Science of the Total Environment 615 (2018) 972–989

Contents lists available at ScienceDirect

Science of the Total Environment

journal homepage: www.elsevier.com/locate/scitotenv

Assessment of groundwater salinity and quality in Gaza coastal aquifer,

Gaza Strip, Palestine: An integrated statistical, geostatistical and
hydrogeochemical approaches study
Madhat Farouk Abu-alnaeem a,b,⁎, Ismail Yusoff b, Tham Fatt Ng b, Yatimah Alias c, May Raksmey d
a
Geology Department, Faculty of Science, University of Al-Azhar, Gaza Strip, Palestine
b
Geology Department, Faculty of Science, University of Malaya, 50603 Kuala Lumpur, Malaysia
c
Chemistry Department, Faculty of Science, University of Malaya, 50603 Kuala Lumpur, Malaysia
d
Faculty of Civil Engineering, University Technology Mara, Malaya, 40450 Shah Alam, Selangor, Malaysia

H I G H L I G H T S G R A P H I C A L A B S T R A C T

• A detailed investigation of factors

controlling GW salinization has been
achieved.
• An integrated approach of statistical and
hydrogeochemical analysis has been
applied.
• Water-rock interaction and multiple
hydrochemical processes govern GW
salinity.
• Seawater intrusion, salt water up coning
and Eocene invasion controlled the
salinity.
• Sewage invasion was distinctive under
urban areas.

a r t i c l e i n f o a b s t r a c t

Article history: A comprehensive study was conducted to identify the salinization origins and the major hydrogeochemical pro-
Received 29 July 2017 cesses controlling the salinization and deterioration of the Gaza coastal aquifer system through a combination ap-
Received in revised form 29 September 2017 proaches of statistical and geostatistical techniques, and detailed hydrogeochemical assessments. These analyses
Accepted 29 September 2017
were applied on ten physicochemical variables for 219 wells using STATA/SE12 and Surfer softwares.
Available online 17 October 2017
Geostatistical analysis of the groundwater salinity showed that seawater intrusion along the coastline, and salt-
Editor: D. Barcelo water up-coning inland highly influenced the groundwater salinity of the study area. The hierarchical cluster
analysis (HCA) technique yielded seven distinct hydrogeochemical signature clusters; (C1&C2: Eocene brackish
Keywords: water invasion, C3 saltwater up-coning, C4 human inputs, C5 seawater intrusion, C6 & C7 rainfall and mixing
Seawater intrusion inputs). Box plot shows a wide variation of most of the ions while Chadha's plot elucidates the predominance
Sewage invasion of Na-Cl (71.6%) and Ca/Mg-Cl (25%) water types. It is found that, the highest and the lowest levels of salinization
Hierarchical cluster analysis and the highest level of nitrate pollution were recorded in the northern area. This result reflects the sensitivity of
Eocene invasion this area to the human activities and/or natural actions. Around 90.4% of the wells are nitrate polluted. The
Saltwater up-coning
main source of nitrate pollution is the sewage inputs while the farming inputs are very limited and restricted
mostly in the sensitive northern area. Among the hydrogeochemical processes, ion exchange process was
the most effective process all over the study area. Carbonate dissolution was common in the study area
with the highest level in clusters 6, 7, 4 and 2 in the north while Gypsum dissolution was significant only
in cluster 1 in the south and limited in the other clusters. This integrated multi-techniques research should be

⁎ Corresponding author at: Geology Department, Faculty of Science, University of Al-Azhar, Gaza Strip, Palestine.
E-mail addresses: [email protected] (M.F. Abu-alnaeem), [email protected] (I. Yusoff), [email protected] (T.F. Ng), [email protected] (Y. Alias).

https://doi.org/10.1016/j.scitotenv.2017.09.320
0048-9697/© 2017 Elsevier B.V. All rights reserved.
 M.F. Abu-alnaeem et al. / Science of the Total Environment 615 (2018) 972–989 973

of benefit for effective utilization and management of the Gaza coastal aquifer system as well as for future work in
other similar aquifers systems.
© 2017 Elsevier B.V. All rights reserved.

1. Introduction densely populated areas all over the world where 1.7 million living in
an area of 365 km2 (PCBS, 2013). It has a characteristically semi-arid
The Gaza Coastal Aquifer which is located in an arid to semi-arid re- Mediterranean Sea climate. The limited annual rainfall (200 mm to
gion within the southwestern part of Palestine has been subject to sig- 400 mm) is the main source of recharge (PWA, 2013). The average an-
nificant groundwater deterioration due to unplanned human activities nual potential evaporation is unsteady during the year and ranges be-
as a result of political conflict and strife in the region. tween 2.1 mm/d in December, to 6.3 mm/d in July. This area is
The groundwater salinization and pollution problems are common generally characterized by frequent droughts and lack of rainfall. More-
critical environmental and socioeconomic issues especially in coastal high- over, the high rate of evaporation decreases the aquifer recharge
ly populated areas like the Mediterranean zone (Maguette et al., 2017; (Lazarou, 2016).
Elyaouti et al., 2009). These problems can be attributed to natural factors The monthly average rainfall and temperature for Gaza Strip from
as the interaction of water with rock, hydrodynamic conditions, impact 1991 to 2015 have been drawn to show the climatic change by month,
of adjacent aquifers and seawater intrusion (Ledesma-ruiz et al., 2015; for rainfall and temperature (Fig. 2). This dataset produced by the Cli-
Elgettafi et al., 2012). Adding to it, anthropic activities such as saline matic Research Unit of East Anglia University (Portal, 2017). The rainfall
water up-coning due to groundwater overexploitation, improper sewage season, as shown in (Fig. 2), extends from October until April with the
disposal, intensive farming activities, unplanned industrialization and highest precipitation in January (around 105 mm). The temperature is
urban expansion (Huang et al., 2013). Thus, numerous studies have inves- generally moderate, ranging between 29 °C in August to around 11 °C
tigated the role of natural and anthropogenic sources on processes of sali- in January.
nization and pollution via hydrogeochemical surveys to develop more
sustainable groundwater management methods (Argamasilla et al., 2017).
2.2. Geological and hydrological settings
In the Gaza Strip, a highly populated and extremely water-starved
coastal area, groundwater is the only source of water for all purposes.
The detailed geological information modified from Sneh et al. (1998)
Aquifers within the Gaza Strip are undergoing a severe salinization
is presented in (Fig. 3). This map shows the geological formations of the
and pollution problems as a result of both intensive human activities
Gaza coastal aquifer that are of Quaternary Pleistocene and Holocene
and natural processes (El-naeem et al., 2010; Vengosh et al., 2005) with-
ages. They consist of calcareous sandstone, sand dunes, silt, clay and
in the region. Decades of increasing groundwater over-exploitation and
gravel with a general slope toward the Mediterranean Sea as shown in
poor management have left the aquifer in a tragic state. The excessive
(Fig. 4).
and continuous abstraction of the groundwater has initiated sea water
Rainfall recharge in the Gaza coastal aquifer occurs through the areas
intrusion from the western side of the study area and saltwater up-
of sand dunes, calcareous sandstone, and sandy loam in the central and
coning and lateral movement of brackish water from the eastern one
western parts of the aquifer while the eastern side of the aquifer is
(Dentoni et al., 2014; Qahman and Larabi, 2006). Moreover, the absence
covered by thick deposits of loess soils (up to 20 m depth) that restrict-
of sewer system infrastructure and the intensive use of agrochemicals
ed rainfall percolation due to their lower hydraulic conductivity
have accelerated groundwater salinization and pollution in the region
(Weinthal et al., 2005).
(Mogheir et al., 2006; Qahman and Larabi, 2006; Saleh, 2007; Shomar
The surface water system in the study area is represented by three
et al., 2010). These poor management conditions has resulted in a seem-
Wadis (water seasonal flow valleys), Israel retained their natural flow
ingly irreversible intensification of groundwater quality (Mogheir et al.,
that are originated from the occupied areas. So, fresh surface water
2005; Vengosh et al., 2005).
here is negligible. Unfortunately, the Wadis are used now for sewage
In Gaza Strip, typically, hydro-geochemical methods are widely used
discharge and solid waste dumping (Rabah, 2013).
to explore and establish the process, the extent and the sources of
Gaza coastal aquifer which is the only source of water in the Gaza
groundwater salinization and pollution. However, the multiplicity of
Strip, is located along the Mediterranean coast from the foothill of
sources of salinization and the influence of hydro-geochemical reactions
Mount Carmel at the southward of Gaza Strip, to the north of the Sinai
and groundwater mixing render a high degree of complexity to the task
Peninsula in Egypt (Zilberbrand et al., 2001). It is a shallow phreatic
for identifying the origin of groundwater salinity.
with a natural dip and a flow that is toward the sea.
Since the groundwater chemistry is differ geographically, application
A schematic hydrogeological cross-section of the Gaza Coastal Aqui-
of statistical and geostatistical tools in conjunction with hydrogeochem-
fer and the surrounding region modified from (Mercado, 1980;
ical approaches is a logical integrated work that can be useful to elucidate
Weinberger et al., 1994; Melloul and Azmon, 1997) is depicted in
the hidden patterns in the physicochemical data (Güler et al., 2012). The
(Fig. 4). Close to the present shoreline, the sequence of the Kurkar
purposes of this integrated work are to understand the salinization ori-
Group attains an average thickness of 180–200 m in the north and
gins and the hydrogeochemical processes controlling the salinization
around 100–120 m in the south, wedging gradually out toward the foot-
and deterioration of the Gaza coastal aquifer system using software
hills of the Judea and Samaria Mountains to a few meters in the east
(Stata/SE 12 and Surfer 13 Golden software Inc.). The importance of
(Vengosh et al., 2005; Zilberbrand et al., 2001). It varies in width from
this study is to improve reliable aquifer management plan for decreasing
7 km in the north to 20 km in the south. The aquifer material consist ba-
the groundwater salinity and pollution problems in the future.
sically of sand (quartz rich), calcareous sandstone, and silt (Kurkar for-
mation) that is interbedded with clays layers and lenses (Vengosh and
2. Study area Keren, 1996). It overlies the layers of chalks of Eocene age at the eastern
side of the study area and thick layers of marls, marine shales, and clays
2.1. Geographical setting (about 400- to 1000-m) of Miocene–Pliocene age (called Saqiya Group)
to the middle and western side of the study area.
The Gaza Strip is located in the southern part of the Mediterranean That groundwater over-exploitation caused a deficiency in water
Coastal Plain of Palestine as shown in (Fig. 1). It is one of the most balance between the total water input (about 55–60 MCM/yr) and the
974 M.F. Abu-alnaeem et al. / Science of the Total Environment 615 (2018) 972–989

