ELECTRONIC OFFPRINT

Use of this pdf is subject to the terms described below

This paper was originally published by IWA Publishing. The author’s right to reuse
and post their work published by IWA Publishing is defined by IWA Publishing’s
copyright policy.

If the copyright has been transferred to IWA Publishing, the publisher recognizes the
retention of the right by the author(s) to photocopy or make single electronic copies
of the paper for their own personal use, including for their own classroom use, or the
personal use of colleagues, provided the copies are not offered for sale and are not
distributed in a systematic way outside of their employing institution. Please note
that you are not permitted to post the IWA Publishing PDF version of your
paper on your own website or your institution’s website or repository.

If the paper has been published “Open Access”, the terms of its use and distribution
are defined by the Creative Commons licence selected by the author.

Full details can be found here: http://iwaponline.com/content/rights-permissions

Please direct any queries regarding use or permissions to [email protected]

 Water Policy 18 (2016) 126–142

A decision tool for sustainable agricultural policies: the case of

water saving scenarios for Apulia Region (Southern Italy)
André Daccachea, Daniela D’Agostinoa, Nicola Lamaddalenaa
and Daniel El Chamib
a
International Centre for Advanced Mediterranean Agronomic Studies (CIHEAM-IAM.B), Via Ceglie 9, Valenzano (BA)
70010, Italy
b
Corresponding author. Department of Agricultural Economics, University of the Free State, PO Box 339, Bloemfontein 9301,
South Africa. E-mail: [email protected]

Abstract

The economy of Apulia Region largely depends on agriculture but the scarce water resources are the main factor
threatening the sustainable production of this sector. This paper describes a geographical information system (GIS)
based water balance tool that integrates maps of crops, climate and soil parameters with various scenarios of crop-
ping pattern and farming practice changes. The aim is to assess the implication of these scenarios on the spatial and
volumetric water needs of the region’s irrigated agriculture. The total net volumetric irrigation needs, under current
land use and full irrigation practices, were estimated on an average year to be 973 million m3. The deficit irrigation
practices currently used in Capitanata water districts can save a volume of 302 million m3 if they are extrapolated
over the entire region. Based on the Common Agricultural Policy (CAP), a replacement of 30% of the actual
tomato areas in Foggia Province with sunflowers (energy crop) or durum wheat (rainfed crop) has potential
water saving of 9 million m3 and 67 million m3, respectively. An additional 103 million m3 of water saving
may be obtained through modernisation of the vineyards’ growing practices. Findings of this paper could be
used to address the agricultural policies towards a sustainable use of the scarce fresh water.

Keywords: Agricultural policies; Apulia Region; Deficit irrigation; GIS; Irrigation requirement; Water
balance; Water scarcity

1. Introduction

Irrigated agriculture in the arid and semi-arid areas of the Mediterranean region provides food and
fibre for the local population and the rest of the world and ensures the livelihood of a large number
of rural dwellers. A dwindling water supply and a prolonged period of drought are largely attributed
to climate change. This combined with the continuous increase in water demand driven by demographic
doi: 10.2166/wp.2015.050

© IWA Publishing 2016

 A. Daccache et al. / Water Policy 18 (2016) 126–142 127

and economic development (Hervieu et al., 2006; Al Khabouri et al., 2007; Wolfe, 2009) are the main
challenges for irrigated agriculture in the Mediterranean (Benoit & Comeau, 2005; El Chami et al.,
2014), especially at the farm level (Berbel et al., 2009). Access to new water sources to cope with
the increase in demand in this water-scarce region, is becoming technically, environmentally and econ-
omically difficult if not impossible (Daccache et al., 2014).
Under such conditions, the effective planning of water resources should take place for a sustainable
strategy (Bazzani et al., 2005; Jacobsen et al., 2012) and should be balanced by programmes mainly on
the demand side (Sjömander Magnusson, 2004; Christian-Smith et al., 2012; Aregay et al., 2013).
In actuality, efforts are mainly focused on agriculture which is the largest water consuming sector
(70% of the global fresh water resources) and has the largest water saving potential (Molden, 2007).
Saving just a small amount of the water used for irrigation and directing it to the urban sector could
improve the living conditions of millions of people (Georgoussis et al., 2009).
The quantification of spatial volumetric irrigation needs is essential to both understand and mitigate
the impact of water scarcity under different climatic, economic, social and environmental scenarios
(Valipour et al., 2015).
In this work, the focus will be on how changes in cropping pattern and on-farm practices affect the
water resources of a highly water-stressed agricultural region. A spatialised decision support system tool
incorporating these changes will allow stakeholders to develop a planning strategy for managing the
Apulian water resources that can ensure and maintain economic and environmental sustainability
(Alemu et al., 2011; Ty et al., 2012; Pierleoni et al., 2014).
A geographical information system (GIS) based computational model performing a monthly water
balance was developed and used to quantify the potential pressure that might cause changes in agricul-
tural policies and farming techniques on the water resources of the region. The obtained results could be
used by local government and stakeholders to address agricultural policies and regulate water abstraction
licensing towards an efficient and sustainable water use strategy.

