a saltmarsh in Barataria Bay, Louisiana. Estuarine,
Coastal and Marine Science 4, 59-64.
Reed, D.J. (1987) Temporal sampling and discharge asymmetry in salt marsh creeks.
Estuarine, Coastal and Shelf Science 25.
Roman, C.T. (1984) Estimating water volume discharge through salt-marsh tidal channels:
an aspect of material exchange. Estuaries 7(3), 259-64.
Sargent, D.M. (1981) The development of a viable method of stream flow measurement
using the integrating float technique. Proc. Inst. Civ. Engrs. 71 (part 2), 1-15.
Sargent, D.M. (1982a) The rising air float technique for the measurement of stream
discharge. Advances in Hydrometry (proceedings of the Exeter Symposium, July
1982). IAHS Publ. no. 134.
Sargent, D.M. (1982b) The accuracy of streamflow measurement using the rising air float
technique. Proc. Inst. Civ. Engrs. 73 (part 2), 179-82.
Scudlark, J.R. & Church, T.M. (1989) The sedimentary flux of nutrients at a Delaware
saltmarsh site - a geochemical prospective. Biogeochemistry 7(1), 55-75.
Teal, J.M. (1962) Energy flow in the saltmarsh ecosystem of Georgia. Ecology 43,614-24.
Tyler, G. (1967) On the effect of phosphorus and nitrogen, supplied to Baltic shore
meadow vegetation. Botanisk Notiser. 120, 433-47.
Valiela, I.B. & Teal, J.M. (1976) Primary production and dynamics of experimentally
enriched saltmarsh vegetation: below ground biomass. Limnology and Oceanog-
raphy 21, 245-52.
Valiela I., Teal, J.M, Volkmann, S, Shafer, D. & Carpenter, E.D. (1978) Nutrient and
particulate fluxes in a salt marsh ecosystem: tidal exchanges and inputs by
precipitation and groundwater. Limnology and Oceanography 23(4), 798-812.
Whiting, G.J, McKellar, H.N, Kjerfve, B. & Spurrier, J.D. (1985) Sampling and
computational design of nutrient flux from a southeastern US saltmarsh.
Estuarine, Coastal and Shelf Science 21, 273-86.
Whiting, G.J, McKellar, H.N, Spurrier, J.D. & Wolaver, T.G. (1989) Nitrogen exchange
between a portion of vegetated salt marsh and the adjoining creek. Limnology
and Oceanography 34(2), 463-73.
Wolaver, T.G, Zieman, J.C, Wetzel, R. & Webb, K.I. (1983) Tidal exchange of nitrogen
and phosphorus between a mesohaline vegetated marsh and the surrounding
estuary in the Lower Chesapeake Bay. Estuarine, Coastal and Shelf Science 16,
321-32.
Wolaver, T.G. & Spurrier, J.D. (1988) The exchange of phosphorus between a Euhaline
vegetated marsh and the adjacent tidal creek. Estuarine, Coastal and Shelf
Science 26, 203-14.
o o o o o o o o o o o o o o o o o o o
Nutrient recycling in intertidal
sediments
S.J. Malcolm and D.B. Sivyer

5.1 Introduction
Intertidal sediments occupy a unique position between land and sea
giving them an importance which is only now becoming apparent. Most
studies have looked in detail at the nature of the biogeochemical processes
(e.g. Klump & Martens, 1989) but have not had the impetus to place the
processes into a quantitative context. The priority and scale of study has
changed due to an awareness that intertidal areas are the front line in
dealing with both the effects of climate change and the movement of
materials from land to sea (Nowicki, 1994). There are now global, regional
and national research programmes examining interactions at the land/sea
boundary (e.g. LOICZ (Pernetta & Milliman, 1995), NECOP (Ortner &
Dagg, 1995), JoNuS (Anon., 1994), LOIS (Anon., 1992)).
Concern about the impact of nutrients on the marine environment is
providing a sharp focus for some of these studies (e.g. Nowicki & Oviatt,
1990; Anon., 1994). Intertidal sediments are a part of the sediment system
that has long helped to maintain the productivity of the coastal zone via
the storage and recycling of nutrients which were imported from offshore
waters (Nixon, 1986; Nixon, 1992). Human population increases, coupled
with aquatic sewerage systems and the development of intensive agriculture,
have resulted in a significant change in the quantities of nutrients supplied
from land sources. Now the main source of nutrients is from the land and
not from the sea (Nixon, Hunt & Nowicki, 1986). What impact has this
change had on the buffering role of intertidal sediments?
Intertidal areas are difficult to study from a practical point of view and
offer conceptual challenges to developing the quantitative budgets
required to address questions about nutrient transport between land and
sea. The methods employed must take account of the key attributes of the
intertidal environment which include variability, patchiness scales and
 Nutrient recycling in intertidal sediments 85

dynamics. The biogeochemical processes that influence the behaviour of

nutrients are reasonably well established but what controls them and how
they interact with the physical environment is not well known.
The purpose of this paper is to outline the physical environmental
factors involved in the functioning of the intertidal system and briefly
review the impact on the biogeochemical processing of nitrogen, phosphorus
and silicon. Consideration of saltmarshes is excluded. Discussion is
focused on the often extensive areas of non-vegetated sand and mud flats
that are uncovered by the tides on a daily basis and for which there is a
limited amount of published information.

5.2 Physical processes impacting intertidal areas

The schematic shown in Figure 5.1 shows the major [biojphysical
disturbances that impact on intertidal sediments. These are tides, waves,
storms, run-off events and sediment movements which result from the
above, bioturbation, bioirrigation and anthropogenic factors such as
fishing and dredging. The disturbances occur over a wide range of time
scales and influence processes over different space scales (Figure 5.2) but
the net effect is to produce an environment that, although broadly
structured, is patchy and dynamic.

Run Off

Salinity Changes

Sediment Accumulation
Bioturbation

Figure 5.1 Schematic showing the [biolphysical impacts on intertidal

sediments.
86 S.J. Malcolm and D.B. Sivyer

5.2.1 Tides
Intertidal sediments are subject to diurnal drying and flooding which
leads to major changes in the local transport of material to and from the
sediments. Clearly when the tide has ebbed there is no overlying water
present and there can be no dissolved output from the sediment. This may
result in increasing concentrations of dissolved substances that have a
source within the sediments, such as ammonium. This pool of high
concentration may then be released when the water returns to flood the
sediment. In contrast, a substance whose source is in the water and is
consumed in the sediment may become depleted when the water is not
present. Examples of these processes are presented below and show that
sediments are both sinks and sources of different nutrients and that the
behaviour changes with the tide.
During the tidal cycle the hydraulic regime in the sediments changes
and advectiveflowalong pressure gradients may occur. Muddy intertidal
sediments have a high porosity but have a low permeability limiting
advective transport. Sandy sediments have a permeability great enough to
allow drainage of the sediments at low water and subsequent saturation as

global diaiige/sca U cl risc/geoiiiorplutlojiy

V irrigation-! sciliiiu'iiliilion/crosioii-
1
j-biogcotlicmisdyH

seconds hours days

Temporal Scales
Figure 5.2 The spatial and temporaral scales of disturbance of the
intertidal environment.
 Nutrient recycling in intertidal sediments 87

the tide advances. The impact of advection and resultant changes, such as
increased penetration of oxygen into the sediments, has yet to be fully
addressed but will have a major impact on redox sensitive processes such
as coupled nitrification and denitrification (Rysgaard et a/., 1994).
Superimposed on the diurnal cycles, the tides exhibit a spring/neap
cycle of changing tidal range which has two effects. Firstly, the area of
sediment exposed and the length of exposure changes as the tide range
alters and, secondly, the velocity of the ebb and flood currents changes.
The sediment source/sink term is area related and the current velocity will
have an impact on disturbance and transport of sediment and on flushing
characteristics (Vorosmarty & Loder, 1994). Therefore, the export and
import of material from intertidal areas is influenced by spring/neap cycles.

5.2.2 Storms/episodic events/waves

The impact of these infrequent events is a significant unknown. Tidal cycle
changes are relatively easy to measure but the influence of events such as
river spate (see below), storms and wind-induced wave action in shallow
areas is more difficult to study. For safety reasons, it is not desirable to
work in estuaries or on intertidal areas during storms and developments
in this area will depend on autonomous measurement devices. Reliable
instrumentation has only recently become available (e.g. Green et a/.,
1992) and it will be some years before our understanding of the overall
impact of episodic events is improved. However, a storm that disturbs
much of the upper 10 cm of sediments over a given intertidal area is likely
to release much of the associated dissolved nutrients which would be
diluted into the water andflushedaway. The loss of nutrient could be very
large and would be different for each nutrient. Nitrogen and silicon would
tend to be enriched in the water while phosphorus would remain attached
to the sediment particles and re-enter the sediment system after the
disturbance. Thus the sediments may accumulate phosphorus relative to
the other nutrients and the concentration will be limited by the pseudo-
equilibrium adsorption/desorption process which buffers phosphorus
concentrations in marine and estuarine waters.

5.2.3 Run-off events

Major run-off events are a feature of most intertidal areas associated with
estuaries. The high flows may limit nutrient processing due to decreased
water residence times; however, if concentration increases with flow,
nutrient processing may increase depending only on the biokinetics of the
process. The concentration of nitrate in river water tends to increase with
88 S.J. Malcolm and D.B. Sivyer

increasing water flow (Figure 5.3) as soluble nitrate is washed from soils by
run-off (Rendell et al, in press). Denitrification rate and extent are
maximal during the winter when flows and resulting concentrations are
high and may similarly be high during single high flow events at any time
of year (Joye & Paerl, 1993).

5.2.4 Sediment transport

Intertidal sediments are generally dynamic. Although intertidal areas are
normally accumulating sediment, the sediments are subject to time
varying patterns of local erosion and deposition. The stability of these
areas will determine the potential for nutrient storage. However, as the
time scales of nutrient regeneration and recycling are generally fast, it may
be possible to ignore sedimentation in the short term. However, leakage
occurs from the surface sediment recycling into the underlying sediments.
A relatively small percentage of the nitrogen and silicon (Brown, 1994) is
incorporated into the sediments while significant amounts of phosphorus
will remain stored (Fox, 1989; de Jonge Engelkes & Bekker, 1993; Prastka
& Jickells, 1995). The extent of the sediment sink is an important factor in

1000-1

800-

600-
o

400-

200-

20 40 60 80 100 120 140 160

Flow (cumecs)

Figure 5.3 The relationship between flow and nitrate concentration in the
river Great Ouse (UK) (JoNuS unpublished data). This relationship is typical
of many European rivers draining agricultural catchments.
 Nutrient recycling in intertidal sediments 89

predicting the future impact of nutrients as measures are taken to control

discharges to sea.

5.2.5 Bioirrigation and bioturbation

The macro-fauna that live in the sediments are one of the main ways of
disturbing and restructuring intertidal sediments (Asmus, 1986). Burrowing
animals will move through the sediment, causing homogenisation to a
characteristic depth. Open burrow structures not only allow the movement
of sediment from one depth to another but also substantially increase the
sediment/water interface. The effect of this is greater oxygen penetration
into the sediments and potentially greater export of regenerated nutrients
(Clavero, Niell & Fernandez, 1994; Marinelli, 1994). Bioturbation has a
significant impact on the rate of denitrification (Pelegri, Nielsen &
Blackburn, 1994).

5.2.6 Human disturbance - fishing, dredging and reclamation

Fishing is thought (North Sea Task Force, 1993) to cause similar
disturbance to that due to the passage of burrowing organisms but on a
greater scale and at a greater rate, while dredging and subsequent disposal
of the dredge spoil can be considered in the same category as a major
storm event with substantial relocation of particulate material. On a
much longer time scale, reclamation and coastal defence engineering
which reduces the intertidal area will have a dramatic effect but this will
not be considered here other than to say that, although there are few
examples of this impact in the literature, it is an area of major importance
as awareness increases of the impact of engineering management on
estuaries and coastal systems.

5.3 Nutrient processing

Nutrients are affected by a variety of processes in intertidal sediments
(Klump & Martens, 1989). Nitrogen is particularly interesting as it exists
in different forms, both organic and inorganic, and may be converted from
one form to another under different conditions. Though subject to
continuing debate, nitrogen is also considered to be the nutrient limiting
primary production in the sea (Gilbert, 1988; Owens, 1993). This has led to
a substantial literature describing aspects of the nitrogen cycle (Blackburn
& Sorensen (1988) and references therein). The behaviour of phosphorus
is dominated by its propensity to associate with particles and iron oxides
90 S.J. Malcolm and D.B. Sivyer

in the estuarine and coastal environment. The fate of phosphorus is bound

up with the movement of sediments but the details of phosphorus cycling
at the local and global level are only now emerging (Berner et a/., 1994).
Silicon geochemistry is considered to be relatively simple; silicon is
removed from water by diatoms and temporarily stored in sediments as
biogenic opal (Berner, 1980; Gehlen & van Raaphorst, 1993). The
dynamics of silicon behaviour have not been well studied and the growth
of diatoms in mats (de Jonge, 1980) at the sediment surface are likely to be
an important component of the silicon story.

5.3.1 Nitrogen
Nitrogen can be present in a large variety of forms in aquatic systems. The
main species of interest are nitrate, nitrite and ammonium together with
organic nitrogen, in both dissolved and particulate form. There is a large
pool of di-nitrogen gas in the atmosphere and dissolved in surface waters
but it is considered that nitrogen fixation is not important in coastal
systems (Seitzinger, 1988). Inputs of nitrogen to intertidal sediments come
from the land, in solution via rivers or groundwater, and in the form of
particulate material following uptake by phytoplankton. Regeneration of
the material that escapes the pelagic system is thought to be the principal
source for the nutrients that support plankton growth through the

]
"'I
I Bioturbation

NH4
Burial

Figure 5.4 Schematic showing the key nitrogen cycle processes in

intertidal sediments.
 Nutrient recycling in intertidal sediments 91

summer when direct input from land is low. Figure 5.4 shows a schematic
for the nitrogen cycle in intertidal sediments. What makes these sediments
different is not the nature of the individual processes but the relative
importance of each and the way they are controlled by the unique physical
environment.
Nitrogen fixation
Nitrogen fixation converts di-nitrogen gas to fixed nitrogen and is the
main return path of nitrogen from the atmosphere to the biosphere. In the
developed parts of the world anthropogenic nitrogen fixation (Kinzig &
Socolow, 1994) dominates the nitrogen cycle. There have been many
studies of the process in the coastal marine environment. However, due to
the ready availability of ammonium and nitrate in intertidal areas, the
nitrogen requirements of plants are not provided by nitrogen fixation
(Capone, 1988). Nitrogen fixation will be important where exogenous
inputs of nitrogen are low or when the pool of readily available nitrogen is
depleted by biological production.
Nitrification and denitrifkation
These processes and their links have been the focus for much scientific
effort (Koike & Sorensen, 1988), not only because they are key nitrogen
transformation processes but also because they provide the most important
route for removal of nitrogen from ecosystems. Nitrification is the
oxidation of ammonium to nitrate, where the ammonium is produced
from organic nitrogen. These are key steps in the regeneration of nitrogen.
Fluxes of both ammonium, which is the preferred source of nitrogen for
phytoplankton, and nitrate from sediments to the water have been
measured. Under appropriate conditions, and there is some debate about
those conditions, nitrate can be converted to gaseous products, principally
di-nitrogen (Seitzinger, 1980). Coupled nitrification/denitrification has
been considered to be the dominant source of gaseous nitrogen products
in sediments (Lloyd, 1993) but recent work (e.g. Rysgaard et al, 1994)
suggests that under high nitrate concentration denitrification direct from
nitrate is the major pathway. This is illustrated in Figure 5.5, using
unpublished data from the JoNuS programme, which shows the progressive
loss of nitrate from sediment interstitial water during exposure of
intertidal sediments. This is particularly significant when considering the
flow of nitrogen from land to sea in systems that are heavily impacted by
nitrate run-off and, serendipitously, provides a mechanism for amelioration
of the impact in coastal seas which may be at risk of problems resulting
from eutrophication.
92 S.J. Malcolm and D.B. Sivyer

Nitrous oxide
Nitrous oxide is a product of both nitrification and denitrification and is
of environmental concern as it is an important greenhouse gas. Law, Rees
& Owens (1992) demonstrate the potential importance of estuaries as a
source of N 2 O to the atmosphere. The yield of nitrous oxide during
denitrification in marine and estuarine systems is of the order of a few per
cent of the total denitrificationflux,much less than the relative yield from
denitrification in fresh water and terrestrial systems (20-50%). Thus in
considering effective management policies, and assuming a desire to avoid
adding more N 2 O to the atmosphere, brackish water and intertidal
environments may offer a better nitrogen removal system rather than
encouraging extensive use of fresh water wetlands (e.g. Fleischer, Stibe &
Leonardson, 1991).

10 15 20 25 30

5-

10-
Q.
0

-t=0 13:10
15-
-t=1 14:20
-t=2 15:45
-t=3 17:00
J
20

Figure 5.5 Time series of interstitial water profiles at an intertidal site in

the Wash (UK), near to the major riverine discharge, after the tide has
ebbed. The inventory of nitrate in the interstitial water decreases with time
due to rapid denitrification. The interstitial water profiles are collected using
an in-situ 'sipping' system consisting of porous probes inserted at fixed
depths in the sediment from which water samples are removed by applying
a vacuum. Samples are preserved using mercuric chloride and returned to
the laboratory for colorimetric analysis.
 Nutrient recycling in intertidal sediments 93

5.3.2 Phosphorus
Compared with nitrogen, the behaviour of phosphorus in intertidal
sediments is less well known. Phosphorus associates with solid surfaces,
particularly the surface of iron oxides in sediments (Prastka & Malcolm,
1994), developing a pseudo-equilibrium which acts as a buffer mechanism
in estuaries (Van Beusekom & de Jonge, 1994). Phosphorus is released
under anoxic conditions by the degradation of organic material and
accumulates in the interstitial water. In intertidal sediments with a low
organic carbon and phosphorus content the behaviour of phosphorus is
dominated by the association with iron oxides (Prastka & Jickells, 1995)
and is also released to the interstitial water under anoxia. However,
remobilisation from the sediments is tempered by the fact that under oxic
conditions, which occur at the sediment/water interface, phosphorus
binds strongly to particles. Thus, to a first approximation, intertidal
sediments act as a significant sink for phosphorus. Disturbance of the
sedimentary regime is the only mechanism that will allow leakage of
phosphorus from the system.

5.3.3 Silicon
Input of silicon to intertidal sediments comes from two sources, the
sedimentation of the tests of pelagic diatoms and in situ uptake by benthic
diatoms from the water. Both processes result in the input of particulate
biogenic silica to the sediment which corrodes in alkaline sea water
(Barker et a/., 1994). Silicon accumulates in the interstitial water which
results in a flux of silicon out of the sediments. The flux may have a
significant seasonal variation due to the growth of benthic diatoms during
the spring/summer period and due to bioirrigation (Marinelli, 1994).
Diatoms have been used to look at the history of changing productivity,
and hence nutrient inputs, in lake systems. The use of diatoms as
palaeoenvironmental indicators in coastal environments is limited because
of the preservation problem (Barker et al, 1994) and the paucity of areas
which are continuously accreting sediment. Brown (1994) studied a
number of sediment cores from the Wash for preservation of diatom
material. The biogenic silica content of the sediments decreases with
depth (Figure 5.6) which may indicate progressive corrosion with depth
and/or increased input of silica to the sediments with time. Gross changes
in the type of diatom present in the sediment may support the latter
conclusion but the work is of a preliminary nature. Further examination
of siliceous palaeoenvironmental indicators in intertidal sediments will be
useful.
94 S.J. Malcolm and D.B. Sivyer

5.4 Benthic mats

An important feature of intertidal sediments is the presence of a mat of
diatoms and associated bacteria in the surface sediment. Benthic production
in clear shallow water can be important, as benthic productivity can equal
or exceed the productivity of the phytoplankton. This will clearly have a
major impact on nutrients by direct uptake, adding to the pool of organic
material available for regeneration as well as modifying the transport of
nutrients between sediment and water (Hopner & Wonneberger, 1985).
The presence of a mat also helps to stabilise the sediment surface making it
less prone to physical disturbance. Mats of benthic microphytes produce
and consume oxygen in the surface sediments, enhancing the penetration
of oxygen into the sediments during the photo period. The alteration of
the oxygen regime has a significant impact on both the pathway and
extent of denitrification (Sundback et al, 1991; Risgaard-Petersen et al.,
1994). The whole balance of this system could therefore change dramatically
between summer when production is occurring and winter when the
organic material produced in the summer is being degraded. This could
enhance the amount of denitrification from nitrate in winter run-off by
bringing the redoxcline nearer to the sediment surface and enhance

Biogenic SiO 2 (weight %)

o.o 0.2 0.3 0.4 0.5

J
40

Figure 5.6 Distribution of biogenic silica in intertidal sediments of the

Wash. Biogenic silica was determined by a wet chemical method.
 Nutrient recycling in intertidal sediments 95

coupled nitrification/denitrification in the summer by increasing the

oxygen supply.

5.4.1 Morphological control of the impact of intertidal areas

If the processes occurring in intertidal sediments have some unique
qualities and result in sometimes unexpected impacts on nutrients, there is
one feature which may dominate the overall importance of intertidal
areas. This is the morphology of the system. For example, the intertidal
sediments of the Wash embayment located at the distal end of the
canalised Gt Ouse Estuary (Figure 5.7a) remove a minimum of 30% of the
incident nitrate by denitrification. This may be contrasted with the
Humber Estuary where the intertidal areas are located marginally (Figure
5.7b) and the overall loss is no more than 10% (unpublished JoNuS data).
Although other factors have a role, it is the location of the intertidal areas
with respect to theflowof water that is critical in determining the impact
of intertidal sediments on the transport of nutrients from land to sea.

5.5 Concluding remarks

Intertidal sediments occupy a unique position to exert a controlling
influence on the transfer and recycling of nutrients between land and sea.
KEY

\ •'•..•] Intertidal areas

River inputs

Figure 5.7 The distribution of intertidal sediments in {a) the Wash and (b)
the Humber Estuary in relation to the main fresh water inputs. The intertidal
in the Wash is at the distal end of canalised estuaries whereas the intertidal
in the Humber tends to be marginal to the flow of water. Diagram not to
scale.
96 S.J. Malcolm and D.B. Sivyer

Prior to anthropogenic change of global nutrient cycles it is likely that

intertidal sediments were importers of nutrients from the sea. The
productivity of coastal waters was maintained by storing and regenerating
nutrients on a seasonal basis allowing the development of significant
fisheries. The main inputs of nutrients are now land derived due to the
impact of human activities. Intertidal areas may now be acting as a buffer
which, while maintaining the productivity of coastal waters, acts to
ameliorate inputs of nutrients to the coastal seas. The capacity of the
buffer system is not known but must be considered when management
measures designed to alleviate problems caused by nutrient input are
being formulated.

Acknowledgements
The authors would like to thank all of the members of the JoNuS Project
Team for interesting discussions about nutrient transformations in
estuaries and coastal environments. JoNuS (Joint Nutrient Study) seeks
to quantify the transport of nutrients from land to sea and is funded by the
Ministry of Agriculture, Fisheries and Food, the Department of the
Environment and the National Rivers Authority (all UK). The views
expressed are those of the authors and do not reflect the policy of the
funding departments.