Fig. 1. Location map of Gaza Strip.

total water output (about 200 MCM/yr for both domestic and agricul- Numerous researches have been discussed the hydrochemistry of
ture use) (PWA, 2014). Consequently, two regional cone of depressions the Gaza aquifer. It is commonly believed that, the natural flow of Eo-
were formed in the southern and northern areas with lowest level of cene brackish water and sea water intrusion, were highly influenced
− 19 m and − 5 m below mean sea level respectively as shown in the salinity of the Gaza aquifer (Ghabayen et al., 2006; Melloul and
(Fig. 5) (PWA, 2014). Collin, 2000; Vengosh et al., 2005). A more recent researches considered
the biggest threats to the groundwater salinity in the study area are sea-
water intrusion and up-coning of saltwater (Al-Yaqubi et al., 2007;
Baalousha, 2011). Other researchers have discussed the role of lack of
sewer system, improper sewage disposal (using cesspits) and intensive
use of agrochemicals that have contributed to the deterioration of
groundwater quality in the region (Al-Agha and El-nakhal, 2004).
Elamassi (2012) noted that three factors controlled the chemistry of
groundwater in Gaza governorate (sea water intrusion, carbonate disso-
lution and nitrate pollution). According to (Vengosh et al., 2005), the
hydrochemistry of the Gaza aquifer was divided into three types. The
high Na+/Cl− ratio chemical signature type is recorded along the eastern
part toward the lateral movement of Eocene brackish water. The second
type is characterized by a Ca-Cl composition with low Na+/Cl− ratio
(b0.86) and recorded in the western part of the aquifer along the coast-
line. The third type is distinguished by pollution with high NO− 3 (up to
450 mg/L). Groundwater salinity is generally lower in the northern
part of the Gaza Strip (Aliewi et al., 2001) while the nitrate pollution is
lower in the southern area and higher in the northern one. Moreover,
Loess soils deposits which contains calcium carbonate as a major mineral
and other minerals as gypsum and halite could be a source of saliniza-
tion. The high sulfate content in groundwater in the southern areas is at-
Fig. 2. The monthly average rainfall and temperature of Gaza Strip. tributed to dissolution of gypsum in the loess soil (Vengosh et al., 2005).
 M.F. Abu-alnaeem et al. / Science of the Total Environment 615 (2018) 972–989 975

Fig. 3. Geological map of Gaza Strip.

3. Methodology (do not show a normal distribution). Therefore, a log transformation was
performed to generate a normal distribution of the data, which is the
3.1. Data collection and preparation precondition of multivariate statistical analysis for improved confidence
in the results and interpretations (Davis and Sampson, 1986; Güler et al.,
This study utilized a comprehensive set of existing data which in- 2002; Wang et al., 2015). Standardization of log-transform data was also
cludes physicochemical water variables that have been obtained from applied prior to the multivariate statistical analysis to maintain that
the Palestinian Water Authority (PWA). These variables for 219 wells every variable has equal weight in the statistical analysis (Davis and
are (TDS, Ca2+, Mg2+, Na+, K+, Cl−, SO2− − −
4 , HCO3 , NO3 and pH). Statis- Sampson, 1986; Güler et al., 2002). This standardization process assem-
tical and geostatistical analysis of the data are performed using software bles the data as unitless values to avoid the effects of value dimensions
(Stata/SE 12 and Surfer 13 Golden software Inc.). (unit consistencies) before statistical analysis (Güler et al., 2002;
Protocol of data collection, preparation and applying of statistical Machiwal and Jha, 2015). For multivariate statistical analysis, Log-
(including univariate and multivariate statistical analysis), geostatistical transformation of data for standardization protocols is a common ap-
techniques and hydrogeochemical approaches is shown in (Fig. 6). For proach (Güler et al., 2002). All calculations and statistical analyses were
charge balance evaluation of chemical analyses the concentrations of conducted using STATA/SE 12 software package.
the various constituents are recalculated from mg/L to meq/L. Only
those water samples whose values of charge balance errors were within 3.2. Statistical analysis
±5% are used in this study to combine the various ions in a chemically
meaningful method (Nwankwoala and Udom, 2011). 3.2.1. Multivariate cluster analysis
All the physicochemical variables (TDS, Ca2+, Mg2+, Na+, K+, Cl−, Multivariate cluster analysis is a statistical technique used to simpli-
− −
SO2−
4 , HCO3 , NO3 ) (except pH) are positively skewed in the distribution fy complex systems and rearrange huge datasets. It is used for getting a
976 M.F. Abu-alnaeem et al. / Science of the Total Environment 615 (2018) 972–989

Fig. 4. A hydrogeological cross-section of the Gaza Coastal Aquifer and the surrounding layer.

comprehensive evaluation and significant information that may be on Euclidian distance together with Ward's method for linkage to
useful for water chemistry assessment (Cloutier et al., 2008). So, many generate the most distinct hydrogeochemical signature clusters for
latent sources and processes in a complex hydrogeochemical environ- water samples.
ment can be deduced through these advanced statistical methods
(Güler et al., 2012). 3.2.2. Geostatistical analysis
Hierarchical cluster analysis (HCA) is one of the most popular multi- The spatial distribution of the TDS, was displayed using kriging spa-
variate cluster analysis technique that can be applied successfully for di- tial interpolation technique in Surfer 13 to show the distribution of this
viding the samples into similar groups (clusters) (Cloutier et al., 2008). variable and predict the values surrounding this measured variable. The
The splitting into clusters is based on their similarity in chemical com- clusters spatial pattern was also displayed as map. The importance of
position within a cluster and dissimilarities between different clusters these maps are: 1) To determine the areas at risk of salinization and pol-
(Güler et al., 2012; Huang et al., 2013). This means that, the chemical lution and the already high salinized and polluted areas 2) To investi-
variables of each water sample of the cluster be similar to its associated gate the factors responsible on these patterns. 3) To display the
water samples more closely than any other water sample from the other relationships between the clusters spatial location and potential influ-
clusters (Güler et al., 2002). The combining process starts with as many ence due to both natural and human factors. 4) To offer important infor-
groups of water samples as individuals. The groups are then linked, ac- mation for best planning and management decisions.
cording to their similarity, until only one remains (Templ et al., 2008).
The result of the multivariate hierarchical cluster analysis can be ob- 3.3. Hydrogeochemical approaches
served as a dendogram (tree diagram) (Davis and Sampson, 1986),
whose roots are all the water samples. The properties of the clusters Hydrogeochemical approaches using various graphical representa-
can be distinguished using descriptive statistics and graphical tions such as Box plot, Pie plot, modified Piper plot (Chadha's plot),
representation. Chloro-alkaline index (CAI) and scatter plots of ions and ionic ratios,
Both of Q or R-modes can be applied for clustering. Q-mode is are useful for visual examination of hydrogeochemical data for recog-
usually applied to reveal the interactions among the studied samples nizing specific patterns and trends. These traditional methods could
while R-mode is used to reveal the interactions among the water quality be highly beneficial in determining hydrogeochemical water types,
variables (Davis and Sampson, 1986). In this research, Q mode is based identifying the origin of salinity and pollution and understanding the
 M.F. Abu-alnaeem et al. / Science of the Total Environment 615 (2018) 972–989 977

hydrogeochemical processes that influence the salinity and quality of

groundwater (Hiscock, 2009). These methods could show better and
more detailed results when they are applied on the clustered data.

3.4. Hydrogeochemical conceptualization of the Gaza coastal aquifer

system

The conceptualization of the typical hydrogeochemical conditions of

the Gaza coastal aquifer in a two-dimensional cross-sectional view,
could be helpful in simplifying and understanding the various sources
and processes of salinization influencing the study area. The hydrogeo-
chemical signature of these factors can be summarized in six zones as
following:

3.4.1. Seawater intrusion

Seawater intrusion (SWI) is the most common and serious saliniza-
tion factor in coastal aquifer areas (Argamasilla et al., 2017) which have
a hydraulic connection to the sea. Naturally, groundwater of coastal
aquifers flows toward the low places as seas and oceans with the
potential of gravity. The hydrological changes in the coastal area
can cause inland seawater movement in a process called seawater
intrusion (Wetzelhuetter, 2013). The chemical signature of seawater
(zone 1) (Fig. 7) is distinguished by very high TDS up to 38,600 mg/L,
Na-Cl water type, Low Na/Cl ratios (=0.86) and Mg/Ca ˃ 5 with relative
abundance of ions as: Cl− N Na+ N SO2− 4 N Mg
2+
N Ca2+ N K+ N HCO− 3
(M&E, 2000; Mercado, 1980; Vengosh et al., 2005).

3.4.2. Saltwater up-coning

One of the greatest threats to coastal aquifers is the up-coning of nat-
ural deep brine water that underlies the fresh groundwater bodies due
to over-exploitation of fresh groundwater in coastal regions (Wang,
Fig. 5. Water level map of Gaza Strip (PWA, 2014). 2011). This source of groundwater salinization has been recorded in
the form of high saline plumes in inland areas for some Mediterranean

Fig. 6. Methodological protocol.