2. The study area

Southern Italy typifies a semi-arid Mediterranean climate characterised by hot and dry summers and
wet winters. The annual precipitation in Apulia is on average 600 mm but ranges spatially from less than
500 mm to more than 900 mm.
Water scarcity is the major constraint to the social and economic development of Apulia. At the
national level, Apulia is the region with the lowest available water resources per capita (IRSA-CNR,
1999). Other than the Ofanto River, the 20,000 km2 region is almost completely deprived of surface
water. For this reason, one of the largest water conveyance and distribution networks in Europe
(Apulia Aqueduct) was built in the 1930s to transfer large water volumes from bordering regions (Cam-
pania and Basilicata) to satisfy the urban and irrigation needs of the region (Figure 1). With the region
highly dependent on imported water, a complex legislative system for water resources transport and
management was enacted to avoid any water disputes with the neighbouring regions.
The flat landscape of Apulia is predominantly (72%) agricultural lands with the largest crops being
cereals (33%) and olive trees (25%) followed by vineyards, fruit trees and vegetable crops. Cereals and
vegetables are mainly grown in the fertile zone of Foggia, where water is made available to the farmers
through a large irrigation infrastructure managed by the Consortium of Capitanata. Olive trees and
128 A. Daccache et al. / Water Policy 18 (2016) 126–142

Fig. 1. The five administrative provinces of Apulia.

vineyards dominate the central and southern parts of the region and also where irrigation relies exclu-
sively on groundwater abstraction. Although, the crop water demand in the dry semi-arid climate of the
region is relatively high, only 20% of the agricultural fields are irrigated and the main limitation for irri-
gation expansion is the lack of water resources. This explains the popularity of the drip irrigation system
that is more often used by far than all other type of irrigation systems in the region.
The imbalance between demand and availability of water in large parts of the region puts lot of
pressure on farmers relying on irrigation for their production and on those looking for irrigation to
expand and develop their businesses. Farms served by collective irrigation networks face the prospect
of not being able to fully irrigate crops while those relying on groundwater will have to deepen their
wells which has serious implications on the environment and on abstraction costs. The groundwater
levels of wells monitored by the Irrigation Development and Land Transformation Agency in Apulia
and Lucania’ have shown a decreasing trend in various parts of the region (De Girolamo et al.,
2002) with an increase in sea water intrusion in coastal areas (Polemio & Limoni, 2001).

3. Datasets

The regional water balance model used for this work was implemented in Visual Basic incorporating
the GIS component of MapWindow. The model extracts pedoclimatic and land cover data from different
 A. Daccache et al. / Water Policy 18 (2016) 126–142 129

input layers to generate volumetric maps of crop water needs across the region. The model will then be
used to assess potential pressures on water resources that may arise from adopting different agricultural
policies or irrigation practices. A summary of the datasets and the model algorithm is presented hereafter.

3.1. Climatic dataset

Monthly gridded (2
2 km) rainfall and reference evapotranspiration (ET0) maps were obtained by
interpolating the long term average data (1950–1992) of the 162 weather stations of the Servizio Idro-
grafico Nazionale. The spherical function of the Kriging method was found to best match the semi-
variance for most of the weather variables of Apulia (Todorovic & Steduto, 2003).
Due to a lack of complete weather data, the empirical formula of Hargreaves–Samani (Hargreaves &
Samani, 1982) has been used to calculate the reference evapotranspiration (ET0) from minimum and
maximum temperatures. Razzaghi & Sepaskhah (2009) have assessed nine different equations for
ET0 estimation using lysimeter data and found that the FAO-Radiation method ( Jensen et al., 1990)
and the Hargreaves–Samani method were the most effective for estimating the mean monthly ET0 in
a semi-arid environment.
Similar findings were reported by Valipour (2015) showing that the modified Hargreaves–Samani
method provides better estimation of evapotranspiration than any other temperature-based models. A
summary of the rainfall and ET0 maps used to run the model are presented in Figure 2.

3.2. Land cover dataset

The SIGRIA project (INEA, 1999)1 used satellite images to develop further the CORINE land cover
map (EEA, 2000) of the region. The main contribution of SIGRIA is the identification of the geographi-
cal location of fields with access to water for irrigation. A summary of SIGRIA outcomes is presented in
Table 1.
Most of the land cover units ‘j’ (LCU) of SIGRIA are not detailed to single crop level. Some LCUs
(i.e. vegetables…/ fruit trees…) encompass a series of individual crops ‘i’ (tomato, potato, lettuce…/
apple, peach…) each with different growing seasons and cropped area. Hence, the water requirement
under the same pedoclimatic conditions of one LCU might differ depending on the proportion of indi-
vidual crops within.
To overcome this problem and for better accuracy in the crop water requirement, agricultural census
data (ISTAT, 2007) showing the cropped area (Ai) of each individual crop ‘i’ by province were extracted
and appended to the land cover unit. Accordingly each LCU ‘j’ is then represented by a fictive crop with
crop coefficient (Kcj ) and root depth (RDj ) weighted by the area of each crop ‘i’ within that unit, as
follows:
PN
Kci
i¼1P
Ai
Kcj ¼ (1)
Ai

1
SIGRA is an Italian abbreviation for ‘Il sistema informativo territoriale per la gestione delle risorse idriche in agricoltura’
(spatial information system for water management in agriculture).
130 A. Daccache et al. / Water Policy 18 (2016) 126–142

Fig. 2. Monthly average rainfall (mm) and average ET0 (mm) of the case study.

Table 1. Area occupied by each land cover unit according to SIGRIA’s land cover map.
Land cover unit Irrigated (ha (10 m4 m2)) Rainfed (ha (104 m2)) Total (ha (104 m2))
Cereals 58,688 639,131 697,819
Olive trees 110,902 371,257 482,159
Fruit trees 25,532 11,317 36,849
Vineyards 100,176 26,518 126,694
Grass 96,816 137 96,953
Vegetables (spring/summer season) 54,583 – 54,583
Vegetables (summer to autumn/spring season) 34,843 – 34,843
Greenhouses 61 – 61

PN
RDi
i¼1P
Ai
RDj ¼ (2)
Ai

where:
Kci Crop coefficient (Allen et al., 1998) of crop ‘i’ within land cover unit ‘j’;
RDi Root depth of crop ‘i’ within the land cover unit ‘j’ (m);
N Number of different crops ‘i’ within the land cover unit ‘j’.