References
Anon. (1992) Land-Ocean Interaction Study, Science Plan. Natural Environment Research
Council Report, 32 pp.
Anon. (1994) Establishing the nutrient status of UK coastal waters: The JoNuS
Programme. DFR, MAFF, UK. 31 pp. (available from the authors).
Asmus, R. (1986) Nutrient flux in short-term enclosures of intertidal sand communities.
Ophelia 26, 1-18.
Barker, P., Fontes, J-C, Gasse, F. & Druart, J-C. (1994) Experimental dissolution of
diatom silica in concentrated salt solutions and implications for palaeoenviron-
mental reconstruction. Limnology and Oceanography 39, 99-110.
Berner, R.A. (1980) Early Diagenesis: A Theoretical Approach. Princeton University Press.
Berner, R., Ruttenberg, K.C., Ingall, E.D. & Rao, J.-L. (1984) The nature of phosphorus
burial in modern marine sediments. In Wollast, R., Mackenzie, F.T. and Chou,
L. (eds.) Interactions ofC, N, P and S Biogeochemical Cycles and Global Change.
NATO ASI Series Vol. 14, pp. 365-78. Springer-Verlag, Berlin.
Blackburn, T.H. & Sorensen, J. (1988) Nitrogen Cycling in Coastal Marine Environments.
SCOPE 33, 451 pp. John Wiley and Sons Ltd, Chichester.
Brown L.J. (1994) The geochemistry of biogenic silica in sediments. MSc. Thesis,
University of Leeds, Leeds, UK.
Capone, D.G. (1988) Benthic nitrogen fixation. In Blackburn, T.H. and Sorensen, J. (eds.)
 Nutrient recycling in intertidal sediments 97

Nitrogen Cycling in Coastal Marine Environments. SCOPE 33, pp. 85-123. John
Wiley and Sons Ltd, Chichester.
Clavero, V., Niell, F.X. & Fernandez, J.A. (1994) A laboratory study to quantify the
influence of Nereis diversicolor O.F. Muller in the exchange of phosphate
between sediment and water. Journal of Experimental Marine Biology and
Ecology 176, 257-67.
Fleischer, S., Stibe, L. & Leonardson, L. (1991) Restoration of wetlands as a means of
reducing nitrogen transport to coastal waters. Ambio 20, 271-2.
Fox, L.E. (1989) A model for inorganic control of phosphate concentrations in river
waters. Geochim. Cosmochim. Acta 53, 417-28.
Gehlen, M. & Raaphorst, W. van (1993) Early diagenesis of silica in sandy North Sea
sediments: quantification of the solid phase. Mar. Chem. 42, 71-83.
Gilbert, P.M. (1988) Primary productivity and pelagic nitrogen cycling. In Blackburn,
T.H. and Sorensen, J. (eds.) Nitrogen Cycling in Coastal Marine Environments.
SCOPE 33, pp. 1-31. John Wiley and Sons Ltd, Chichester.
Green, M.O., Pearson, N . D , Thomas, M.R., Rees, C D , Rees, J.R. & Owen, T.R.E. (1992)
Design of a data logger and instrument platform for seabed sediment transport
research. Continental Shelf Research 12, 543-62.
Hopner, T. & Wonneberger, K. (1985) Examination of the connection between the
patchiness of benthic nutrient efflux and epiphytobenthos patchiness on intertidal
flats. Netherlands Journal of Sea Research 19, 277-85.
Jonge, V.N. de (1980) Fluctuations in the organic carbon to chlorophyll ratios for
estuarine benthic diatom populations. Mar. Ecol. Prog. Ser. 2, 345-53.
Jonge, V.N. de, Engelkes, M.M. & Bakker, J.F. (1993) Bio-availability of phosphorus in
sediments of the western Dutch Wadden Sea. Hydrobiologia 253, 151-63.
Joye, S.B. & Paerl, H.W. (1993) Contemporaneous nitrogen fixation and denitrification in
intertidal microbial mats: rapid response to runoff events. Marine Ecology
Progress Series 94, 267-74.
Kinzig, A.P. & Socolow, R.H. (1994) Human impacts on the nitrogen cycle. Physics Today
46,24-31.
Klump, J.V. & Martens, C.S. (1989) The seasonality of nutrient regeneration in an organic
rich coastal sediment: kinetic modeling of changing pore-water nutrient and
sulfate distributions. Limnology and Oceanography 34, 559-77.
Koike, I. & Sorensen, J. (1988) Nitrate reduction and denitrification in marine sediments.
In Blackburn, T.H. and Sorensen, J. (eds.) Nitrogen Cycling in Coastal Marine
Environments. SCOPE 33, pp. 251-73. John Wiley and Sons Ltd, Chichester.
Law, C.S., Rees, A.P. & Owens, N.J.P. (1992) Nitrous oxide: estuarine sources and
atmospheric flux. Estuarine, Coastal and Shelf Science 35, 301-14.
Lloyd, D. (1993) Aerobic denitrification in soils and sediments: from fallacies to facts.
TREE 8, 352-6.
Marinelli, R.L. (1994) Effects of burrow ventilation on activities of a terebellid polychaete
and silicate removal from silicate pore waters. Limnology and Oceanography 39,
303-17.
Nixon, S.W. (1986) Nutrient dynamics and the productivity of coastal waters. In Halwagy,
R., Clayton, D. and Behbehani, M. (eds.) Marine Environment and Pollution,
pp. 97-115. The Alden Press, Oxford.
Nixon, S.W. (1992) Quantifying the relationship between nitrogen input and the
productivity of marine ecosystems. Proceed. Adv. Mar. Tech. Conf, Tokyo 5,57-83.
Nixon, S.W., Hunt, C D . & Nowicki, B.L. (1986) The retention of nutrients (C, N, P), heavy
metals (Mn, Cd, Pb, Cu) and petroleum hydrocarbons in Narragansett Bay. In
98 S.J. Malcolm and D.B. Sivyer

Lasserre, P. and Martin, J.-M. (eds.) Biogeochemical Processes at the Land/Sea

Boundary, pp. 99-122. Elsevier, London.
North Sea Task Force (1993) North Sea Quality Status Report. Oslo and Paris
Commissions, London. Olsen and Olsen, Fredensborg, Denmark, 132 pp.
Nowicki, B.L. (1994) The effect of temperature, oxygen, salinity and nutrient enrichment
on estuarine denitrification rates measured with a modified nitrogen gas flux
technique. Estuarine, Coastal and Shelf Science 38, 137-56.
Nowicki, B.L. & Oviatt, C.A. (1990) Are estuaries traps for anthropogenic nutrients?
Evidence from estuarine mesocosms. Marine Ecology Progress Series 66,131^46.
Ortner, P.B. & Dagg, M.J. (1995) Nutrient-enhanced coastal ocean productivity explored
in the Gulf of Mexico. EOS, Transact. Amer. Geophs. Union 76, 98-109.
Owens, N.J.P. (1993) Nitrate cycling in marine waters. In Burt, T.P., Heathwaite, A.L. and
Trudgill, S.T. (eds.) Nitrate: Processes, Patterns and Management, pp. 169-209.
John Wiley and Sons Ltd., Chichester.
Pelegri, S.P., Nielsen, L.P. & Blackburn, T.H. (1994) Denitrification in estuarine sediment
stimulated by the irrigation activity of the amphipod Corophium volutator.
Marine Ecology Progress Series 105, 285-90.
Pernetta, J.C. & Milliman, J.D. (1995) Land Ocean Interactions in the Coastal Zone.
Implementation Plan. I GBP Global Change Report 33, 215 pp. International
Council of Scientific Unions.
Prastka, K.E. & Jickells, T.D. (1995) Sediment/water exchange of phosphorus at two
intertidal sites on the Great Ouse estuary, S.E. England. Netherlands Journal of
Aquatic Ecology 29, 245-55.
Prastka, K.E. & Malcolm, S.J. (1994) Particulate phosphorus in the Humber estuary.
Netherlands Journal of Aquatic Ecology 28, 397-403.
Rendell, A.R., Horrobin, T.M., Jickells, T.D., Edmunds, H.M., Brown, J. & Malcolm, S.J.
(in press) Nutrient cycling in the Great Ouse Estuary and its impact on nutrient
fluxes to The Wash, England. Estuarine, Coastal and Shelf Science.
Risgaard-Petersen, N., Rysgaard, S., Nielsen, L.P. & Revsbech, N.P. (1994) Diurnal
variation of denitrification and nitrification in sediments colonized by benthic
microphytes. Limnology and Oceanography 39, 573-9.
Rysgaard, S., Risgaard-Petersen, N., Sloth, N.P., Jensen, K. & Nielsen, L.P. (1994) Oxygen
regulation of nitrification and denitrification in sediments. Limnology and
Oceanography 39, 1643-52.
Seitzinger, S.P. (1980) Denitrification and N 2 O production in nearshore marine sediments.
Geochim. Cosmochim. Ada. 44, 1853-60.
Seitzinger, S.P. (1988) Denitrification in freshwater and coastal marine ecosystems:
ecological and geochemical significance. Limnology and Oceanography 33,702-24.
Sundback, K., Enoksson, V., Graneli, W. & Pettersson, K. (1991) Influence of sublittoral
microphytobenthos on the oxygen and nutrient flux between sediment and
water: a laboratory continuous flow study. Marine Ecology Progress Series 74,
263-79.
Van Beusekom, J.E.E. & de Jonge, V.N. (1994) The role of suspended matter in the
distribution of dissolved inorganic phosphate, iron and aluminium in the Ems
estuary. Netherlands Journal of Aquatic Ecology 28, 383-95.
Vorosmarty, C.J. & Loder, T.C. (1994) Spring-Neap tidal contrasts and nutrient dynamics
in a marsh-dominated estuary. Estuaries 17, 537-51.
o o o o o o o o o o o o o o o o o o o
An overview of carbon and sulphur
cycling in marine sediments
G. Ruddy

6.1 Introduction
In most marine-type sediments (including intertidal ones) organic carbon
is the only reducing agent to enter a sediment column. The remainder of
the sediment load arrives in its oxidised form, and, with the exception of
straightforward compaction, early diagenesis (i.e. the process of change
during burial) results directly or indirectly from the flow of electrons. The
initial source of the electrons (organic matter) is sequentially oxidised in
microbially mediated reactions, using a range of available oxidising
agents, and results in some degree of vertical zonation in sediment
chemistry (Richards et a/., 1965; Froelich et al, 1979). This is because
microbial communities outcompete each other for organic carbon (Stumm
& Morgan, 1970; Claypool & Kaplan, 1974). However, the resolution of
these sediment horizons is likely to be poor because:

(a) although the most important substrate shared by each community

is organic carbon, it has been suggested that different communities
metabolise different fractions of it, each with a first order rate
constant (Berner, 1977). This would produce overlapping or
coexistence of zones, unless the competition between communities
was for another shared substrate (e.g. nitrate).
(b) the 'dead cat in the mud' effect (Coleman, 1985), where the
substrates for microbial metabolism do not enter the system as
disseminated and reactive particles to give a homogeneous
reaction mixture, but rather as poorly sorted, highly concentrated
sources of metabolites that may survive beyond zones where they
would have been exhausted in a simple layered model. For solid
substrates such as organic carbon, iron oxides and manganese
oxyhydroxides, this may be an important control on their rate of
supply to the reaction.
100 G. Ruddy

(c) the rate of supply of soluble oxidising agents (i.e. oxygen, nitrates
and sulphate) and the removal of reaction products (e.g. H 2 S,
F e 2 + , Mn 2 + , HCO 3 ~) will depend not only on the rate of burial
and diffusion (Lerman, 1979), but also on the modifying effects of
bioturbation (e.g. Aller, 1980, 1984) and the regeneration of
oxidising potential in oxidising conditions (Goldhaber & Kaplan,
1974; Jorgensen, 1977; Swider & Makin, 1989). Biomixing (e.g.
Berner & Westrich, 1985; Hines & Jones, 1985; Sharma et a/.,
1987) and tortuosity (the lengthening of diffusion paths in
sediment pores relative to open solutions, e.g. Li & Gregory, 1974)
may therefore have important consequences for the arrangement
of the diagenetic zones.
(d) seasonality will cause the sinusoidal migration of zones upwards
and downwards from an average or net level due to factors which
vary seasonally such as bioturbation and temperature (Jorgensen,
1977; Abdollahi & Nedwell, 1979; Troelson & Jorgensen, 1982),
and hence each reaction will move constantly through the
diagenetic effects of other reactions. The resolution of boundaries
may be lowered further by different rates of equilibrium attainment
for the different components of each zone.

Where the sedimentation rate is low, or other flux controls such as the
rate of diffusion are high, the observed zonation may depend very little on
the burial fluxes of reactants, and the chemical system may be considered
relatively open to exchange with its surroundings. The greater this degree
of openness is, the greater the perturbation of a layered sequence of
diagentic reactions is likely to be. This is likely to be particularly true of
intertidal sediments and has important consequences for the interpretation
of observations.
What follows is a discussion of some of the carbon and sulphur
mechanisms and ideas contained within the diagenetic literature. It is
intended to give the reader both a general background against which to
interpret observations and also to emphasise the importance of micro-
structure in understanding natural systems.

6.2 Carbon cycling

The rate of organic carbon remineralisation by microbial oxidation has
been empirically shown by Middleburg (1989) to follow a first order decay
curve. This relationship accounts for carbon diagenesis over about eight
orders of magnitude in the carbon influx, covering very diverse environments.
 Overview of carbon and sulphur cycling 101

The relationship can be stoichiometrically related to rates of electron

acceptor reduction and has the advantage over previous, more complicated
models, such as Berner's 'multi-G' model (Berner, 1978) in that no
distinction between different types of organic carbon with different
reactivities, is required. If the relationship is valid, it suggests that carbon
diagenesis rates are determined by one common reaction. The suggestion
is that this reaction is the bacterial fermentation of bio-polymers,
producing metabolisable organic carbon for microbial communities
using electron acceptors like oxygen and sulphate. This step produces
amino acids, fatty acids, sugars, etc. (collectively known as monomers)
from the proteins, lipids and carbohydrates which comprise the labile
biopolymers (Welte, 1973). This transformation is similar to methanogenesis
in that no external oxidising agent is involved (i.e. it is a form of chemical
disproportionation).
The rates of monomer metabolism also appear to be first order with
respect to the concentration of the monomer (rather than second order
with respect to monomer and oxidising agent concentration), and
therefore the concentration of the oxidising agent in question can be
regarded as being in excess within its 'zone' (Devol et a/., 1984; Martens &
Klump, 1984; Kuivila, Murray & Devol, 1989). This suggests that
microbial communities do indeed compete for metabolisable organic carbon.
In Table 6.1 a series of idealised decomposition reactions are listed in
order of their energy yields per mole of carbon, and therefore, broadly
speaking, their diagenetic sequence. The relative importance of each
reaction to the overall carbon budget depends on the relative supply of
each reaction's electron acceptor and the relative supply of each reaction's
metabolisable carbon. For carbon supply, the rate of sedimentation is

Table 6.1 Idealised decomposition reactions in order of their energy yields per
mole of carbon from an idealised organic molecule (after Richards et al., 1965)

Aerobic respiration: (CI^OyNHa) (H 3PO4)Z 4- (x 4- 2y)O2 =>

xCO 2 + (x + ly)H 2 O 4- yHNO 3 4- zH 3 PO 4
Manganese reduction: (CH 2 O) X (NH 3 )/H 3 PO 4 ) Z 4- 4xMnOOH 4- 7xCO 2 4- xH 2 O=>
8xHCO 3 ~ + 4xMn 2 + + yNH 3 4- zH 3 PO 4
Nitrate reduction: (CH 2O);c(NH3)>!(H3PO4)z + 4xNO 3 " =>
xCO 2 4- 3xH 2 O 4- 4xHCO 3 " + 2xN 2 + 5j;NH3 + 5zH 3 PO 4
Iron reduction: (CH2O);c(NH3)>.(H3PO4)z + 4xFe(OH) 3 4- 7xCO 2 =>
8xHCO 3 " 4- 3xH 2 O + 4xFe 2 + + yNH 3 + zH 3 PO 4
Sulphate reduction: 2(CH 2 O) x (NH 3 yH 3 PO 4 ). + x S O 4 2 " =>
2xHCO 3 " 4- xH 2 S + 2yNH 3 + 2xH 3 PO 4
Methaneproduction:2(CH 2 0) x (NH 3 yH 3 P0 4 ).=»xC0 2 + xCH 4 4- 2yNH 3 + 2zH 3 PO 4
102 G. Ruddy

important, since a rapid accretion rate will tend to increase the flux of
labile carbon to lower 'zones' in the sediment column and therefore
enhance their contribution to overall remineralisation. Clearly the size of
the influx will also be important in this respect, since a low flux will tend to
be used up more aerobically than a bountiful one for a given accretion
rate. Rates of supply of the various electron acceptors vary depending on:
(a) their rate of burial
(b) particulate nature
(c) molecular diffusion coefficients
(d) resupply by biomixing from above
(e) regeneration of their oxidising potential by migration into more
oxidising environments.
Regeneration processes, in particular, are considered to be critical
during early diagenesis, since they will usually account for the bulk of
potential energy release (e.g. Ruddy, 1993). It is in this respect that
micro-environments have an important role to play in sediment chemistry.
A mass balance model for carbon based on the observations of many
workers in this field is shown in Figure 6.1. It is designed to give some
indication of the types of profiles that might be expected in a typical
sediment column from a mass balance point of view, rather than to

refractory
depth biopolymers buried carbonates

suboxic zone

sulphate reduction authigenic

zone carbonates

methanogenesis
, zone

% Total Carbon Present

0 100
Figure 6.1 Notional carbon speciation in sediments. Profiles of carbon
reservoirs observed in sediments, combined to give a qualitative picture of
carbon cycling processes.
 Overview of carbon and sulphur cycling 103

quantify each carbon reservoir. In the oxic layer, decomposition proceeds

aerobically. Aller (1982) suggests that biomixing of labile organic carbon
downwards will result in a 'convex up' profile in this zone, and also
prevents the build up of alkalinity due to the enhanced upward diffusion
of dissolved carbon dioxide. If the oxic layer is well established then much
of the diffusive flux of carbon dioxide out of the sediment column will
originate from this zone (Janke, Emerson & Murray, 1982; McNichol, Lee
& Druffel, 1988).
Suboxic metabolism, which includes nitrate, manganese and iron
reduction, does not generally seem to form a discrete layer due to the
relatively low levels of nitrate available under natural conditions and the
insoluble nature of manganese oxyhydroxides and iron oxides. These
reactions appear to be restricted, where they are apparent at all, to
micro-environments, but since, of the reduction reactions in Table 6.1,
they are the only ones to generate alkaline conditions, they may be
important for the precipitation and preservation of carbonate minerals.
Sulphate reduction in marine sediments is often the dominant form of
carbon remineralisation, and contributes most of the alkalinity observed
in the sediment pore-water. When concentrations of sulphate fall below 35
to 40 |iM, carbon dioxide reduction by methanogens may begin, whereas
sulphate levels below about 30|iM are required before the onset of
acetate-type reduction reactions that dominate methanogenesis (Kuivila
et a/., 1989). Carbon dioxide reduction is usually only evident in the
sulphate reduction zone, where bicarbonate is produced, but it can
contribute about 65% of the methanefluxat its peak (Crill & Martens, 1986).
Under certain circumstances, the bulk of the methane flux leaving the
sediment does so as gas bubbles, which can form established tubes
through the sediment in summer months, provided the hydrostatic
pressure is low enough (Martens & Klump, 1984). There have also been
many observations of a secondary sulphate reduction rate peak at the
sulphate reduction-methanogenesis interface (e.g. Crill & Martens, 1987).
In this case the effect appeared to occur as summer blooms of short
duration, in close association with hydrogen-oxidising communities in
the sulphate reducing zone, and may have accounted for as much as 30%
of the integrated yearly sulphate reduction where it was present.
At steady state, the carbon leaving the system as carbon dioxide and
methane fluxes into the overlying water or air, will be equivalent to the
amount of carbon remineralised in the column below, integrated over its
depth. Dissolved organic carbon (DOC) exchange from pore-water to sea
water is thought to have a minor effect on the mass balance (Martens &
104 G. Ruddy

Klump, 1984). Howes, Dacey & King (1984) have shown that saltmarsh
diagenesis of organic carbon produced broadly similar carbon re-
mineralisation and oxygen uptake rates (not because the carbon was
aerobically remineralised, but because the oxygen uptake also integrated
the oxidation of reduced products and therefore anaerobic remineralisation
as well). This budget, however, incorporated some important seasonal
fluctuations, where during the summer, carbon remineralisation rates
were higher than oxidation rates and resulted in the build up of reduced
end products. During the winter the trend was reversed with a corresponding
increase in the system's oxidising potential.
The reverse situation is observed at Stiffkey marsh, North Norfolk, UK
(Ruddy, 1993). Here, the steady state decomposition of organic carbon
(Figure 6.2a), principally by sulphate reduction, forms a layer of solid
sulphides immediately below the surface of the sediment (see Figure 6.4c).
During the winter, there is a net increase in the concentration of these
sulphides as a result of relatively limited oxygen influx (Figure 6.2b). The
situation in summer appears to change due to enhanced sediment
irrigation by burrowing infauna. This supplies enough oxygen to not only
oxidise the increased production (due to higher temperatures) of the
reduced species (e.g. H2S), but also oxidise the solid sulphides accumulated
over the winter. The implication is that both oxic and anoxic bacterial
reactions are present in spatially related micro-environments within the
sediment column, and the degree of anoxia observed is the balance of the two.

6.3 Sulphur cycling

Figure 6.3 shows a mass balance model for sulphur speciations. Once
other more thermodynamically amenable electron acceptors have been
depleted, sulphate is used by sulphate reducing bacteria, such as Desulfovibrio
sp. This occurs until the sulphate concentration falls to a point where it is
outcompeted for the organic substrate by methanogenic bacteria. The
sulphide produced by sulphate reduction is then removed or recycled by
different processes depending on the environment of deposition.
Variations in the sulphate reduction rate have been shown to depend
on four factors:

(a) temperature (which controls the rate of bacterial metabolism)

(b) pressure (which affects the removal of dissolved gases such as H2S,
CO2)
(c) the concentration of sulphate
(d) the concentration of reactive carbon.
 Overview of carbon and sulphur cycling 105

Since the work of Boudreau & Westrich (1984) and Kuivila et al. (1989)
has shown that the sulphate concentration only becomes rate limiting at
concentrations well below 10% that of sea water, and changes in pressure
in tidal systems are fairly small, the two important controls on marine and

10 -

5 20

30

40
0.5 1 1.5 2
organic carbon (dry weight %)
(a)

25

M Oxygen Demand
• Oxygen Supply
20

V 15
•Li

nn

I
Nov Dec
Ik
Jan Feb Mar Apr May
-L
Jun Jul Aug Sep
(b)
Figure 6.2 Carbon decomposition at Stiffkey, North Norfolk, UK
(1990-1991) (Ruddy, 1993). (a) Organic carbon concentration with depth
shows progressive steady state depletion due to decomposition, (b)
Seasonal oxygen supply and demand estimated from modelling of carbon
and sulphide oxidation rates.
106 G. Ruddy

intertidal sediment chemistry will be temperature, which will result in the

seasonality of the cycle, and the reactive organic carbon concentration.
Maximum rates of reduction, and therefore maximum rates of supply of
reduced soluble sulphides and alkalinity, occur where the rate of supply of
reactive organic carbon is highest (i.e. at the surface of the sulphate
reduction 'zone'), and when the temperature is highest. The sulphate
reduction rate will follow the carbon remineralisation rate over the same
zone, unless, as mentioned earlier, secondary sulphate reduction of the
upwardly migrating products of deeper processes is also occurring.
The removal of reduced end products will also be important, since any
build up of these products will inhibit the forward reaction and may reach
toxic concentrations. This removal can occur in two ways: firstly, by
precipitation of relatively insoluble reduced sulphur, and, secondly,
diffusion away from the site of production and subsequent reoxidation.
This division is not a simple one, however, since the two processes
interact. The precipitation of pyrite, for instance, can be regarded as one
step in the reoxidation of reduced sulphides.
Observed authigenic reduced sulphide minerals include amorphous
FeS, mackinawite (FeS), greigite (Fe3S4), pyrite (FeS2) and to a much
lesser extent minerals which incorporate other cations, notably Mn(II).
Alabanite (MnS), however, is very rare, and has only been found
associated with sedimentary systems where Mn/Fe ratios are exceptionally

depth polysulphides
0
oxic zone thiols

elemental
suboxic zone sulphur

sulphate reduction
zone organic
sulphur

methanogenesis

% Total Sulphur Present

0 100
Figure 6.3 Notional sulphur speciation in sediments. Profiles of sulphur
reservoirs observed in sediments, combined to give a qualitative picture of
sulphur cycling processes.
 Overview of carbon and sulphur cycling 107

high (e.g. in parts of the Baltic (Boesen & Postma, 1988)). Amorphous FeS,
mackinawite and greigite are metastable (and more soluble) with respect
to pyrite so they are generally not great repositories of reduced sulphur,
whereas pyrite comprises the greater part of the total sulphur retained by
the sediment.
The solubility products of the metastable iron sulphides are also not
usually exceeded, except where sulphate reduction is at its peak, whereas
pyrite is usually supersaturated in pore-waters throughout the year. This
implies that to some extent it might form directly, without the precursor
FeS thought to be necessary for framboid formation (cf. work by Luther et
aU 1992, discussed below). Lord & Church (1983) found that sulphide
precipitation seemed to occur in two separate ways within the sulphate
reduction zone. Pyrite concentrations, as single crystals rather than
framboids, increased rapidly at the top of the sulphate reduction zone (to
form about half of the pyrite buried), before sulphide became detectable in
the pore-waters, whereas within the zone of pore-water sulphide
accumulation, rates of formation were much slower. The sulphur in pyrite
has an average oxidation state of — I, rather than — II, so an oxidising
agent is obviously required at some stage in the process, and since it seems
likely that pyritisation occurs by the addition of sulphur (rather than the
removal of iron), it is reasonable to suggest that the oxidising agent might
be an oxidised sulphur species such as elemental sulphur. Elemental
sulphur often shows a broad peak just above the sulphate reduction zone,
implying its formation by oxidation with either Fe(III) minerals, Mn(IV)
minerals or indeed oxygen brought down by irrigation processes, but it is
also found higher in the sediment profile, possibly associated with
particular micro-environments. The labelled incubation studies of Howarth
& Jorgensen (1984) show that sulphate reduction does occur within the
oxic zone, further supporting this idea.
It may be that this elemental sulphur is the source of polysulphides
(S^2"), which were proposed by Rickard (1975) to react with iron
monosulphides and directly precipitate pyrite in euhedral form. This
polysulphide is thought to result from the reaction of elemental sulphur
and HS" (Teder, 1971; Gigenbach, 1972). Luther (1991) suggests that the
conversion from FeS to euhedral FeS2 occurs via a complex formed by
FeS and S52~, since the step requires both FeS to be solubilised and a
change in the electron configuration of the Fe(II). These laboratory
studies seem to confirm Rickard's original mechanism.
The upper zone of sulphide precipitation can therefore be regarded as
being formed by the reaction of FeS with elemental sulphur, which is
108 G. Ruddy

known to nucleate FeS precipitation (Boulegue, Lord & Church, 1982),

and is also thought to be available at the oxic-anoxic boundary. Since
pore-water sulphide does not accumulate in this zone, the limiting factor
on the formation of pyrite at this level would seem to be the availability of
S( — II) rather than Fe(II) or S(0). In the lower zone, concentrations of
metastable iron sulphides are low, implying rapid conversion to pyrite. It
is therefore the formation of the metastable iron sulphides that becomes
limiting to the process. Since dissolved sulphide is accumulating at this
level, it is Fe(II) availability that is considered to be the limiting factor. The
lower concentrations of more oxidised forms of sulphur (e.g. elemental
sulphur) may also limit this process, since the zone is also characterised by
greigite (Fe 3S4), perhaps initially forming from iron monosulphide
enriched pyrite - Fe(II)3,S(0),S( — II) 3 - that eventually forms pyrite
framboids by internal nucleation (Sweeney & Kaplan, 1973). The
preservation of metastable iron sulphides can be expected where the
availability of reduced sulphur is low, and they are often found associated
with siderite cements, themselves apparently diagnostic of a relatively
high Fe/S ratio (Pye et a/., 1990). It may be that the pyrite formed in this
zone forms directly by reaction between H 2 S 2 (itself formed by the
reaction of S( - II) and S(0)) and Fe(II).
An alternative mechanism is suggested by Drobner et al. (1990), where
the reaction between FeS and H 2 S to produce pyrite and H 2 (where H + is
the oxidant) was observed under anaerobic conditions in laboratory
experiments. This ties in with the observation that H 2 might constitute an
important source of energy for some bacterial communities (Crill &
Martens, 1987).
The integrated sulphate reduction rates of coastal marine sediments far
exceed the amount of reduced sulphur actually buried (e.g. Jorgensen,
1977), and assuming steady state this could occur either by loss of the
sulphur in reduced form (e.g. the volatile gas dimethyl sulphide, DMS) or
by loss of the sulphur during regeneration processes during mixing at the
surface. The flux of reduced sulphides from the sediment by ebullition
(transport of gas as bubbles), perhaps associated with fairly stable gas
bubble tubes formed during the summer, was measured by Chanton,
Martens & Goldhaber (1987) in shallow marine sediments, and was found
to be a very small part of the sulphur budget. As far as regeneration of
sulphate is concerned, sulphate/chloride ratios often exceed those of sea
water in the oxic zone, which suggests that reoxidation may enhance
sulphate concentrations and thereby total sulphate reduction. The
reduced sulphur species involved may be either upwardly migrating
 Overview of carbon and sulphur cycling 109