978 M.F. Abu-alnaeem et al. / Science of the Total Environment 615 (2018) 972–989

Fig. 7. Conceptual hydrogeochemical model of the Gaza coastal aquifer in a cross-sectional view showing the various sources and processes of salinization.

coastal aquifers as in Palestine (Al-Yaqubi et al., 2007; Melloul concentration in NO−

3 and lower in Cl
−
may indicate farming inputs
−
and Collin, 2000). The chemical signature of saltwater up-coning commonly without Cl component (Wang et al., 2016).
(zone 2) (Fig. 7) is different by higher TDS up to 115,000 mg/L TDS up
to 38,600 mg/L, Ca-Cl water type, Na/Cl b 0.86, Mg/Ca ˃ 1, K/Cl b 0.019,
3.4.5. Rainfall recharge
SO4/Cl = 0.05 (Vengosh et al., 2005; M&E, 2000; Mercado, 1980). Rela-
In coastal regions, rainfall contents are similar to the components of
tive to seawater characteristics, deep saltwater is higher in Ca content
seawater (Na-Cl water type with same ionic ratios of seawater) but
and lower in K. The higher Ca content is produced by reverse ion ex-
TDS is very low. Hence, the expected chemical signature of rainfall
change process with aquifer clay material and carbonate dissolution
(zone 6) (Fig. 7) could be Na-Cl water type with Na/Cl lower than unity.
process (Vengosh and Rosenthal, 1994). The lower K content in the
deep saltwater may be caused by adsorption of K+ ions on aquifer
clay material. 4. Results and discussion

4.1. Groundwater chemistry

3.4.3. Eocene brackish water
The presence of adjacent brackish aquifer in a hydraulic connection
A statistical summary of the physicochemical variables of the
with fresh groundwater can be considered as a natural threat to the
groundwater wells is shown in (Table 1). It is found that groundwater
quality of freshwater aquifers. In the study area, the Eocene chalk and
was approximately neutral to slightly alkaline with average pH 7.6
limestone aquifer is located in the eastern side of the Gaza coastal aqui-
and ranges from 6.7 to 8.3. This alkalinity was derived from the carbon-
fer. The flow of Eocene water toward the Gaza aquifer increased the sa-
ate minerals dissolution in the form of bicarbonate (HCO− 3 ). Based on
linity of the groundwater in its eastern parts (Ghabayen et al., 2006;
World Health Organization (WHO) standards (WHO, 2011), pH values
Qahman and Larabi, 2006). The chemical signature of the Eocene
of all the wells are within the WHO standards. The importance of
water (zone 3) (Fig. 7) is Ca-rich water type where the aquifer material
pH value comes from its sharing role to determine which minerals
is Chalk and Limestone, medium TDS (up to 5000 mg/L), Na/Cl = 0.86,
will dissolve, precipitate or remain on the groundwater system
Mg/Ca b 1, K/Cl b 0.019 and SO4/Cl = 0.05 (Aliewi et al., 2001; Vengosh
(Vinayachandran & Narayana, 2014). The electrical conductivity
et al., 2002).
(EC) ranges between 597 and 30,400 μS/cm with an average value of
4245 μS/cm. A wide difference in TDS values recorded as 370 to
3.4.4. Human inputs 18,848 mg/L with an average of 2634 mg/L with the highest values
Anthropogenic inputs that contribute to groundwater salinization close to the coastal line in the north and the lowest values also in the
and pollution include sewage and farming inputs. Absence of proper north but inland position.
wastewater treatment and sewage disposing in open areas as in Wadi Among the ions, Cl− ion has the highest variations in concentrations
Gaza could be considered as one of the main sources of groundwater sa- (78 to 10,318 mg/L with an average of 1101 mg/L) with 80% of the wells
linization and pollution in the Gaza Strip (M&E, 2000). The chemical sig- exceeded the WHO standards. Na+, the second major ion, ranges from
nature of sewage input (zone 4) (Fig. 7) are characterized by NO3 rich 41 to 5400 mg/L with an average of 621 mg/L. Around 75.4% of
water type, low TDS (up to 1200 mg/L), Na/Cl = 1.1 and high SO4/Cl the wells are above the standards. Mg2 + and Ca2 + vary from 23 to
ratio (0.09) (Vengosh and Rosenthal, 1994; Vengosh et al., 1999). The 665 (average 109 mg/L) and from 25 to 657 (average 133 mg/L) respec-
chemical signature of farming inputs (zone 5) (Fig. 7) are distinguished tively. The percentages of the wells with high concentration above the
−
by unique Ca-Mg-NO3 water type (Vengosh et al., 2005), low TDS, Na/Cl standards are 17.8% and 15.5% respectively. SO2−4 and HCO3 concentra-
ratio ˃ 1, high SO4/Cl ratio due to the use of gypsum fertilizers (Vengosh tions vary from 8 to 1604 (average 257 mg/L) and 101 to 1280 (average
et al., 2002). Areas of sewage inputs could be differentiated from the 278 mg/L) with 40% and 20.5% of the wells exceeded the standards re-
other farming inputs by Cl− and NO− 3 concentrations. The area with spectively. NO− 3 is high in the research area and ranged from 18 to
higher concentration in Cl− and lower in NO− 3 may suggest the influ- 496 (average 123 mg/L) where 90.4% of the wells were nitrate polluted.
ence of municipal sewage and animal manure, while others with higher K+ is distinguished by low concentrations relative to other cations;
 M.F. Abu-alnaeem et al. / Science of the Total Environment 615 (2018) 972–989 979

Table 1
Statistical summary of the ten physicochemical parameters.

2013/2014

Statistical TDS pH Ca2+ Mg2+ Na+ K+ Cl− NO−

3 SO2−
4 HCO−
3

Mean 2634 7.6 133 109 621 10.9 1101 123 257 278
Max 18,848 8.3 657 665 5400 155 10,318 496 1604 1280
Min 370 6.9 25 23 41 1.4 78 18 8 101
%NWHO 78.1 0 15.5 17.8 75.4 7.3 80 90.4 40 20.5

All units in mg/L except pH (unitless).

ranging from 1.4 to 155 with an average of 10.9 mg/L. 7.3% of the wells the direct impact of rainfall. This pattern reveals the sensitivity of the
are above the permissible limit. northern area to the influence of the external factors as seawater intru-
sion, saltwater up-coning and the rainfall recharge (sand dunes area).
4.1.1. Spatial distribution of groundwater salinity Scattered plumes of high TDS have been recorded inland with values
Geostatistical analysis through constructing maps is a common exceeding 6000 mg/L in the north and 7000 mg/L in the south. These
method to understand the spatial pattern of the salinity in the research plumes may be formed by up coning of saltwater in a highly populated
area. The constructed TDS map (Fig. 8), using kriging method in Surfer and urbanized region of heavy groundwater pumping. Moreover, the
13 software, shows a very wide spatial difference in TDS values (370 over-exploitation increased the hydraulic gradient from the eastern
to18848 mg/L) that reflects the occurrence of a severe salinization pro- side especially in the southern part and accelerated the invasion of Eo-
cess in some parts of the study area. The highest TDS values have been cene brackish water from the eastern part of the study area. Finally,
recorded along the coastline especially in the north between Gaza city the impact of intensive human activities such as sewage infiltration
and Jabalia and in the south between Khan Younis city and Dier Al and farming inputs may contribute to the deterioration of groundwater
Balah. These high records under urban areas can be attributed mainly quality and increase the salinity in the study area.
to seawater intrusion and saltwater up-coning in the study area. As Based on the TDS classification by (Rabinove et al., 1958), water sa-
the highest values were recorded in the north (along the coastline), linities are classified into four groups of freshwater (TDS b 1, 000 mg/L),
the lowest ones were also recorded in the north (inland areas) under slightly saline (1000 to 3000 mg/L), moderately saline (brackish) (3000

Fig. 8. Spatial pattern of TDS (mg/L) in Gaza coastal aquifer.

980 M.F. Abu-alnaeem et al. / Science of the Total Environment 615 (2018) 972–989

to10,000 mg/L) and very saline (TDS N 10,000 mg/L). In the study area, it 4.2.1. Clusters mapping
is found that 21.9% of the wells are freshwater, 58.5% slightly saline, The spatial distribution of the seven clusters with Pie plots represen-
15.4% brackish and 4.2% highly saline. tation of the relative percentage of their average major ions (in meq/L)
for every cluster are shown on the map (Fig. 10). This map can be help-
ful to determine the location and general chemical signature of the wells
4.2. Hierarchical cluster analysis of hydrochemistry of each cluster. Consequently, we can correlate between their chemistry
and the type of human activities and/or natural actions that predomi-
The results of the multivariate hierarchical cluster analysis can be nant in that location. Therefore, each cluster or a set of clusters should
observed in the dendrogram (tree diagram), whose roots are all the be impacted hydrogeochemically by the particular activity or action.
219 water samples (Fig. 9). The seven branches (clusters) of the dendro- The distribution of clusters specific with the differences in their Pie
gram could be increased or decreased by moving the phenon line down plots reflects variation in their hydrogeochemical signatures. The pres-
or up on the dendrogram (the higher the phenon value, the lower the ence of cluster 1 and 2 at the eastern side facing the lateral movement
number of the branches). The seven clusters are yielded after selecting of Eocene brackish water differs from the grouping of wells of cluster
the phenon line with a linkage distance of 25. Thus, water samples 5 in the north and close to the coastline within and adjacent to a highly
with a linkage distance below 25 are grouped into the same cluster. populated and urbanized region of heavy groundwater pumping. Also,
The phenon value selection is based on getting the lowest number of the distinct location of cluster 4 under the greatest population density
clusters that clarify most satisfactory results of the difference in hydro- (Gaza City) differs from the wells of cluster 6 which are located inland
geochemical characters of the water samples (Cloutier et al., 2008; in the northern portion of the study area (sandy and rainy area). The
Monjerezi, 2012). Each cluster represents a distinct hydrogeochemical direct impact of recharge via rainfall are expected to have a greater in-
signature (facies) that reflects the effect of a hydrogeochemical process fluence on the chemical signatures of wells of cluster 6. Moreover, the
or a combined of hydrogeochemical processes on the samples of this wells near the coastline (i.e. cluster 3) are expected to be associated
cluster. with heavy groundwater pumping can initiate saltwater up-coning
The level of dissimilarity in chemical signature between the seven and may denote the stark differences from the scatter distribution of
clusters can be deduced from the linkage distance between the clusters wells within cluster 7 that are located inland under a mixing influence
(Fig. 9). The increase in the linkage distance reveals a higher hydrogeo- of anthropogenic and natural processes.
chemical difference and a wider variation in the hydrogeochemical pro- The yielded seven clusters have different physicochemical charac-
cesses in the study area. Samples from C6 and C7 are linked to the other ters that are shown descriptively (Tables 2) and graphically as Pie
five clusters at a high linkage distance, reveal that these clusters plots in meq/L (Fig. 10) and Box plots in mg/L (Fig. 11). The box plots de-
hydrogeochemically differ from the other five clusters. Moreover, C6 pict the variation of the 8 groundwater quality variables (Ca2+, Mg2+,
and C7 have the lowest linkage distance between the seven clusters, Na+, K+, Cl−, SO2− − −
4 , HCO3 and NO3 ) within the seven clusters using
and thus, have the lowest dissimilarity among all clusters. It can thus the minimum, the quartiles and the maximum values of these variables
be concluded that the hydrochemistry of C6 water samples would in meq/L (Fig. 12) and in mg/L (Fig. 11) along with the permissible stan-
have similarities with the ones of C7. Among the remaining five clusters, dards for drinking water prescribed by WHO. The physicochemical
C5 is the highest in dissimilarity with the other four clusters C1–C4 as it characterization (signatures) of the seven clusters is discussed as
has a high linkage distance to them. The degrees of hydrogeochemical following:
dissimilarities between the clusters in these pairs (C3 and C4) and
(C1 and C2) are middle (moderate linkage distance). These wide varia- 4.2.1.1. Cluster 1 (Eocene, carbonate and urban inputs). Cluster 1 includes
tions in the linkage distance between clusters reflect a wide variation in a high number of wells (40) that are located mostly in the south toward
the hydrogeochemical processes in the study area. the eastern boundary along the lateral invading Eocene seawater region