3.3. Soil dataset

The soil database of Apulia was obtained by analysing circa 4,000 soil samples collected from the
entire region (Steduto et al., 1999; Steduto & Todorovic, 2001). Each sample contains general infor-
mation on the site, such as slope, stoniness, morphology, texture, parent material, soil colour, layer
 A. Daccache et al. / Water Policy 18 (2016) 126–142 131

depth and permeability. This information was necessary to develop a US Department of Agriculture
(USDA) and Food and Agriculture Organization of the United Nations (FAO) soil classification map
at a scale of 1:1,000,000.
Using part of these samples, Cainarca (1998) validated different pedotransfer functions with labora-
tory tested soil hydraulic parameters (Gupta & Larson, 1979; De Jong, 1982; Rawls et al., 1982, 1991;
Saxton et al., 1986; Vereecken et al., 1989). Accordingly he recommended the use, for each soil type, of
the appropriate pedotransfer function. This work was used to estimate the water holding capacity of the
4,000þ soil samples from texture, bulk density and organic matter content. The spatial distribution of
soil hydraulic characteristics was then obtained by taking the average value of the total soil samples
within each soil classification unit (Figure 3).

4. Water balance model

The previously described climate, land cover and soil datasets are used as input for a one dimensional,
monthly, soil water balance in which the soil system is divided into two connected subsystems. The
upper subsystem represents the water dynamics at the root depth, while the second represents the natural
groundwater recharge. This simplistic water balance approach was used since the aim of this work is not
to accurately model the physical process but to present a decision support tool based on crop water
needs for sustainable agricultural policies.
The monthly soil moisture content variation is summarised hereafter:

dwi ¼ Pi  ETci  ROi þ Ii (3)

where:
Pi Precipitation (mm) in month ‘i’;
ETci Crop evapotranspiration (mm) in month ‘i’;

Fig. 3. Water holding capacity per 1 meter soil depth and location of soil samples collected by ACLA 2 project.
132 A. Daccache et al. / Water Policy 18 (2016) 126–142

ROi Surface runoff (mm) in month ‘i’;

Ii Irrigation applied (mm) in month ‘i’.

According to the methodology described by Dean (1970) the effective rainfall for month (Peffi ) that
will contribute to the natural water supply of the crop is obtained through:
 
Pi
Peffi ¼
(125  0:2Pi ) for Pi , 250 mm (4)
125

Peffi ¼ 125 þ 0:1Pi for Pi . 250 mm: (5)

The difference between the monthly rainfall (Pi ) and the net water infiltration (Peffi ) is the surface
runoff (ROi ) for month ‘i’. Due to the flat landscape of Apulia, the hydrological fluxes through the
soil surface were considered vertical and the water table interaction with the unsaturated zones is ignored
due to the deep aquifers of the region.
The infiltration below the root zone (In fi ) is then separated into groundwater recharge (GWi ) and sub-
surface runoff (ROsubi ) using the coefficient of potential infiltration (/) of the geological layers of the
region (Celico, 1986), which depends on the composition and the morphology of the parent material.
Consequently:

GWi ¼ /
In fi (6)

ROsubi ¼ (1  /)
In fi : (7)

Within a drainage basin, sub-surface runoff will spill away and will not contribute to the groundwater
recharge, while within an endorheic basin, ROsubi will be forced to re-infiltrate in the soil rather than
being lost to the sea.
Crop evapotranspiration (ETc ) is defined as the water flux to the atmosphere through soil evaporation
and plant transpiration. In this work, crop evapotranspiration (ETc ) was estimated using the Allen et al.
(1998) methodology where reference evapotranspiration (ET0 ) is adjusted by a correction factor known
as crop coefficient (Kc ):

ETc ¼ Kc
ET0 (8)

Kc depends mainly on the crop type, variety and growth stages. For each individual crop, Kc values
were obtained from a long series of field experiments conducted at the Mediterranean Agronomic Insti-
tute of Bari (Caliandro et al., 1990) and, where experimental data were not available, from the literature
(Allen et al., 1998).
Evaporation from bare soil was calculated by adjusting the reference evapotranspiration (ET0 ) with a
correction coefficient that depends on soil texture, quantity of infiltrated water and on the number of
days between consecutive rainfall events (Allen et al., 1998). The latter were obtained from the analysis
of daily weather data and the average number of consecutive rainy days then the rainfall events were
spatially distributed using Thiessen polygons techniques.
 A. Daccache et al. / Water Policy 18 (2016) 126–142 133

The water balance model starts in January assuming the soil water content is at field capacity. To
reduce any error that might occur from such assumption, the model iterates three times with the
same climatic data and the results of the third iteration are used in this paper. The outcomes of the
model based on the climatic, land cover and farming practices scenarios consist of monthly irrigation
need, runoff and groundwater recharge.