sulphide ions or upwardly mixed (or stationary if the zones shift

seasonally) solid iron sulphides, which are converted by both microbially
mediated and abiotic oxidation to S(0), polysulphides and ultimately
sulphate. Jorgensen (1990), using isotopically labelled incubations of fresh
water sediments, showed that the majority of this anoxic sulphide
oxidation proceeded by way of thiosulphate. Two-thirds of the initial
product of this process was released into the pore-waters as thiosulphate,
the remainder being sulphate. This thiosulphate was then transformed
microbially into SO 4 2 ~, reduced to form S2~, or disproportionated to
produce SO42~ and S2~. Overall, in this anoxic sediment, 72% of the
thiosulphate sulphur became S2~ and the remainder SO 4 2 ~. Fossing &
Jorgensen (1990) confirmed thesefindingson estuarine sediments, observing
a very similar pattern.
The equilibrium of the resulting species and the reduced species they
formed from will depend on the redox conditions at the time, and will
probably involve those reactive iron(III) minerals that are present; hence
the equilibrium is likely to vary from environment to environment.
Davis-Colley, Nelson & Williamson (1985) have shown that this balance
may control the behaviour of the sediment as a sink or source of
pollutants since many form polysulphide-type complexes which can
greatly enhance the solubility of some otherwise insoluble trace elements.
Since pyrite is thought to be an important sink for trace metals (e.g.
Raiswell & Berner, 1986; Morse & Cornwell, 1987), the cycling of these
elements may be intimately linked to the cycling of sulphur. However, as
Howarth & Jorgensen (1984) have pointed out, these relatively oxidised
sulphur phases are very dynamic and can undergo a plethora of inorganic
and organic redox reactions, usually in association with organic sulphides,
so quantifying this part of the sulphur cycle is extremely difficult.
Luther et al. (1983) looked at organic sulphur cycling in saltmarsh
pore-waters, and found that during cold periods organic sulphur
compounds were nearly absent from the system. The large concentrations
of tetrathionate in the oxic layer also suggested that much of this
sulphur was converted to inorganic sulphur during the cold period. The
most important organic sulphur species they found, which could reach
concentrations equivalent to reduced inorganic sulphur in the sulphate
reduction zone, was a thiol, possibly glutathione. These thiols are known
to form under acid conditions by the reaction of organic molecules with
sulphur anions, but they also form where pyrite production is at its
maximum. This suggests that pyrite may be a starting point for their
formation. Moreover, this reaction, by microbial carbon reduction,
110 G. Ruddy

oxidised pyrite to form the thiol and sulphate in the absence of oxygen
and without generating acidity, which fits the observations of relatively
constant pH at this level in the sediment.
At the surface of some marine sediments, organic sulphur can comprise
as much as 50% of the total sulphur present (Francois, 1987) due to
biosynthesis which incorporates sulphur of all oxidation states, but also,
because of the reactivity of sulphides and polysulphides, by chemical
addition. There is usually an increasing S/C ratio with depth in sediments,
partly associated with humic substances, and most of this increase occurs
in the oxic and suboxic zones. This organic repository may be the source
of the sulphur required to convert metastable iron sulphides, formed in
the lower part of the sulphur reduction zone, to framboidal pyrite, which
is often found closely associated with organic matter.
The availability of iron partly controls pyritisation processes, which in
turn is an important control on the cycling of sulphur. Sulphur retention
by the sediment system will depend on the presence of a layer of reactive
iron oxides at the surface to precipitate upwardly migrating S( — II) (or
available sulphides) which may in turn control trace metal mobilisation,
sediment productivity and community metabolism. Free iron, which is
the dominant cation in most of the diagenetic mineral phases, will likewise
control the precipitation of environmentally diagnostic mineral assemblages,
which suggests that, as an oxidant, Fe(III) accounts for a relatively large
proportion of the cycling of electrons. Iron cycling may therefore
constitute a major part of the budgeting of sulphur (and other elements) in
coastal marine sediments. The processes involved are not well understood
because of the complexity of the iron minerals and the poorly understood
mechanisms of pyrite precipitation.
Luther et al. (1992) have suggested that in saltmarsh sediments, where
primary productivity is very high, the production of organic ligands by
plants and bacteria may form complexes with Fe(III) and Fe(II) and may
therefore have a role to play in the cycling of iron. Their observations
suggest that there are four important steps that account for its mass
transfer. Firstly, the solubilisation of Fe(III) by organic ligands, which
enhances this otherwise slow reaction, but on the other hand suggests a
biotic seasonality. Secondly, this mobile oxidising agent will then be
reduced by reduced species in the sediment, such as pyrite (thereby
oxidising it). Thirdly, the reduced iron complex formed will be oxidised
principally by Fe(III) minerals, which Luther et al. (1992) considered to be
the dominant oxidant. The net effect of this is the transfer of electrons from
the organic matter to Fe(III) minerals via a cyclic organo-Fe complex.
 Overview of carbon and sulphur cycling 111

Where the sulphate reduction rate is high, producing an excess of H 2 S,

reduced iron will precipitate to give rise to a reduced sulphur pool, which
comprises the fourth step. This reduced pool may then be reoxidised when
H 2 S levels fall off, either with depth or with seasonal change. Luther et al.
(1992) considered that Mn(IV), being less abundant than Fe(III), did not
form a significant reservoir for these electrons. In terms of electron flow,
however, it should be noted that the size of the potential reservoir is only
one of the important factors. The rate of the reactions involved and their
free energy changes may offset any relative scarcity.
It is worth noting that although the specific mechanisms of the sulphur
cycle are of critical importance to understanding marine-type sediment
diagenesis, it is also true that the mechanisms themselves form a part of a
wider dynamic system capable of responding to changes in conditions.
For instance, the saltmarsh sediment at Stiffkey, North Norfolk, shows an
interesting pattern of iron, manganese and sulphur accumulation with
depth (Figure 6.4a,fc,c) as well as seasonally. Pore-water concentrations of
iron and manganese are high at the surface and during the winter, where
the sediment is relatively anoxic. When oxygen supply is high relative to
demand (i.e. below about 2 cm and during the summer), pore-water
concentrations of these metals fall and solid sulphides are slowly oxidised.
The net result of this is a system which preserved very little of its reducing
potential (i.e. electrons). It results from rapid manganese and iron redox
cycling (hourly and daily turnover of reservoirs) and longer term sulphur
cycling (seasonal and yearly reservoir turnover), which form dynamic
intermediate electron transfer cycles between the carbon source and the
oxygen sink (Ruddy, 1993).

6.4 Conclusion
A summary of the diagenetic reactions concerning the transfer of electrons
during burial is shown in Figure 6.5. Each line represents a flux of
electrons from one reservoir to another. Each environment will show a
different balance of these reservoirs, since each biodiagenetic regime will
be dominated by different fluxes used to accommodate the flow of
electrons from the organic matter. Ultimately, these electrons will be
either buried, as authigenic minerals or as geopolymers, or leave the
system by reaction with oxygen. It is well established that even in the most
reducing environments (i.e. where anaerobic metabolism accounts for the
bulk of the carbon degradation), only a small proportion of the electron
flux that leaves the organic material is actually buried (principally as
 0.4 0.6 0.8
(c) Total Sulphur (dry weight %)

Figure 6.4 Iron, manganese and sulphur at Stiffkey, North Norfolk

(1990-1991) (Ruddy, 1993). (a) Pore-water iron profiles, (b) Pore-water
manganese profiles, (c) Solid sulphide (predominantly pyrite) profile.
Overview of carbon and sulphur cycling 113

CH2O.N.P

0> MnO,

pore-water
reservoir
electron flux

electron flux on pot

Figure 6.5 Electron transfer reactions for sediments (Ruddy, 1993).

CH2O.N.P (organic matter) is transformed by bacterial decomposition
reactions to bicarbonate in the pore-waters. This is the primary source of
electrons (and therefore energy) for the remainder of the sediment redox
chemistry. Most of the primary flux of electrons may pass through the
sulphur, iron and manganese cycles, but will eventually react with oxygen.
Only a small part of the total electron flux will ultimately be buried as
reduced minerals.
114 G. Ruddy

pyrite), and since the terminal electron acceptor for the remainder of the
flux will be oxygen, any anaerobic metabolism of organic matter that
occurs can be regarded as indirect aerobic metabolism. Over Phanerozoic
time (the last 570 million years), the relatively constant oxygen concentration
in the atmosphere (Berner, 1982, 1989; Garrels & Lerman, 1984) testifies
to the fact that the burial flux of organic matter and pyrite must be
equivalent to the flux of reduced weathering products back to the surficial
environment, even though, at a point in time and space, this steady state
may not exist.
For marine-type sediments, the important fluxes during the adjustment
of incoming 'energy' or diagenesis, are likely to be sulphate reduction (cf.
Jorgensen, 1982; Howarth & Jorgensen, 1984) and the reoxidation of its
products by more thermodynamically favourable electron acceptors (e.g.
Howarth & Teal, 1979; Giblin & Howarth, 1984). For any element
capable of undergoing redox reactions, the same pattern of cyclical
reduction and oxidation will be evident. The sediment system can
therefore be described in terms of a series of redox 'cells', operating in
parallel. These cells accommodate the flux of electrons from the organic
matter to their ultimate sinks, in such a way that they minimise the
potential difference between the reducing and oxidising agents, and
ultimately the potential difference between the sediment system and its
environment. The majority of these transformations are likely to be
microbially mediated because this energy can be harnessed for their
processes. Any burial flux of electrons represents lost energy and an excess
of atmospheric oxygen.
These redox cells can operate on a number of scales that depend on the
length of the diffusion path from the point that the oxidised form becomes
reduced to the point where it reduces another sediment constituent. In
some pelagic cores these diffusion paths can be observed in linear portions
of the pore-water profiles (e.g. Sawlan & Murray, 1983). Here the
sedimentation rate and the carbon burial rate are sufficiently low, relative
to diffusion, to extend the processes of early diagenesis over tens of metres
into the sediment. In coastal environments the sedimentation rate and the
concentration and reactivity of the organic matter is often high, which
results in a much more complex pattern. In this case, the distances
between the cells are much shorter, since by definition the adjustment
must occur more rapidly. Like laminar and turbulent flow, there may
come a point where the flow of electrons downwards is better dispersed
through 'eddies', which in this case are transitory micro-environments
with small-scale three dimensional diffusion, rather than more stable
 Overview of carbon and sulphur cycling 115

zones separated by large scale one dimensional diffusion. The ubiquitous

nature of bacteria ensures that the microbial ecology at a point in the
sediment can keep pace with the changing conditions, so one can envisage
centres of primary microbial activity surrounded by areas of redox
transfer in a shifting pattern with time (and therefore depth). At a point in
time, therefore, the 'zones' of electron acceptor use will be arranged
radially about these centres of activity, with the redox cells transferring
the electrons outwards towards areas where the most thermodynamically
favoured electron acceptors are still relatively available.
The important point is that no redox reaction operates in isolation;
they form chains of electron transfer that release energy and drive
sediment diagenesis. The bulk chemistry observed during sampling may
mask the very dynamic nature of some of the elements, since a sample will
'average out' the differences that may be present over very small distances.
Connections between different parts of sediment diagenesis (e.g. between
microbial redox reactions and authigenic minerals), that may be very
direct at the microscale, may not therefore be apparent. Conversely,
factors such as bioturbation and bioirrigation may determine the length
of diffusion paths between redox cells, and these need to be sampled on a
larger scale.

Acknowledgements
This work was funded by a NERC PhD studentship at the University of
East Anglia, Norwich, UK. I should like to thank Julian Andrews, Tim
Jickells and Carol Turley for critical reviews of the manuscript.

References
Abdollahi, H. & Nedwell, D.B. (1979) Seasonal temperature as a factor influencing
bacterial sulfate reduction in a salt marsh. Microb. Ecol. 5, 73-9.
Aller, R.C. (1980) Diagenetic processes near the sediment-water interface of Long Island
Sound 1. Decomposition and nutrient element in geochemistry (S, N, P). In
Estuarine Physics and Chemistry: Studies in Long Island Sound (ed. B. Saltzman),
Adv. in Geophysics 22, 238-350. Academic Press, New York.
Aller, R.C. (1982) The effects of macrobenthos on the chemical properties of marine
sediments and overlying water. In Animal-Sediment Relations: The Biogenic
Alteration of Sediments (eds. P.L. McCall & M.J.S. Tevesz), Plenum Press, New
York, 53-102.
Aller, R.C. (1984) The importance of relict burrow structures and burrow irrigation in
controlling sedimentary solute distributions. Geochim. Cosmochim. Acta 48,
1929-34.
Berner, R.A. (1977) Stoichiometric models for nutrient regeneration in anoxic sediments.
Limnol. Oceanog. 22, 781-6.
116 G. Ruddy

Berner, R.A. (1978) Sulfate reduction and the rate of deposition of marine sediments. Earth
Planet. Sci. Letters 37, 492-8.
Berner, R.A. (1982) Burial of organic matter and pyrite sulfur in the modern ocean: its
geochemical and environmental significance. Am. Jour. Sci. 282, 451-73.
Berner, R.A. (1989) Biogeochemical cycles of carbon and sulfur and their effect on
atmospheric oxygen over Phanerozoic time. Palaeogeog. Palaeoclim. Palaeoecol.
(Global & Planetary change section) 75, 95-122.
Berner, R.A. & Westrich, J.T. (1985) Bioturbation and the early diagenesis of carbon and
sulfur. Am. Jour. Sci. 285, 193-206.
Boesen, C. & Postma, D. (1988) Pyrite formation in anoxic environments of the Baltic. Am.
Jour. Sci. 288, 575-603.
Boudreau, B.P. & Westrich, J.T. (1984) The dependence of bacterial sulfate reduction on
sulfate concentration in marine sediments. Geochim. Cosmochim. Acta 48,2503-16.
Boulegue, J., Lord, CJ. & Church, T.M. (1982) Sulfur speciation and associated trace
metals (Fe, Cu) in the pore waters of Great Marsh, Delaware. Geochim.
Cosmochim. Acta 46, 453-64.
Chanton, J.P., Martens, C.S. & Goldhaber, M.B. (1987) Biogeochemical cycling in an
organic-rich coastal marine basin. 7. Sulfur mass balance, oxygen uptake and
sulfide retention. Geochim. Cosmochim. Acta 51, 1187-99.
Claypool, G. & Kaplan, I.R. (1974) The origin and distribution of methane in marine
sediments. In Natural Gases in Marine Sediments (ed. I.R. Kaplan), Plenum
Press, New York, 99-139.
Coleman, M.L. (1985) Geochemistry of diagenetic non-silicate minerals: kinetic consider-
ations. Phil. Trans. R. Soc. London A 315, 39-56.
Crill, P.M. & Martens, C.S. (1986) Methane production from bicarbonate and acetate in
an anoxic marine sediment. Geochim. Cosmochim. Acta 50, 2089-97.
Crill, P.M. & Martens, C.S. (1987) Biogeochemical cycling in an organic-rich coastal
marine basin. 6. Temporal and spatial variations in sulfate reduction rates.
Geochim. Cosmochim. Acta 51, 1175-86.
Davis-Colley, R.J, Nelson, R.D. & Williamson, K.J. (1985) Sulflde control of cadmium and
copper concentrations in anaerobic estuarine sediments. Mar. Chem. 16,173-86.
Devol, A., Anderson, J., Kuivila, K. & Murray, J. (1984) A model for coupled sulfate
reduction and methane oxidation in the sediments of Saanich Inlet. Geochim.
Cosmochim. Acta 48, 993-1004.
Drobner, E., Huber, H., Wachtershauer, G., Rose, D. & Stetter, K.O. (1990) Pyrite
formation linked with hydrogen evolution under anaerobic conditions. Nature
346, 742-4.
Fossing, H. & Jorgensen, B.B. (1990) Oxidation and reduction of radiolabelled inorganic
sulfur compounds in an estuarine sediment, Kysing fjord, Denmark. Geochim.
Cosmochim. Acta 54, 2731-42.
Francois, R. (1987) A study of sulfur enrichment in the humic fraction of marine sediments
during early diagenesis. Geochim. Cosmochim. Acta 51, 17-27.
Froelich, P., Klinkhammer, G., Bender, M., Luedtke, N., Heath, G., Cullen, D., Dauphin,
P., Hammond, D., Hartman, B. & Maynard, V. (1979) Early oxidation of organic
matter in pelagic sediments of the eastern equatorial Atlantic: suboxic diagenesis.
Geochim. Cosmochim. Acta 41, 1075-90.
Garrels, R.M. & Lerman, A. (1984) Coupling of the sedimentary sulfur and carbon cycles -
an improved model. Am. Jour. Sci. 284, 989-1007.
Giblin, A.E. & Howarth, R.W. (1984) Pore water evidence for a dynamic sedimentary iron
cycle in salt marshes. Limnol. Oceanog. 29, 47-63.
 Overview of carbon and sulphur cycling 117

Gigenbach, W. (1972) Optical spectra and equilibrium distribution of polysulfide ions in

aqueous solution at 20°C. Inorg. Chem. 11, 1724-30.
Goldhaber, M. & Kaplan, I. (1974) The sulfur cycle. In The Sea, Vol. 5 (ed. E.D. Goldberg).
Wiley, New York, 569-655.
Hines, M.E. & Jones, G.E. (1985) Microbial biogeochemistry and bioturbation in the
sediments of Great Bay, New Hampshire. Estuar. Coast. Sh. Sci. 20, 729-42.
Howarth, R.W. & Jorgensen, B.B. (1984) Formation of 35 S labelled elemental sulfur and
pyrite in coastal marine sediments (Limfjorden, Denmark) during short term
35
SO 4 reduction measurements. Geochim. Cosmochim. Ada 48, 1807-18.
Howarth, R.W. & Teal, J.M. (1979) Sulfate reduction in a New England salt marsh. Limnol.
Oceanog. 24, 999-1013.
Howes, B.L., Dacey, J.W.H. & King, G.M. (1984) Carbon flow through oxygen and sulfate
reduction pathways in salt marsh sediments. Limnol. Oceanog. 29, 1037-51.
Janke, R., Emerson, S. & Murray, J. (1982) A model of oxygen reduction, denitrification
and organic matter mineralization in marine sediments. Limnol. Oceanog. 27,
610-23.
Jorgensen, B.B. (1977) The sulfur cycle of a coastal marine sediment (Limfjorden,
Denmark). Limnol. Oceanog. 5, 814-32.
Jorgensen, B.B. (1982) The ecology of the bacteria of the sulfur cycle with special reference
to anoxic-oxic interface environments. Phil. Trans. R. Soc. London B 298,543-61.
Jorgensen, B.B. (1990) The sulfur cycle in freshwater sediments: role of thiosulfate. Limnol.
Oceanog. 35, 1329-42.
Kuivila, K., Murray, J. & Devol, A. (1989) Methane production, sulfate reduction and
competition for substrates in the sediments of Lake Washington. Geochim.
Cosmochim. Ada 53, 409-16.
Lerman, A. (1979) Geochemical Processes'. Water and Sediment Environments, Wiley, New
York.
Li, Y-H. & Gregory, S. (1974) Diffusion of ions in sea water and deep-sea sediments.
Geochim. Cosmochim. Ada 38, 703-14.
Lord, CJ. & Church, T.M. (1983) The geochemistry of salt marshes: sedimentary ion
diffusion, sulfate reduction, and pyritization. Geochim. Cosmochim. Ada 47,
1381-91.
Luther, G.W. (1991) Pyrite synthesis via polysulfide compounds. Geochim. Cosmochim.
Acta 55, 2839-49.
Luther, G.W., Church, T.M., Scudlark, J.R. & Cosman, M. (1983) Inorganic and organic
sulfur cycling in salt marsh pore waters. Science 232, 746-9.
Luther, G.W., Kostka, J.E., Church, T.M., Sulzberger, B. & Stumm, W. (1992) Seasonal
iron cycling in the salt-marsh sedimentary environment: the importance of
ligand complexes with Fe (II) and Fe (III) in the dissolution of Fe (III) minerals
and pyrite, respectively. Mar. Chem. 40, 81-103.
Martens, C.S. & Klump, J.V. (1984) Biogeochemical cycling in an organic-rich coastal
marine basin. 4. An organic carbon budget for sediments dominated by sulfate
reduction and methanogenesis. Geochim. Cosmochim. Acta 48, 1987-2004.
McNichol, A., Lee, C. & Druffel, R. (1988) Carbon cycling in coastal sediments: 1. A
quantitative estimate of the remineralization of organic carbon in the sediments
of Buzzards Bay, MA. Geochim. Cosmochim. Acta 52, 1531-43.
Middleburg, J.J. (1989) A simple rate model for organic matter decomposition in marine
sediments. Geochim. Cosmochim. Acta 53, 1577-81.
Morse, J. & Cornwell, J. (1987) Analysis and distribution of iron sulfide minerals in recent
anoxic marine sediments. Mar. Chem. 22, 55-69.
118 G. Ruddy

Pye, K., Dickson, J., Schiavon, N., Coleman, M. & Cox, M. (1990) Formation of
siderite-Mg-calcite-iron concretions in intertidal marsh and sandflat sediments,
north Norfolk, England. Sedimentology 37, 325-43.
Raiswell, R. & Berner, R.A. (1986) Pyrite and organic matter in Phanerozoic normal
marine shales. Geochim. Cosmochim. Acta 50, 1967-76.
Richards, F.A., Cline, J.D., Broenkow, W.W. & Atkinson, L.P. (1965) Some consequences
of the decomposition of organic matter in Lake Nitinat, an anoxic fjord. Limnol.
Oceanog. Suppl. 10, R185-R201.
Rickard, D.T. (1975) Kinetics and mechanism of pyrite formation at low temperatures.
Am. Jour. Sci. 275, 636-52.
Ruddy, G. (1993) Microenvironmental modelling of redox chemistry in salt marsh
sediments. PhD Thesis. University of East Anglia, Norwich, UK.
Sawlan, W. & Murray, J.W. (1983) Trace metal remobilization in the interstitial waters of
red clay and hemipelagic marine sediments. Earth Planet. Sci. Letters 64,213-30.
Sharma, P., Gardner, L., Moore, W. & Bollinger, M. (1987) Sedimentation and bioturbation
in a salt marsh as revealed by 2 1 0 Pb, 137 Cs and 7Be studies. Limnol. Oceanog. 32,
313-26.
Stumm, W. & Morgan, J.J. (1970) Aquatic Chemistry, Wiley, New York.
Sweeney, R.E. & Kaplan, I.R. (1973) Pyrite framboid formation: laboratory synthesis and
marine sediments. Econ. Geol. 68, 618-34.
Swider, K.T. & Makin, J.E. (1989) Transformations of sulfur compounds in marsh-flat
sediments. Geochim. Cosmochim. Acta 53, 2311-23.
Teder, A. (1971) The equilibrium between elementary sulfur and aqueous polysulflde
solutions. Acta Chem. Scand. 25, 1711-28.
Troelson, H. & Jorgensen, B.B. (1982) Seasonal dynamics of elemental sulfur in two coastal
marine sediments. Estuar. Coast. Sh. Sci. 15, 255-66.
Welte, D. (1973) Recent advances in organic geochemistry of humic substances and
kerogen. A review in Advances in Organic Geochemistry (ed. J.P. Riley and F.
Bienner). Editions Technip, Paris, 4-13.
o o o o o o o o o o o o o o o o o o o
Microbial activity and diagenesis in
saltmarsh sediments. North Norfolk,
England
K. Pye, M.L. Coleman and W.M. Duan

7.1 Introduction
The relationship between microbial processes, diagenetic reactions and
mineralisation within intertidal sediments, including the development of
diagenetic carbonate concretions, tubules and localised cemented layers,
has attracted major interest in recent years. Such phenomena are
relatively common in intertidal sediments around the British Isles and on
other European coasts (see e.g. Pye, 1981; Andrews, 1988; Al-Agha et al,
1995; Duck, 1995). The processes associated with their formation are
important in relation to the cycling of carbon, sulphur and iron, as well as
the fate of industrial and other pollutants. Early diagenesis within
saltmarshes in particular is of importance because of the high biological
productivity which is characteristic of such environments, and their
position at the interface between marine and terrestrial systems (see
Vernberg, 1993, for a recent review of saltmarsh processes).
The role of microbial activity in early diagenesis has been recognised for
some time, and a variety of approaches has been used to investigate these
processes, including, for example, studies of C-S-Fe interactions (Berner
& Westrich, 1985), rates of sulphate reduction (Jorgensen, 1978; Jorgensen
& Bak, 1991), nitrate reduction (Sorenson, 1987) and Fe/Mn reduction
(Lovley, 1991). Recently, significant progress has been made in quantifying
the rates of microbial reactions and the interactions between biotic and
abiotic factors (e.g. Blackburn & Blackburn, 1993; Parkes et a/., 1993a,b),
partly as a result of the application of biochemical techniques which allow
detailed specification of microbial biomass, community structure and
nutritional status (e.g. White, 1993; Eglinton, Parkes & Zhao, 1993).
Important advances in understanding have also been made as a result of
experimental investigations which have sought to define the effects of
various limiting environmental factors on rates of microbial activity (e.g.
120 K. Pye, M.L. Coleman and W.M. Duan

Lovley, 1991; Coleman et a/., 1993), and through the development of

improvedfieldmethods of micro-scale sampling and subsequent laboratory
analysis (e.g. Davison et a/., 1991).
This contribution provides an illustration of the application of biochemical
and geochemical techniques in the context of an ongoing investigation
into the inter-relationships between microbial activity and sediment
diagenesis within saltmarsh and intertidal flat sediments near Warham on
the north Norfolk coast, eastern England (Figure 7.1). This investigation,

GENERAL AREA MAP

0 I5ki

INTER-TIDAL SAND FLAT

She ashin
> - \ * - • : *• * J " Q|e p
eroded marsh
v*c"*V
v
lL>
. Site
^ ' - ^ * c - : Nature ^^94090
» ^ <* r "»" Reserve ,?