Fig. 9. Dendrogram shows how the seven clusters (C1-C7) are linked hierarchically. The linkage distance elucidate the degree of the dissimilarity between the clusters. The red line
(phenon line) cut the dendrogram at the desired level 25, then each connected group of wells forms cluster.
 M.F. Abu-alnaeem et al. / Science of the Total Environment 615 (2018) 972–989 981

Fig. 10. Spatial distribution of the seven clusters with their Pie plot of the major ions and Pie plot for relative % of the average TDS in a location map of the Gaza Strip which includes the
three Wadis and the spatial distribution of the main land uses areas. Hydrogeochemical Characterization of the Clusters.

as shown in (Fig. 10). The wells within this cluster are generally charac- groundwater chemistry in the area has been impacted by carbonate dis-
terized by high concentrations of Cl−, Na+, SO24 − and NO− 3 (NWHO solution and associated increased level of HCO−
3 .
standards) (WHO, 2011) while Ca2 +, Mg2 +, HCO− 3 and K
+
were
˂WHO standards (Table 2; Fig. 11). The salinity ranged from 1515 to 4.2.1.3. Cluster 3 (saltwater up-coning and human inputs). Cluster 3 is
3536 mg/L with an average of 2492 mg/L. The highest abundance of sul- comprised of a relatively low number of wells (24) that are located
fate is originated predominantly from the Eocene aquifer and the gyp- near from the coastline (Fig. 10). This cluster is specifically distin-
sum dissolution in the overlying loess soil (Vengosh et al., 2005) and guished by the high salinity where TDS values range between 2220
from the infiltrated sewage and farming inputs (Abu-Jabal et al., 2014; and 8187 mg/L with an average of 4208 mg/L (Table 2). The concentra-
Marghade et al., 2012). tions of Na+, Cl−, SO24 − and NO− 3 were above WHO standards while
HCO− +
3 and K values were below these standards. Values of Ca
2+
and
4.2.1.2. Cluster 2 (Eocene invasion, carbonate dissolution and human in- Mg2+ were above and below the standards. Such groundwater chemi-
puts). Cluster 2 contains (35) wells that are located mostly in the cal signatures reflect the potential presence and impact of saltwater
north toward the eastern boundary where lateral invading Eocene up-coning on the on the salinity of wells of this area.
water occurs (Fig. 10). The wells within this cluster are generally char-
acterized by high concentrations of Cl−, Na+, HCO− −
3 and NO3 (N WHO 4.2.1.4. Cluster 4 (mixing human and natural inputs). Cluster 4 contained
standards) while Ca2 +, Mg2 +, SO24 − and K+ are ˂WHO standards as (30) wells that are located in the urbanized area of Gaza City with the
shown in (Fig. 11). The salinity ranged from 985 to 3078 mg/L) with highest population density within the study area (Fig. 10). They are
an average of 1825 mg/L. Cluster 2 is higher in TDS and in SO2− 4 than characterized by concentration of Cl−, Na+ and NO− 3 which exceed
cluster 1 and lower in HCO− −
3 ions. The HCO3 in cluster 2 reflects the im- WHO standards, while Ca2+, Mg2+, and predominantly K+ concentra-
properly treated waste water associated with infiltration basins of the tions are below WHO standards. Values of SO24 − and HCO− 3 were
North Gaza Wastewater Treatment Plant (NGWWTP). As a result, the above and below the standards (Fig. 11). Nitrate values within this
982 M.F. Abu-alnaeem et al. / Science of the Total Environment 615 (2018) 972–989

Table 2
Statistical summary of the variables of the seven clusters.

Statistic TDS pH Ca2+ Mg2+ Na+ K+ Cl− NO−

3 SO2−
4 HCO−
3

All (N = 219 wells) mean 2634 7.6 133 109 621 10.9 1101 123 257 278
mx 18,848 8.3 657 665 5400 155.5 10,318 496 1604 1280
min 370 6.915 25 22.75 41.2 1.3 77.7 17.8 8 101.5
Cluster 1 (N = 40 wells) mean 2492 7.8 89 76 648 6.1 943 120 353 229
mx 3536 8.2 225 164 1073 14.2 1318 323 791 359
min 1515 7.5 26.6 30.3 353 2.7 578 17.8 166 125
Cluster 2 (N = 35 wells) m 1852 7. 5 60 64 497 5 654 81 180 460
mx 3078 7.77 93 119 905 16.25 1201 138.4 425.5 1280
min 985 7.01 26 26 205 2.4 305 41.3 30.1 247
Cluster 3 (N = 24 wells) mean 4208 7.5 240 180 946 9.3 1864 138 471 241
mx 8187 8.1 630 344 1725 19.8 4279 240.7 1017 364
min 2220 7.02 106 80 426 5.7 865 53 144 145
Cluster 4 (N = 30 wells) mean 2010 7.4 134 96.7 403 15.7 667 219 186 370
mx 5549 7.7 484 292 1116 87.2 2937 496 508 995
min 1125 7.11 83.5 41.5 134 2.9 244 124 73.7 244
Cluster 5 (N = 17 wells) mean 10,813 7.2 479 450 2748 60 5639 114 849 216
mx 18,848 7.68 657 665 5400 155.5 10,318 184 1604 285
min 5434 6.9 277 164.5 1345 6.9 2396 56 377 118
Cluster 6 (N = 24 wells) mean 1015 7.4 115 63.3 123 4 295 133 61.8 232
mx 2067 7.74 220 118 290 7.3 941 225 129 301
min 617 7.2 57.5 39 43 1.9 89 77.3 33 151
Cluster 7 (N = 49 wells) mean 877 7.8 58.3 45.2 167 3.36 261 88 62.1 196
mx 1494 8.3 104.7 83 350 7.5 554 254.7 203 380
min 370 7.5 25 22.7 41 1.3 77.7 25.7 8 101

All concentrations are reported in mg/L except for pH unit-less. The red color values indicate the highest values while the blue for the lowest ones.

cluster specifically, were the highest recorded for the study area indicat- study area. All of the wells exceeded WHO standards for all chemical
ing the significant impacts from human activities primarily due to species (variables) with the exception of HCO− 3 (Fig. 11). Such ground-
sewage-related infiltration. These results reflect a mixing influence of water chemical signatures reflect the extreme impact (presence) of sea-
salinization and pollution and revealing the primary processes water intrusion on the wells within this area.
impacting groundwater vulnerability in this northern portion of the
study area. 4.2.1.6. Cluster 6 (rainfall dilution with human inputs). Cluster 6, repre-
sented by a relatively low number of wells (24), is located in the north-
4.2.1.5. Cluster 5 (seawater intrusion and human inputs). Cluster 5, associ- ern portion of the study area (Fig. 10). In general, most of the chemical
ated with the lowest number of wells (17), is located along the northern species measured (i.e. Ca2+, Mg2+, HCO− 2−
3 , SO4 and K+) were below
study area close to the coastline (Fig. 10). The wells within this cluster −
WHO standards with the exception of Cl and NO− 3 which were above
are characterized by salinity where TDS values range between 5434 these standards (Fig. 11). Na+ values (concentrations) were either ele-
and 18,848 mg/L with an average of 10,813 mg/L (Table 2). This cluster vated or low, ranging between the low and high range of WHO concen-
is characterized by the highest values of, TDS Ca2 +, Mg2 +, Na+, Cl−, tration standards. The lowest concentrations of SO2− 4 and Na+ and the
SO2− +
4 and K with the lowest values of pH (less alkaline) over the entire highest concentrations of Mg2+ ions were recorded in this cluster. The

Fig. 11. Box plot diagram of the major ions in mg/L in the seven clusters.
 M.F. Abu-alnaeem et al. / Science of the Total Environment 615 (2018) 972–989 983

Fig. 12. Box plot of ions in (/L) for the seven clusters.