5. Water saving scenarios

5.1. Restricted irrigation practices

The crop evapotranspiration (ETc ) used in the model corresponds to the maximum evapotranspiration
that a given crop can reach when planted and grown under optimal conditions without any water short-
age, nutrient, pest or diseases stresses (Allen et al., 1998). In the field, these conditions are rarely
achieved technically and sometimes economically and so the difference between the actual (ETa ) and
potential crop evapotranspiration (ETc ) is proportional to the level of stress exerted on the plant. Ciollaro
et al. (1993), Lamaddalena et al. (1995) and Khadra & Lamaddalena (2006) all compared the actual
water consumed by the farmers in an irrigation district managed by the Consortium of the Capitanata
(Foggia) with that estimated using the standard crop evapotranspiration methods. They found that farm-
ers use less water than that needed to maintain the maximum evapotranspiration rate. Hence, they
identified a series of irrigation reduction factors (Kr ) for different plant stages in order to reflect the
actual irrigation practices conducted by local farmers (Table 2).
The irrigation season for olive trees in the Mediterranean extends from May to September. Olive is a
drought tolerant crop and this explains the dominance of rainfed production in Apulia. However, numer-
ous studies have shown that high quality fruits and minimisation of the alternate bearing phenomenon
require supplemental irrigation (Goldhamer et al., 1994; Çiçek et al., 2014). Oil content in rainfed trees
was similar or even greater than in irrigated trees (Goldhamer et al., 1994). Therefore, the best manage-
ment practices for olive irrigation consist of maintaining mild to moderate levels of water stress at
different growing stages (Ben-Gal et al., 2009). According to Capitanata farmers’ practices, a deficit
irrigation of 60% during the post-flowering and pre-maturity period (June–July) is well accepted by
olive trees growers (Table 2). In May and August, which correspond to the flowering and filling
stages, the plant is more sensitive to water stress and this is consistent with the findings of Çiçek
et al. (2014) and Goldhamer et al. (1994).
Fruit trees and vegetables are much more water sensitive when compared to olive crops. Also, these
are considered as cash crops and hence the money value obtained from each unit of water applied (m3) is
potentially much higher than that applied on olive trees (D’Agostino et al., 2014; Scardigno et al.,

Table 2. Reduction coefficients (Kr) used to reflect Capitanta farmers’ practices (Khadra & Lamaddalena, 2006).
Crop April May June July August September
Vegetables 0.8 0.85 0.9 0.9 0.7 0
Fruit trees 0.6 1 1 1 0.75 0
Olive trees 0 0.6 0.4 0.4 0.6 0.4
Vineyards 0 0 0.7 0.75 0.5 0.15
134 A. Daccache et al. / Water Policy 18 (2016) 126–142

2014). This further explains Captinata farmers’ behaviour where almost-full irrigation is practised
during May, June and July in an attempt to boost the yield and quality of these high value crops
(Table 2).
Farmers in the Consortium of Capitanata (Foggia) are grouped into water users associations to
manage a large infrastructure of pressurised water distribution systems. Because they share the same
water source, these farmers are well trained to efficiently control their water consumption. Knowledge
sharing, training, water pricing and electronic devices for water abstraction monitoring and control have
played an important role in making these farmers efficient irrigators. Comparatively, farmers with pri-
vate water sources (almost all farmers outside Foggia region), without any restrictions on water access or
any water pricing regulations in place, have a tendency to over-irrigate. In this ‘efficient irrigators’ scen-
ario, it is assumed that all of the irrigators in the region will adopt the same water restriction practices as
those used by Capitanata farmers. It is notable that water restriction is also practised for quality purposes
especially on high quality crops such as olives for oil production and vineyards.

5.2. Land use change scenarios

The proposed land use change scenarios are based on the Common Agricultural Policy (CAP) that
aims to safeguard the environmental and socio-economic aspects of European rural areas. In particular,
the EC regulation n. 1782/2003 introduced direct support for different crops including durum wheat and
energy crops. Under this regulation, sunflower and oilseed rape (suitable for biofuel production) are
identified as the most likely to expand in Italy (Faini et al., 2004) and rainfed wheat is promoted as
a replacement for irrigated agriculture in an attempt to reduce the pressure on the water resources
especially in water stressed regions.
In this context, land use change scenarios will focus on the large tomato production area in Foggia
that is most likely to be affected by the CAP. In the other provinces, perennial crops (mainly vineyards
and olive trees) dominate the landscape, and are less likely to be replaced for technical, economic and
traditional reasons. Therefore, the two cropping pattern change scenarios proposed and applied are the
following:

1. Scenario tomato–sunflower. This scenario assumes a replacement of 30% of the tomato growing area
with sunflower, a representative energy crop.
2. Scenario tomato–durum wheat. In this scenario, 30% of the tomato area will be replaced by rainfed
durum wheat, an important rainfed crop in the region grown largely for pasta production.

5.3. Modernisation technique scenarios

In Apulia, vineyards for table grapes are traditionally grown on overhead trellises characterised by
lush foliage completely shading the ground. Vertical trellises are modern training systems where the
vineyard canopy can be better managed to eliminate the excessive shading that impedes grape ripening,
encourages disease development but has no effect on photosynthesis. As crops predominately lose their
water through their leaves’ stomata, restricted canopy development will have a direct impact on crop
transpiration and hence on water demands.
 A. Daccache et al. / Water Policy 18 (2016) 126–142 135

With drip irrigation prevalent on vineyard irrigation, evapotranspiration (ETr ) under vertical trellis
was adjusted using the ground cover (GC) percentage from the fully ground covered crop evapotrans-
piration (ETc ) as follows (Sharples et al., 1985):

ETr ¼ ETc
(0:1
(GC)0:5 ): (9)

The proposed scenario requires switching all of the table grape vineyards in the region from the domi-
nant, traditional overhead trellis to modern and potentially less water demanding vertical trellis systems.