0 400
i I I I I I
metres

Figure 7.1 Location of the study area and sampling locations.

 Microbial activity and diagenesis in saltmarshes 121

which involves inter-disciplinary collaboration between complementary

research teams at the Universities of Reading, Tennessee, Leeds and the
United States Geological Survey, uses a number of investigative approaches,
including studies of solid phase mineralogy, texture and chemistry,
pore-water sampling and analysis, laboratory and field experimentation,
and characterisation of recent microbial activity through the application
of signature lipid biomarker techniques. This contribution is concerned
mainly with the information provided by biomarker and associated
geochemical analysis of sediments from two marshes of differing age and
recent sedimentation history. The general nature of the information which
can be provided by biomarkers is briefly reviewed below, but for more
detailed information the reader should refer to other recent reviews by
White (1993) and Eglinton et al (1993).

7.2 Nature of the study area

The Warham saltmarshes form part of an extensive belt which extends for
a distance of 35 km along the coast of north Norfolk. The general setting
and geomorphological character of the marshes have previously been
reviewed by Pye (1992). At Warham there are two distinctive marsh
morpho-stratigraphic units, the Upper Marsh and the Lower Marsh, to
seaward of which lies a wide intertidal sand flat. The surface of the Upper
Marsh lies at c. 2.90 m O.D. (Ordnance Datum) at its seaward end,
declining to 2.73 m O.D. near the landward limit. The level of mean high
water spring tides (MHWST) decreases eastwards along the coast from
3.00 m O.D. at Wells Bar, approximately 3 km to the west of the study site,
to 2.65 m O.D. at Blakeney Bar, approximately 4 km to the east. The
respective heights for the predicted Highest Astronomical Tide are 3.10 m
and 2.75 m O.D. Consequently the Warham Upper Marsh is covered only
by the highest normal spring tides and periodic storm surges (Pye, 1992).
The Upper Marsh is separated from the Lower Marsh by a low dune ridge
which has a crest height of 3.5 to 3.6 m O.D. and a maximum width of
c. 20 m. Levees, which have a maximum elevation of 3.4 m, but more
typically lie in the range 3.1 to 3.2 m O.D., occur along the banks of the
major transverse creek (Stonemeal Creek) which separates the Upper
Marsh into inner and outer sections. Smaller levees flank the secondary
creeks, forming discrete areas of low-lying backmarsh which is prone to
waterlogging after high tides or heavy rain. Linear salt pans, which
represent the abandoned remnants of former tidal creeks, are numerous
(Figure 7.2). The remaining active creeks show a high degree of sinuosity
and lateral stability.
122 K. Pye, M.L. Coleman and W.M. Duan

Drilling investigations, palaeoecological analyses and the results of

radiocarbon dating have shown that marshes began to form along the
coastal fringe between Morston and Wells at least 6600 14C yr ago
(Funnell & Pearson, 1989). At Stiffkey, located just to the east of the
Warham marshes, sediments interpreted as saltmarsh deposits extend to a
depth of approximately — 7m O.D. Archaeological remains of Romano-
British age have been identified on several parts of the inner marshes
along the coast, testifying to an age of > 2000 years. Studies at Warham
have also shown that industrial pollutant metals are restricted to the
uppermost 5-10 cm of the sediment sequence on the Upper Marsh (Pye et

Figure 7.2 Oblique air photograph of the Warham Upper Marsh, showing
the main transverse creek (Stonemeal Creek) and the Cocklestrand path
(running left to right across the lower part of the photograph).
 Microbial activity and diagenesis in saltmarshes 123

a/., unpublished data), indicating only 5-10 cm of accretion in the past

100-150 years. This marsh can be regarded as mature in the sense that it
has achieved a surface level which is in equilibrium with the local tidal
frame. In the context of the microbial and geochemical processes
considered here, its sedimentological and geomorphological regime can
be regarded as essentially 'steady'.
The upper 2 m of sediments on the Upper Marsh consist mainly of fine
silts with localised sandier facies along the courses of the larger tidal
creeks (both active and inactive). The surface is generally well-vegetated
by halophytic plants. The higher, better drained areas along creek levees
and the margins of the seaward dune ridge are vegetated chiefly by
Halimione portulacoides, Sueda spp., Puccinellia maritima and other grass
species such as Festuca spp., and Juncus spp. Areas of intermediate
elevation are vegetated mainly by Aster tripolium, Limonium vulgare,
Halimione portulacoides and Salicornia perennis. Waterlogged areas and
the margins of some salt pans are often characterised by the presence of
patches of Salicornia spp. and Spartina spp. The vegetation forms a
continuous sward with a dense root mat in the uppermost 20 cm of
sediment. During the eighteenth and nineteenth centuries attempts were
made to enclose part of the marshes between Warham and Wells, just to
the west of the study area, but these attempts failed and were abandoned
(Purchas, 1965). The marshes at Warham itself have never been embanked
and have been modified by man-made drainage diversions to a relatively
minor degree. In recent decades the marsh has experienced only very light
to no grazing pressure and its principal usage has been for wildfowling
and as a nature reserve. Between the late 1930s and early 1950s the
marshes and sandflats to seaward were used for military target practice
with firing positions located at Warham and Stiffkey.
The Warham Lower Marsh is a much younger feature which developed
rapidly after the Second World War, and particularly from the mid 1950s
onwards, when vegetation colonised the higher part of the intertidal sand
flats to seaward of the dune ridge that forms the seaward boundary of the
Upper Marsh. During approximately the past 45 years the rate of vertical
accretion on the Lower Marsh has been rapid, averaging c. 1.5 cm y" 1 .
The landward part of the Lower Marsh currently has a surface elevation
of 2.30 to 2.35 m O.D., falling to c. 2.20 to 2.25 m O.D. near the seaward
edge. At present the marsh is covered by approximately 20% of all high
tides. The vegetation cover on the Lower Marsh is dominated by lower
saltmarsh communities which include species such as Aster, Spartina,
Limonium and Halimione spp. Weakly developed levees are present along
124 K. Pye, M.L. Coleman and W.M. Duan

some of the larger creeks. Much of the intervening marsh is poorly drained
and remains waterlogged for considerable periods after tidal flooding.
Channel pans are less common than on the Upper Marsh and the density
of active tidal creeks is higher. Particularly in the outer part of the Lower
Marsh, many of the tidal creeks display an anastomising character and
have frequently shifted their lateral position. The seaward margin of the
marsh has suffered episodic wave erosion in the past ten years, forming
areas of devegetated but root-bound muddy sand which have been
partially buried by landward moving sand lobes originating from the
intertidal flats. The intertidal flats themselves have an elevation ranging
from 1.5 to 1.9 m O.D. Higher areas are represented by sand bars and
localised accumulations of shingle and shells (mainly Cerastoderma,
Ostrea and Mytilus spp.).
Within the Lower Marsh sediment sequence there is a sharp texture
boundary at a depth of 45-70 cm, depending on the surface elevation
(Figure 7.3). The sudden change from sandy muds to relatively clean,
medium sands reflects the onset of marsh vegetation colonisation and mud
accumulation during the early 1950s. A permanent groundwater table is
present at a depth of approximately 0.8-0.9 m, although its precise level
varies diurnally and seasonally in response to tidal forcing and weather
conditions (Allison & Pye, 1994). Carbonate concretions, many of which
contain shrapnel and other debris, are widespread in the sediments close to
the water table and occasionally at other levels (Figure 7.3). The majority of
the concretions, which may attain a maximum dimension of 25 cm, are
cemented mainly by siderite, Mg-calcite and iron monosulphide (Pye,
1981,1984; Pye e£ a/., 1990). Localised beds of siderite and siderite-goethite
cemented root tubules, together with patches of cemented creek bed
sediments, are also found. Field experiments on the Lower Marsh showed
that small concretions and coatings can form around various nucleii within
as little as 12 months (Pye et a/., 1990; Allison & Pye, 1994). Biogenic
remains within the concretions have been extensively replaced by authigenic
minerals. Replacement of plant material includes (a) coating and replacement
of cells walls only, (b) wall replacement by siderite and cell infilling with
siderite or iron monosulphide, and (c) infilling of roots and stem moulds
with acicular and/or botryoidal siderite. The occurrence of siderite within
the concretions and bioclast coatings is particularly interesting. Previous
studies have associated the formation of siderite with the degradation of
organic matter by anaerobic, methanogenic bacteria, typically in fresh
water, brackish or deep-burial environments where limited supply of
sulphate restricts production of H 2 S by sulphate-reducing bacteria (SRB).
 Microbial activity and diagenesis in saltmarshes 125

However, the results of water chemistry and solid phase stable isotope
analyses indicate that the Warham siderite is forming under very different
conditions. Analysis of pore-waters and surface waters has demonstrated
that fresh water dilution is very localised, being restricted mainly to zones of
groundwater seepage. The majority of concretions occur in sediments
whose pore-waters have a similar composition to sea water, or which are
slightly supersaline due to evapotranspiration (Pye et a/., 1990; Allison &
Pye, 1994). The siderite concretions have markedly negative 5 13C values,
averaging — 5.9%o PDB, suggesting that the siderite is not formed under
methanogenic conditions. Bacterial assays have shown that SRB are
abundant and have high activities in the Lower Marsh, and siderite
formation is unlikely to be related to a shortage of H 2S.

Figure 7.3 A typical section through the Lower Marsh, showing the upper
laminated mud unit, the intermediate zone of brownish sands, and the
lower zone of black, monosulphide stained sands containing an in situ
concretion.
126 K. Pye, M.L. Coleman and W.M. Duan

Concretions similar to those found in the Lower Marsh sediments

occur locally along the banks and beds of creeks on the Upper Marsh, and
in some saltmarsh pans, but are otherwise absent. Both in situ and
reworked concretions occur in the intertidal flat sediments just beyond
the seaward edge of the Lower Marsh. The reworked concretions are
often partly oxidised and contain a range of secondary minerals including
akaganeite, gypsum and goethite (Pye, 1988).
The contrasting ages and sedimentation histories of the two Warham
marshes provide a useful framework for a comparative investigation of
the geochemical environmental conditions prevailing and their relationship
to microbial processes and mineral authigenesis. With this in mind,
several trenches and pits were dug on the Upper Marsh and Lower Marsh
during the period March 1993 - May 1995 (Figure 7.4). Samples of

t$*i -A

Figure 7.4 Trench section at Site 9409 on the Lower Marsh.

 Microbial activity and diagenesis in saltmarshes 127

sediment were collected at closely spaced depth intervals for examination

of the solid phase mineralogy and chemistry, microbial activity and
community structure, and pore-water composition. Detailed discussion
of the results will be presented elsewhere, but a summary of the analytical
approach and some typical results are presented below.

7.3 Lipid biomarkers as indicators of microbial

activity and community structure
The signature lipid biomarker (SLB) technique is based on the premise
that all micro-organisms have cell membranes composed of lipids, in
particular phospholipids. By extracting and characterising the fatty acids
associated with these phospholipids (PLFA), an assessment of the viable
microbial biomass, community structure and nutritional status can be
made. Owing to the labile nature of phospholipids, viable microbial
biomass can be determined by quantifying the total concentration of
PLFA in a sample. Patterns of PLFA recovered from individual isolates
of bacteria can be used to characterise and identify bacteria at the species
level and to estimate community compositions (Tunlid & White, 1992).
Ratios of specific PLFA have also been demonstrated to relate to the
nutritional or metabolic status of certain bacteria (Guckert et a/., 1985). In
the past decade the SLB technique has been applied in a variety of marine
and estuarine contexts (e.g. Baird et a/., 1985; Findlay et al, 1990; Parkes
et a/., 1993a,b), but the work outlined here represents the first known
attempt to apply it in the context of saltmarshes.

7.4 Microbial activity in the Warham Upper Marsh

7.4.1 Investigative methods
A 1.55 m deep trench, measuring 3 m long and 1 m wide, was dug in the
outer part of the Upper Marsh in October 1993 (site UM1 in Figure 7.1).
Sediment samples were collected at 3-5 cm intervals down to the water
table for determination of grain size by laser granulometry, major and
trace element composition by X-ray fluorescence spectrometry (XRF),
bio-assay and more specific studies of Fe, C and S species. In the trench
section at this site the top 20 cm of sediment was dark-grey brown in
colour with abundant plant roots, grading downwards into slightly
mottled, dark grey-brown mud (profile 9211U, Figure 7.5). Below 110 cm
the sediment colour changed to dark grey, but no black monosulphide
zone or concretions were encountered. Excavation of a number of other
128 K. Pye, M.L. Coleman and W.M. Duan

trenches and shallow boring investigations during the period 1993-95

have shown that this pattern of sediment zonation is widespread in the
Upper Marsh. Areas of black, reduced sediment are typically restricted to
the muddy banks and bottoms of small creeks and salt pans which contain
standing water for long periods or which are regularly flushed by tidal
waters. Such a pocket of soft, black mud occurred within the creek
adjacent to the sampled trench section. Samples were also collected from a
pit dug in these deposits (profile UCRK, Figure 7.5) for comparative analysis.
Samples for geochemical and microbial analysis were collected using a
spatula, sealed in plastic bags, and placed in dry ice for transportation to
the laboratory. A 20-30 g sub-sample from each sampling interval was
treated in situ by adding zinc acetate (10% v/v) to fix easily oxidisable
sulphide. After freeze drying and grinding, the samples were split for
geochemical, mineralogical and microbial analyses. Figure 7.6 presents a
summary of the sequence of procedures used. Organic carbon was
determined using a modified Walkley-Black method (Gaudette et a/.,
1974). Reduced sulphur species (acid-volatile sulphur AVS and pyrite
sulphur) were determined on zinc acetate fixed, freeze-dried sediment
using a sequential digestion method adapted from Canfield et al. (1986).
Reactive iron was determined in the Upper Marsh samples using two
extractants, sodium dithionite in a buffer of 0.35 M acetic acid and 0.2 M

station 9211U station UCRK

Vegetated surface

depth 155 cm

jjgg^j] Brownish grey, root-rich, silty sediment

fcj^jj Brown silty sediment, no roots

§H|§ Black, anoxic, soft mud

| | Dark grey, silty sediment

Figure 7.5 Schematic diagram showing the sediment sequence in the

Upper Marsh pit (site UM1) and creek bank (UCRK) profiles.
 Microbial activity and diagenesis in saltmarshes 129

sodium citrate at pH 4.8 (Mehra & Jackson, 1960; Canfield, 1989), and
0.2 M ammonium oxalate and 0.1 M oxalic acid solution at pH 2
(McKeague & Day, 1966).
For the microbial assay, lipids were extracted using a single phase
chloroform/methanol/water extraction after Bligh & Dyer (1959) as
modified by White et al (1979) to include a phosphate buffer. The total
lipid was fractionated into neutral lipid, glyco-lipid and polar lipid
fractions by silicic acid column chromatography (SACC) (Parker et al,
1982; White et al.9 1983) using the three solvents of increasing polarity
(chloroform, acetone and methanol). The polar lipid fraction (PLFA) was

Fresh Sediment) ZnAc 10% h-( AVS&Pyr. S

Figure 7.6 Flow diagram summarising the sequence of analytical

techniques used.
The abbreviations used in the diagram:
AVS: Acid-volatile sulphides PL: polar lipid
Pyr. S: pyrite sulphide TLC: thin layer chromatography
OC: organic carbon MAH&M: mild alkaline hydrolysis and
Reactive Fe: reactive iron methanolysis
B/D: Bligh/Dyer extraction SAH&M: strong acid hydrolysis and
Resid. LPS: residual methanolysis
lipopolysaccharide fatty acids DG: digliceride fatty acid
TL: total lipid Quin: quinones
NL: neutral lipid FAME: fatty acid methyl esters
SACC: silicic acid column LPS: lipopolysaccharide fatty acids
chromatography GC/MS: gas chromatography/mass
GL: glycolipid spectroscopy
130 K. Pye, M.L. Coleman and W.M. Duan

treated with a mild alkaline methanolysis (MAM) to cleave the glyceride

fatty acids and replace their glycerol bond with methyl esters, creating
fatty acid methyl esters (FAME). After further processing and separation,
the fatty acid methyl esters were quantified by capillary gas chromatography
and the peak identifications verified by gas chromatography/mass
spectrometry (GC/MS). Diglyceride fatty acids (DGFA) and quinones
were recovered from the chloroform-extracted fraction using a thin layer
chromatography (TLC) technique for further elutriation and separation
as applied to the PLFA. Lippolysaccharide hydroxy fatty acids (LPS
OHFA) were released from the lipid-extracted residues by strong acid
methanolysis and hexane extraction (Hedrick, Guckert & White, 1991).
After removal of the solvent, the hydroxy fatty acid methyl esters were
derivatised and analysed as for PLFA.

7.4.2 Results
Laser granulometry indicated little variation in the particle size with
depth in the trench section, the median size being 12-15 |im. On average
the bulk sediments were found to contain approximately 20% clay
( < 2 jim), 70% silt (2-63 |im) and < 10% sand ( > 63 |im). Bulk sediment
XRF analyses also showed that the major element proportions are also
relatively constant with depth (Figure 7.7). X-ray diffraction (XRD)
analyses showed some variability in the relative proportions of clay
minerals present in the < 2 j^m fraction, but all samples were dominated
by illite, smectite, kaolinite and chlorite. The calcium carbonate content
was low ( < 1 %), much of the detrital material apparently having been
dissolved. The sediment pH ranged from 6.2 to 6.7.
Organic C content decreased sharply from 9% to 1% in the top few
centimetres of the profile. Between 15 cm and 60 cm depth the organic C
content fluctuated between 1 and 4%, but below 60 cm it remained
constant at c. 1 % (Figure 7.8a). Total Fe content remained fairly constant
with depth at 4-5%. The dithionite-extractable Fe content, which
includes crystalline and amorphous iron oxides (Canfield, 1989), indicates
that the sediments contain significant amounts of Fe oxide phases (Figure
7.8c). However, in virtually all samples ammonium oxalate-extractable Fe
contributes more than half of the dithionite-extractable Fe, indicating
that amorphous and poorly crystalline hydrated Fe oxides are dominant.
In the lower, greyish part of the sequence the dithionite and oxalate-
extractable Fe contents are slightly lower, probably due to iron reduction
in the zone of fluctuating groundwater. Concentrations of elemental S,
AVS and pyrite S between the surface and 125 cm depth were found to be
 Microbial activity and diagenesis in saltmarshes 131

A B C
c
% elemental Vo elemental o elemental
composition Cimposition composition
0 20 40 60 80 0 10 20 3040506070 0 20 40 60 80
5- AAI 1 1 1 U 1 I c . L 1 j _ _
1 1
I5HD
15-

25-
\ I5-. >ll

35- V 1 •
E
CJ
E
o 45- I1 \? E

1
1T
sz sz 25-^ II [J

r
Q. Q. Q.
CD CD 55- (D
Q

r1
a
T
Q
65-
35-< )ll C
75- v ' n

85- a
Y'
95- 6 i J 454

% Fe 2 O 3 % K2O % SiO 2 % AI 2 O 3

A
% elemental % elemental
B c
% elemental
composition composition composition
I 2 3 0 2 4 6
4J
15-
25-
35-
15-
45-
E
o o

1L 25-
CD ci)
Q Q

35-

45-1

- + - % CaO % MgO % Na2O

Figure 7.7 Variation in major element composition with depth at Site
UM1. Profiles A and B represent the eastern and western ends of the trench
section (UM1); Profile C is the nearby creek bank profile (UCRK).
 Org. C % Total PLFA(pmol/g) Fe% RS%

0 2 4 6 10 0 100000 200000 300000 0 1 2 3 4 5 0.00 0.01 0.02 0.03 0.04 0.05

0
Depth
(cm)
20- 20- 20" 20"

40- 40- 40" 40"

60- 60- 60" 60"

80" 80- 80" 80-

100- 100 -d 100- 100-

120" 120" 120- Oxalate-Fe 120-

140-
Dithio-Fe
140- 140" 140"

160 160 J 160 J 160 J

(a) (b) (c) (d)

Figure 7.8 Depth variation in the Upper Marsh trench section UM1 (Profile A) of: (a) % organic carbon, {b) viable microbial biomass
indicated by total PLFA (pmol / g of dry sediment), (c) % reactive iron and % total iron, and id) total % reduced sulphur. Analytical errors
associated with the data points are < 5% for HCI-extractable Fe, 8-10% for oxalate- and dithionite-extractable Fe, 5-10% for sulphur species,
and c. 4% for organic carbon values.
 Microbial activity and diagenesis in saltmarshes 133

below the detection limit (Figure 7.8d). Between 125 and 155 cm elemental
S and AVS were also undetectable, but total reduced S represented
0.045% of the dry sediment. This suggests that sulphate reduction is a
relatively minor process in these sediments at the present time and may
never have been very important.
The total PLFA content was found to decrease rapidly with depth in
the uppermost 40 cm, showing a markedly steeper decline than organic C
(Figure 7.8fc). This suggests that organic matter becomes less readily
metabolisable with depth. Six groups indicative of community structure
were identified on the basis of PLFA GC/MS signatures: normal
saturates, terminally branched saturates, mid-branched saturates, mono-
enoics, branched monoenoics, and polyenoics. Normally saturated and
unsaturated monoenoic PLFA were found to be dominant. The two
groups show inverse abundance with depth, unsaturated monoenoics
becoming less abundant down the profile (Figure 1.9a & b\ possibly due
to a change in membrane structure from unsaturated to saturated. The
distribution of hydroxy fatty acids (OHFA) and the ratio OHFA/PLFA
were used to estimate gram negative bacteria contributions to the total
microbial community (cf. Parker et al, 1982). The ratio of OHFA/PLFA
increased throughout the sediment profile to a point where three times
more OHFA were recovered than were PLFA (Figure 7.9c). These
observations indicate that, although the total number of viable microbial
cells decreased with depth, the proportion of gram negative bacterial cells,
relative to the total viable biomass, increased with depth. The biomarker
for Desulfobacter (10Mel6:0) shows a peak just below the top of the
profile, indicating maximum absolute abundance at this level, and
generally declined with depth (Figure 7.9d). It shows consistently greater
abundance than the biomarker for Desulfovibrio (il7: Iw7c), which also
shows greatest abundance in the upper 30-40 cm (Figure 7.9e).
Nutritional stress in bacterial communities is commonly associated
with the accumulation of energy-reverse polymers, and a measure of
nutritional status can be made by measuring the ratios of lipid storage
polymers to cellular biomass. In some bacterial species the ratio of 19
carbon cyclopropyl (cy 19:0) to its monoenoic precursor (18:lw7c)
increases with nutrient stress (Tunlid & White, 1992). In the sediment
profile studied, this ratio increases rapidly with depth in the upper 40 cm,
but falls again in the lower, greyish part of the section (Figure 7.10a). A
similar trend is shown by other nutritional stress indicators, such as
cyl7 = 0/16:lw7c, 18: Iw7t/18: Iw7c and 16: Iw7t/16: Iw7c (Figures
T.lOb-d). This pattern with depth may reflect the fact that nutrients are
 Mono., mol % N. Satur., mol ' ratio, OHFA/PLFA 10Mel6:0, mol< il7:lw7c, mol %
10 20 30 40 50 10 20 30 40 0 2 4 6 8 10 12 0 1 2 3
1 1 1 1 1 1
o- 0 0
Depth
(cm)
20- 20- 20- 20 20

40" 40- 40" 40" 40

60- 60- 60" 60- 60-

80- 80- 80- 80" 80"

100- 100- 100" 100" 100"

120" 120- 120" 120" 120"

140" 140" 140" 140- 140-

160 J 160 J 160 J 160 J 160 J

(a) (b) (d) (e)

Figure 7.9 Depth variation in microbial community structure in the Upper Marsh trench section (Profile A) indicated by changing
abundances of: (a) monoenoics, {b) normal saturates, (c) ratio of OHFA to PLFA, {d) 10Me16:0 (biomarker for Desulfobacter), and (e) i17:1w7c
(biomarker for Desulfovibrio).
 Ratio Ratio Ratio Ratio
1 2 3 4 0.0 0.2 0.4 0.6 0.0 0.1 0.2 0.3 0.00 0.05 0.10 0.15
0
Depth
(cm)
20 20" 20 20"

40" 40" 40" 40

60- 60" 60" 60-

80- 80" 80" 80"

100- 100- 100- 100"

120- 120- 120" 120"

140" 140- 140" 140"

160 J 160 J 160 J 160 J

(a) (b) (c) id)
Figure 7.10 Depth variation of indicators of nutritional stress in the Upper Marsh trench section (Profile A, indicated by ratios of: (a)
cy19:0/18:1wc, (b) cy17:0/16:1wc, (c) 18:1w7t/18:1w7c and (d) 16:1w7t/16: 1w7c.
136 K. Pye, M.L. Coleman and W.M. Duan

supplied to the sediment both from the surface by sea water flooding and
organic matter addition from the vegetation, and from groundwater.
Sediments at a depth of 100-110 cm apparently provide the most difficult
environment for bacteria to thrive in this profile.
These data are consistent with the picture of the Upper Marsh
experiencing slow vertical accretion in response to a very gradual rise in sea
level. Throughout the time period represented by the sediment profile
investigated (estimated to be 1600 to 3200 years, assuming an average
accretion rate similar to that during the past 100 years), predominantly oxic
conditions appear to have prevailed throughout the upper part of the
sediment column. Anaerobic bacteria, including sulphate reducing bacteria,
are poorly represented in this profile; although they show highest
abundance in the upper, more organic-rich part of the profile, they become
relatively more important as a proportion of all bacterial types with depth.
Pore-water analyses have shown that the groundwater and interstitial
waters above the water table are not depleted in sulphate, and therefore the
activity of sulphate reducing bacteria may have been restricted principally
by the availability of suitable reactive organic matter and/or by the
relatively high levels of oxygenation which allow aerobic bacteria to thrive.
The soft black mud in profile UCRK (Figure 7.5) contained approximately
2% organic C with little variation with depth except at the very bottom of
the profile (Figure 7.1 la). This mud is interpreted to be a relatively recent
deposit brought in by tides which still regularly fill the creeks. Levels of
total Fe in the black mud were similar to those in the nearby trench
section, but the proportions of dithionite-extractable and oxalate-extractable
Fe were slightly higher and showed a higher degree of variation with
depth, suggesting greater mobility under more pronounced anaerobic
conditions (Figure 7. lib). Concentrations of reduced sulphur species were
much higher in this profile than in the trench profile (Figure 7.11c),
indicating significant sulphate reduction. Microbial assays were not
undertaken on these samples, however, and the activity of sulphate-reducing
bacteria has not been confirmed by lipid biomarker techniques. Based on
the evidence available, it seems likely that the high levels of bacterial
sulphate reduction are related chiefly to the availability of easily
metabolisable organic matter in the recently deposited mud.

