Introduction

Sustainable development is necessary to transform the fast and overheated growth into a long and steady development suited for both the environment and the economy. Giving the planning authorities guidelines is an important milestone on the way to sustainable planning. The area of study (93.5 km2) is located at Al Arish area, North Sinai, Egypt (Fig.1). The area locates within 31°01′29″ and 31°09′02″ longitudes 33°45′11″ and 33°51′58″ latitudes, respectively. The study area located in the semi-arid belt, where annual rainfall value is less than 100 mm/year and the evaporation value is >1,641 mm/year (Egyptian Metrological Authority 1996). This area is characterized by its shallow Quaternary sands aquifer, which is highly vulnerable to pollution. The aquifer is recharged mainly by rainfall with minor recharges from underlying aquifers. Intensive agricultural practices in the delta of Wadi Al Arish combined with rainfall and well-drained soils (El Shazly and Abdel-Gaphour 1990) make the unconfined Quaternary aquifer susceptible to nitrate contamination.

Fig. 1
figure 1

Location map of the study area (shading of the digital elevation model)

Full size image

The objective is to provide current baseline groundwater quality data throughout the area and to address the potential pollution sources. Maps containing simplified quality categories were carried out to evaluate the groundwater pollution. A plan was established that placed greater emphasis on determining trends in groundwater quality and correlating water quality with possible contributing factors such as location, land-use, and aquifer depth.

Approach

The following approach is proposed to help achieve the objective; develop hydrogeologic cross-sections, collect water-level data and determine aquifer properties, collect water quality data from operating wells, and evaluate groundwater contaminant loads to the Quaternary sands aquifer.

Materials and methods

The study area (around Al Arish city) was chosen because it is characterized by intensive agricultural and human activities. The high load of different contaminants, the shallow water table, and the coarse type of soil sediments contribute to the high vulnerability of the aquifer. Groundwater quality samples were collected from wells completed in only the Quaternary aquifer. For this study, 21 wells (currently in use) were selected based on the preliminary field survey (Fig. 1). The selected wells are distributed over the study site and are used for domestic, drinking, and agricultural purposes. Groundwater samplings were carried out in April 2002, where water samples were collected after stagnant water had been pumped from the well casing. Electrical conductivity (EC), oxidation–reduction (Eh), pH, and temperature were measured directly in the field. Samples for laboratory analysis were immediately filtered in the field through 0.45-µm cellulose membrane filter. Groundwater was sampled and analyzed to determine the major components (Na, K, Ca, Mg, Cl, NO3, SO4, and HCO3) in 500-ml polyethylene bottles after three times of washing with the sample water. Collected samples for trace elements were acidified with several drops of ultra-pure hydrochloric acid. Different determination methods were used for the different elements such as titrimetric (HCO3 and CO3); Ion-Chromatography (Combination Merck/Hitachi, D6000A) for Na, K, Ca, Mg, Cl, SO4, and NO3; inductively coupled plasma mass spectrometer (ICP-MS) for metals and trace elements; and Spectophotometric (HACH Spectrophotometer DR/2000) for PO4 (Table 1).

Table 1 Chemical analyses of selected constituents in groundwater from wells sampled during investigation of the Quaternary aquifer in Al Arish area, April, 2002
Full size table

All the constructed maps and their data-base were carried out using the Geographic Information System technique (TNTmips, GIS software; Microimages 2004). The maps were constructed using the integrated geomorphologic and hydrogeologic and land-use aspects of the study area. In this regard, a total of 21 groundwater samples were collected in April, 2002, and analyzed for major, minor, and trace elements (Table 1).

Hydrogeologic setting

Shata (1959) and Morsy et al. (1995) studied the geologic units within the study area. Holocene sand sheets, sand dunes, and alluvial and beach deposits cover the study area surface. The sand sheets move freely especially during storm periods. The modern sand dunes in the coastal belt form movable elongated ridges. While the stabilized sand dune ridges are dominant in the southern part of the study area, beach deposits are made up of loose sand and calcareous sandstone rich in shell fragments. The modern alluvial deposits are composed of sand, silt, and loam. Pleistocene deposits are classified into three main units: old alluvial deposits, old beach deposits, and Kurkar deposits (Fig. 2a, b). Old alluvial deposits are composed of three series; the upper one is composed of medium to coarse sands intercalated with silt; the middle one is composed of alternation of sand and calcareous clay; while the lower one is composed of alternating gravel and coarse sands interbedded with calcareous clay. Old beach deposits are composed of well-sorted sands and sandstone intercalated with clay. Kurkar deposit is represented by calcareous sandstone; it is divided into two series, upper continental Kurkar and lower marine Kurkar (Fig. 2a, b).