6. Results and discussion

More than 80% of the Apulian territory is used for crop production. However, while hot and dry sum-
mers characterise the Apulian climate, on average, only 20% of agricultural fields are irrigated, Brindisi
having the lowest percentage of irrigated land and Foggia the highest (Table 3). Water scarcity in the
region is the main factor limiting the development and intensification of crop production, especially
in the central and southern provinces that rely almost exclusively on groundwater.
The total volumetric irrigation need of the entire region, under current cropping patterns and climatic
conditions, was estimated to be 973 million m3 a year, 41% of which is concentrated in Foggia Province
alone, followed by Bari (19%) and Lecce (17%) (Table 3). On average and based on land use and the
pedoclimatic conditions of the region, 2,480 m3 is needed yearly for each irrigated hectare (104 m2).
Most of the irrigated farms in Foggia are within the administrative boundaries of the Consortium of
Capitanata; a water user association that operates and manages a complex infrastructure of water distri-
bution systems designed for on-demand access to pressurised water. The maximum water storage
capacity of the collective systems is 140 million m3 per year which is only 37% of the total irrigation
demand of the province (Table 3). Hence, to satisfy the remaining 63% of irrigation need in the pro-
vince, farmers must depend on groundwater abstraction from private wells (Table 3).
A consortium (water users association) that manages a similar collective water distribution system in
the Province of Taranto is Storna e Tara. Its maximum storage capacity per year is 53 million m3 or 41%
of the total irrigation demand, so the remaining 59% of the total volume can only be abstracted from the
private groundwater wells (Table 3). Therefore, even if the storage capacity of the system is greatly
increased, groundwater will remain the primary source of water because the collective water distribution
systems are only capable of supplying the few privileged irrigators that are located in the geographical
area of the consortium.
The Terre d’Apulia consortium has the capacity to store 16 million m3 of Ofanto River water or 7% of
Bari’s irrigation need, but this volume is for the urban water supply as well. So in times of scarcity,
domestic use is always the priority for the stored water. Further, the irrigation needs of Lecce and Brin-
disi, where surface water (lakes, rivers, etc.) is completely absent, were estimated to be 165 million m3
and 95 million m3 per year, respectively.
As the data indicate (Table 3), groundwater is the main source of water for irrigation (77%) in Apulia
Region. Therefore appropriate management of groundwater resources is crucial, especially in coastal
areas where over-abstraction will lead to sea water intrusion.
Olive orchards and vineyards dominate the landscape of the Apulia Region. These are relatively
drought tolerant crops and are widely grown as rainfed crops across the region. Hence, a large
 136
A. Daccache et al. / Water Policy 18 (2016) 126–142
Table 3. Annual irrigation water requirements (IWR) and estimated annual groundwater abstraction (GWA) for each province of Apulia under full and
controlled irrigation management.
Available
Irrigated area surface water IWR GWA GWA/IWR
Full Controlled Full Controlled
Full Controlled
4 2 3 3 3 3
Province Ha (10 m m ) % Million m % Million m % Million m % Million m % Million m3 % % %
Bari 90,073 23 13 6 184 19 123 18 172 23 110 24 93 90
Brindisi 41,489 11 0 0 95 10 66 10 95 12 66 15 100 100
Foggia 139,585 36 150 69 401 41 291 43 251 33 142 31 63 49
Lecce 75,199 19 0 0 165 17 102 15 165 22 102 22 100 100
Taranto 45,672 12 53 25 129 13 89 13 76 10 35 8 59 40
Total 392,018 100 216 100 973 100 671 100 758 100 455 100 78 60
 A. Daccache et al. / Water Policy 18 (2016) 126–142 137

margin of water saving can be obtained with deficit irrigation without significant compromise on yield
and quality.
This explains a 302 million m3 difference in the volumetric irrigation need of the region between
adopting full and partial crop water need practices that are based on the irrigation practices of Capita-
nata’s farmers as previously described in Table 2. Such deficit practices could be the result of limited
access to water and/or measures for boosting crop quality. The implications of irrigation deficit on yield
and quality were not assessed in this work but we recommend full investigation on these aspects in any
future work.
The results of the scenarios in Foggia that replace 30% of the irrigated tomato cropped land with sun-
flower (energy crop) or durum wheat (rainfed crop) are presented in Table 4. Annually, 9 million m3 and
67 million m3 of water respectively could be saved under the hypothesis of a full irrigation regime. If
deficit irrigation is considered, water saving is fairly lower but remains highly substantial. The values
show the large amount of water consumed by tomato and the large potential water saving at regional
scale if new cropping pattern changes are adopted.
Based on the cropping pattern and climatic demand, the potential water saving varies from 0 to
300 mm spatially across the region (Figure 4). The most remarkable change was observed in Lecce
region with the highest percentage change in water saving (38% saving compared to full irrigation
water requirements (IWR)) (Table 5). Foggia, the province with the largest concentration of irrigated
crops in Apulia (69%), can achieve a water saving of 110 million m3 yearly but this, in terms of percen-
tage of water saving, is the lowest among the provinces (Table 5) that can be explained by the
dominance of relatively water stress sensitive vegetables.
In summary, results in Table 6 show that a switch from horizontal to vertical trellis vineyards across
the region is estimated to save 103 million m3 yearly, with Foggia Province, alone, accounting for
almost half of the savings (46 million m3 per year). From Table 6, the modernisation of Apulia’s vine-
yards will have a significant positive impact not only on yield quality but also on water consumption.