7.5 Microbial activity in the Lower Marsh

Two pits were dug on the Lower Marsh in October 1993 in order to allow
sampling of sediments and concretions for comparison with the data from
 OC% RS%

0 1 2 0.0 0.1 0.2 0.3 0.4 0.5

0
Depth marsh land surface
(cm)
10- 10-

20- 20-

30- creek base surface 30-

40"

50-

60-

70-

80-

90
Total-Fe%
Dith-Fe%
Ox-Fe%

(a) (b) (c)

Figure 7.11 Depth variation within the Upper Marsh Creek section (Profile UCRK): (a) % organic carbon; {b) % reactive iron and total iron,
and (c) % reduced sulphur species.
138 K. Pye, M.L. Coleman and W.M. Duan

the Upper Marsh. Site 93LD was located in the mid part of the Lower
Marsh to the east of the Cocklestrand footpath (Figure 7.1), approximately
15 m away from the nearest creek. Site 93CB was located 100 m further
seaward on the bank of one of the larger creeks. During September 1994
two further trenches were dug at a site (Site 9409) approximately 100 m to
the east of Site 93LD. A further shallow pit was also excavated at this time
in the area of eroded marsh which forms the boundary between the Lower
Marsh and the intertidal sand flat (Site 9409a, Figure 7.1).
At Site 93LD the top 45 cm of sediment consisted of mottled,
brownish-grey mud, containing abundant roots in the top 25-30 cm
(Figure 7.12). A narrow transition zone composed of sandy mud between
45 and 48 cm was underlain by grey-brown and grey sand to a depth of
80 cm. Sands below the water table at 80 cm were stained black by iron
monosulphides. A concretion containing a corroded metal nucleus was
recovered at a depth of 85 cm. The upper 52 cm of sediment in section
93CB consisted of greyish-brown mud overlying brown and greyish-brown
sand and, beneath the water table, black monosulphide-stained sand and
grey sand (Figure 7.12). A shrapnel-containing concretion was also
recovered from this profile, just below the water table. Sub-samples of the
sediment and concretions were collected for analysis in the same manner
as samples from the Upper Marsh. However, an additional Fe extraction
stage using cold 0.5N HC1, prior to successive oxalate and dithionite
extractions, was performed on the Lower Marsh samples.
The organic carbon content of the muddy sediments in the upper 45 cm
at Site 93LD attained a maximum of 3.34% near the surface, decreasing
with depth (Figure 7.13a). Total viable microbial biomass decreased with
depth at a slightly sharper rate than the decline in organic carbon (Figure
7.13&). The greatest fall in biomass occurred at the boundary between the
upper muddy and lower sandy horizons. A small increase occurred at
85 cm depth in the reduced sediments around the concretion. Levels of
total extractable Fe (i.e. combined HC1 + oxalate + dithionite-extractable
Fe) in the upper muddy sediments were found to be similar to those found
in the Upper Marsh (c. 2%). Levels were significantly lower in the
underlying brownish muddy sands ( < 0.5%) and increased in the
underlying dark grey sands before dropping again in the black sands near
the groundwater table (Figure 7.13c). Since grain size and mineralogical
analyses provided no evidence of significant differences in the detrital
composition of the brown, grey and black sands, the observed variations
in extractable Fe content are inferred to reflect diagenetic remobilisation.
Levels of total reduced sulphur species were relatively low but measurable
 Site 93CB Site 93LD

Marsh Surface (40 cm) Marsh Surface

cm 93LD1:0-3cm Grey mud, many roots

93LD2.3-6 cm Grey mud, many roots
10-
93LD3:6-9 cm Grey mud, many roots
52 cm
20 93LD4:12-15 cm Grey mud, some roots

30- 93LD5:20-23 cm Brown-grey mud, with roots

93LD6:30-33 cm B-G mud, scattered roots

\*y Creek Bank (5 m wide) 40 -
93LD7:45-48 cm Grey-brown muddy sand
LCB 1:52-55 cm Clean brown sand 50
LCB2:60-63 cm Grey-brown sand 93LD8:52-55 cm Dark grey sand
60-
LCB3:67-70 cm Brown-grey (B-G) sand 93LD9:62-65 cm Dark grey sand
70
LCB4:74-77 cm Black sand 93LD10:72-75 cm Dark grey sand

80 93LD11:83-87 cm Black sand

LCB5:85-90 cm B-G sand, above concretion Z
Concretion occured at depth of 81-90 cm
LCB6:90-100 cm B-G sand, adjacent to concretion 93LD12:85 cm Black sand,
(LCB9:97 cm Black sand, surrounding concretion) 90
closely surrounding concretion.

LCB7:100-103 cm B-G sand, below concretion

LEGEND: !*>>>;] Greyish mud, rich in roots Greyish sand Concretion

I I Brownish sand Black sand Ground water table

Figure 7.12 Diagram showing the sediment profiles at Site 93CB and Site 93LD on the Lower Marsh.
 OC ( Total PLFA (pmol/g) Fe% TRS%

2 10000 20000 30000 0.0 0.1 0.2

0
Depth
(cm)

20- 20" 20-

40- 40" 40"

60- 60" 60"

80" 80" Dithionite 80"

Oxalate
• 0.5NHC1

100 J 100 J 100 J

(a) (c) (d)

Figure 7.13 Depth variation in geochemical parameters and indicators of microbial activity at site 93LD: (a) % organic carbon; {b) viable
microbial biomass, indicated by total PLFA, (c) % total reactive iron determined by sequential extraction, (d) % total reduced sulphur.
 Microbial activity and diagenesis in saltmarshes 141

in the upper part of the profile, showing a progressive reduction with

depth and a sudden increase in the black sands adjacent to the concretion
(Figure 7.13d).
The bioassay results showed that, in general, the proportion of
terminally branched saturates increased slightly, and that of monoenoics
decreased, with depth in the profile (Figure 7.14a and b). The OHFA/PLFA
ratio increased with depth in the muddy part of the profile, suggesting
more significant gram negative populations (Figure 7.14c). Levels of
OHFA fell again in the upper, brownish-grey, sandy sediments and
showed a small secondary peak in the zone of black sands. The
biomarkers for Desulfobacter (10Mel6:0) and Desulfovibrio (il7:lw7c)
showed the former in greater relative abundance throughout the profile
(Figure 7.14d and e), as was observed on the Upper Marsh. Desulfobacter
are apparently relatively most abundant in the upper muddy sediments
and the grey sands, while Desulfovibrio become relatively more significant
in the black sands around the concretion. An increase in the ratio of
cyl9:0/18: Iw7c with increasing depth was also observed, suggesting an
increase in nutritional stress of the microbial community (Figure 7.14/).
This may indicate the presence of an aged microbial population and/or an
increase in anaerobic respiration. A slight reduction in apparent stress is
indicated in the mid part of profile 93LD, suggesting increased availability
of nutrients and/or oxygen levels within the uppermost sandy sediments.
Stress levels then increase again in the underlying grey sands which are
low in reactive Fe and organic C, before decreasing again in the black,
reactive Fe and reduced sulphur-rich sands close to the water table.
In profile 93CB, organic carbon content increased dramatically with
depth below the surface of the creek bank (Figure 7.15a). Total microbial
biomass showed a reverse trend, however (Figure 7.15fc). Levels of total
extractable Fe were lower than in profile 93LD and were relatively
constant with depth (Figure 7.15c). The abundance of total reduced
sulphur species increased dramatically with depth, attaining a maximum
value in the black sands surrounding the concretion (Figure 7A5d).
Indicators of community structure show a slight decrease in the abundance
of monoenoics and an increase in terminally branched saturates with
depth (Figure 7.16a andfc),as also seen in profile 93LD. The OHFA/PLFA
ratio showed a slight increase in the lower part of the black sand zone
(Figure 7.16c), suggesting relatively more important gram negative
populations. The biomarker for Desulfobacter showed a progressive
increase in abundance with depth, but that for Desulfovibrio peaked in the
black sands above the concretion (Figure 7A6d and e).
 Mono. Mol % Bra. Satur., mol% OHFA/PLFA 10Mel6:0,mol% il7:lw7c, mol % Cy 19:0/18:lw7c
20 30 40 50 10 20 30 40 0.00 0.02 0.04 0.06 0 2 4 6 8 10 0 1 2 3 4 5 0 1 2

20- 20" 20" 20- 20"

40" 40" 40" 40" 40"

60- 60" 60" 60" 60"

80" 80- 80" 80- 80-

100-1 100-1 100 J 100 J 100 J

100 J (b) id) (e)

Figure 7.14 Indicators of community structure and environmental stress in profile 93LD indicated by (a) abundance of monoenoics, {b)
abundance of branched saturates, (c) the ratio of OHFA to PLFA, (of) the biomarker for Desulfobacter (10Me16:0), (e) the biomarker for
Desulfovibrio (i17:1w7c), and (/) the ratio of cy 19:0/18:1w7c.
 PLFA (pmol/g) Fe% TRS%
10000 20000 QQ Q2 o.4 0.6 0.8 1.0 0.02 0.06 0.10 0.14
o- 0

10

20
20- M Dith,
g | Ox-Fe%
• 0.5N-HC1 30
30-

40'
40"

50'
50-

60"
60"

70-
70"

80"
80

90-
90

100-
100

no-1 110-
110 (c) (d)
Figure 7,15 Depth variation in geochemical parameters and indicators of microbial activity in profile 93CB: (a) % organic carbon, {b) viable
microbial biomass, indicated by total PLFA, (c) % reactive Fe, and (d) % total reduced sulphur.
 Mono. Mol % TB. Satur., mol% OHFA/PLFA 10Mel6:0, mol % il7:lw7c, mol % Cy 19:0/18:lw7c
20 30 40 0 10 20 30 o.OO 0.01 0.02 0.03 0 1 2 3 4 5 0.2 0.4 0.6 0.8 1.0 0.0 0.1 0.2 0.3
0 0 0
Depth
(cm)
10 10- 10" 10- 10- 10-

20 20" 20" 20" 20" 20-

30" 30- 30" 30- 30" 30-

40" 40- 40" 40" 40- 40-

creek base creek base creek base creek base creek base
50" 50- 50" creek base 50- 50" 50-

60- 60" 60" 60" 60-

60-

70-
70- 70- 70" 70" 70-

80" 80" 80- 80" 80-

80"

90- 90- 90" 90" 90-

90"

100- 100" 100" 100- 100-

100-

110" 110" 110" 110" 110-

(a) (b) (c) id) (e)

Figure 7.16 Indicators of community structure and environmental stress in profile 93CB: (a) abundance of monoenoics, {b) abundance of
branched saturates, (c) OHFA/PLFA ratio, (d) the biomarker for Desulfobacter (10Me16:0), (e) the biomarker for Desulfovibrio 017:^70), and
{D the ratio of cy 19:0/18: Iw7c.
 Microbial activity and diagenesis in saltmarshes 145

These results suggest that anaerobic microbial activity, including

sulphate reduction, typically increases with depth in the Lower Marsh
muddy sediments but remains of relatively limited importance compared
with aerobic microbial processes. Anaerobic activity decreases in the
relatively well-drained brown and greyish-brown sands which represent
the former intertidal flat, but increases again in the black and grey sands
which lie below and immediately above the water table. In the upper
muddy sediments the vertical variation in degree of anaerobic activity is
probably controlled by the balance between organic carbon content and
oxygen availability, both of which decline with depth. In the underlying
sandy sediments sea water saturation maintains anaerobic conditions
below the water table, but rates of anaerobic microbial decomposition are
limited mainly by the availability of suitable organic carbon sources.
In the creek bank sediments, profile 93CB, levels of organic carbon are
much lower than in the saltmarsh near-surface sediments, profile 93LD,
but levels of reduced sulphur species are significantly higher while total
viable microbial biomass is of the same approximate magnitude. This
suggests that the form in which the organic carbon is present is of major
significance in controlling the rate and nature of microbial activity. Algae,
diatoms and other marine species which live in the near-surface creek
sediments, and which are regularly introduced to the system by tidal
flushing, are probably of great significance in this respect. Analyses of
pore-water samples have indicated that levels of sulphate are high
throughout the Warham marsh system, and sulphate availability is
therefore unlikely to be a limiting factor in sulphate reduction. However,
there is clearly a greater opportunity for maintenance and replenishment
of the dissolved sulphate source within the near-surface creek-bank
sediments and below the saline groundwater table.
In comparing the biomarker signatures of the Upper Marsh and Lower
Marsh, the most distinguishing feature is the different rate of decline of
total viable microbial biomass with depth in the upper part of the profiles.
In the Upper Marsh an order of magnitude decrease in the viable biomass
occurred in the upper 20 cm while in the Lower Marsh profile a similar
order of magnitude decrease occurred over a vertical distance of c. 40 cm.
This difference reflects the varying rates of sediment accretion on the two
marshes.
Analysis of samples taken from different parts of the carbonate
concretions, and from the adjacent host sediment at site 93LD, showed
that organic carbon content is higher within the concretions than in the
surrounding sediments (Figure 7.17a). Little difference was found between
146 K. Pye, M.L. Coleman and W.M. Duan

20-

0.0
host sedi-far

40
Fe%
• 0.5NHC1
30- 0 Oxalate
H Dithionite

20-

10-

0
Fe core outer host
ib)

Fe core outer host mud

(c)

Figure 7.17 Variation in geochemical parameters and indicators of

microbial activity within different parts of the concretion and host sediment
from site 93LD: (a) % organic carbon and viable microbial biomass, {b)
different forms of extractable iron, (c) % acid volatile sulphide (AVS) and %
total reduced sulphur (TRS).
 Microbial activity and diagenesis in saltmarshes 147

the inner and outer portions of the concretion collected from profile
93CB, but no organic carbon was determined in the iron-rich core.
Microbial biomass was found to attain a maximum in the sediment
immediately surrounding the concretion, but was also high in the outer
part of the concretion. Levels of extractable iron were found to decline
systematically from the core of the concretion towards the rim (Figure
7.176). Total reduced sulphur levels were found to be high in both inner
and outer parts of the concretion compared with the host sediment
(Figure 7.17c). In terms of community structure, monoenoics were again
found to be the dominant group in all parts of the concretion and in the
surrounding sediment, with the relative abundance of this group achieving
a maximum in the outer part of the concretion (Figure 7.18a). The
indicators of nutritional stress showed maximum stress conditions in the
inner part of the concretion, decreasing to a minimum in the surrounding
sediment (Figure 7.18b). The biomarkers for Desulfobacter and Desulfovibrio
suggested that the former are relatively more abundant within the
concretion and in the sediment immediately adjacent to it, whereas the
latter are more abundant in the sediment some distance away at the same
stratigraphic level (Figure 7.18c). Within the concretion the relative
abundance of Desulfobacter was found to increase from the core towards
the rim, whereas that of Desulfovibrio declined. The observed dominance
of Desulfovibrio in the iron-rich core of the concretion is consistent with
the finding that these bacteria are capable of reducing ferrous iron when
sulphate supply is limited (Coleman et al, 1993).

7.6 Conclusions
Lipid biomarker techniques have proved to be a powerful tool which, in
combination with more conventional geochemical and mineralogical
analyses, has shed new light on the diagenetic reactions taking place
within the north Norfolk saltmarshes. The results obtained from the most
recent sampling on the Lower Marsh and intertidal flats have still to be
fully interpreted, but it is clear that there are major differences between the
diagenetic regimes of the Upper Marsh and the Lower Marsh which
reflect their contrasting accretion histories and sedimentological charac-
teristics.

Acknowledgements
We thank Nigel Pontee, Stephen Crooks and Tim Steel of Reading
University for assistance with field sampling, and David White, David
148 K. Pye, M.L. Coleman and W.M. Duan

TTBS
MBS
Poly
Total Mono
• BMono

• Cy 17/16: lw
Cy 19/18: lw
16:lw,t/c
18:lw,t/c

Fe core outer host sedi-far

mol%

4- 10Mel6:0
il7:lw7c

3-

2-

1-

(c)
Fe core outer host far-away
Figure 7.18 Variation in microbial community structure and indicators of
environmental stress within different parts of the concretion and host
sediment from profile 93LD: (a) community structure-TBS = terminally
branched saturates, MBS = mid-branched saturates; Poly = polyenoics, Total
Mono = total monoenoics, BMono = branched monoenoics. Sedi-far
represents the host sediment at some distance from the concretion but at
the same stratigraphic level; {b) indicators of nutritional stress; (c)
biomarkers for Desulfobacter (10Me16:0) and Desulfovibrio (i17:1w7c).
 Microbial activity and diagenesis in saltmarshes 149

Hedrick and David Ringelberg of CEB, Knoxville, Tennessee, for assistance

with the bioassays. Fieldwork on the Warham marshes is undertaken by
permission of English Nature. The work was supported by NERC Grants
GR3/5476 and GR3/5476A. This paper represents University of Reading
PRIS Contribution No. 414.

References
Al-Agha, M.R., Burley, S.D., Curtis, C D . & Esson, J. (1995) Complex cementation textures
and authigenic mineral assemblages in recent concretions from the Lincolnshire
Wash (east coast, UK) driven by Fe(0) to Fe(II) oxidation. Journal of the
Geological Society, London, 152, 157-71.
Allison, P.A. & Pye, K. (1994) Early diagenetic mineralization and fossil preservation in
modern carbonate concretions. Palaios 9, 561-75.
Andrews, J.E. (1988) Methane-related Mg-calcite cement in recent tidal flat sediments
from the Firth of Forth. Scottish Journal of Geology 24, 233^4.
Baird, B.H., Nivens, D.E., Parker, J.H. & White, D.C. (1985) The biomass, community
structure, and spatial distribution of the sedimentary macrobiota from a high
energy area of the deep sea. Deep Sea Research 32, 1089-99.
Berner, R.A. & Westrich, J.T. (1985) Bioturbation and the early diagenesis of carbon and
sulfur. American Journal of Science 285, 193-206.
Blackburn, T.H. & Blackburn, N.D. (1993) Rates of microbial processes in sediments.
Philosophical Transactions of the Royal Society of London A, 344, 49-58.
Bligh, E.G. & Dyer, W.J. (1959) A rapid method of total lipid extraction and purification.
Canadian Journal of Biochemistry and Physiology 37, 911-17.
Canfield, D.E. (1989) Reactive iron in marine sediments. Geochemica Cosmochimica Acta
53, 619-32.
Canfield, D.E., Raiswell, R., Westrich, J.T., Reaves, C M . & Berner, R.A. (1986) The use of
chromium reduction in the analysis of reduced inorganic sulfur in sediments and
shales. Chemical Geology 16, 59-62.
Coleman, M.L., Hedrick, D.B., Lovley, D.R., White, D.C. & Pye, K. (1993) Reduction of
Fe(III) in sediments by sulphate reducing bacteria. Nature 361, 436-8.
Davison, W., Grimes, G.W., Morgan, J.A.W. & Clarke, K. (1991) Distribution of dissolved
iron in sediment pore waters at submillimetre resolution. Nature 252, 323-5.
Duck, R.W. (1995) Subaqueous shrinkage cracks and early sediment fabrics preserved in
Pleistocene calcareous concretions. Journal of the Geological Society, London,
152, 151-6.
Eglinton, G., Parkes, R.J. & Zhao, M. (1993) Lipid biomarkers in biogeochemistry: future
roles? Marine Geology 113, 141-5.
Findlay, R.H., Trexler, M.B., Guckert, J.B. & White, D.C. (1990) Laboratory study of
disturbance in marine sediments: response of a microbial community. Marine
Ecological Progress Series 62, 121-33.
Funnell, B.M. & Pearson, I. (1989) Holocene sedimentation on the north Norfolk barrier
coast in relation to relative sea level change. Journal of Quaternary Science 4,25-36.
Gaudette, H.E., Flight, W.R., Toner, L. & Folger, D. (1974) An inexpensive titration
method for the determination of organic carbon in recent sediments. Journal of
Sedimentary Petrology 44, 249-53.
Guckert, J.B., Antworth, C.P., Nichols, P.D. & White, D.C. (1985) Phospholipid,
ester-linked fatty acid profiles as reproducible assay for changes in prokaryotic
150 K. Pye, M.L. Coleman and W.M. Duan

community structure of estuarine sediments. FEMS Microbiology Ecology 31,

147-58.
Hedrick, D.B., Guckert, J.B. & White, D.C. (1991) Archaebacterial ether lipid diversity:
analysis by supercritical fluid chromatography. Journal of Lipid Research 32,
656-66.
Jorgensen, B.B. (1978) A comparison of methods for the quantification of bacterial sulfate
reduction in coastal marine sediment. II. Estimation from chemical and
bacteriological field data. Geomicrobiology 1, 45-64.
Jorgensen, B.B. & Bak, F. (1991) Pathways and microbiology of thiosulfate transformations
and sulfate reduction in a marine sediment (Kattegat, Denmark). Applied and
Environmental Microbiology 57, 847-56.
Lovley, D.R. (1991) Dissimilatory Fe(III) and Mn(IV) reduction. Microbiology Reviews 55,
259-87.
McKeague, J.A. & Day, J.H. (1966) Dithionite and oxalate extractable Fe and Al as aids in
differentiating various classes of soils. Canadian Journal ofSoil Science 46,13-22.
Mehra, O.P. & Jackson, M.L. (1960) Iron oxide removal from soils and clays by a
dithionite-citrate system buffered with sodium bicarbonate. Clays and Clay
Minerals 7, 317-27.
Parker, J.H., Fredrickson, H.L., Vestal, J.R. & White, D.C. (1982) Sensitive assay, based on
hydroxy-fatty acids lipopolysaccharide lipid A for gram negative bacteria in
sediments. Applied and Environmental Microbiology 44, 1170-7.
Parkes, R.J., Cragg, B.A., Getliff, J.M., Harvey, S.M, Fry, J.C., Lewis, C.A. & Rowland, SJ.
(1993a) A quantitative study of microbial decomposition of biopolymers in
recent sediments from the Peru Margin. Marine Geology 113, 55-66.
Parkes, R.J., Dowling, N.J.E., White, D.C, Herbert, R.A. & Gibson, G.R. (1993b)
Characterization of sulphate-reducing bacterial populations within marine and
estuarine sediments with different rates of sulphate reduction. FEMS Microbiology
Ecology 102, 235-50.
Purchas, A.W. (1965) Some History of Wells-next-the-Sea and District. East Anglian
Magazine Ltd, Ipswich, 140 pp.
Pye, K. (1981) Marshrock formed by iron sulphide and siderite cementation. Nature 294,
650-2.
Pye, K. (1984) SEM analysis of siderite cements in intertidal marsh sediments, Norfolk,
England. Marine Geology 56, 1-12.
Pye, K. (1988) An occurrence akaganeite (p-FeO-OH.Cl) in recent oxidized carbonate
concretions, Norfolk, England, Mineralogical Magazine 52, 125-6.
Pye, K. (1992) Saltmarshes on the barrier coastline of north Norfolk, eastern England. In
Saltmarshes: Morphodynamics, Conservation and Engineering Significance, ed.
J.R.L. Allen & K. Pye. Cambridge University Press, Cambridge, pp. 148-78.
Pye, K., Dickson, J.D., Schiavon, K , Coleman, M.L. & Cox, M. (1990) Formation of
siderite-Mg calcite-iron sulphide concretions in intertidal marsh and sandflat
sediments, north Norfolk, England. Sedimentology 37, 325-43.
Sorenson, J. (1987) Nitrate reduction in marine sediment: pathways and interaction with
iron and sulfur cycling. Geomicrobiology Journal 5, 401-21.
Tunlid, A. & White, D.C. (1992) Biochemical analysis of biomass, community structure,
nutritional status, and metabolic activity of the microbial community in soil. In
Soil Biochemistry, Volume 7, eds. G. Stotzky & J.M. Bollag, Marcel Dekker, New
York, pp. 229-62.
Vernberg, F.J. (1993) Salt-marsh processes: a review. Environmental Toxicology and
Chemistry 12, 2167-95.
 Microbial activity and diagenesis in saltmarshes 151

White, D.C. (1993) In situ measurement of microbial biomass, community structure and
nutritional status. Philosophical Transactions of the Royal Society of London A,
344, 59-67.
White, D.C, Davis, W., Nickels, J., King, J.D. & Bobbie, R.J. (1979) Determination of the
sedimentary microbial biomass by extractable lipid phosphate. Oceanologia 40,
51-62.
White, D.C, Smith, G.A., Gehron, M.J., Parker, J.H., Findlay, R.H, Martz, R.F. &
Fredrickson, H.L. (1983) The ground water aquifer microbiota: biomass,
community structure and nutritional status. Developments in Industrial Microbi-
ology 24, 201-11.
 8
o o o o o o o o o o o o o o o o o o o
The behaviour of radionuclides in the
coastal and estuarine environments of
the Irish Sea
P. McDonald and S.R. Jones

8.1 Introduction
The focus of this chapter is on the behaviour of radionuclides in the
coastal and intertidal sediments of the Irish Sea, in particular those in the
north-east. A review of previous research concerned with the fate of
radionuclides in and around the Irish Sea is presented along with some
recent research the authors have performed.