Fig. 2
figure 2

a Hydrogeologic cross-section A–A′ in Al Arish area. b Hydrogeologic cross-section B–B′ in Al Arish area

Full size image

The Quaternary sands aquifer within the study area is characterized by intensive exploitation for domestic and agricultural as well as by variety of natural factors affecting groundwater chemistry. This aquifer includes different lithologic units (sand, sandstone, and calcareous sandstone), which are thought to be connected according to their hydrogeological situation (Fig. 2a, b). The principal source of recharge is percolation of rainfall and flash flood water through permeable soil and sands. Also the aquifer is connected to the underlying deeper aquifers (Cretaceous aquifers) as a result of expected faults south and east of Al Arish (Dames and Moore 1984; Fig. 2a).

Since the middle of the last century, the continuous over-pumping has exploited the aquifer, where water table decreasing dramatically with time. According to several hydrogeologic studies (Dames and Moore 1984; El Bihery and Lachmar 1994; El Alfy and Merkel 2006), groundwater table measurements indicate that there is a noticeable rising during the winter rainfall (recharge season) and slight lowering during summer (the dry season). Groundwater flow is affected by rainfall recharge, excessive pumping, evapotranspiration, and the hydraulic connection with the deeper aquifers (El Bihery and Lachmar 1994). El Alfy and Merkel (2006) carried out a water table contour map for the Quaternary aquifer in May, 2002, and found a general trend of flow from south northwards towards the coast although this direction is dissected by numerous local flow directions towards the groundwater abstraction depressions (Fig. 3). Extensive agricultural activities within the area had resulted in several groundwater depression cones which are represented by three main cones. The largest one is located in the southern part with a diameter of approximately 50 km (El Alfy and Merkel 2006). Figure 4 shows the depth to water in the northern part is <10 m, while it increases south ward to >70 m. This reflects the high vulnerability of the Quaternary aquifer especially in the northern part more than the southern part.

Fig. 3
figure 3

Water table of the Quaternary aquifer in Al Arish area in 2002 (El Alfy and Merkel 2006)

Full size image
Fig. 4
figure 4

Depth to water contour map of Al Arish area (2002)

Full size image

Land-use pattern and urbanization

Different types of land-uses are recognized within the study area. The northern part is represented by Al Arish city (urban built-up and rural areas), while the central and southern parts are prevailed by agricultural development. The main pollution sources concerning cultivated land are the excessive application of fertilizers and pesticides (Gehad 2003) as well as unsewered rural and urban lands of long-term influence, while pollution sources connected with farming are mainly from intensive livestock farms.

Results and discussion

The Quaternary aquifer supports extensive irrigation activity and is subjected to contamination by nitrate, with concentrations ranging 2–500 mg/l and the median and average values are 58 and 107 mg/l, respectively (Table 1). The lowest concentration value (2 mg/l) represents a likely background concentration. According to the classification of Madison and Brunett (1985), 33.3% of the samples have a nitrate concentration <44.3 mg/l, while 66.7% of the water samples have nitrate concentrations values >44.3 mg/l (Table 2). The spatial distribution of nitrate is interpolated using the kriging method and its fitted variogram, it shows that the southern, central, and eastern parts of the study area generally have the lower nitrate concentrations (Fig. 5). These lower concentration levels point to the limited agricultural activities and the relative deepness of the water table (Fig. 4). The northern and western parts have sites with noticeably higher concentration of nitrates (>100 mg/l). These higher values of nitrate relate to the shallow water table (<10 m) and infiltration of fertilizers used in this district together with seepage of sewage and dairy and poultry farming wastes practiced near the Al Arish city (Fig. 1). Since the concentration pattern of NO3 suggests that the aquifer is already affected by the infiltration of pollutant agents from the surface, so care should be taken in the future to preserve these high-potential pollution areas for supplying waters. Smith and Lerner (2007) proposed framework to aid assessors at the early stages of environmental risk assessment to determine the likely significance of cation exchange, sorption to organic matter, and acid buffering processes in aquifers.