7. Conclusions

The region of Apulia is under severe water stress conditions as are many semi-arid Mediterranean
regions. However, the cropping pattern of the region, dominated by water stress tolerant olive trees,
vineyards and cereals, gives Apulia great water saving potentiality and high adaptability to deficit irri-
gation practices. The deficit irrigation practices already adopted by farmers in the northern part of the
region showed annual water saving of around 300 million m3 when compared to full irrigation. The
modernisation of the vineyards growing techniques would also save up to 103 million m3 annually of

Table 4. Annual irrigation water requirement in the province of Foggia under actual and different land use change scenarios.
Irrigation water requirement (million m3)
Province of Foggia
Scenario Full Controlled % of change
Actual situation 401 291 27
Tomato → Sunflower 392 284 27
Tomato → Durum wheat 333 237 29
 138
A. Daccache et al. / Water Policy 18 (2016) 126–142
Fig. 4. Spatial volumetric irrigation needs (left) and potential water saving (right) using controlled irrigation scenario.
 A. Daccache et al. / Water Policy 18 (2016) 126–142 139

Table 5. Annual water saving resulted from the modernisation of the vineyards growing techniques.
Actual situation of
vineyards Scenario vineyard
Irrigation water Irrigation water
requirement (Million m3) requirement (Million m3)
Province Full Controlled % change Full Controlled % change
Bari 184 123 33 165 112 32
Brindisi 95 66 30 87 62 29
Foggia 401 291 27 355 267 25
Lecce 165 102 38 149 93 38
Taranto 129 89 31 114 81 29
Total 973 671 31 870 615 29

Table 6. Water savings per scenario.

Water savings scenarios (Million m3)
Province Current (Million m3) Controlled Vineyard Sunflower Durum wheat
Bari 184 62 19 – –
Brindisi 95 29 8 – –
Foggia 401 109 46 – 68
Lecce 165 63 15 – –
Taranto 129 40 15 – –
Total 973 303 104 9.2 68

valuable fresh water. The CAP could also play an important role in water saving if tomato growing in
Foggia Province is replaced by energy or rainfed crops.
Groundwater remains the major source of water for agriculture in most of the region due to the
absence of surface water. This means that major control of coastal abstractions is needed in order to
avoid major environmental problems. Pressurised water distribution systems have played an important
role in agricultural intensification with positive impacts on the socio-economic aspects of the rural areas
of Foggia and Taranto.
Still, the methodology described in this study has some limitations. The use of monthly water balance
hides the effect of the temporal distribution of rainfall on surface runoff calculation and consequently on
effective rainfall estimation. The same observation applies to temperature, where daily variation may
affect the cumulative plant evapotranspiration rates and, consequently, the irrigation water requirement.
Nonetheless, as the aim of this work is to assess the agricultural water consumption and groundwater
exploitation of the entire region in terms of quantity, spatial and temporal distribution in an average cli-
matic year, the estimations may be considered adequate and the simplifications adopted in this paper can
be justified (Prajamwong et al., 1997).
The results of this work may be used as a decision support system for water managers and policy-
makers to improve the sustainable management of the territory, to cope with the continuous increase
in water demand by the different user sectors and to manage drought and water quality deterioration.
Given the important role that climate changes play in exacerbating water demand, drought conditions
140 A. Daccache et al. / Water Policy 18 (2016) 126–142

and water quality deterioration, further studies have to be focused on the assessment of the impact of
climate changes in Apulia.

Acknowledgements

The authors would like to acknowledge Apulia Region, which funded this research by means of the
‘Riorientamenti produttivi del territorio agricolo Pugliese per uno sviluppo rurale sostenibile’ project.
They also would like to express a great appreciation to Ms D. F. Glasgow for the constructive sugges-
tions and the proofreading provided.

References
Al Khabouri, A., Muyibi, A. S., Ahmed, M. T. & Kabbashi, A. N. (2007). Integrated water resource management plan
(IWRMP) in Oman: way forward. Water Policy 9(5), 457–473. DOI: 10.2166/wp.2007.021.
Alemu, E., Palmer, R., Polebitski, A. & Meaker, B. (2011). Decision support system for optimizing reservoir operations using
ensemble streamflow predictions. Journal of Water Resources Planning and Management 137(1), 72–82. DOI: 10.1061/
(ASCE)WR.1943-5452.0000088.
Allen, R. G., Pereira, L. S., Raes, D. & Smith, M. (1998). Crop evapotranspiration: guidelines for computing crop water
requirements. FAO Irrigation and Drainage Paper No. 56. FAO, Rome.
Aregay, F. A., Minjuan, Z. & Bhutta, Z. M. (2013). Irrigation water pricing policy for water demand and environmental man-
agement: a case study in the Weihe River basin. Water Policy 15(5), 816–829. DOI:10.2166/wp.2013.005.
Bazzani, G., Di Pasquale, S., Gallerani, V. & Viaggi, D. (2005). Water framework directive: exploring policy design issues for
irrigated systems in Italy. Water Policy 7(4), 413–428.
Ben-Gal, A., Agam, N., Alchanatis, V., Cohen, Y., Yermiyahu, U., Zipori, I., Presnov, E., Sprintsin, M. & Dag, A. (2009).
Evaluating water stress in irrigated olives: correlation of soil water status, tree water status, and thermal imagery. Irrigation
Science 27(5), 367–376. DOI: 10.1007/s00271-009-0150-7.
Benoit, G. & Comeau, A. (eds) (2005). A sustainable future for the Mediterranean. The Blue Plan’s Environment and Devel-
opment Outlook. Part 2, Six Sustainability Issues. Earthscan, London.
Berbel, J., Viaggi, D. & Manos, B. (2009). Estimating demand for irrigation water in European Mediterranean countries
through MCDM models. Water Policy 11(3), 348–361. DOI: 10.2166/wp.2009.043.
Cainarca, C. G. (1998). Verifica dell’applicabilità delle funzioni di pedotrasferimento per la determinazione delle caratteristiche
idrauliche dei suoli. MSc thesis. CIHEAM-IAMB, Bari, Italy.
Caliandro, A., Catalano, M., Rubino, P. & Boari, F. (1990). Research on the Suitability of Some Empirical Methods for Esti-
mating the Reference Evapotranspiration in Southern Italy. European Society of Agronomy, Paris.
Celico, P. (1986). Prospezioni Idrogeologiche. Liguori Editions, Naples, vol. I–II.
Christian-Smith, J., Cooley, H. & Gleick, P. H. (2012). Potential water savings associated with agricultural water efficiency
improvements: a case study of California, USA. Water Policy 14(2), 194–213. DOI: 10.2166/wp.2011.017.
Çiçek, Y., Acar, B. & Direk, M. (2014). Irrigation effect on olive yield and some recommendations for well incomes of pro-
ducers. Journal of International Scientific Publications – Agriculture and Food 2, 106–110.
Ciollaro, G., Lamaddalena, N. & Altieri, S. (1993). Analisi comparativa fra consumi idrici stimati e misurati in un comprensorio
irriguo dell’Italia meridionale. Associazione Italiana di Genio Rurale. Rivista di Ingegneria Agraria 4, 234–243.
D’Agostino, D. R., Scardigno, A., Lamaddalena, N. & El Chami, D. (2014). Sensitivity analysis of coupled hydro-economic
models: quantifying climate change uncertainty for decision-making. Water Resource Management 28(12), 4303–4318.
DOI: 10.1007/s11269-014-0748-2.
Daccache, A., Ciurana, J. S., Rodriguez Diaz, J. A. & Knox, J. W. (2014). Water and energy footprint of irrigated agriculture in
the Mediterranean region. E"nvironmental Research Letters 9(12), 124014. DOI: 10.1088/1748–9326/9/12/124014.
Dean, R. (1970). Irrigation Water Requirements. Technical Release n. 21, USDA-SCS Engineering Division. Available from:
http://www.nrcs.usda.gov/Internet/FSE_DOCUMENTS/nrcs142p2_037289.pdf (accessed 18 August 2014).
 A. Daccache et al. / Water Policy 18 (2016) 126–142 141