8.1.1 Boundaries of the Irish Sea

The Irish Sea is a semi-enclosed body of water, bounded by the eastern
coast of Ireland and the coasts of north Wales, north-west England and
south-west Scotland. It connects with the Atlantic Ocean through St
George's Channel in the south and the North Channel to the north. Water
enters the Irish Sea both through St George's Channel and the North
Channel, although the former dominates, and leaves via the North Channel
(Howorth, 1984; Dickson, 1987). The flowrate of water out through the
North Channel is about 5 x 10 9 m 3 d~ 1 and the residence time of the water
in the northern Irish Sea is about one year (Pentreath et a/., 1984).
In relation to radionuclide behaviour, the distribution of bed sediments
according to grain size is of particular importance. Figure 8.1 shows the
disposition of bed sediments according to the grain size description of
Pantin (1977, 1978). The two main areas of muddy sediments, one
extending southwards from St Bees Head, Cumbria, towards Morecambe
Bay and the other lying between the Isle of Man and the north-east coast
of Ireland, are of particular significance.
The dynamic behaviour of bed sediments has been the subject of much
research and debate. Some authors (Mauchline, 1980; Pantin, 1991)
consider that both the main areas of muddy sediment are actively
 Radionuclides in the Irish Sea 153

accreting whilst others (Kirby et a/., 1983; Kershaw & Young, 1988)
consider that the net accumulation rate is zero or very low (about 0.1 mm
y" 1 ). Despite this uncertainty, the sediment is by no means static.
Extensive vertical sediment mixing occurs through the action of biota,

55°

54C

53C

52C

Figure 8.1 Distribution of bed sediments by grain size (Pantin, 1977,

1978).
154 P. McDonald and S.R. Jones

principally the Echurian worm Maxmulleria lankesteri and the Thalassinid

shrimp Callinassa subterranea (Kershaw et a/., 1984; Swift & Kershaw,
1986). Sediment resuspension arises from localised erosion, bioturbation
and human activities such as trawling, and is balanced on average by
deposition.

8.1.2 Radionuclide inputs and total discharge inventories

The Irish Sea receives inputs of radionuclides, either directly or via
discharge to rivers, from nuclear installations including the nuclear fuel
reprocessing plant at Sellafield in Cumbria, the Magnox power stations at
Trawsfynydd and Wylfa in Wales and at Chapelcross in south-west
Scotland, the uranium enrichment plant at Capenhurst, near Chester and
the uranium fuel fabrication plant at Springfields, near Preston (Ministry
of Agriculture, Fisheries and Food, 1972-1993). In addition, the phosphoric
acid production plant at Marchon, near Whitehaven in Cumbria, has
been responsible for significant inputs of uranium series radionuclides
(McCartney, Kershaw & Allington, 1990).
Generally, inputs from Sellafield have dominated the inventories of
fission products and actinides in the Irish Sea; the inventory of uranium
series radionuclides is dominated by naturally occurring material with the
emissions from Marchon and Springfields dominating the anthropogenic
input and enhancements over natural levels being localised to areas close
to the discharges (Hamilton, 1980; McCartney et al, 1990). This chapter
concentrates on the long lived fission product 137 Cs and the actinides
238
Pu, 2 3 9 ' 2 4 0 p u ? 2 4 1 Pu and 241 Am since they have been discharged into
the Irish Sea from Sellafield in quantities large by comparison with
weapons testing fallout and deposition from the Chernobyl accident, and
their long half-lives create the possibility of accumulation and/or a
retained inventory from past discharges. Time series environmental
monitoring data and sediment deposition chronologies obtained from
core samples in Maryport harbour, West Cumbria (Kershaw et al, 1990)
are consistent with the Sellafield discharge chronologies (Gray et a/.,
1995). The annual discharges of 137 Cs, 2 4 1 Pu, Pu alpha (i.e. 2 3 8 Pu +
239,24oPu) a n d 2 4 i A m f r o m Sellafield are shown in Figure 8.2. The
dramatic reductions in discharges since the 1970s reflect improvement in
effluent treatment at the site (British Nuclear Fuels pic, 1971-1993).
The cumulative discharges as at 1992, allowing for radioactive decay
since the date of discharge and for the ingrowth of 2 4 1 Am from discharged
241
Pu, are shown in Figures 8.3 and 8.4. These cumulative discharges
represent the current global inventories resulting from the discharges; the
 Radionuclides in the Irish Sea 155

actual inventories in the Irish Sea are lower, reflecting losses from the Irish
Sea due to transport processes as later discussed.

8.2 Radionuclide behaviour: sediment-water interactions,

sediment behaviour
The transport of radionuclides within the Irish Sea and further afield, and
the development of the total inventory of the Irish Sea as a function of
time, are determined by interactions of the radionuclides with the water

6000 Beta gamma emitters

137
Cs discharge
5000 241
Pu discharge
90
Sr discharge

1950 1955 1960 1965 1970 1975 1980 1985 1990 1995
Year

Alpha emitters

s
JS

1950 1955 1960 1965 1970 1975 1980 1985 1990 1995
Year
Figure 8.2 Discharges to the Irish Sea from Sellafield (British Nuclear
Fuels pic, 1971-1993).
 156 P. McDonald and S.R. Jones

and sediment phases and by the transport processes affecting sea water
and sediments. Caesium exhibits straightforward chemical behaviour as a
metal from Group I, existing only in a single oxidation state. The
environmental chemistry of the actinides is more complex as they can
exist in a number of different oxidation states. Thus, plutonium can be
present as Pu(III), Pu(IV), Pu(V) or Pu(VI). 2 3 9 ' 2 4 0 p u present in filtered
sea water from the Irish Sea contains a mixture of these oxidation states

137
40000 i Csand 241 Pu

137
Cs cumulative discharge
137
Cs inventory
30000 " 241
Pu discharge
cr
PQ 24l
?u inventory

© 20000 "

10000 "

1950 1955 1960 1965 1970 1975 1980 1985 1990 1995
Year
137 241
Figure 8.3 Cs and Pu inventories in the Irish Sea.

238
Pu cumulative discharge
238
Pu inventory
1000 -
239/240 p u cumulative discharge
900 " 239/240 p u inventory
24
800 " 'Am cumulative discharge
a* 241
Am inventory
700 "

600 -

500 "

400 "
'S 300 -
200 -
100 "

1950 1955 1960 1965 1970 1975 1980 1985 1990 1995

Year

Figure 8.4 Actinide inventories in the Irish Sea.

 Radionuclides in the Irish Sea 157

(Nelson & Lovett, 1978; Pentreath, Harvey & Lovett, 1985) with more
than 90% in the oxidised forms (V + VI) and Pu(V) probably predominating
(Nelson & Orlandini, 1979). Less is known about the environmental
chemistry of 2 4 1 Am but it appears to exist almost entirely as Am(III).
The most important aspect of radionuclide behaviour affecting transport
and subsequent distribution is their propensity to adsorb onto sediments.
This is conventionally characterised by reference to an equilibrium
distribution coefficient, KD, between sediment and water phases. Typical
values for this coefficient are shown in Table 8.1 (Pentreath et al, 1984;
International Atomic Energy Agency (IAEA), 1985; Kershaw et a/., 1986;
Howorth & Kirby, 1988). The coefficients given in Table 8.1 are the ratio
of total radionuclide concentration in the sediment phase to that in the
water phase, and may mask differing behaviour according to oxidation
state. Thus, KD values for Pu(III + IV) are of the order 106 whereas those
for Pu(V + VI) are less than 104 (Nelson & Lovett, 1978). Accordingly,
different oxidation states of Pu predominate in water and sediment phases
and the apparent KD values may vary with distance from the discharge
point due to preferential scavenging of the reduced forms by sedimentation
(Sholkovitz, 1983; Pentreath et a/., 1985).
Caesium is said to behave 'conservatively'; that is, the bulk of the
radionuclide inventory is associated with the water phase and so
transport processes are dominated by the bulk movement of sea water.
Plutonium and americium, on the other hand, behave 'non-conservatively';
the bulk of their inventory is associated with sediments and the transport
processes affecting sediments are very important to their behaviour. The
proportion of each nuclide present in the water column as suspended
particulate is a simple function of KD value and suspended sediment load
(Sholkovitz, 1983), as indicated in Table 8.2. Thus, for the full range of
sediment loadings, water column inventories of 137 Cs are dominated by

Table 8.1 Sediment/water distribution coefficients

(Pentreath et ai, 1984; IAEA, 1985; Kershaw et ai,
1986; Howorth & Kirby, 1988)

Radionuclide Distribution coefficient (KD)

137
Cs 103
239,240p u
3 x 105
Pu{III,IV} 106
Pu{V,VI} 104
241
Am 2 x 106
158 P. McDonald and S.R. Jones

material in solution whereas those for 241 Am are dominated by material

adsorbed to suspended sediments; water column inventories of 239 ' 240 Pu
are also dominated by suspended sediments at loadings above about 4 mg
r1.
The adsorption of radionuclides to sediments is strongly affected by
particle size and mineralogy (Aston, Assinder & Kelly, 1985; Ramsay &
Raw, 1987; Livens & Baxter, 1988a). In particular, fine grained material
accumulates higher concentrations of all radionuclides and the presence
of clay minerals also enhances adsorption.
The net effect of the above processes is that the bulk of the Cs inventory
is transported with water movements with a small fraction being adsorbed
to suspended particulate and deposited in bed sediments, whereas a large
fraction of the Pu and Am inventory is adsorbed to suspended sediments
and deposited in the bed sediments of the eastern Irish Sea. The behaviour
of Pu and Am deposited in bed sediments is then determined by the
processes of sediment mixing (in which bioturbation is very important),
resuspension, and remobilisation of adsorbed radionuclides into the
solution phase.

8.2.1 Intertidal areas: beaches, harbours, estuaries, saltmarshes

Because of the strong inverse association between sediment grain size and
radionuclide concentrations, it is the areas in which finer sediments
accumulate that are of greatest significance. The disposition of coastal
sediment types is shown in Figure 8.5. Exposed beaches exhibit coarser
sediments, typically sand or shingle, because of the energy input from
wave action. The radionuclide concentrations in these areas are generally
low; however, because wave action mixes radionuclides to a substantial
depth and because they represent a large proportion of the intertidal area,

Table 8 . 2 Fraction of total water column inventory carried by

suspended sediment (Sholkovitz, 1983)

Fraction present as suspended sediment at stated

sediment loading
Nuclide *D lmgl"1 lOmgT1 lOOmgP 1
137
Cs 1 X 103 0.001 0.01 0.09
239,240p u
3 x 105 0.23 0.75 0.97
241
Am 2 x 106 0.67 0.95 0.99
 Radionuclides in the Irish Sea 159

they contain a large proportion of the total inventory of radionuclides in

the Irish Sea (Eakins et a/, 1988,1990; Garland et al, 1989; Carpenter et
aU 1991).
Estuaries provide areas of lower energy where the finer grained
sediments (silts and muds) can deposit. The estuary at Ravenglass, which
is some 9 km from Sellafield and formed at the confluence of the rivers Irt,
Mite and Esk, has been extensively studied by many researchers (e.g.
Hetherington, 1976; Aston & Stanners, 1982a,b,c; Hamilton & Clarke,
1984; Aston et a/., 1985; Assinder, Kelly & Aston, 1985; Burton, 1986).
Measurements have also been reported in estuaries from south-west
Scotland to north Wales (Aston et a/., 1981; MacKenzie & Scott, 1982,
1993; McDonald et al.9 1992; Jones et a/, 1994).
Estuaries are complex and dynamic environments, with sediments
subject to both erosion and deposition at different times and to varying
water chemistry, particularly in salinity, pH and suspended sediment
load. These all influence the behaviour of radionuclides (Clifton &
Hamilton, 1982; Kelly et a/., 1991). A single estuary is likely to contain
sedimentary deposits ranging from coarse sands to fine muds with a
correspondingly large range of radionuclide concentrations. Most actively

Figure 8.5 Coastal sediments types.

160 P. McDonald and S.R. Jones

accreting sediment deposits also show a variation in radionuclide content

with depth which can be correlated with the chronology of discharges
from Sellafield and used to yield estimates of sedimentation rate (e.g.
Aston & Stanners, 1979). However, the interpretation of such profiles is
complicated by the fact that sedimentation rates at a particular site are
unlikely to be constant with time, and may indeed vary from net erosion to
net deposition (Clifton & Hamilton, 1982; MacKenzie & Scott, 1993).
Most of the estuarine conditions discussed above have been derived
through study of radionuclides in the estuary at Ravenglass. However,
because of the attention given to this estuarine area regarding radionuclide
behaviour in an intertidal environment it is appropriate to provide more
details of such research here.
Hamilton & Clarke (1984), in describing the recent sedimentation
history at the Esk Estuary, found that the central regions of the estuary
sediments were being deposited at a rate of 3 to 4 cm y ~ *; along the banks
lower rates of between 1 and 2cm y " 1 were found, while the saltmarsh
proper was probably emerging at a rate of- 0.5 c m y " 1 . These differences
of sedimentation rate were directly related to grain size of sediments and
biological activity, the higher rates being associated with the deposit of
coarse sands and the lower rates being associated with deposition offinesilts.
Aston & Stanners (1981,1982b), when investigating the distribution of
radionuclides in intertidal, saltmarsh and tidal inlet samples, concluded
that grain size distributions within local areas seemed to control radionuclide
activity concentrations. Also observed was the high variability of
radionuclide concentration within small sampling areas. Plutonium
studies revealed that deposition of particulate material was the principal
control of its input to the estuary and that it was geochemically associated
with the non-detrital iron/manganese phase in the oxic sediment layer.
Also 2 3 8 p u / 2 3 9 ' 2 4 0 p u activity ratios indicated that post-depositional
migration of Pu at that time was negligible.
Assinder et al (1985) reported that the behaviour of radionuclides in the
Esk Estuary was determined by its physical characteristics and by its
location close to the Sellafield discharge point. Relatively high specific
activities of radionuclides in the dissolved phases of waters were encountered.
They found that sediment reworked from earlier deposits within the
estuary formed a major part of the sediment load of the tidal waters and
therefore provided a significant part of the total activity of the particulate
phase. The temporal and spatial variation of the total water activity for
radionuclides such as Pu follows the pattern for suspended sediment
concentration whereas, for conservative radionuclides (e.g. 137 Cs), the
 Radionuclides in the Irish Sea 161

distribution of total water activity resembles that for salinity. Assinder et

al. (1985) also reported that the particulate phase was dispersed more
effectively within the estuary than the dissolved phase leading to
proportionally greater contamination of the upper reaches at high water
and over most of the estuary at low water.
Detailed investigations on the effect of low pH and salinity (Burton,
1986) showed that at low tide and when fresh water was backed up by the
incoming tide, remobilisation of actinides into solution occurred in the
estuary. These actinides, however, could then be readsorbed by the
inactive river sediment load and redispersed around the estuary. This
activity, however, may be lost from the estuary with the suspended
sediment load on the ebb tide.
Although Sellafield discharges have been greatly reduced since 1980,
Ravenglass is still a focus of attention due to the redispersion of
historically deposited radionuclides and the intertidal processes that
influence them.
Most estuaries, including Ravenglass, contain both exposed sediments
and saltmarsh, which consists of sediment covered by salt tolerant plant
species such as Spartina anglica, Puccinellia maritima, Agrostis stolonifera,
Festuca rubra and others. Saltmarshes build up by continuing accretion of
sediment, mostly as a consequence of the trapping of suspended sediment
by vegetation during tidal inundations; the growth rate of the marsh
decreases as the elevation above mean sea level increases, because the
frequency of inundation decreases (Jones et a/., 1994). A saltmarsh will
typically show a gradation from seaward to landward ends, the landward
end being older and higher above mean sea level. Vegetation also shows a
gradation with the less salt tolerant species limited to the higher areas
where inundation is less frequent, and the diversity of vegetation greater.
Saltmarsh vegetation is an effective trap for the deposition of fine
sediment, so that saltmarshes usually contain silts and muds rather than
coarser grained material, and so contain the highest concentrations of
radionuclides in the estuary (Emptage & Kelly, 1990; Jones et al, 1994).
Saltmarsh sediments are stabilised by the vegetation and consequently
often show well developed profiles of radionuclides with depth with
subsurface maxima which correspond to the deposition of sediment at the
time of maximum discharges from Sellafield (Bonnet, Appleby & Oldfield,
1988; Jones et al, 1994). The depth of penetration of radionuclides into the
profile is usually determined by the sedimentation rate over the period of
highest discharges from Sellafield, i.e. 1965-1985. Areas of marsh which
were in active growth at this time can have radionuclide profiles extending
162 P. McDonald and S.R. Jones

to 1 m or more in depth; areas of marsh which were mature at this time

may have essentially all the radionuclides in the upper 0.2 m or less.

8.2.2 Radionuclide loss rates and total sediment inventories

Of the total available material (i.e. cumulative discharges corrected for
radioactive decay and/or ingrowth) the fractions retained in the Irish Sea
for 137 Cs ? 239 > 240 Pu and 241 Am are approximately 10%, 90% and 95%
respectively (MacKenzie et a/., 1994). The inventories for 137 Cs, 2 3 8 Pu,
239,240pu a n d 24ip u a r e c u r r e n t i y reducing at rates of 160,1.5,0.8 and 430
TBq y~ * respectively, as a result of advection through the North Channel
and radioactive decay exceeding the current input rates from Sellafield,
whilst the 241 Am inventory is increasing at 12 TBq y ~x largely because of
ingrowth from the remaining inventory of 2 4 1 Pu.
Total inventories of 2 3 9 ' 2 4 0 p u and 241 Am in the Irish Sea have also
been estimated from environmental measurements, particularly in sediments.
Pentreath et a\. (1984) have assessed the distribution in seabed sediments
from surveys made in 1977 and 1978 as shown in Figure 8.6. They
concluded that the total sediment inventory amounted to 280 TBq of

Figure 8.6 Plutonium inventory (kBq rrr2) in top 0.3 m of bed sediments
(Pentreath et a\., 1984).
 Radionuclides in the Irish Sea 163

239,240pu ancj 34Q j g q of 24i^ m ^ most o f w hi c h w a s contained in a

coastal strip some 30 km wide extending from Kirkudbright Bay in the
north to the Ribble Estuary in the south. This strip corresponds to an area
of the seabed with a high proportion offinegrained sediment (Figure 8.1).
The intertidal areas on the north-west coast of England and the
south-west coast of Scotland also contain significant inventories of
radionuclides associated with sediments. Surveys made in the 1980s
(Eakins et al, 1988,1990; Garland et al, 1989; Carpenter et a/., 1991) have
indicated total inventories of 36 TBq of 239 ' 240 Pu and 50 TBq of 241 Am in
the intertidal area between south-west Scotland and the coast of north
Wales, as summarised in Table 8.3.
The inventory summarised above may be an underestimate, because
the surveys covered largely exposed mud and sand and, although the
total area covered by saltmarsh is lower (Carpenter et al, 1991), the
inventories per unit area for saltmarsh can be higher (Oldfield et al,
1993). The measured inventories may be compared with the inventories
estimated for other compartments of the Irish Sea as summarised in
Table 8.4.
It can be seen that about 70% of the calculated inventory of 2 3 9 ' 2 4 0 p u
and 241 Am are accounted for by the measurements. It seems unlikely that
underestimate of the loss through the North Channel could account for
the discrepancy. It is likely that Pu and Am penetrating below the 0.3 m
core depth sampled by Pentreath et al. (1984), and perhaps additional

Table 8.3 Inventories of actinides in the intertidal areas

of north-west England and south-west Scotland (Eakins
etal, 1988,1990; Garland etai, 1989; Carpenter ef al, 1991)

Radionuclide inventory (TBq)

Area 238p u 239,240p u 241
Am
Wigtown Bay 0.4 1.9 2.9
Inner Solway* 1.0 4.0 6.0
Cumbrian coast 1.2 5.7 8.9
Morecambe Bay 2.2 10.8 15.6
Lancashire coast 1.8 11.6 12.7
Ribble Estuary 0.2 1.1 3.6
Mersey Estuary 0.02 0.02 0.2
Dee Estuary 0.2 0.9 1.0
Total 7 36 51

*Estimated
164 P. McDonald and S.R. Jones

material in the intertidal areas not yet comprehensively surveyed, will

account for the 'missing' Pu and Am.

8.2.3 Geochemical associations of radionuclides

It is now well established that particle transport is the dominant
mechanism of supply of Sellafield-derived radionuclides to the Solway
region (MacKenzie, Scott & Williams, 1987; McDonald et al, 1990).
Onshore transfer of this sediment from further south in the Irish Sea is
obvious upon consideration of the high sedimentation rates in these areas
and the fact that there is no significant fluvial input (McDonald, Cook &
MacKenzie, 1993). It is of importance to establish whether changes in
speciation and availability of radionuclides occur upon transport from
the highly saline, alkaline and well oxygenated marine environment to
terrestrial areas where the ground water is brackish, pH values range from
slightly alkaline to slightly acidic and oxygen depleted conditions can
occur at depth due to degradation of organic matter. Described below are
some aspects of recent studies of the mechanism of radionuclide transport
to the Solway Coast and of the geochemical behaviour of radionuclides in
Solway floodplain sediments.
A transect of five surface sediments (cores A-E, Table 8.5) from the
north-east Irish Sea revealed the specific activities of 137 Cs, 241 Am and
Pu isotopes to be of the same magnitude over a 60 km transect (except for
core C, in which activities were an order of magnitude lower) (McDonald
et a/., 1990). The size fractionation results (Table 8.6) revealed an inverse
correlation between particle size and radionuclide concentration, con-
firming the well documented preferential association of radionuclides

Table 8.4 Actinide inventory estimates for the Irish Sea

(referred to 1977/78) (Willans, Smith & Jones, 1994)

Compartment Inventory, TBq

239, 240p u 241Am

Seabed sediments 280 340

Intertidal sediments 36 51
Water column 5 5
Total measured 320 390
Discharge less loss 450 600
through North Channel
 Radionuclides in the Irish Sea 165

with finer size fractions (Hetherington, 1976). These data indicate that
the pool of radionuclides in the surface offshore sediment of the north-east
Irish Sea is spatially well mixed over the area studied (cf. cores A and E)
and that radionuclide concentrations are subject to greater control by
particle size composition rather than distance from Sellafield (McDonald
et al9 1990).
The geochemical behaviour of Cs and Pu in Solway floodplain cores,
collected from Southwick Water, south-west Scotland, has been studied
by Allan et al. (1991) by employing a sequential leaching scheme (Cook et
al, 1984; Livens & Baxter, 1988b; McDonald et al, 1990). This scheme
incorporates solutions of calcium chloride (readily available sites), acetic
acid (specific adsorption sites), tetra-sodium pyrophosphate (organically
associated), ammonium oxalate/oxalic acid, (Fe and Mn secondary
minerals), dilute nitric acid (dilute acid soluble sites) and nitric acid/
hydrofluoric acid (residual fraction). This technique has been subject to

Table 8.5 Surface radionuclide concentrations of north-east Irish Sea

sediments (Bq kg-1 ± 2a counting error) (McDonald et al., 1990)

Distance from
Depth Sellafield
137 238 239, 240p u 241
Core (m) (km) Cs Pu Am
A 37 14 596 + 26 62 + 7 328 + 32 392 + 26
B 35 20 229 + 10 46 + 7 205 + 26 205 + 14
C 40 29 69 + 4 7±1 38 + 3 55 + 4
D 45 49 262 + 12 49 + 6 233 + 24 236 + 15
E 15 63 462 + 22 61 + 20 297 + 26 309 + 18

Table 8,6 Particle size fractionation of north-east Irish Sea surface

sediments (McDonald et al., 1990)

% Coarse % Fine % Coarse

% Organic sand sand silt % Fine silt % Clay
Core carbon ( > 160urn) (50-160urn) (20-50jam) (2-20urn) (<2um)

A 8.0 2.9 46.0 14.8 22.9 13.4

B 3.6 27.3 53.3 4.8 3.2 9.3
C 2.0 84.3 10.3 0 0 5.4
D 2.9 68.1 16.1 2.3 4.3 11.4
E 5.2 3.2 68.3 11.7 3.3 13.4

Particle size distributions calculated from total mineral content. Organic fraction
determined from sample weight loss at 500 °C.
166 P. McDonald and S.R. Jones

recent general criticism (Khoeboian & Bauer, 1987) but remains a

valuable method of investigating the ease of removal of radionuclides and
other metals from soils and sediments, and in the present approach
sequential leaching is regarded as selectively dissolving radionuclides in
operationally defined, notionally discrete components of the sample.
It is evident from Table 8.7 that, independent of the depth (or age) of the
sediment, negligible quanties of 137 Cs were dissolved by reagents selected
to remove the available, exchangeable, organic bound and secondary
iron/manganese bound components. In each case, at least 80% of the
137
Cs was tightly bound in the residual phase and between 11 and 15%
was associated with the fraction extracted by 1MHNO 3 . This is consistent
with the observation by Allan et a\. (1991) of a KD of about 105 applicable
to desorption of 137 Cs from this sediment.
Associations of 2 3 9 ' 2 4 0 p u in surface samples revealed 4 1 % in the
'organic' phase, 39% in the iron/manganese extract and 16% in the
residual fraction, with minor quantities in the other compartments (Table
8.8). A systematic variation in these associations was observed as a
function of depth with the 'organic' component containing progressively
less 2 3 9 ' 2 4 0 p u and the iron/manganese and residual fractions containing
correspondingly more. In no instance was there any significant release of
239,240pu j n faQ available or exchangeable extractions.
These trends, in conjunction with an observed decrease in organic
content from 5% at the surface to 3% at depth, suggest that, as organic
matter is degraded, some of the plutonium which it originally contained is
released but is immediately taken up by other components of the sediment
(i.e. Fe/Mn secondary minerals). This observation is compatible with

Table 8.7 Sequential 137Cs leaching from Southwick Water floodplain

(% distribution) (Allan et al., 1991)

Depth % Organic
(cm) carbon RA EX OR Fe/Mn DAS RE
0-5 5.2 BDL BDL BDL 2 12 86
20-25 5.1 BDL BDL 0.4 4 15 81
40-45 4.8 2 BDL BDL 3 12 83
55-60 4.2 BDL BDL BDL 2 14 84
65-70 3.8 BDL 4 BDL 1 11 84

RA - readily available; EX - exchangeable; OR - organically bound; Fe/Mn - Fe/Mn

secondary minerals; DAS - dilute acid soluble; RE - residual; BDL - below detection limits
 Radionuclides in the Irish Sea 167

recent work on this sediment (Graham, Livens & Scott, 1993) showing
that the abundance of functional groups in humic acids (in particular -
COOH) decrease with depth and hence remove potential radionuclide
binding sites.
The results of the sequential leaching experiments for plutonium in a
Solway soil, an intertidal sediment and two Irish Sea sediments are
presented in Table 8.9 (McDonald et a/., 1990). It is evident that, for all
samples analysed, almost of all of the plutonium ( > 98%) is associated in
the more strongly bound fractions (organic, Fe/Mn secondary minerals
and residual phases) consistent with the results for floodplain sediments.
In the marine and intertidal sediments, the general order of geochemical
association of plutonium is: Fe/Mn secondary minerals > organic >
residual > specific adsorption > readily available. In the soil, the order
of association is: organic > Fe/Mn secondary minerals > residual > readily
available, specific adsorption. The soil sample was collected from just

Table 8.8 Sequential 239,24opu leaching from Southwick Water floodplain

(% distribution) (Allan et a/., 1991)

Depth % Organic
(cm) carbon RA EX OR Fe/Mn DAS RE
0-5 5.2 BDL BDL 41 39 2 16
20-25 5.1 BDL 0.3 43 40 0.5 16
40-45 4.8 BDL 1 24 51 2 22
55-60 4.2 BDL BDL 34 46 1 19
65-70 3.8 BDL BDL 17 57 2 24

RA - readily available; EX - exchangeable; OR - organically bound; Fe/Mn - Fe/Mn

secondary minerals; DAS - dilute acid soluble; RE - residual; BDL - below detection limits

Table 8.9 Sequential 239'240Pu leaching from a selection of Solway sediments

and soil (% distribution^ 1a counting errors) (McDonald et al.f 1990)

% Organic
Sample carbon RA EX OR Fe/Mn RE
Core A 8.0 <0.5 0.50 + 0.05 31 + 3 47 + 4 22 + 2
CoreE 5.2 <0.05 0.70 + 0.04 40+ 1 37 + 2 23 + 2
Kippford 5.3 <0.05 <0.05 32 + 2 56 + 3 11 + 1
Meickle-
Richorn soil 8.3 0.10 + 0.02 <0.05 56 + 3 32+ 1 11 + 1

RA - readily available; EX - exchangeable; OR - organically bound; Fe/Mn - Fe/Mn

secondary minerals; RE - residual
168 P. McDonald and S.R. Jones

above the mean high water mark at a location subject to periodic tidal
inundation, and the isotope activity ratio results confirm that the
plutonium is from Sellafield waste. The plutonium associations, however,
suggest a redistribution from the Fe/Mn secondary mineral and perhaps
residual phases to the organic phase upon transition from the aquatic to
the terrestrial environment.
Finally another area of interest is the 'mud-patch' off Sellafield acting as
a source of radionuclides to the Irish Sea environment. From a core taken
there in 1992, McDonald et al. (1993) found that the 137Cs/241Am activity
ratio decreased systematically from 6.60 at a depth of 34-36 cm to 1.36 at
the surface (Figure 8.7). While these variations are small relative to those
of the annual discharge (7.0 to 1218), they indicate that, in addition to
mixing (evident from the constancy of Pu/Am activity ratios down the
core), there has been a loss of 137Cs relative to 241Am from this sediment.
The results suggest that radiocaesium redissolution is significant to a
depth of 25 cm or more which is in contrast to the results of McCartney et
al (1994) who estimated that 10 cm was the effective limit for this process.
The major source of 137Cs to the Irish Sea was predominantly from
Sellafield. Although 137Cs discharges are still performed, their magnitude
compared to previous years is very small. The re-dissolution 137Cs from
historical deposits, however, is now a significant source of 137Cs to the
intertidal areas of eastern Irish Sea. Such a source term has to be
considered when interpreting 137Cs behaviour in these intertidal areas.