Table 2 Nitrate concentration in groundwater of the Quaternary aquifer according to the classification of Madison and Brunett (1985)
Full size table
Fig. 5
figure 5

Areal distribution of nitrate (NO 3 ) concentrations (mg/l) of Quaternary aquifer. Variogram (Gaussian, anisotropy ratio = 2.0 and angle = 79°)

Full size image

Application of phosphate fertilizer to soil can increase phosphate (PO4) as well as release some other metal such as arsenic, where phosphate is adsorbed on the surface of soil grains (Pfeifer et al. 2004). The concentration values of PO4 in the collected water samples vary between 40 to 120 µg/l and its median and average values are 60 and 70 µg/l, respectively. According to TrinkwV (2001), the limitation of water quality for the domestic uses of phosphate is 6.7 mg/l and EU Drinking Water Standard for potable supply is about 2 mg/l. However, all the water samples were laid under this limit. Generally, the phosphate concentrations are increased in the southern part (120 µg/l) to the center of the area (100 µg/l), while in the northern part they are relatively small (40 µg/l), (Fig. 6). However, the land-use shows that the northern district is covered by urban built-up area and the agriculture areas are limited. Therefore, the land-use is related to groundwater pollution resulting in risk to ecology.

Fig. 6
figure 6

Areal distribution of phosphate (PO 4 ) concentrations (µg/l) of Quaternary aquifer. Variogram (rational quadratic, anisotropy ratio = 2.0 and angle = 40°)

Full size image

Vanadium (V) is a rare element naturally abundant in rocks, soil, surface and groundwater, the air, petroleum, and in crustaceans etc. (Winter 2004). The solubility of V in oxidized conditions is relatively high whereas it decreases drastically in reducing conditions (Hem 1970). Vanadium is found in different oxidation states, the most common being V5+ which has aneugenic effects. The vanadium concentrations found in the collected water samples vary between 6 and 17 µg/l, while the average and median values are 12 µg/l (Table 1). The aerial distribution of vanadium was carried out using its fitted variogram; it shows an increase in the northern part near Al Arish city, while the smaller values are recorded in the southern part (Fig. 7). This refers to the high human activities in the northern part; petroleum and carbon consumption as close to a highway with heavy traffic, use of phosphorites and increase of organic content. All water sample have vanadium concentration <100 µg/l, which is the limit for irrigation waters (Ayers and Westcot 1987) and for cattle drinking water (Kenneth and Neeltje 2002).

Fig. 7
figure 7

Areal distribution of vanadium (V) concentrations (µg/l) of Quaternary aquifer. Variogram (rational quadratic (R = 2.0 and A = 35°), nugget effect (E = 1.0 and M = 1.0))

Full size image

Chromium (Cr) can exist in water as Cr3+ cations or as anions in which the oxidation state is Cr6+. Concentrations of chromium in natural waters that have not been affected by waste disposal are commonly less than 10 µg/l (Hem 1992). In this study, the chromium concentrations vary between 4 and 107 µg/l with an average of 42 µg/l and median value of 31 µg/l (Table 1). Figure 8 shows the spatial distribution of chromium, where chromium increases in the northern part beneath and near El Arish city (>50 µg/l were attributed to the high human activities such as waste disposal), while the lower concentrations are restricted to the southern part (low human activities as well as near to recharge area). A maximum permissible concentration of 50 µg/L total chromium has been included in US mandatory drinking water standards (US Environmental Protection Agency 1976; WHO 2006), according to this, 33% of the water samples exceed this limit.

Fig. 8
figure 8

Areal distribution of chromium (Cr) concentrations (µg/l) of Quaternary aquifer. Variogram (rational quadratic, anisotropy ratio = 1.72 and angle = 131.5°)