De Girolamo, A. M., Limoni, P. P., Portoghese, I. & Vurro, M. (2002). Il bilancio idrogeologico delle idrostrutture pugliesi.
Sovrasfruttamento e criteri di gestione. L’Acqua 3, 33–45.
De Jong, R. (1982). Assessment of empirical parameters that describe soil water characteristics. Canadian Biosystems Engin-
eering 24, 65–70.
EEA (2000). CORINE land cover technical guide. Technical Report n. 40, European Environment Agency (EEA), Copenha-
gen, Denmark.
El Chami, D., Scardigno, A. & Khadra, R. (2014). Equity for an integrated water resources management of irrigation systems in
the Mediterranean: The case study of South Lebanon. New Medit 13(4), 39–45.
Faini, A., Nocentini, G., Bonari, E., Fraga, A., Galli, M., Guidi, W., Picchi, G., Piccioni, E., Sabbatini, T., Villani, R., Pampana,
S., Ginanni, M., Mainardi, M. & Spinelli, R. (2004). Le Colture dedicate ad uso energetico: il progetto Bioenergy Farm.
ARSIA-Agenzia Regionale per lo Sviluppo e l’Innovazione nel settore Agricolo-forestrale, Firenze, Italy. Quaderno
ARSIA n.6: 167.
Georgoussis, H., Babajimopoulos, C., Panoras, A., Arampatzis, G., Hatzigiannakis, E., Ilias, A. & Papamichail, D. (2009).
Regional scale irrigation scheduling using a mathematical model and GIS. Desalination 237(1–3), 108–116. DOI: 10.
1016/j.desal.2007.12.027.
Goldhamer, D. A., Dunai, J. & Ferguson, L. (1994). Irrigation requirements of olive trees and responses to sustained deficit
irrigation. Acta Horticulturae 356, 172–175.
Gupta, S. C. & Larson, W. E. (1979). Estimating soil water retention characteristics from particle size distribution, organic
matter content, and bulk density. Water Resources Research 15, 1633–1635. DOI: 10.1029/WR015i006p01633.
Hargreaves, G. H. & Samani, Z. A. (1982). Estimating potential evapotranspiration. Tech. Note. ASCE Journal of Irrigation
and Drainage Engineering 108(3), 225–230.
Hervieu, B., Capone, R. & Abis, S. (2006). Responsible and sustainable development in the Mediterranean. Analytic Note n.
15. CIHEAM. Available from: http://ciheam.org/images/CIHEAM/PDFs/Observatoire/NAN/nan15_en.pdf (accessed 18
August 2014).
INEA (1999). SIGRIA Sistema Informativo per la Gestione delle Risorse Idriche in Agricoltura. See: http://www1.inea.it/irri/
azione1.cfm (accessed 18 August 2014).
IRSA-CNR (1999). Un futuro per l’acqua in Italia. Quaderno n. 109, Consiglio Nazionale Ricerche, Rome.
ISTAT (2007). Agricultural Census. Italian National Institute of Statistics. See: http://www.istat.it/agricoltura/datiagri/coltiva-
zioni (accessed 18 August 2014).
Jacobsen, S. E., Jensen, C. R. & Liu, F. (2012). Improving crop production in the arid Mediterranean climate. Field Crops
Research 128, 34–47. DOI: 10.1016/j.fcr.2011.12.001.
Jensen, M. E., Burman, R. D. & Allen, R. G. (1990). Evapotranspiration and irrigation water requirements. American Society
of Civil Engineers, Manual of Practice, New York, NY, p. 70.
Khadra, R. & Lamaddalena, N. (2006). A simulation model to generate the demand hydrographs in large-scale irrigation sys-
tems. Journal of Biosystems Engineering 93(3), 335–346. DOI: 10.1016/j.biosystemseng.2005.12.006.
Lamaddalena, N., Ciollaro, G. & Pereira, L. S. (1995). Effect of changing irrigation delivery schedules during periods of lim-
ited availability of water. Journal of Agricultural Engineering Research 61(4), 261–266. DOI: 10.1006/jaer.
Molden, D. (2007). Water for food, water for life: a comprehensive assessment of water management in agriculture. Earthscan,
London and IWMI, Colombo, Sri Lanka.
Pierleoni, A., Camici, S., Brocca, L., Moramarco, T. & Casadei, S. (2014). Climate change and decision support systems for
water resource management. Procedia Engineering 70, 1324–1333. DOI: 10.1016/j.proeng.2014.02.146.
Polemio, M. & Limoni, P. P. (2001). L’evoluzione dell’inquinamento salino delle acque sotterranee della Murgia e del Salento.
Memorie della Società Geologica Italiana 56, 327–331.
Prajamwong, S., Merkley, G. P. & Allen, R. G. (1997). Decision support model for irrigation water management. Journal of
Irrigation and Drainage Engineering 123(2), 106–113. DOI: 10.1061/(ASCE)0733-9437(1997)123:2(106)).
Rawls, W. J., Brakensiek, D. L. & Saxton, K. E. (1982). Estimation of soil water properties. Transactions of the ASAE 25(5),
1316–1320 and 1328.
Rawls, W. J., Gish, T. J. & Brakensiek, D. L. (1991). Estimating soil water retention from soil physical properties and charac-
teristics. Advances in Soil Science 16, 213–234. DOI: 10.1007/978-1-4612-3144-8_5.
Razzaghi, F. & Sepaskhah, A. R. (2009). Assessment of nine different equations for ET0 estimation using lysimeter data in a
semi-arid environment. Archives of Agronomy and Soil Science 56, 1–12. DOI: 10.1080/03650340902829180.
142 A. Daccache et al. / Water Policy 18 (2016) 126–142