Cs/Am

7-,
6
•
5 • •
4 •
3
2 • •••
1
0
C) 10 20 30 40

Depth (cm)
Figure 8.7 Activity ratio profile in Irish Sea sediment (McDonald, Cook &
MacKenzie, 1993).
 Radionuclides in the Irish Sea 169

References
Allan, R.L., Cook, G.T., MacKenzie, A.B. & Pulford, I. (1991) Vertical distribution and
geochemical associations of radionuclides in Solway Firth saltmarsh sediment.
In Heavy Metals in the Environment, ed. J.G. Farmer, CEP Consultants Ltd
Publishers, 445-8.
Assinder, D.J., Kelly, M. & Aston, S.R. (1985) Tidal variations in dissolved and particulate
phase radionuclide activities in the Esk Estuary, England, and their distribution
coefficients and particulate activity fractions. Journal of Environmental Radioactivity
2, 1-22.
Aston, S.R. & Stanners, D.A. (1979) The determination of estuarine sedimentation rates by
Cs-134/Cs-137 and other artificial radionuclide profiles. Estuarine and Coastal
Marine Science 9, 529-41.
Aston, S.R. & Stanners, D.A. (1981) Americium in intertidal sediments from the coastal
environs of Windscale. Marine Pollution Bulletin 12, 149-53.
Aston, S.R. & Stanners, D.A. (1982a) Gamma emitting fission products in surface
sediments of the Ravenglass Estuary. Marine Pollution Bulletin 13, 135-8.
Aston, S.R. & Stanners, D.A. (1982b) Local variability in the distribution of Windscale
fission products in estuarine sediments. Estuarine and Coastal Shelf Science 14,
167-74.
Aston, S.R. & Stanners, D.A. (1982c) The transport to, and deposition of, americium in
intertidal sediments of the Ravenglass estuary and its relationship to plutonium.
Environmental Pollution 3, 1-10.
Aston, S.R., Assinder, D.J. & Kelly, M. (1985) Plutonium in intertidal coastal and estuarine
sediments in the northern Irish Sea. Estuarine and Coastal Shelf Science 20,761-71.
Aston, S.R., Assinder, D.J., Stanners, D.A. & Rae, J.E. (1981) Plutonium occurrence and
phase distribution in sediments of the Wyre estuary, north-west England.
Marine Pollution Bulletin 12, 308-14.
Bonnett, P.J.P., Appleby, P.G. & Oldfield, F. (1988) Radionuclides in coastal and estuarine
sediments from Wirral and Lancashire. Science ofthe Total Environment 70,215-36.
British Nuclear Fuels pic. (1971-1993) Radioactive discharges and monitoring of the
environment. Published annually, British Nuclear Fuels pic, Warrington, Cheshire.
Burton, P.J. (1986) Laboratory studies on the remobilisation of actinides from Ravenglass
estuary sediment. Science of the Total Environment 52, 123-45.
Carpenter, R.C., Burton, P.J., Strange, L.P. & Pratley, F.W. (1991) Radionuclides in
intertidal sands from Morecambe Bay to the Dee Estuary. Report AERE-R
13803, AEA Technology, Harwell, Oxfordshire.
Clifton, R.J. & Hamilton, E.I. (1982) The application of radioisotopes in the study of
estuarine sedimentary processes. Estuarine and Coastal Shelf Science 14,438-46.
Cook, G.T., Baxter, M.S., Duncan, H.J., Toole, J. & Malcolmson, R. (1984) Geochemical
association of plutonium in the Caithness environment. Nuclear Instruments and
Methods in Physics 223, 517-22.
Dickson, R.R. (ed.) (1987) Irish Sea status report of the Marine Pollution Monitoring
Management Group. Aquatic Environment Monitoring Report, MAFF Directorate
of Fisheries Research Lowestoft 17, 1-83.
Eakins, J.D., Morgan, A., Baston, G.M.N., Pratley, F.A., Yarnold, L.P. & Burton, P.J.
(1988) Studies of environmental radioactivity in Cumbria. Part 8. Plutonium and
americium in the intertidal sands of north-west England. Report AERE-R12061,
AEA, Harwell.
Eakins, J.D., Morgan, A., Baston, G.M.N., Pratley, F.W., Strange, L.P. & Burton, P.J.
(1990) Measurement of ot-emitting plutonium and americium in the intertidal
170 P. McDonald and S.R. Jones

sands of West Cumbria, UK. Journal of Environmental Radioactivity 11, 37-54.

Emptage, M. & Kelly, M. (1990) Gamma-dose rates in the Esk Estuary, Department of the
Environment Report DOE/RW/90/074.
Garland, J.A., McKay, W.A, Cambray, R.S. & Burton, P.J. (1989) Man-made radionuclides
in the environment of Dumfries and Galloway. Nuclear Energy 28, 369-92.
Graham, M.C., Livens, F.R. & Scott, R.D. (1993) Associations of actinides with soil humic
substances in saltmarsh sediments in south-west Scotland. In Heavy Metals in
the Environment, eds. R.J. Allan and J.O. Nriagu, CEP Consultants Ltd
Publishers, 239^2.
Gray, J., Jones, S.R. & Smith, A.D. (1995) Discharges to the environment from the
Sellafield site, 1952-1992. Journal of Radiological Protection 15, 99-131.
Hamilton, E.L (1980) Concentration and distribution of uranium in Mytilus edulis and
associated materials. Marine Ecology Progress Series 2, 61-73.
Hamilton, E.I. & Clarke, K.R. (1984) The recent sedimentation history of the Esk Estuary,
Cumbria, UK: the application of radiochronology. Science of the Total Environment
35, 325-86.
Hetherington, J.A. (1976) The behaviour of plutonium nuclides in the Irish Sea. In
Environmental Toxicity of Aquatic Radionuclides, eds. M.W. Miller and J.N.
Stannard, 81-106. Ann Arbor Science Publishers Inc., Ann Arbor, Michigan.
Howorth, J.M. (1984) Currents in the eastern Irish Sea. Oceanographic Marine Biology,
Annual Review 22, 11-53.
Howorth, J.M. & Kirby, C.R. (1988) Studies of environmental radioactivity in Cumbria.
Part 11 - Modelling the dispersion of radionuclides in the Irish Sea. Report
AERE-R11734, AEA, Harwell.
IAEA (1985) Sediment XDs and concentration factors for radionuclides in the marine
environment. Technical report series no. 247, International Atomic Energy
Agency, Vienna.
Jones, S.R., Rudge, S.A., Murdock, R.N. & Johnson, M.S. (1994) The dynamics of
vegetation contamination by radionuclides on a tide washed pasture in the
Mersey estuary. Water, Air and Soil Pollution 75, 265-75.
Kelly, M., Emptage, ML, Mudge, S, Bradshaw, K. & Hamilton-Taylor, J. (1991) The
relationship between sediment and plutonium budgets in a small macrotidal
estuary: Esk Estuary, Cumbridge, UK. Journal of Environmental Radioactivity
13, 55-74.
Kershaw, P.J. & Young, A. (1988) Scavenging of 2 3 4 Th in the eastern Irish Sea. Journal of
Environmental Radioactivity 6, 1-23.
Kershaw, P.J., Swift, D.S., Pentreath, R.J. & Lovett, M.B. (1984) The incorporation of
plutonium, americium and curium into the Irish Sea seabed by biological
activity. Science of the Total Environment 40, 61-81.
Kershaw, P.J., Pentreath, R.J., Harvey, B.R., Lovett, M.B. & Boggis, S.J. (1986) Apparent
distribution coefficients of transuranium elements in UK coastal waters. In
Application of Distribution Coefficients to Radiological Assessment Models, eds.
T.H. Sibley and C. Myttenaere, 227-87. Elsevier Applied Science Publishers,
London and New York.
Kershaw, P.J, Woodhead, D.S, Malcolm, S.J, Allington, D.J. & Lovett, M.B. (1990) A
sediment history of Sellafield discharges. Journal of Environmental Radioactivity
12, 201-41.
Khoeboian, C. & Bauer, C.F. (1987) Accuracy of selective extraction procedures for metal
speciation in model aquatic sediments. Analytical Chemistry 59, 1417-23.
Kirby, R, Parker, W.R, Pentreath, R.J. & Lovett, M.B. (1983) Sedimentation studies
 Radionuclides in the Irish Sea 171

relevant to low-level radioactive effluent dispersal in the Irish Sea. Part III. An
evaluation of possible mechanisms for the incorporation of radionuclides into
marine sediments. Report from the Institute of Oceano graphic Sciences, Wormley,
(178). 1-63.
Livens, F.R. & Baxter, M.S. (1988a) Particle size and radionuclide levels in some West
Cumbrian soils. Science of the Total Environment 70, 1-17.
Livens, F.R. & Baxter, M.S. (1988b) Chemical associations of artificial radionuclides in
Cumbrian soils. Journal of Environmental Research 7, 75-86.
MacKenzie, A.B. & Scott, R.D. (1982) Radiocaesium and plutonium in intertidal
sediments from southern Scotland. Nature 299, 613-16.
MacKenzie, A.B. & Scott, R.D. (1993) Sellafield radionuclides in Irish Sea sediments.
Environmental Geochemistry and Health 15, 173-84.
MacKenzie, A.B., Scott, R.D. & Williams, T.M. (1987) Mechanisms for northward
dispersal of Sellafield waste. Nature 329, 42-5.
MacKenzie, A.B., Scott, R.D., Allan, R.L., Ben Shaban, Y.A, Cook, G.T. & Pulford, I.D.
(1994) Sediment radionuclide profiles: implications for the mechanisms of
Sellafield waste dispersal in the Irish Sea, Journal of Environmental Radioactivity
23, 39-69.
Mauchline, J. (1980) Artificial radioisotopes in the marginal seas of north western Europe.
In The North-west European Shelf Sea: the Sea Bed and the Sea in Motion. II.
Physical and Chemical Oceanography and Physical Resources, eds. F.T. Banner,
M.B. Collins and K.S. Massie, 517-42. Elsevier. Amsterdam.
McCartney, M., Kershaw, P.J. & Allington, D.J. (1990) The behaviour of 2 1 0 Pb and 2 2 6 Ra
in the eastern Irish Sea. Journal of Environmental Radioactivity 12, 243-65.
McCartney, M. Kershaw, P.J., Woodhead, D.S. & Denoon, D.C. (1994) Artificial
radionuclides in the surface sediments of the Irish Sea, 1968-1988. Science of the
Total Environment 141, 103-38.
McDonald, P., Cook, G.T., Baxter, M.S. & Thompson, J.C. (1990) Radionuclide transfer
from Sellafield to south-west Scotland. Journal of Environmental Radioactivity
12, 258-98.
McDonald, P., Cook, G.T., Baxter, M.S. & Thompson, J.C. (1992) The terrestrial
distribution of artificial radioactivity in south-west Scotland. Science of the Total
Environment 111, 59-82.
McDonald, P., Cook, G.T. & MacKenzie, A.B. (1993) The dispersal of radionuclides in the
Irish Sea. In Heavy Metals in the Environment, eds. R.J. Allan and J.O. Nriagu,
CEP Consultants Ltd Publishers, 239-42.
Ministry of Agriculture, Fisheries & Food (1972-1993) Radioactivity in the coastal and
surface waters of the UK. Published annually, MAFF Directorate of Fisheries
Research, Lowestoft.
Nelson, D.M. & Lovett, M.B. (1978) Oxidation states of plutonium in the Irish Sea. Nature
276, 599-601.
Nelson, D.M. & Orlandini, K.A. (1979) Identification of Pu(V) in natural waters. Report
ANL-79-65, part III, 57-9, Argonne National Laboratory.
Oldfield, F., Richardson, N., Appleby, P.G. & Yu, L. (1993) 241 Am and 137 Cs activity in
fine grained saltmarsh sediments from parts of the N. E. Irish Sea shoreline.
Journal of Environmental Radioactivity 19, 1-24.
Pantin, H.M. (1977) Quaternary sediments of the northern Irish Sea. In Quaternary
History of the Irish Sea, eds. C. Kidson and M. J. Tooley, 27-54. Seel House Press,
Liverpool.
Pantin, H.M. (1978) Quarternary sediments from the northeast Irish Sea: Isle of Man to
172 P. McDonald and S.R. Jones

Cumbria. Bulletin of the Geological Survey of Great Britain 64, 1-43.

Pantin, H.M. (1991) The sea-bed sediments around the United Kingdom. Research Report
of the British Geological Survey, Offshore Geological Services, Keyworth, SB/90/1.
Pentreath, R.J., Harvey, B.R. & Lovett, M.B. (1985) Chemical speciation of transuranium
nuclides discharged into the marine environment. In Speciation of Fission and
Activation Products in the Environment, eds. R.A. Bulman and J.R. Cooper,
312-25, Elsevier, London.
Pentreath, R.J., Lovett, M.B., Jefferies, D.F., Woodhead, D.S., Talbot, J.W. & Mitchell,
N.T. (1984) The impact on public radiation exposure of transuranium nuclides
discharges in liquid wastes from fuel element reprocessing at Sellafield, UK. In
Radioactive Waste Management, Proc. Symp., Seattle, 1983, 5, 315-329,
International Atomic Energy Agency, Vienna.
Ramsay, J.D.F. & Raw, G. (1987) Physical characteristics of colloids and particulates in
coastal sediments and seawater. Report AERE R-12086, AEA, Harwell.
Sholkovitz, E.R. (1983) The geochemistry of plutonium in fresh and marine water
environments. Earth Science Reviews 19, 95-161.
Stather, J.W, Dionian, J , Brown, J , Fell, J.P. & Muirhead, C.R. (1986) The risks of
leukaemia and other cancers in Seascale from radiation exposure. NRPB-R171
Addendum, HMSO, London.
Swift, D.J. & Kershaw, P.J. (1986) Bioturbation of contaminated sediments in the
north-east Irish Sea. International Council for the Exploration of the Sea,
Copenhagen, CM 1986/E: 18.
Willans, S.M, Smith, A.D. & Jones, S.R. (1994) Development validation of the Sellafield
Environmental Assessment Model (SEAM). IRPA Regional Congress on Radio-
logical Protection, Portsmouth 6-10 June 1994. Proceedings of the 17 th Congress,
243-6. Nuclear Technology Publishing.
o o o o o o o o o o o o o o o o o o o
The sorption of hydrophobic pyrethroid
insecticides to estuarine particles: a
compilation of recent research
J.L. Zhou and S.J. Rowland

9.1 Introduction
Intertidal sediments, especiallyfine-grainedestuarine sediments, may be a
net sink, or at least a transient repository of hydrophobic organic
chemicals (HOC) from a number of sources. HOC have a strong, though
variable, affinity for sediment minerals, especially for clays coated with
natural organic matter (OM). Measurement of this adsorptive capacity is
thus an important goal for geochemists wishing to study the budget and
fate of HOC in the intertidal environment.
Although a considerable literature of adsorption measurements exists,
the cocktail of HOCs which reaches estuarine waters is so diverse and
originates from so many sources (including from municipal discharges,
from effluents from industrial works sited on estuaries, and from riverine
inputs) that inevitably there are important gaps in our knowledge.
Included in this is information about a number of important agrochemical
HOCs which may enter estuaries, via riverine transport. This chapter
focuses particularly on one group of HOCs (pyrethroids), the chapter by
Turner and Tyler discusses the behaviour of some other groups.
The pyrethroid agrochemicals represent a quarter of the world insecticides
market and are widely used in the UK, continental Europe, the USA and
elsewhere as household, public health and agricultural insecticides (Hill,
1989; Perrior, 1993). The chemicals are described by the generic name
'pyrethroids' because they are all man-made analogues of the esters of the
natural cyclopropanoic acid, chrysanthemic acid, found in the plant
family pyrethrums, but in fact they exhibit a range of physico-chemical
properties in the environment. For example, the high octanol-water
partition constants (Xow 1260000 to 10000000; Figure 9.1) of the
pyrethroids reflect a high affinity for organic matter such as that found in
soils. Indeed, this adsorption onto (or partitioning into) organic matter
174 J.L. Zhou and S.J. Rowland

may to a large extent immobilise the chemicals at the site of application.

This restriction on transport is important since pyrethroids have a high
toxicity to some aquatic organisms (Hill, 1985). Nonetheless small but
measurable concentrations of pyrethroids are found in river waters (NRA,
1992), possibly due to physical transport of the soil-sorbed chemicals, for
example, during periods of heavy rainfall. Once in river water, sorption

Permethrin, Kow = 1 260 000

C
K CH3
C =CH C0.0CH
Cl
CH3

Tefluthrin, Kow = 3 160 000

F F
CH3
C =CH C0.0CH 2
Cl
F F

Cypermethrin, Kow = 4 000 000

CN
Cl

C =CH C0.0CH
Cl
CHU

Lambdacyhalothrin, Kow = 10 000 000

CN
CHo
C = C0.0CH
Cl
CH,

Figure 9.1 Chemical structure of the four pyrethroids and their respective
Kow values.
 Sorption of pyrethroid insecticides: recent research 175

processes may again be important since the sorbate-sorbent equilibrium

is now disturbed, so desorption and even re-adsorption processes might
occur.
Presumably river-borne pyrethroids may eventually be exported to
estuaries, although few published data exist to test this theory. Indeed,
prior to our recent studies, very few published data were available to allow
an assessment of even the above sorption processes to be made. It is
therefore the purpose of this short review to compile the results of a 5 year
programme of research into the sorptive properties of pyrethroids onto
minerals, soil and estuarine particles. These have been published previously
as individual studies in diverse sources which may not be widely read by
sedimentary geochemists. We felt, therefore, that it would be useful to
compile such data.

9.2 Materials and methods

The details of the experimental methods have been published in full
previously and readers are referred to these studies for a comprehensive
description of methodologies (Zhou et a/., 1994; 1995a-c). A brief
description is given below.

9.2.1 Sorbates
Radiolabelled (14C) synthetic pyrethroids (permethrin, tefluthrin, cyper-
methrin and lambdacyhalothrin; Zeneca) were purified by radio thin layer
chromatography (TLC) or radio high performance liquid chromatography
(HPLC) to > 96% purity and dissolved in hexane as stock solutions.

9.2.2 Sorbents
Three minerals (montmorillonite, aluminium oxide and kaolinite) of
known particle size and specific surface areas (Zhou et al, 1994) and a
sterilised silty clay loam (Champaign, USA) were used as sorbents, along
with suspended particulate matter (SPM) collected in polythene carboys
from several sites in the Tees Estuary, UK. SPM was collected by high
speed continuous flow centrifugation and stored in 0.1M NaHCO 3 at
4°C.Traces of organics were removed from the three minerals by washing
with 0.01M NaOH.

9.2.3 Natural organic matter

+
H saturated humic, fulvic and hydrophilic macromolecular natural
organic matter was isolated from the river Dodder, Ireland, by adsorption
and cation exchange chromatography (Hayes, personal communication).
176 J.L. Zhou and S.J. Rowland

Elemental compositions were measured directly on P 2 O 5 -dried samples

(oxygen by difference of ash-free material) and aromaticity by 13 C NMR
(Zhou et aU 1994).

9.2.4 Adsorption
(a) Natural organic matter onto minerals
Adsorption was carried out in conical flasks or centrifuge tubes. A
weighed amount of mineral (20-200 mg) was added to a flask or tube, to
which 50 ml of a dilute solution of OM was added. Solutions were
adjusted to desired pH, purged with nitrogen, sealed and shaken for 2
days. Solutions were then filtered or centrifuged and supernatant examined
by UV-visible spectrophotometry at 350 nm. The amount of OM on the
minerals was then calculated by mass balance. Control experiments
showed no adsorption to the glassware.
(b) Pyrethroids onto minerals and minerals coated with OM
Sorption was studied by a batch isotherm method using 30 ml centrifuge
tubes. Radio-labelled pyrethroid in hexane was added to minerals at the
required pH with and without OM (see (a) above). Solvent was removed
by nitrogen purge. Tubes were capped, shaken for 24 hr and centrifuged.
Radioactivity in the supernatant was measured by scintillation counting
following scintillant addition.
Glass-sorbed pyrethroid was measured in the same way after washing
off with dichloromethane three times. Control experiments without
minerals allowed glass sorption to be independently measured. All
experiments were made at least in duplicate. Average recovery was 90%.
(c) Pyrethroid onto soil
Mixtures of radiolabelled tefluthrin (6 jag), water (30 ml) and soil (2.0 g)
were placed in centrifuge tubes, shaken (24 hr), the supernatant removed
by centrifugation and radioactivity determined as above. The radioactivity
of the filtered solids was determined after filtering and treatment with a
cocktail specifically formulated to dissolve membrane filters (Zhou et a/.,
1995c). All experiments were carried out in triplicate and control tubes
were used without soil for mass balance and assessment of glass-sorbed
tefluthrin.

9.2.5 Desorption
Desorption was studied by three methods but the data from only one are
reported here. After adsorption equilibrium, the contents of one tube were
 Sorption of pyrethroid insecticides: recent research 177

poured into a 51 glass beaker containing 3.51 of Millipore grade water.

The solution was stirred constantly with a magnetic stirring bar. Samples
(25 ml) were taken at selected time intervals and filtered through a 0.45 |im
membrane filter. The soil particles and filter paper were mixed with 10 ml
of a specially formulated scintillation cocktail, shaken and counted. The
amount of non-desorbed tefluthrin was calculated.

9.3 Results and discussion

9.3.1 Simple sorption theory
The distribution or partitioning of chemicals between solid and aqueous
phases is often described by the distribution ratio or coefficient KD, where

C s
K -

where

Cs = concentration on solid phase

Cw = concentration in water phase

However, the Cs value for HOCs is controlled most by the adsorption

onto (and/or absorption into) natural organic matter associated with the
solid phase.
Thus the organic matter-water (or organic carbon-water) partition
coefficient X om (or Xoc) is most useful for description of the sorption
behaviour of HOCs where:

where

/om = weight fraction of solid which is natural organic matter

foc = weight fraction of solid which is natural organic carbon

A somewhat related 'surrogate' measure of the partitioning behaviour of

HOCs between aqueous and organic phases is the so-called octanol-
water partition coefficient, Kow, where:

K = ^
OW /~*
178 J l . Zhou and S.J. Rowland

where

Co = concentration of HOC in n-octanol phase

Cw = concentration of HOC in aqueous phase

A large database of measured and calculated Kow values is available (e.g.

Leo, Hansch & Elkins, 1971) and free energy relationships between X ow
and Kom have been established.
Readers are referred to the excellent text by Schwarzenbach, Gschwend
& Imboden (1993) for a fuller explanation.