chloride and sodium were dominant in all clusters except this cluster 6 of Eocene groundwater which is rich in Ca2+, predominantly impacted
where the water type is (Cl-Ca-Na-Mg-HCO3 type). These results can be these clusters and triggered the base ion exchange process in this area. A
attributed to mixing impacts of direct rainfall recharge with human ac- low Na+/Cl− ratio (i.e. Na+ deficiency) is mostly recorded in clusters
tivities in this highly vulnerable sand dunes area. 3–7. The highest deficiency in Na+ is recorded in clusters 5 and 3
which are located in close proximity to the coastline, revealing that sea-
4.2.1.7. Cluster 7 (human inputs with rainfall dilution). Cluster 7 is associ- water intrusion and saltwater up-coning are predominantly impacted
ated with the highest number of wells (49) (Fig. 10). Compared to all these clusters and triggered the reverse ion exchange process in this
other clusters, the hydrogeochemical values/concentrations within region.
this cluster were lowest for TDS, Ca2 +, Mg2 +, Cl−, K+, HCO− 3 and The highest nitrate pollution is recorded in clusters 4 and 6 which
SO2−
4 . However, the highest values of pH were recorded in this cluster. are located under the urbanized areas in the northern study area. The
In general, the water chemical species including SO24 −, HCO− 3 , Ca
2+
, highest abundance of sulfate was recorded in cluster 1 in the south pri-
Mg2+ and K+ were below WHO standards while the NO− 3 concentra- marily due to the Eocene invasion and gypsum dissolution in the cover-
tions exceeded these standards (Fig. 11). Na+ and Cl− have values ing loess soil (Vengosh et al., 2005), as well as the infiltration of sewage
above and below the standards. These results reveal that this cluster, and farming/agricultural-related inputs (and activities) in the region
which is located predominantly in the northern rainy area toward in- (Abu-Jabal et al., 2014; Marghade et al., 2012). The highest values for
land, has been impacted by the direct rainfall recharge and dilution ef- Mg2+ ions concentration was recorded cluster 6 may be attributed to
fects. Moreover, the high vulnerability of the groundwater in this dolomite dissolution and human inputs in this rainy area. Correspond-
region (northern portion of study area) is demonstrated by all wells in ingly, seawater invasion in clusters 5 and 3 generally exhibited an in-
this cluster being contaminated with nitrate as a result of anthropogenic crease in Mg2+ and SO2− 4 concentrations in these areas.
activities (i.e. sewage disposal and agricultural inputs). The highest nitrate pollution is recorded in clusters 4 and 6 which
are located under the urbanized areas in the northern study area. The
4.2.2. Relative ions abundance of the clusters highest abundance of sulfate was recorded in cluster 1 in the south pri-
The relative ion abundance (chemical signatures) of the seven clus- marily due to the Eocene invasion and gypsum dissolution in the cover-
ters are represented in (Table 3) and as Box plot (Fig. 12). They show ing loess soil (Vengosh et al., 2005), as well as the infiltration of sewage
that chloride and sodium were the dominant ions in all clusters except and farming/agricultural-related inputs (and activities) in the region
cluster 6 along the northern boundary (Cl-Ca-Na-Mg-HCO3 type). The (Abu-Jabal et al., 2014; Marghade et al., 2012). On the other hand, the
high Na+/Cl− ratio (i.e. Na+ surplus) is mostly recorded in clusters 1 highest concentration of Mg2+ in cluster 6 suggests the dolomite disso-
and 2. This Na-rich groundwater indicates that, the lateral movement lution source in this rainy area with human inputs. Moreover, seawater
invasion in clusters 5 and 3 raised up the Mg2+ and SO2− 4 concentra-
Table 3
tions in these areas.
The relative abundance of the seven clusters ions.
4.3. Hydrogeochemical appraisals from ionic relations
Cluster Relative abundance of ions (2013/2014)

C (1) Na+ N Cl− N N Mg2+ N SO2−

4 N Ca
2+
N HCO− −
3 N NO3 N K
+
4.3.1. Hydrogeochemical facies
C (2) Na+ N Cl− N N HCO−3 N Mg
2+
N Ca2+ N SO2− −
4 N NO3 N K
+
Groundwater can be categorized into different water types based on
C (3) Cl− N N Na+ N N Mg2+ N Ca2+ N SO2− −
4 N HCO3 N NO3 N K
− +

C (4) Cl− N Na+ N N Mg2+ N Ca2+ N HCO− − 2−

3 N NO3 N SO4 N K
+ the relative percentage of its chemical components. The different water
C (5) Cl− N N Na+ N N Mg2+ N Ca2+ N SO2− −
4 N HCO3 N NO3 N K
− + types reflect the natural and anthropogenic factors and hydrogeochem-
C (6) Cl− N N Ca2+ N Na+ N Mg2+ N HCO− − 2−
3 N NO3 N SO4 N K
+
ical processes that influence on the groundwater chemistry in the study
C (7) Cl− N Na+ N N Mg2+ N Ca2+ N HCO− − 2−
3 N NO3 N SO4 N K
+
area. Chadha's plot used to determine the major hydrogeochemical pro-
Total Cl− N Na+ N N Mg2+ N Ca2+ N HCO− 3 N SO2−
4 N NO −
3 N K+
cesses controlling the chemistry of groundwater and to compare the
984 M.F. Abu-alnaeem et al. / Science of the Total Environment 615 (2018) 972–989

water types of the clusters. In Chadha's plot, the X-axis represents the When there is a softening process (i.e. sequestered of Ca2 + and
difference between (Ca2 + + Mg2 +) and (Na+ + K+) in meq/L% and Mg2+ of the water and releasing of Na+ and K+ from the aquifer mate-
the Y axis represent the difference between (HCO− −
3 ) and (Cl + SO4 )
2−
rial), the CAI index values tend to decrease (negative values)
in meq/L %. The four hydrogeochemical processes and the water types (Argamasilla et al., 2017; Elgettafi et al., 2012). This hydrogeochemical
suggested by Chadha (1999) are shown in the quadrants of the plot exchange process is called base ion exchange process. On contrast, if
(Fig. 13). there is a hardening process (i.e. sequestered of Na+ and K+ on the
It is found that most of the clusters wells located in Na-Cl (71.6%) aquifer material and releasing of Ca2 + and Mg2 +), CAI index values
while the wells of cluster 6 and half of cluster 7 are located in Ca/Mg- tend to increase (positive values) and this hydrogeochemical process
Cl (25%) field. This result reveals that groundwater in the Gaza coastal is called reverse ion exchange process (Argamasilla et al., 2017; Zaidi
aquifer is mostly saline. Generally, most of the wells tend to locate at et al., 2015; Taylor, 2016). CAI values which are close to zero reveal
higher values on (y-axis) and lower values of (X-axis). It can be noted the absence of ion exchange process while the higher CAI values either
that the surplus of Cl in clusters 5 and 3 appeared from the locations positive (until +1) or negative (until −1) reflect a higher level of this
of their wells within the field of Na-Cl near to 100% of Cl− and SO24 − hydrogeochemical process.
on (Y-axis) and b50% for the Na+ + K+ on (X-axis). On the other Commonly, the low-salinity areas had negative CAI, demonstrating
hand, cluster 1 and 2 are also located in Na-Cl\\ field but with higher that the groundwater in these areas is subject to freshening processes,
values on (X-axis) and lower values on (Y-axis) revealing a surplus in accompanying with Na+ being released from the aquifer material
Na in these clusters due to the dominance of base ion exchange process (Taylor, 2016) and Ca2 + fixation from the groundwater. In coastal
in these areas. Generally, these results indicate that high degree of salin- regions, rainfall contents resemble to the components of seawater (i.e.
ity has occurred and the reverse ion exchange process was dominant. Na-Cl water type) but with very low TDS. So, when the rainfall re-
Moreover, the salinization process is very rapid in all the study area charges the aquifer in this area, the expected hydrogeochemical reac-
that prevents the formation of the freshwater area even in the rainy tion between this water type and the aquifer carbonate material is
area in the north. This result can be attributed to the presence of reverse ion exchange process (i.e. positive CAI values). In addition, at
multi-sources of salinization that impacts completely the groundwater the sewage invasion areas, the Na+/Cl− ratio is higher than unity. So,
chemistry and changes it to saline water. the expected hydrogeochemical process in the invaded area by waste-
water is reverse ion exchange process.
As shown in (Fig. 14), 43.8% of the wells are below zero value with
4.3.2. Water-rock interaction process
the lowest value − 0.51 while 56.2% of the wells are above zero with
Water-rock interaction is one of the major natural providers to the
the highest value 0.59. Nearly all of the wells from cluster 5 (CAI:
groundwater chemical components (Tóth, 1999). In coastal aquifers,
0.08–0.36) and cluster 3 (CAI: − 0.05–0.53) are above zero likely due
the groundwater chemistry may expose to modify and hide its original
to the influence of seawater intrusion and saltwater up-coning respec-
composition by a diversity of rock-water interaction processes as cation
tively. The highest positive CAI values in Cluster 6 (CAI: − 0.08–0.59)
exchange and soluble mineral dissolution processes (Monjerezi, 2012).
which located in the northern rainy area, reflect the direct impact
of rainfall on this area with the influence of sewage inputs. Clusters 7
4.3.2.1. Chloro-alkaline index (CAI). The chloro-alkaline indices, which (CAI: − 0.51–0.44), cluster 4 (CAI: − 0.34–0.41) and cluster1 (CAI:
proposed by Schoeller (1965), can be used for measuring the degree −0.45–0.45) show equal positive and negative CAI values. The positive
of ion exchange reactions between the aquifer minerals and groundwa- CAI values of the cluster 7 may suggest direct impact of rainfall and
ter. It is commonly used to determine the ion exchange processes regu- sewage inputs on this area. The other clusters (i.e. 4 and 1) can be attrib-
lating the chemistry of groundwater. The CAI is calculated using the uted mainly to the impact of anthropogenic activities in these highly
following formula: [Cl− − (Na+ + K+)]/Cl− where the values are in populated urban areas. The negative CAI values of these clusters (i.e. 7,
meq/L. 4 and 1) with all the wells of cluster 2 may suggest that base ion

Fig. 13. Chadha's Plot for the seven clusters.