Full size image

Arsenic (As) is an important water trace contaminant as it is one of the few substances shown to cause cancer in humans through regular consumption of drinking water. As(III) is more toxic than As(V), though the latter can be often reduced to the former in the human body. It is thought that the greater toxicity of As(III) is due to its ability to retain in the body longer since it becomes bound to dulfhydryl groups. The arsenic main application is through agriculture, where arsenic-bearing fungicides and insecticides are widely distributed (Nriagu 1994). The common arsenic-based pesticides are the insecticide lead arsenate, Pb3(AsO4)2, and calcium herbicide arsenate, both of which contain As(V); and the herbicide sodium arsenate, Na3AsO3 and Paris Green, Cu3(AsO3)2, both of which contain As(III). The spatial distribution map of arsenic in groundwater was carried out using its fitted variogram (Fig. 9). Dissolved As concentrations in the collected groundwater samples ranged from 11 to 36 µg/l with an average value of 22 µg/l (Table 1 and Fig. 9). The highest values are recorded in the central part, which is characterized by its high agricultural activities, therefore high potentiality to use arsenic fungicides and insecticides. Also the use of phosphate fertilizer plays a clear role in the releasing of arsenic into water. Application of phosphate fertilizer to soil in Southern Switzerland has also been shown to release As (Pfeifer et al. 2004). While arsenic concentrations decrease in the southern part, where the depth to water is relatively high and the agriculture activities are relatively limited. Drinking water upper level of As is 10 µg/l (WHO 2006; US Environmental Protection Agency 2001), where it now sets this value as maximum contamination level goal. All of the water samples exceed this limit, therefore due to the proved toxicology of arsenic, some restrictions and limitations should be fixed. It is technically feasible to achieve arsenic concentrations of 5 µg/l or lower using any of several possible treatment methods (e.g., electrodialysis or reverse osmosis). According to the limitations of watering animals attained by Matthess (1990; less than 200 mg/l), all of the groundwater samples within the area could be used safely.

Fig. 9
figure 9

Areal distribution of arsenic (As) concentrations (µg/l) of Quaternary aquifer. Variogram (rational quadratic (R = 2.0 and A = 142.3°), nugget effect (E = 1.0 and M = 4.0))

Full size image

Selenium (Se) concentrations in groundwater are typically of few micrograms per liter level. Selenate and selenite are the dominant forms of Se in water (Cutter and Bruland 1984; Presser 1994), but selenate has been identified as the predominant dissolved Se species (>90%) in shallow wells and drain waters (Deveral et al. 1994). Mikkelsen et al. (1988) investigated the effect of level and source of salinity on absorption of Se, where it is higher in the presence of sulfate salinity than in the presence of chloride salinity. Also Se is As-antidote so higher Se in high As water is a good sign. Selenium content of groundwater in the study area ranged between 50 and 136 µg/l with an average value of 86 µg/l. The aerial distribution of the Se in the studied area shows an increasing trend from south to northwest (Fig. 10). This is possibly related to high-agricultural practice in the northern part of the study area or increase of SO4 from the dissolution of Gypsum, also the human activities at and near Al Arish city play a significant role. According to water quality guidelines for Se of the (NAS–NAE 1973), the maximum contamination level of Se in water for drinking purposes is 10 µg/l and the maximum permissible level (MPL) for water used for irrigation is 20 µg/l. Unfortunately, all the studied water samples have higher Se more than the MPL (20 µg/l).

Fig. 10
figure 10

Areal distribution of selinum (Se) concentrations (µg/l) of Quaternary aquifer. Variogram (Gaussian (R = 3 and A = 60°), nugget effect (E = 100 and M = 0))

Full size image

Lead (Pb) tends to maintain low concentration levels in surface and groundwater as its mobility is low. Lead is present in tap water as a result of its dissolution from household plumbing systems (pipes, solder, and fittings in old buildings). The amount of lead dissolved from the plumbing system depends on pH, temperature, water hardness, and standing time of the soft and acidic water, since lead is oxidized by oxygen in acidic environments. Certain of the uses of lead by humans have tended to disperse lead widely through the environment such as addition of tetraethyl lead to promote more efficient combustion of gasoline used in automobile engines. In the study area, the concentrations of lead in the different water samples range between 0.5 and 4.3 µg/l with an average and median values of 1.2 and 0.6 µg/l (Table. 1). Although the concentration of lead all over the study area is below the guideline value of drinking water (10 µg/l), its areal distribution shows relatively higher concentration in the northern part near El Arish city (Fig. 11).

Fig. 11
figure 11

Areal distribution of lead (Pb) concentrations (µg/l) of Quaternary aquifer. Variogram (rational quadratic, anisotropy ratio = 2.1 and Angle = 255°)

Full size image

Uranium is present in the environment as a result of leaching from natural deposits, release in mill tailings, the combustion of coal and other fuels, and the use of phosphate fertilizers that contain uranium. Levels in drinking water are generally less than 1 µg/l, although concentrations as high as 700 µg/l have been measured in other geographic localities. Thorium concentration values are under detection limit of ICP (<0.5 µg/l), where Th cannot be dissoluted easily in water (Table 1). Total dissolved U concentrations are relatively high in the oxygenated waters, while lower concentrations occur generally under the reducing conditions (Hem 1970). Where the U concentration values vary between 1.7 and 14.6 µg/l with an average and median values of 6.7 and 6.3 µg/l, respectively (Table 1 and Fig. 12). The spatial distribution of U in the study area is examined using the fitted variogram. It shows that U concentration has small values in the southern part but it increases north word, especially in northeastern part. This is attributed to the high agriculture practice and the excessive use of phosphate fertilizers, where U is found as a by-product. In the addendum to the Guidelines, published by WHO (1998), a health-based guideline value of 20 µg/l was established. Most of the studied water samples have values significantly less than these guidelines.