9.3.2 Experimental challenge of pyrethroid studies

The extreme hydrophobicity of pyrethroids, as reflected in the high X ow
values (Figure 9.1) is reflected in a strong adsorption to many solid
surfaces including experimental apparatus such as glass, PTFE and
filtration membranes (House & Ou, 1992). Probably this partly accounts
for the lack of Kom (Koc) values available prior to our studies. Indeed the
problem is illustrated by the data shown in Figure 9.2 which shows the
adsorption of four pyrethroids, permethrin, tefluthrin, cypermethrin and
lambdacyhalothrin onto the glass walls of experimental apparatus at
equilibrium. Clearly, in the absence of competing sorbents such as soil, a
major proportion ( > 60%) of the pyrethroids is not present in free
solution but is sorbed to the walls.
Such phenomena dictated two major philosophies for our work with
pyrethroids: (a) a mass balance approach was taken to account for all
losses in the experiments; (b) as long as the pyrethroids maintained their
chemical integrity (did not hydrolyse or otherwise degrade) then use of
radiolabelled analytes was desirable. This allowed the low concentrations
of the pyrethroids in the aqueous phase to be accurately and reproducibly
determined and the mass balance to be computed. The chemical integrity
of the radiochemicals was determined by radio-chromatographic techniques
(radio-TLC and radio-HPLC).

9.3.3 Adsorption of pyrethroids to minerals and soil

The strong adsorption of pyrethroids for minerals, especially minerals
coated with natural organic matter such as are found in soils, is illustrated
in our studies by the behaviour of tefluthrin. Thus Figure 9.3 shows
sorption of tefluthrin onto organic-free kaolinite, alumina (aluminium
oxide) and montmorillonite. Clearly the sorptive capacity of the minerals
varies in the order montmorillonite > alumina > kaolinite but all three
 Sorption of pyrethroid insecticides: recent research 179

isotherms are linear over the concentration range investigated. Values of

KD varied from 92ml g" 1 (montmorillonite) to 50 (alumina) to 7.6
(kaolinite). This corresponds to only low coverage of the surface of the
minerals (Zhou et a/., 1995b) and, as expected, also to the specific surface
area of the minerals measured by the BET and nitrogen isotherm methods

0.012
0
0.94 cm
/

0.045
/o '

0.030 •

/o
0.015 -
/o

(a)
/

0<
0 15 30 45 60 0 3 6 9 12 15
Pyrethroid in solution (jjg I"1) Pyrethroid in solution (pg r 1 )

0.0036
u.uuou
Kw= 0.91 cm

/o

0.0045
o /

0.0030 -

0.0015 -

(c)
/ . .
0<
0 1.5 3.0 4.5 6.0 1.0 2.0 3.0 4.0

Pyrethroid in solution (pg l~1) Pyrethroid in solution {\iq

Figure 9.2 Adsorption of synthetic pyrethroids onto glass walls in the

absence of sorbents: (a) permethrin, (b) tefluthrin, (c) cypermethrin and (d)
lambdacyhalothrin.
180 J.L. Zhou and S.J. Rowland

(Zhou et al, 1994). When the minerals are coated with natural organic
matter such as riverine aquatic humic substances, sorption increases
several fold due to the strong partitioning of the pyrethroid into the OM
(Figure 9.4). The isotherms are still linear (r2 > 0.98) but partitioning
increases, with increases in the proportion of organic coatings on the
mineral (foc). The influence of the OM can be obtained by deducting the
effect of adsorption of tefluthrin onto clean mineral (Kmin):

Joe

where Kt is the slope of the isotherms for organic-coated minerals and S is

the exposed surface area of the mineral (m2 g~ *). The value of S can be
calculated from the measured total surface area, the radius of gyration and
average molecular weight of the natural organic matter. As shown in

i i 1

(A) Montmorillonite

(•) Aluminium oxide

(•) Kaolinite
i
1.5 -

1.0

A
0.5

—*
0 I
0 5 10 15 20
1
Aqueous concentration (ug l~ )

Figure 9.3 Adsorption of tefluthrin by 'clean' mineral particles.

 ID

foc values are: foe values are: oc values are

(A) 0.0012 (o) 0.0022 T (•) 0.0065

(A) 0.0015
. (D) 0.0007
I (o) 0.0021
(•) 0.0002
(•) 0.0007
(•) 0.0002 / 0.0012
• (a) 0.0006
f ^ /
/ 1 AI 0.0003

o 4 /
A
/ /
o /

ion
y
CO

c 3
CO
4-J O / /
£Z
CD
8
/ / •

// //
O
c •
o
o // A/
/
"O
CD
/
a
o
co 4 I /Ik / i

0 0 I
0 5 10 15 0 5 10 15 20 25 0 5 10 15 20
1 1 1
Aqueous concentration (jug I" ) Aqueous concentration (jug I" ) Aqueous concentration (jug I" )

(a) (b) (c)

Figure 9.4 Sorption of tefluthrin by montmorillonite coated with (a) hydrophilic macromolecular acid, {b) fulvic acid and (c) humic acid.
 182 J.L. Zhou and S.J. Rowland

Figure 9.5, the Koc values obtained in this way vary not only with the
extent of natural organic matter coating (foc) but also with the character of
the OM. This corresponds also to increased adsorption with increased
aromaticity of OM. Since the aromaticity of terrestrial-derived (i.e. soil)
OM is likely to be high due in part to a preponderance of lignin residues,
then soils can be expected to have a high affinity for pyrethroids (Zhou et
al, 1994).
Indeed when the sorption of tefluthrin onto a typical mid-west silty clay
loam (consisting of 4% coarse sand, 29% fine sand, 36% silt, 31% clay
and 3.3% OM) from Champaign, Illinois, USA was investigated (Zhou et
al.9 1995c) about 95% of the tefluthrin was soil-sorbed at equilibrium. The
KD and Koc values were determined (assuming OC = OM/1.724; Zhou et
al, 1995c) and found to be 545 + 15 and 28454 ± 790 respectively. Repeat
experiments gave KD 469 + 15 and Koc 24490 + 761. Thus good precision

800 000 1 1
' i

• •
Humic acid

600 000

O
O
A

A
A
400 000 A Fulvic acid -

Hydrophilic macromolecular acid

o 200 000

°oo o
Organic coatings on Tees estuarine particulates

i i

0 0.005 0.010 0.015 0.020 0.025

Mass fraction organic carbon (foc)

Figure 9.5 Variation of kow for tefluthrin with the degree and nature of
organic coatings.
 Sorption of pyrethroid insecticides: recent research 183

was obtained (RSD 2.8-3.0%). The lower Koc values obtained for soil (c.
27 000; n = 5) compared with minerals coated with the most aromatic
natural OM available to us (humic acids: c. 800000) are probably a
function of the far higher concentration of soil sorbent used than minerals
(the so-called 'sorbent concentration effect'), kinetic effects (Van Hoof &
Andren, 1991) and the presence of non-settling colloidal material from the
soils which may raise the apparent Cw value and hence lower soil KD and Koc.
Nonetheless, the soil data show that pyrethroids such as tefluthrin can
be expected to be strongly adsorbed to soils. What happens however when
this equilibrium is disturbed by physical washing of soil-pyrethroid
particles into a river or lake?

9.3.4 Desorption of pyrethroids from soil

Somewhat surprisingly, given its hydrophobicity, our studies suggest that
tefluthrin desorbs quite rapidly from soil when presorbed material is
transferred to a water body - represented in our experiments by a large
(-3.51) beaker of fresh water. Indeed, the desorption appears to follow
first-order kinetics (r2 = 0.98) with a typical rate constant and half-life of
about 0.002 hr" 1 and 5hr respectively. Desorption appeared to be
independent of pH and salinity of the water but dependent on temperature
(all at 95% confidence levels) as shown in Figures 9.6 to 9.8. Calculation of
thermodynamic parameters for the desorption process from the van't
Hoff equation (Zhou et al, 1995c) showed that the process was exothermic
(AH = - 56 kJmol" 1 ) and spontaneous(AG « - 2 2 to - 2 3 k J m o r 1 ) .
This is consistent with a partition mechanism for tefluthrin (Zhou et al,
1995c).
It is important to note that our experiments were carried out with pure
tefluthrin, not the formulated material sold for agricultural use which was
not available to us in radiolabelled form. Furthermore, our soil-sorbed
tefluthrin samples had not been 'aged'; desorption was studied after only
24 hr adsorption equilibrium.
Nonetheless, these experiments indicate that pyrethroids may re-enter
the dissolved phase under some conditions.

9.3.5 Re-adsorption of pyrethroids onto estuarine particles

Assuming that pyrethroids may reach estuaries by riverine transport
(budgets of this phenomenon are currently being assessed; LOIS, 1994)
the question arises as to whether re-adsorption onto estuarine particulate
matter is possible. Given thefindingsabove, this seems entirely reasonable
184 J.L. Zhou and S.J. Rowland

but no published data appeared to exist prior to our studies (Zhou et al,
1995a?b).
Adsorption of tefluthrin, cypermethrin and lambdacyhalothrin onto
estuarine particles collected from the Tees is illustrated in Figure 9.9
(Zhou et a/., 1995a-c) along with data for the same chemicals sorbed onto
hydrophilic natural organic matter coated onto montmorillonite. Values
of Koc ranged from 120000 to about 450000 for all three compounds
depending on the/ oc (Figure 9.9). A possible explanation for the decrease
in Koc with increase in/ oc is that, at the lowest/oc values, all natural OM
sites on the particle surfaces are accessible to the sorbate (pyrethroid) but
as foc increases the OM adopts a different interfacial configuration,
reducing the availability of sorption sites. Such a phenomenon is not
without precedent for other HOCs (Garbarini & Lion, 1986; Schlautman
& Morgan, 1993).

3.0
(A) pH 5.2
(•) pH 5.9
(•) pH 6.5

2.0

1.0

0 200 400 600 800 1000 1200 1400 1600

Time of sampling (min)

Figure 9,6 Effect of solution pH on tefluthrin desorption kinetics.

 Sorption of pyrethroid insecticides: recent research 185

9.4 Conclusions
Sorption studies of a variety of widely used hydrophobic pyrethroid
insecticides have shown:

(a) that the extreme hydrophobicity of the chemicals requires great

care with experimentation in order to quantify losses onto
glassware and other apparatus. A mass balance approach to
experiments and use of radiolabelled sorbates is recommended.
(b) adsorption of pyrethroids to a variety of minerals is significant,
and at the concentrations studied, linear. Values of KD for clean
minerals (Kmin) ranged from about 100 to about 8 ml g" 1 .
(c) adsorption to minerals coated with natural organic matter
increased severalfold relative to uncoated minerals. Values of KD
for sorption of tefluthrin to a US soil were about 500 ml g" 1 .

3.0

(•) Ionic strength 0

(D) Ionic strength 0.02
(A) Ionic strength 0.08

2.0

1.0

0 200 400 600 800 1000 1200 1400 1600

Time of sampling (min)

Figure 9,7 Effect of ionic strength on tefluthrin desorption kinetics.

 186 J.L. Zhou and S.J. Rowland

(d) desorption of tefluthrin adsorbed onto soil was rapid and

spontaneous ( — AG « 22-23 kJ mol~ *) and followed first-order
kinetics (k « 0.002 h~ *; t± « 5 h) when the soil was transferred to
a vessel containing a relatively large volume of fresh water.
(e) adsorption of a variety of pyrethroids onto estuarine particles
from the Tees was strong with Koc values ranging from about
120000 to 300000. Values of Koc decreased with increased foc
possibly due to reorientation of the natural organic matter at high
foc values.

Acknowledgements
We are grateful to the publishers of Water Research, Environment
International and the International Journal of Environmental Analytical
Chemistry and the British Crop Protection Council for permission to

IP

t
(A) 9°C

\ (•) 15°C
(•) 20°C

2.0 ye -
3.

A A
• \ #
1.0

n
0 200 400 600 800 1000 1200 1400 1600

Time of sampling (min)

Figure 9.8 Effect of temperature on tefluthrin desorption kinetics.

 Sorption of pyrethroid insecticides: recent research 187

publishfiguresfrom the articles cited. The review presented here represents

but a summary of the results of a major input of knowledge and experience
from colleagues at Plymouth Marine Laboratory, University of Plymouth,
Brixham Environmental Laboratory (Zeneca) and Zeneca Agrochemicals
Ltd, UK, especially Professor R.F. Mantoura, Dr J. Braven, Brian
Harland and Mike Lane. We are grateful to Zeneca for funding the
research and for permission to publish. We are also grateful to Dr Mike
Hayes, University of Birmingham, for the supply of natural aquatic humic
material isolated from the River Dodder, Ireland.

500 000
(*) Lambdacyhalothrin on hydrophilic OM
(A) Lambdacyhalothrin on Tees particles
(•) Cypermethrin on hydrophilic OM
(•) Cypermethrin on Tees particles
400 000 (•) Tefluthrin on hydrophilic OM
(o) Tefluthrin on Tees particles

• •
•

300 000

•
200 000 •

O O
o> 100 000
6

0.005 0.010 0.015 0.020 0.025

Mass fraction organic carbon (fot

Figure 9.9 Summary of the organic carbon normalised partition

coefficients for pyrethroids.
188 J.L. Zhou and S.J. Rowland

References
Garbarini, D.R. & Lion, L.W. (1986) Influence of the nature of soil organics on the
sorption of toluene and trichloroethylene. Environ. Sci. Technol. 20, 1263-9.
Hill, I.R. (1985) Effects on non-target organisms in terrestrial and aquatic environments. In
The Pyrethroid Insecticides. Leahey, J.P. ed. Taylor & Francis, London. 151-262.
Hill, I.R. (1989) Aquatic organisms and pesticides. Pest. Sci. 27, 429-65.
House, W.A. & Ou, Z. (1992) Determination of pesticides on suspended solids and
sediments: investigations on the handling and separation. Chemosphere 24,819-32.
Leo, A., Hansch, C. & Elkins, D. (1971) Partition coefficients and their uses. Chem. Rev. 71,
525-53.
LOIS (1994) The first LOIS RACS (C) workshop. Burwalls, Bristol, 28-30 March 1994.
Copies from P. Donkin, J. Harris, P. Watson or R. Willows. Plymouth Marine
Laboratory, Plymouth, Devon.
NRA (National Rivers Authority) (1992) Summary of chemical data for Yorkshire rivers,
1990-1992. Available from NRA, Yorkshire.
Perrior, T.R. (1993) Chemical insecticides for the 21st century. Chem. Ind. 22, 883-7.
Schlautman, M.A. & Morgan J.J. (1993) Binding of a fluorescent hydrophobic organic
probe by dissolved humic substances and organically coated aluminium oxide
surfaces. Environ. Sci. Technol. 27, 2532.
Schwarzenbach, R.P., Gschwend, P.M. & Imboden, D.M. (1993) Environmental Organic
Chemistry. John Wiley & Sons, New York.
Van Hoof P.L. & Andren, A.W. (1991) In Organic Substances and Sediments in Water.
Baker, R.A. ed., Lewis publishers, Michigan, USA. Volume 2, 149-67.
Zhou, J.L., Rowland, S., Mantoura, R.F.C. & Braven, J. (1994) The formation of humic
coatings on mineral particles under simulated estuarine conditions - a mechanistic
study. Water Research 28, 571-9.
Zhou, J.L., Mantoura, R.F. & Rowland, S.J. (1995a) Partition of synthetic pyrethroid
insecticides between dissolved and particulate phases. Water Research 29,1023-31.
Zhou, J.L., Rowland, S.J., Braven, J., Mantoura, R.F.C. & Harland, B.J. (1995b) Tefluthrin
sorption to mineral particles: role of particle organic coatings. Int. J. Env. Analyt.
Chem. 58, 275-85.
Zhou, J.L., Mantoura, R.F.C, Lane, M. & Rowland, S.J. (1995c) Sorption and desorption
of tefluthrin insecticide by soil under simulated run-off systems. In Pesticide
Movement to Water BCPC monograph No. 62. (A. Walker et al, eds.) BCPC,
Farnham, Surrey.
 Index
o o o o o o o o o o o o o o o o o o o

acid methanolysis 130 caesium 45, 47-9, see also 137Cs

actinides 154, 156, 161 Callinassa subterranea 154
adsorption Capenhurst 154
of pyrethroids 176, 178-83 carbon
of radionuclides 158 cycling 99-104
adsorption modelling 51-4 diagenesis 100
advective flow 86, 87 labile 102, 103
advective transport 86, 87 mass balance model 102, 103
Agrostis stolonifera 161 microbial reduction 109
air float technique 69, 70, 81 reactive 105, 106
akaganeite 126 remineralisation 104, 106
alkalinity 102, 103, 106 supply 102
aluminium oxide 178-9 carbon dioxide 103, 104
241
Am 154, 155, 156, 157, 158, 162-3, 168 carbonate 103
ammonification 65 concretions 124
ammonium 7, 86, 90, 91 minerals 103
anoxia 104 Chapelcross 154
Aster tripolium 123 chemical bombs 11
atmosphere 92 chlorinated organic compounds 46, 47, 49
authigenic minerals 111,115,124 clay minerals 158
climate change 3
bacteria 94, 110, 115, 136 Clyde estuary 43-5, 48, 51, 55
communites 108 coastal defence 89
metabolism 104 coastal protection 1
species 133 community budgeting 61-2
bacterial fermentation 101 computer modelling 20
bacterial reactions 101 concretions 124, 125, 126, 136^47
137
barium 11-12 Cs 154, 155, 156, 158, 160, 162-3, 166, 168
bioassay 141 cypermethrin 175, 184, 187
bioirrigation 89, 115
bioturbation 89, 100, 115 degradation of organic matter 164, 166-7
branched monoenoics 133 deltas
buffer system 96 Indus 6
Mississippi 6
cadmium 45, 47-9, 51, 54-6 Nile 6
190 Index

denitrification 62, 65, 88, 89, 91, 92, 94, 95 nutrient 59-82
depth profiles framboid formation 107, 108
radionuclides 161, 168
trace metals 27-36 gas chromatography/mass spectrometry
desorption (GO/MS) 130
of pyrethroids 176-7, 183 global warming 3
of radionuclides 166 glutathione 109
of trace metals 26 goethite 126
desorption modelling 50-1 gram negative bacteria 133, 141
desulfobacter 133, 141, 142, 144, 147 grain size 17, 19, 23, 24, 28, 33, 152, 158, 160,
Desulfovibrio sp 104, 133, 141, 142, 144, 147 164-5
diagenetic history 29-30 grain size proxy 34
diagenetic reactions 100, 101, 111 gram negative bacteria 133, 141
diagenetic sequence 101 greenhouse effect 4
diagenetic zones 100 greenhouse gas 92
diatoms 90, 93, 94 gypsum 126
diffusion 106, 114, 115
diffusion chamber studies 61, 62, 63-5, 81 Halimione portulacoides 123
diffusive flux 103 hexane extraction 130
diglyceride fatty acids 130 historical pollution trends 27-36
dimethyl sulphide 10, 108 Humber Estuary 43-5, 51, 95
discharge volume measurement 68, 69, 72, 74-6 humic substances 110
dissolved organic carbon 103 pyrethroid sorption onto 180-3
distribution coefficient see KD
distribution coefficient 43-56 insecticide 173-88
dredging 89 integrating rising air float technique 69-70, 81
drying and flooding 86 interstitial water 93
inventories 162-4
early diagenesis 34, 99, 102, 114, 119 Irish Sea 152-68
electron iron
acceptor 101, 102, 104, 114 accumulation 111
flux 111 cycling 110
transfer 111 depth variation 137, 143, 140
engineering management 89 levels 136, 138
episodic events 87 mobility 136
Esk Estuary 160 iron/manganese phase 20, 160, 166, 167, 168
eutrophication 91 iron monosulphide 124
extraction techniques 20 iron oxides 89, 93, 99, 103, 110, 130
iron sulphide 107-9
fatty acid methyl esters (FAME) 130 isotherms 178-81
Fe and Mn oxides 29, 34, 49, 51
Festuca rubra 161 Juncus spp 123
Festuca spp 123
field sampling 60-74 KD 157, 177, 179, 182, 183, 185
fishing 89 Koc 177, 182-4
fission products 154 Kow 173, 177-8, 182
flume studies 61, 62-3, 64, 65, 81 kaolinite 178-9
flux
American studies 65 lambdacyhalothrin 175, 184, 187
calculations 74-80 laser granulometry 130
measurement 59-82 Limonium vulgar'e 123
 Index 191

lipid biomarker techniques 121, 127, 136 storage 88

lippolysaccharide hydroxy fatty acids nutritional stress 133, 135 (Fig. 7.10), 141
(LPSOHFA) 130 North Inlet marsh 64

management 89, 92, 96 organic carbon 7, 93, 99, 100, 101

mangal forest 1, 4, 6, 9 analysis 128
mangal swamp 3, 6, 11 content 130, 136, 138, 141, 145
manganese decomposition 104, 105
(IV) 111 depth variation 29, 137 (Fig. 7.11), 140
accumulation 111 (Fig. 7.13), 143 (Fig. 7.11)
minerals 107 diagenesis 104
oxyhydroxides 99, 103 organic material
Maxmulleria lankesteri 154 degradation 93, 94
Marchon 154 organic matter
marsh see saltmarsh anaerobic metabolism of 114
methanogenesis 101, 103 associations 110
methanogenic bacteria 104, 124 burial flux 114
microbial activity 119, 120, 121, 127-47 decomposition 7, 101, 136
assay 129, 136 degradation 124
biomass 127, 138, 140, 143, 146 reactivity 114
microbial communities 8, 99, 101, 134, 141, transfer of electrons from 111
142 (Fig. 7.14) oxic-anoxic boundary 108
microbial metabolism 99 oxic zone 110
microbial oxidation 100 oxygen 94, 101, 107, 111
micro-environment 9, 102, 103, 104, 107, 114 availability 145
monoenoics 133, 141, 142, 144, 147 in atmosphere 114
montmorillonite 175-81, 184 influx 104
morphology 95
Morston 122 paniculate phosphorus 65
mud flats 85 partition coefficient 43-56
permethrin 175, 179
neap to spring cycle 72-6, 87 pH 10
nitrification 62, 91, 95 phosphate 65, 77, 78, 80
nitrogen 7, 64, 65, 87, 88, 89, 90 phospholipids (PLFA) 127, 129, 130, 133, 140,
cycle 89, 91 142, 143, 144
fixation 90, 91 phosphorus 85, 87, 88, 89, 90, 93
forms 90 photogrammetry 70
regeneration 91 physical disturbances 85
nitrous oxide 92 phytoplankton 90, 94
nitrate 87,88,91, 103, 119 pollutant 109, 122
nitrite 87-8, 90 polyenoics 133
Norfolk coast 120, 121 polysulphides 107, 109, 110
North Channel 152, 162 pore-water
nutrient composition 125, 127, 136
flux 59-82 F e a n d M n 111
impact 84, 89 profiles 114
limited 66, 89 saltmarsh 109
recycling 84-96 sulphide 108
regeneration 88, 89, 96 supersaturation (sulphide) 107
seasonal variation 72-4 primary production 66, 89
source 66, 84, 90 productivity 7, 82, 84, 93, 94, 110
192 Index

238
Pu 154, 155, 156, 160, 162-3 spatial variability 66-7
239(24oPu 1 5 4 > 1 5 5 ) 156i 157> 15g5 160> 162_35 Springfields 154
166-7 St George's Channel 152
241
Pu 154, 155, 156, 162-3 Stiffkey 69, 78, 104, 105, 111, 122
Puccinellia maritima 123, 161 storms 87
pyrethroid 173-88 storm events 6, 7
pyrite 106, 107, 108, 109, 110, 114 suboxic metabolism 103
pyritisation 107, 110 sub-oxic zone 110
Sueda spp 123
radiotracer 45 sulphate 100, 101, 103, 104, 105
radon 11-12 formation 109
Ravenglass 159, 160, 161 regeneration 108-9
reactive iron supply 124
analysis 128 sulphate reducing bacteria (SRB) 104, 124, 125,
reclamation 89 136
redox sulphate reduction 102, 103, 104, 106, 107, 108,
cells 114, 115 111, 114, 119, 133, 136, 145
conditions 10, 109 sulphide
equilibria 8 flux 108
reactions 9, 109, 114 ions 109
research programmes 84 minerals 106
Ribble Estuary 163 organic 109
river spate 87-8 oxidation 109, 111
run-off events 87-8 precipitation 107, 108, 110
reactivity 110
Salicornia perennis 123 solid 104, 106
San Francisco Bay 26 sulphur
saltmarsh 1, 31-2, 60, 66, 70, 76, 81, 104, 109, accumulation 11
110, 121, 122, 123, 124, 127, 145, 161 acid volatile sulphur (AVS) 128, 130, 133
saturates budget 108
branched-depth distribution 142 (Fig. 7.14), cycle 109
143 (Fig. 7.16) cycling 99-100, 104-11
mid-branched 133 elemental 107, 108, 130, 133
normal 133 organic 109, 110
terminally branched 133, 141 pyrite 128, 130
sea level rise 1, 5, 11 reduced 143 (Fig. 7.15)
seasonality 9, 106, 110 reduced (pool) 111, 140 (Fig. 7.13)
sediment cores 27-36 sulphur species 10
sediment depth profiles 27-36 depth variation 137 (Fig. 7.11)
sediment transport 88 mass balance model for 104, 105
sedimentation rate 101, 114, 160 organic 109
Sellafield 154, 160, 161, 162, 168 surface area 17, 179, 180
sequential leaching of radionuclides 165 suspended particulate matter (SPM) 56, 158,
Severn Estuary 25, 26, 31-6 175
shrapnel 138 suspended sediment 67, 160, 161
siderite
cements 108, 124, 125 Tees Estuary 175, 182, 184, 186
concretions 125 tefluthrin 176-87
silicon 85,87,88,90,93 temporal variability 66
Sol way 164-8 terminal electron acceptors 8
Spartina anglica 161 thin layer chromatography 130
 Index 193

thiol 109 Trawsfynydd 154

thiosulphate 109 turbidity maximum 55
tides 86-7
tidal asymmetry 74-7 uranium series radionuclides 154
tidal creek 61, 62, 64
hydrodynamics 68 velocity meter 69, 70, 75
measurements 61, 62, 64, 68, 69, 70, 81
tidal cycle 6 Warham 121, 122, 123, 125
tidal flux see flux Wash 92, 93, 95
trace metals 18-41 water velocity 68
bioavailability 17, 21, 22 waves 87
mobilisation 110 Wells 122
partitioning 17 wildlife habitats 3
sink for 109 wildlife sanctuaries 3
spatial variation 23 Wylfa 154
speciation 20
temporal fluctuations 26 zinc 31-6