 M.F. Abu-alnaeem et al. / Science of the Total Environment 615 (2018) 972–989 985

4.3.2.3. Gypsum dissolution. Gypsum dissolution can be assessed based

on the Ca2+/SO2− 4 ratio in meq/L (Li et al., 2015). If the source of these
ions is Gypsum dissolution, the Ca2 +/SO24 − should be close to unity
(Elgettafi et al., 2012). The scatter plot (Fig. 16) of Ca2 +/SO2− 4 versus
Cl− (meq/L) shows that most of the wells locate above the unity line.
These wells includes all the wells of cluster 6, most of cluster 7 and 5,
halves of clusters 3 and 4 and a few wells in cluster 1. The excess
of Ca2+ may be attributed to reverse ion exchange process during sea-
water intrusion (cluster 5) and saltwater up-coning (cluster 3), carbon-
ate dissolution during rainfall recharge (cluster 6 and half of 7) and
nitrification process in nitrate polluted areas (cluster 4) which acceler-
ates the dissolution of carbonate aquifer material.
On the other hand, a few of wells plot close to the unity line (clusters
1, 2, 3, and 5) revealing the limited role of gypsum dissolution in the
groundwater chemistry of the study area. The surplus of SO2− 4 in these
clusters may suggest the fixation of calcium ion after gypsum dissolu-
tion or may indicate the presence of another source of sulfate in these
areas. The fixation may be due to base ion exchange reaction during
Fig. 14. Scatter plot of CAI versus Cl (meq/L) for the seven clusters. the invasion of Eocene water in these eastern areas. The other sources
of the excess of SO24 − could arise from the human activities in these
exchange process takes place in these areas as a result of Eocene brack- highly populated urban areas in the form of sewage and fertilizers in-
ish water invasion. puts. Moreover, coastal regions sediments usually contain pyrite miner-
al (Morse et al., 2007). Oxidation of sulfide minerals as pyrite could
increase in sulfate concentration of groundwater (Elgettafi et al., 2012).
4.3.2.2. Carbonate dissolution. Carbonate dissolution can be evaluated
based on the Ca2+/HCO− 3 ratio in meq/L. If the source of these ions is cal- 4.3.3. Na+/Cl− versus Cl and salinization sources
cite dissolution, the Ca2+/HCO− 3 should be close to unity and if it is do- The relationship between Na+/Cl− versus Cl− is important to identi-
lomite, Ca2 +/HCO− 3 should be close to 0.5. The ionic ratio diagram fy the salinization sources. If the source of these ions is halite dissolu-
(Fig. 15) of Ca2 +/HCO− 3 versus Cl
−
(meq/L) shows that most of the tion, the Na+/Cl− should be equal one (Basins and Geo, 2004) but if
wells plot are above the unity line especially those belong to clusters 5 this ratio deviates from unity it will reflect other sources as seawater in-
and 3. The excess of Ca2+ may be attributed to reverse ion exchange trusion, saltwater invasion and water-rock interaction.
process during seawater intrusion (cluster 5) and saltwater up-coning As shown in (Fig. 17), most of the wells fall below one in Na+/Cl−
(cluster 3). Most of the wells of clusters 6, 7 and 4 are located between ratio revealing that these waters may be influenced by other sources.
the unity line of calcite dissolution and the other line of dolomite disso- Additionally, this relationship is useful in estimating the type and de-
lution. The process of carbonate dissolution could be happened in these gree of cation exchange process in groundwater. In (Fig. 17), the ratio
areas as a result of rainfall influence and/or the acidic medium in nitrate of Na+/Cl− is lower than one in clusters 5 and 3, suggesting that reverse
polluted areas. ion exchange process between Na+ and Ca2+, and Mg2+ takes place in
Some of the wells of cluster 2 plot below the 0.5 line revealing an ex- groundwater and contribute to groundwater salinization. Most of the
cess in HCO− 3 ions in these wells. This result may suggest the fixation of wells of clusters 1and 2 show higher ratio of Na+/Cl− than unity
calcium ion after carbonate dissolution during the invasion of Eocene reflecting that base ion exchange processes occurred and have likely
water in these eastern areas or may indicate the presence of another modified the salinity and chemical signatures in these areas. Thus, the
source of bicarbonate in these areas. The other sources of the extra of data suggests that hydrogeochemical processes due to seawater intru-
HCO− 3 could arise from the dissolution of carbon dioxide from the atmo- sion and saltwater up-coning are occurring in cluster 3 and 5, while Eo-
sphere during rainfall or result from the oxidation process of organic cene brackish water may be invading the wells of clusters 1 and 2. Most
matter, which is a common pollutant in coastal regions (Carol and of the wells within clusters 6 and 7 exhibit freshwater chemical signa-
Kruse, 2012). tures with TDS values b 1000 mg/L. Nearly all of the wells of cluster 6

Fig. 15. Scatter plot of Ca2+/HCO−

3 ratio in meq/L versus Cl (meq/L) for the seven clusters. Fig. 16. Scatter plot of Ca2+/SO2− −
4 ratio in meq/L versus Cl (meq/L) for the seven clusters.
986 M.F. Abu-alnaeem et al. / Science of the Total Environment 615 (2018) 972–989

With the exception of one well in the northern study area (C/167),
all wells within the study area were recharged by waters of varying sa-
linity. Wells impacted by severe salinization include all of those in clus-
ter 5 and half within cluster 3. The remaining wells in cluster 3, although
not as severe, are influenced by highly saline water. The wells in cluster
1 are impacted by extremely high salinities, posing health hazards in
drinking water, likely due to the Eocene invasion with sewage invasion
in this south area.

4.3.5. Mg2+/Ca2+ versus Cl− and salinization sources

In groundwater, Mg2+/Ca2+ ratio is usually less than unity while in
sea water the Mg2+/Ca2+ ratio greater than one. Thus, higher ratios of
Mg2+/Ca2+ can be found in seawater intrusion zones and similarly in
areas experiencing of dolomite dissolution. It should be noted that
81.3% of the wells exhibit Mg2+/Ca2+ ratios N1, indicating that seawa-
ter intrusion, saltwater up-coning, reverse cation exchange, dolomite
dissolution processes or sewage invasion may play a significant role
on the groundwater chemistry signatures in the study area.
Fig. 17. Na+/Cl− ratio versus Cl− relationship for the seven clusters. Based on a study by Mayo and Loucks (1995), a distinction between
the areas of calcite and dolomite dissolution can be identified by calcu-
and more than half of the wells in cluster 7 exhibit Na+/Cl− ratio values lating the Mg2+/Ca2+ molar ratios of these cations. If the molar ratio is
less than one, revealing that carbonate dissolution processes may be rel- less than one, the dissolution of calcite has greater contribution to the
atively significant in the northern areas that receives a higher propor- water chemistry, whereas greater prevalence of dolomite dissolution
tion of recharge via precipitation in this region. This is consistent with occurs when the Mg2 +/Ca2 + molar ratio is equal or N 1. Katz et al.
lower pH precipitation water that infiltrate under short residence (1997) describe that lower molar ratios (i.e. ˂0.5) indicate silicate-
times. Chemical ratio signatures of the wells of cluster 4 fall above and weathering sources and/or processes are occurring. As shown in
below the equiline suggesting the impact of various hydrogeochemical (Fig. 19), only one well exhibited Mg2 +/Ca2 + a ratio lower than 0.5,
processes as sewage invasion which is rich in Na+ relative to Cl− which reveal that negligible silicate-weathering processes were occur-
(Vengosh and Keren, 1996), seawater intrusion, saltwater up-coning, ring in the study area.
Eocene brackish invasions and carbonate dissolution. The average of Mg2+/Ca2+ molar ratio for wells within all clusters
are N1, with the exception of cluster 6. This result reveals that calcite
4.3.4. Degree of salinization (inverse of Simpson's ratio) dissolution may be occurring (and predominant) in cluster 6 (under
The relationship between HCO− − −
3 /Cl ratio with Cl for the majority the direct impact of rainfall), while the other clusters are impacted by
of wells (as a whole), exhibited a pronounced negative slope indicating seawater and saltwater intrusion (i.e. clusters 3 and 5) or influenced
the reverse relationship between carbonate (represents freshening) by dolomite dissolution (i.e. clusters 1, 2, 4, and 7). It can be concluded
and Cl− (represents salinization) (Fig. 18). Based on the inverse of that dolomite dissolution is the predominant hydrogeochemical process
Simpson's ratio, the influence of saline water can be classified into six in comparing with calcite dissolution in the study area.
classes. In general, a HCO− −
3 /Cl ratio N2 indicates freshwater recharge Relative to the wells of cluster 5 which located close to the shoreline,
whereas a ratio less than unity indicates higher degrees (or potential) the wells of cluster 3 shows lower Mg2+/Ca2+ molar ratio and lower
of salinization (Chaudhuri and Srinivasulu, 2014; October et al., 2013). Cl− concentration. This is a chemical signature that can be used to

Fig. 18. Scatter plot of HCO− − −

3 /Cl ratio versus Cl for the seven clusters.
 M.F. Abu-alnaeem et al. / Science of the Total Environment 615 (2018) 972–989 987

based on the chemical ratio signatures (Fig. 20). The wells with higher
concentration in Cl− and lower in NO− 3 may suggest the influence of
municipal sewage and animal manure, while those wells with higher
concentration in NO− 3 and lower in Cl
−
may indicate farming inputs
−
which are commonly without Cl component (Fig. 20) (Wang et al.,
2016). The fourth group with very low NO− 3 /Cl
−
ratio and very high
Cl− can be considered to be predominantly influenced by seawater in-
trusion processes.
Cluster 1 and 2 wells plot near to the domestic sewage inputs reveal-
ing that wastewater infiltration was the major source of nitrate. The
wells within clusters 6 and 7 were influenced by mixing inputs of
both of fertilizers and domestic sewage due to its location under
urban areas and cultivated areas. Due to the intensive farming inputs
in cluster 6 area, the nitrate concentration in wells is much higher
than in cluster 7 which is characterized by higher sewage inputs than
farming inputs. The wells of cluster 3 also seems to be influenced signif-
icantly by mixing inputs of domestic sewage and seawater intrusion. In
cluster 5, all of the wells are generally located in the area of dominant
Fig. 19. Scatter plot of Mg2+/Ca2+ versus Cl− for the seven clusters. seawater intrusion. The wells within cluster 4 plot in a region of the
graph that indicated that the sewage inputs is the major source of ni-
distinguish between seawater intrusion area and saltwater up-coning trate under this urban area with some wells influenced by mixing of
area. Meanwhile, the higher Mg2+/Ca2+ ratio and lower Cl− concentra- sewage and farming inputs.
tion are characterized the wells of cluster 2 in the NE part of the study The influence of farming practices on area of permeable sand dunes
area when comparing them with the wells of cluster 1 in the SE part. soil at the north appeared better in the water samples of the clusters 6
The higher Mg2+ concentration in the north can be attributed to the dis- and 7 than the other clusters. Generally, farming inputs were limited
solution of carbonate matrix of sandstone which is rich in Mg (Gavish compared with sewage inputs in spite of the intensive farming practices
and Friedman, 1969). This dissolution is a result of acidic medium of in Gaza Strip. This is due to the fact that most agricultural areas are lo-
the nitrate pollution which is higher in the eastern side of the northern cated on impermeable to semi-permeable soils along and toward the
area. Moreover, among the seven clusters, the highest Mg2+/Ca2+ ratio eastern boundaries of the study area. These farming conditions prevent
is recorded for the wells of cluster 2 then for clusters 1 and 5. This is a or at least retard the transfer of farming inputs to groundwater while
clear indication that the eastern clusters 1 and 2 were influenced by Eo- sewage inputs are common and cover a wide spatial extent in which
cene brackish water which is characterized by Na-Mg-Cl water type groundwater is more vulnerable to sources and transfer of sewage
(Vengosh et al., 2005). inputs.