Saxton, K. E., Rawls, W. J., Romberger, J. S. & Papendick, R. I. (1986). Estimating generalized soil water characteristics from
texture. Soil Science Society of America Journal 50, 1031–1036. DOI: 10.2136/sssaj1986.03615995005000040039x.
Scardigno, A., D’Agostino, D., El Chami, D. & Lamaddalena, N. (2014). Impacts of climate change on agricultural water
management: application of an integrated hydrological-economic modelling tool in a semi-arid region. In: Agricultural
Cooperative Management and Policy. Zopounidis, C., Kalogeras, N., Mattas, K., van Dijk, G. & Baourakis, G. (eds).
Springer, New York, pp. 199–222. DOI: 10.1007/978-3-319-06635-6_11.
Sharples, R. A., Rolston, D. E., Biggar, J. W. & Nightingale, H. I. (1985). Evaporation and soil water balances of young trickle-
irrigated almond trees. In: Proceedings of the 3rd International Drip/Trickle Irrigation Congress, November 18–21, Fresno,
CA (USA), pp. 792–797.
Sjömander Magnusson, T. (2004). Household responsiveness to water demand management incentives in Windhoek, Namibia.
Water Policy 6(5), 453–471.
Steduto, P. & Todorovic, M. (2001). The Agro-ecological characterisation of Apulia region (Italy): Methodology and experi-
ence. In: Soil resources of Southern and Eastern Mediterranean countries. Options Méditerranéennes. Zdruli, P., Steduto,
P., Lacirignola, C. & Montanarella, L. (eds). Série B: Etudes et Recherches 34. CIHEAM Publishing, pp. 143–158. See:
http://ressources.ciheam.org/om/pdf/b34/01002091.pdf (accessed 18 August 2014).
Steduto, P., Todorovic, M., Caliandro, A., Catalano, M., Quaglino, E., Rusco, E., Stellati, M., Nerilli, E., De Santis, S., Madaro,
A., Pezzato, A., Filomeno, O., D’Amico, O., Fanelli, D. & Cainarca, C. (1999). Agro-ecological characterization of the
Apulia Region: GIS applications and modelling. In: Emerging Technologies For Sustainable Land Use And Eater Manage-
ment. Musy, A., Pereira, L. S. & Fritsch, M. (eds). 2nd Inter-Regional Conference on Environment-Water, Lausanne
(Switzerland), 1–3 September 1999, PPUR. European Society of Agricultural Engineering, p. 13. CD ROM publications.
Todorovic, M. & Steduto, P. (2003). A GIS for irrigation management. Physics and Chemistry of the Earth 28, 163–174. DOI:
10.1016/S1474-7065(03)00023-8.
Ty, T. V., Sunada, K. & Ichikawa, Y. (2012). Water resources management under future development and climate change
impacts in the Upper Srepok River Basin, Central Highlands of Vietnam. Water Policy 14(5), 725–745. DOI: 10.2166/
wp.2012.095.
Valipour, M. (2015). Temperature analysis of reference evapotranspiration models. Meteorological Applications 22(3), 385–
394. DOI: 10.1002/MET.1465.
Valipour, M., Ahmadi, M. Z., Raeini-Sarjaz, M., Sefidkouhi, M. A. G., Shahnazari, A., Fazlola, R. & Darzi-Naftchali, A.
(2015). Agricultural water management in the world during past half century. Archives of Agronomy and Soil Science
61(5), 657–678. DOI: 10.1080/03650340.2014.944903.
Vereecken, H., Maes, J., Feyen, J. & Darius, P. (1989). Estimating the soil moisture retention characteristic from texture, bulk
density, and carbon content. Soil Science 148(6), 389–403. DOI: 10.1097/00010694-198912000-00001.
Wolfe, S. E. (2009). What’s your story? Practitioners’ tacit knowledge and water demand management policies in southern
Africa and Canada. Water Policy 11(4), 489–503. DOI: 10.2166/wp.2009.059.

Received 19 February 2015; accepted in revised form 18 May 2015. Available online 13 July 2015