4.3.6. NO− − −
3 /Cl versus Cl plot and nitrate sources 5. Conclusion
The scatter plot of NO− − −
3 /Cl ratio versus Cl in meq/L can be used to
distinguish between nitrate sources and whether farming inputs and/or In this study, an integration of statistical, geostatistical and hydro-
municipal sewage inputs play significant roles to the observed ground- geochemical approaches were applied for ten physiochemical water
water chemistry (Murgulet and Tick, 2013; Wang et al., 2016). The data chemistry parameters for 219 municipal wells using STATA/SE12 soft-
are plotted and show that four distinct groups of wells are identified ware and Surfer 13 Golden software Inc. to understand the salinization

Fig. 20. Scatter plot of NO− −

3 /Cl versus Cl for the seven clusters.
988 M.F. Abu-alnaeem et al. / Science of the Total Environment 615 (2018) 972–989

origins and the hydrogeochemical processes controlling the salinization Carol, E.S., Kruse, E.E., 2012. Hydrogeochemical characterization of the water resources in
the coastal environments of the outer Río de la Plata estuary, Argentina. J. S. Am.
and deterioration of the Gaza coastal aquifer system. Earth Sci. 37, 113–121.
Geostatistical analysis of the groundwater salinity showed that sea- Chadha, D.K., 1999. A proposed new diagram for geochemical classification of natural wa-
water intrusion along the coastline, and saltwater up-coning inland ters and interpretation of chemical data. Hydrogeol. J. 7 (5), 431–439.
Chaudhuri, S., Srinivasulu, A., 2014. Long term (1960–2010) trends in groundwater con-
highly influenced the groundwater salinity of the study area. HCA tech- tamination and salinization in the Ogallala aquifer in Texas. J. Hydrol. 513, 376–390.
nique enabled the definition of seven distinct hydrogeochemical signa- Cloutier, V., Lefebvre, R., Therrien, R., Savard, M.M., 2008. Multivariate statistical analysis
ture clusters (C1&C2: Eocene brackish water invasion, C3 saltwater up- of geochemical data as indicative of the hydrogeochemical evolution of groundwater
in a sedimentary rock aquifer system. J. Hydrol. 353, 294–313.
coning, C4 human inputs, C5 seawater intrusion, C6 & C7 rainfall and Davis, J.C., Sampson, R.J., 1986. Statistics and Data Analysis in Geology. John Wiley and
human inputs). Box plot shows a wide variation of most of the ions Sons, New York.
while Chadha's plot elucidates the predominance of Na-Cl (71.6%) and Dentoni, M., Deidda, R., Paniconi, C., Qahman, K., Lecca, G., 2014. A simulation/optimiza-
tion study to assess seawater intrusion management strategies for the Gaza Strip
Ca/Mg-Cl (25%) water types. It is found that, the highest and the lowest
coastal aquifer (Palestine). Hydrogeol. J. 23, 249–264.
levels of salinization and the highest level of nitrate pollution were re- Elamassi, K.S., 2012. Assessment of groundwater quality using multivariate and spatial
corded in the northern area. This result reflects the sensitivity of this analyses in Gaza governorate-Palestine. M.Sc. Thesis. Islamic University, Gaza Strip,
area to the human activities and/or natural actions. In spite of the Palestine.
Elgettafi, M., Elmandour, A., Himi, M., Casas i Ponsatí, A., Elhaouadi, B., 2012. Messinian sa-
wide farming practices in the Gaza Strip, the hydrogeochemical results linity crisis impact on the groundwater quality in Kert aquifer NE Morocco: hydro-
reveal the limited impact that the irrigation return flow contributes to geochemical and statistical approaches. Int. J. Water Resour. Environ. Eng. 4, 11,
groundwater pollution. 339–351 2012.
El-naeem, M.F., Heen, Z.A., Tubail, K., 2010. Factors behined groundwater salinization in
The main hydrogeochemical processes that controlled the salinity north governorates of Gaza Strip (1994–2004). Fourteenth International Water Tech-
and quality of the clusters are ion exchange processes, nitrification pro- nology Conference, IWTC 14 2010, Cairo, Egypt, pp. 893–907.
cess, carbonate and gypsum dissolution processes. Among the hydro- Elyaouti, F. El, Mandour, A. El, Khattach, D., Benavente, J., Kaufmann, O., 2009. Salinization
processes in the unconfined aquifer of Bou-Areg (NE Morocco): a geostatistical, geo-
geochemical processes, it is found that the ion exchange process was chemical, and tomographic study. Appl. Geochem. 24, 16–31.
effective process all over the study area. This founding was confirmed Gavish, E., Friedman, G.M., 1969. Progressive diagenesis in Quaternary to Late Tertiary
through CAI index calculations and Na/Cl ratio estimation. Base ion ex- carbonate sediments: sequence and time scale. J. Sediment. Petrol. 39, 980–1006.
Ghabayen, S.M.S., McKee, M., Kemblowski, M., 2006. Ionic and isotopic ratios for identifi-
change process was the dominant in the eastern areas in clusters 1, 2 cation of salinity sources and missing data in the Gaza aquifer. J. Hydrol. 318,
and limited in clusters 4 and 7 as a result of Eocene brackish water inva- 360–373.
sion. The reverse ion exchange process was dominant in the remaining Güler, C., Thyne, G.D., McCray, J.E., Turner, K.A., 2002. Evaluation of graphical and multi-
variate statistical methods for classification of water chemistry data. Hydrogeol. J.
clusters and was the highest in cluster 6 due to rainfall recharge of low
10, 455–474.
salinity Na-Cl water type and in clusters 5 and 3 due to seawater intru- Güler, C., Kurt, M.A., Alpaslan, M., Akbulut, C., 2012. Assessment of the impact of anthro-
sion and saltwater up-coning respectively. The nitrification process was pogenic activities on the groundwater hydrology and chemistry in Tarsus coastal
predominant in the study area. Around 90.4% of the wells are nitrate plain (Mersin, SE Turkey) using fuzzy clustering, multivariate statistics and GIS tech-
niques. J. Hydrol. 414, 435–451.
polluted. The hydrogeochemical investigation reveals that the main Hiscock, K.M., 2009. Hydrogeology: Principles and Practice. John Wiley & Sons, New York.
source of nitrate pollution is the sewage inputs while the farming inputs Huang, G., Sun, J., Zhang, Y., Chen, Z., Liu, F., 2013. Impact of anthropogenic and natural
are very limited and restricted mostly in the sensitive northern area processes on the evolution of groundwater chemistry in a rapidly urbanized coastal
area, South China. Sci. Total Environ. 463–464, 209–221.
within clusters 6, 7 and 2. Carbonate dissolution was common in the Katz, B.G., Coplen, T.B., Bullen, T.D., Davis, J.H., 1997. Use of chemical and isotopic tracers
study area with the highest level in clusters 6, 7, 4 and 2 and almost ab- to characterize the interactions between ground water and surface water in mantled
sent in clusters 5 and 3. Gypsum dissolution process was significant only karst. Ground Water 35 (6), 1014–1028.
Lazarou, E., 2016. Water in the Israeli-Palestinian conflict. EPRS/European Parliamentary
in cluster 1 and limited in the other clusters. It could be attributed to the Research Service, pp. 1–8.
dissolution of gypsum from the loess soil and the sewage infiltration Ledesma-Ruiz, R., Pastén-Zapata, E., Parra, R., Harter, T., Mahlknecht, J., 2015. Investigation
from the common use cesspits in this southern area. of the geochemical evolution of groundwater under agricultural land: a case study in
northeastern Mexico. J. Hydrol. 521, 410–423.
The integrated approach of statistical, geostatistical and hydrogeo- Li, Wei, Wang, M.-Y., Liu, L.-Y., Yan, Y., 2015. Assessment of long-term evolution of
chemical study has proved to be potential and effective for investigating groundwater hydrogeochemical characteristics using multiple approaches: a case
the salinization origins and the hydrogeochemical processes controlling study in Cangzhou, Northern China. Water 7 (3), 1109–1128.
Machiwal, D., Jha, M.K., 2015. Identifying sources of groundwater contamination in a
the salinization and deterioration of the Gaza coastal aquifer system in
hard-rock aquifer system using multivariate statistical analyses and GIS-based
detail. This integrated multi-approaches study within the Gaza coastal geostatistical modeling techniques. J. Hydrol. Reg. Stud. 4, 80–110.
aquifer system can be useful for effective utilization and management Maguette, N., Orban, P., Otten, J., Stumpp, C., Faye, S., Dassargues, A., 2017. Temporal changes
of this vital resource and for future work in similar aquifer systems in groundwater quality of the Saloum coastal aquifer. J. Hydrol. Reg. Stud. 9, 163–182.
Marghade, D., Malpe, D.B., Zade, a.B., 2012. Major ion chemistry of shallow groundwater
under high anthropogenic pressures. of a fast growing city of Central India. Environ. Monit. Assess. 184, 2405–2418.
Mayo, A.L., Loucks, M.D., 1995. Solute and isotopic geochemistry and ground water flow
in the central Wasatch Range, Utah. J. Hydrol. 172, 31–59.
References Melloul, A., Azmon, B., 1997. A graphic expression of salinization and pollution of ground-
water the case of Israel's groundwater. Environ. Geol. 30, 126–136.
Abu-Jabal, M.S., Abustan, I., Rozaimy, M.R., El-Najar, H., 2014. Groundwater beneath the Melloul, A., Collin, M., 2000. Sustainable groundwater management of the stressed Coast-
urban area of Khan Y ounis City, southern Gaza Strip (Palestine): hydrochemistry al aquifer in the Gaza region. Hydrol. Sci. J. 45, 147–159.
and water quality. Arab. J. Geosci. 8, 2203–2215. Mercado, A., 1980. Groundwater Salinity in the Yarkon-Taninim Beer Sheva Basin: Statis-
Al-Agha, M.R., El-nakhal, H. a, 2004. Hydrogeochemical facies of groundwater in the Gaza tical Models for the Simulation of the Salinity Regime in the Taninim Springs. TAHAL,
Strip, Palestine. Hydrol. Sci. 49, 359–372. Tel Aviv (Rep. 01/80/37 47).
Aliewi, A.S., Mackay, R., Jayyousi, A., Nasereddin, K., Mushtaha, A., Yaqubi, A., 2001. Nu- Metcalf & Eddy Inc, 2000. Integrated Aquifer Management Plan: Final Report. Gaza Coast-
merical simulation of the movement of saltwater under skimming and scavenger al Aquifer Management Program USAID Contract No. 294-C-00-99-00038-00.
pumping in the Pleistocene aquifer of Gaza and Jericho Areas, Palestine. Transp. Po- Mogheir, Y., De Lima, J.L.M.P., Singh, V.P., 2005. Assessment of informativeness of ground-
rous Media 43, 195–212. water monitoring in developing regions (Gaza Strip case study). Water Resour.
Al-Yaqubi, A., Aliewi, A., Mimi, Z., 2007. Bridging the domestic water demand gap in Gaza Manag. 19, 737–757.
Strip-Palestine. Water Int. 32 (2), 219–229. Mogheir, Y., Singh, V.P., de Lima, J.L.M.P., 2006. Spatial assessment and redesign of a
Argamasilla, M., Barberá, J.A., Andreo, B., 2017. Factors controlling groundwater saliniza- groundwater quality monitoring network using entropy theory, Gaza Strip, Palestine.
tion and hydrogeochemical processes in coastal aquifers from southern Spain. Sci. Hydrogeol. J. 14, 700–712.
Total Environ. 580, 50–68. Monjerezi, M., 2012. Groundwater Salinity in Lower Shire River Valley (Malawi) Hydro-
Baalousha, H.M., 2011. Mapping groundwater contamination risk using gis and groundwa- geochemical and Isotope Constraints on Sources and Evolution. University of Oslo,
ter modelling. A case study from the gaza strip, palestine. Arab. J. Geosci. 4, 483–494. Norway.
Basins, C.R., Geo, J.E., 2004. Identification and evolution of hydrogeochemical processes of Morse, J.W., Thomson, H., Finneran, D.W., 2007. Factors controlling sulfide geochemistry
groundwater environment in an area of the Palar and identification and evolution of in sub-tropical estuarine and bay sediments. Aquat. Geochem. 13, 143–156.
hydrogeochemical processes in the groundwater environment in an area of the Palar Murgulet, D., Tick, G.R., 2013. Understanding the sources and fate of nitrate in a highly de-
and Cheyyar River Basins. Environ. Geol. 46, 47–60. veloped aquifer system. J. Contam. Hydrol. 155, 69–81.
 M.F. Abu-alnaeem et al. / Science of the Total Environment 615 (2018) 972–989 989

Nwankwoala, H.O., Udom, G.J., 2011. Hydrogeochemical facies and ionic ratios of ground- Vengosh, A., Rosenthal, E., 1994. Saline groundwater in Israel: its bearing on the water cri-
water in Port Harcourt, Southern Nigeria. Res. J. Chem. Sci. 1, 87–101. sis in the country. J. Hydrol. 156 (1), 389–430.
October, V.I., Annapoorani, A., Murugesan, A., Ramu, A., Renganathan, N.G., 2013. Hydro- Vengosh, A., Spivack, A.J., Artzi, Y., Ayalon, A., 1999. Geochemical and boron, strontium,
geochemical characteristics of groundwater quality in Chennai City using gis — a case and oxygen isotopic constraints on the origin of the salinity in groundwater from
study. Res. J. Chem. Environ. 3 :pp. 39–47 Available online at:. www.ijrce.org. the Mediterranean coast of Israel. Water Resour. Res. 35 (6), 1877–1894.
Palestinian Central Bureau of Statistics (PCBS), 2013. Projected Population in the Palestin- Vengosh, A., Gill, J., Davisson, M.L., Hudson, G.B., 2002. A multi-isotope (B, Sr, O, H, and
ian Territory, Establishments Report, Ramallah, Palestine. p. 54. C) and age dating study of groundwater from Salinas Valley, California:
Palestinian Water Authority (PWA), 2013. Seasonal Rainfall Assessment in Gaza Strip Re- hydrochemistry, dynamics, and contamination process. Water Resour. Res. 38 (1),
port Prepared by PWA, Prepared by: Mahmoud Abdel Latif, Director of Water Infor- 1–17.
mation Bank Department. Gaza, Palestine. Vengosh, A., Kloppmann, W., Marei, A., Livshitz, Y., Gutierrez, A., Banna, M., Raanan, H.,
Palestinian Water Authority (PWA), 2014. Water Resources Status Summary Report/Gaza 2005. Sources of salinity and boron in the Gaza strip: natural contaminant flow in
Strip. Unpublished Report by Water Resources Directorate, Gaza, Palestine. the southern Mediterranean coastal aquifer. Water Resour. Res. 41 (1).
Portal, Climate Change Knowledge, 2017. The World Bank Group. http:// Vinayachandran, N., Narayana, A.C., 2014. Hydrogeology and hydrochemistry of the aqui-
sdwebx.worldbank.org/climateportal/index.cfm/, Accessed date: 14 September 2017. fer systems of Kuttanad area, Kerala: Their role in understanding the evolution of
Qahman, K., Larabi, A., 2006. Evaluation and numerical modeling of seawater intrusion in groundwaters Doctoral dissertation, Cochin University of Science and Technology.
the Gaza aquifer (Palestine). Hydrogeol. J. 14, 713–728. Wang, 2011. Isotopic and hydrogeochemical studies of the coastal aquifer-aquitard sys-
Rabah, F.K.J., 2013. Impact assessment of wastewater and solid waste disposal in to Wadi tem in the Pearl River Delta, China. Ph.D. Thesis. The University og Hong Kong, China.
Gaza on groundwater, soil, and seawater. Int. J. Emerg. Technol. Adv. Eng. 3 (9), Wang, H., Jiang, X.W., Wan, L., Han, G., Guo, H., 2015. Hydrogeochemical characterization
71–81. of groundwater flow systems in the discharge area of a river basin. J. Hydrol. 527,
Rabinove, C.L., Longford, R.H., Brookhart, J.W., 1958. Saline water resources of North Da- 433–441.
kota. US Geographical Survey Water Supply Paper. 364, p. 1418. Wang, W., Song, X., Ma, Y., 2016. Identification of nitrate source using isotopic and geo-
Saleh, A., 2007. Impact of Pumping on Saltwater Intrusion in Gaza Coastal Aquifer, Pales- chemical data in the lower reaches of the Yellow River irrigation district (China). En-
tine. M.Sc. Thesis. An-Najah National University, Palestine. viron. Earth Sci. 75, 1–13.
Schoeller, H., 1965. Qualitative evaluation of groundwater resources. Methods Tech. Weinberger, G., Rosenthal, E., Ben-Zvi, A., Zeitoun, D.G., 1994. The Yarkon-Taninim
Groundw. Investig. Dev. UNESCO, pp. 54–83. groundwater basin, Israel hydrogeology: case study and critical review. J. Hydrol.
Shomar, Basem, Abu-Fakher, S., Yahya, A., 2010. Assessment of groundwater quality in the 161, 227–255.
Gaza strip, Palestine using GIS mapping. J. Water Resour. Prot. 2, 93–104. Weinthal, E., Vengosh, A., Marei, A., Gutierrez, A., Kloppmann, W., 2005. The water crisis
Sneh, A., Bartov, Y., Weissbrod, T., Rosensaft, M., 1998. Geological map of Israel, 1:200,000. in the Gaza strip: prospects for resolution. Groundwater 43 (5), 653–660.
Isr. Geol. Surv. (shapefile version). http://www.arcgis.com/home/item.html?id= Wetzelhuetter, C., 2013. Groundwater in the Coastal Zones of Asia-Pacific. Vol. 7. Spring-
ee0b0e242dc54e20bd5d953c7781f5b3, Accessed date: August 2017. er, Perth, WA, Australia: Foundation Inc. (CERF). https://doi.org/10.1007/978-94-
Taylor, S., 2016. The hydrogeochemistry of the shallow groundwater in western Sydney: a 007-5648-9.
study of salinisation processes in shallow groundwater. Ph.D Thesis. University of WHO, 2011. Guidelines for Drinking-water Quality. Vol. 38. WHO chronicle.
Sydney, New South Wales, Australia. Zaidi, F.K., Nazzal, Y., Jafri, M.K., Naeem, M., Ahmed, I., 2015. Reverse ion exchange as a
Templ, M., Filzmoser, P., Reimann, C., 2008. Cluster analysis applied to regional geochem- major process controlling the groundwater chemistry in an arid environment: a
ical data: problems and possibilities. Appl. Geochem. 23 (8), 2198–2213. case study from northwestern Saudi Arabia. Environ. Monit. Assess. 187 (10), 607.
Tóth, J., 1999. Groundwater as a geologic agent: an overview of the causes, processes, and Zilberbrand, M., Rosenthal, E., Shachnai, E., 2001. Impact of urbanization on hydrogeo-
manifestations. Hydrogeol. J. 7, 1–14. chemical evolution of groundwater and on unsaturated-zone gas composition in
Vengosh, A., Keren, R., 1996. Chemical modifications of groundwater contaminated by re- the coastal city of Tel Aviv, Israel. J. Contam. Hydrol. 50 (3–4), 175–208.
charge of treated sewage effluent. J. Contam. Hydrol. 23, 347–360.