Fig. 12
figure 12

Areal distribution of uranium (U) concentrations (µg/l) of Quaternary aquifer. Variogram linear (slope = 0.037, R = 2.0 and A = 113.6°)

Full size image

Statistical analysis

Geostatistics were carried out for the studied groundwater samples since it is difficult to pay attention to the different factors affecting groundwater contamination simultaneously. As the aquifer consist of relatively three similar units (sand, sandstone, and calcareous sandstone) and they are hydraulically connected, it is assumed that it presents one aquifer. The variables used in calculations were: depth, pH, Eh, EC, TDS, K, Na, Mg, Ca, Cl, NO3, SO4, HCO3, PO4, V, Cr, Ni, Cu, As, Se, Pb, and U, while Co, Cd, Hg, and Th variables were excluded because most of the values are under detection limits. The statistical software package Statistica 6 (StatSoft Inc., 2004) was used for the calculations. Descriptive statistics were calculated (minimum, maximum, mean, median, standard deviation, and standard error) and the different parameters were tested for normality using the nonparametric one-sample Kolmogorov–Smirnov test. The Kolmogorov–Smirnov (K-S) statistic tests the hypothesis that the data are normally distributed. A low significance value (generally less than 0.05) indicates that the distribution of the data differs significantly from a normal distribution (SPSS 2007). Frequency histograms were calculated for all parameters and then tested for normal distribution with the help of the nonparametric one-sample Kolmogorov–Smirnov test for goodness-of-fit. The theoretical distribution is set normal when using the K-S test for normality testing. A calculated significance value higher than the proposed significance level (α) indicates a normal distribution of the studied variable. In this study (α) is always equal to 0.05.

Cluster analysis

Cluster analysis was used to test water quality data and determine the hydrochemical groups of distinct populations that may be significant in the geologic context (Suk and Lee 1999; Farnham et al. 2000; Swanson et al. 2001; Davis 2002; El Alfy and Merkel 2006). The cluster analysis was carried out for the analyses results of the different water samples collected in April–May 2002. Prior to the analysis, the initial data set was standardized with criteria presented by Davis (2002). As the data matrix combines different variables (physical, chemical, and others), a standardization of the data must be achieved in order to overcome the problem of using different units in measurements of the variables. Cluster analysis of the z-standardized input data matrix of 21 cases using the hydrochemical composition of 22 variables (Table 1). Figure 13 shows that there are three clusters, which is interpreted at similarity level with a similarity linkage distance of 12.

Fig. 13
figure 13

Areal distribution of the groundwater clusters of the Quaternary aquifer

Full size image

The first cluster (I) has high values of EC, TDS, Na, Mg, Cl, V, Cr, Se, and U and moderate values of pH, Ca, SO4, Cu, As, and Pb and low values of Eh, K, NO3, HCO3, PO4, and Ni (Table 3). In case of Cl-rich groundwater, its chemistry is highly influenced by mineral weathering associated with cation-exchange reactions. This cluster is located in the northeastern part of the study area. It suggests the increasing of salinity of the Quaternary aquifer as a result of longer groundwater flow path and the deposition of calcite and aragonite minerals, as well as increase the probability of upward leakage of saline water from deeper aquifers (El Alfy and Merkel 2006). On the other hand, this cluster delineates areas with the minimum human activities in the study area.

Table 3 The variables average values of the three clusters of the Quaternary aquifer in Al Arish area, Egypt
Full size table

The second cluster (II) has high values of pH, K, Ca, NO3, SO4, PO4, Ni, Cu, As and Pb; moderate values of depth, Eh, EC, TDS, Na, Mg, Cl, HCO3, Cr, Se, and U; and low values of V (Table 3). This cluster is located in the central part of the studied area extending from south to north (Fig. 13). This group shows moderate flow path and percolation of groundwater, it is characterized by the mixing of recharge sources (rainfall and runoff). This is confirmed with the moderate negative saturation indices of some minerals (El Alfy and Merkel 2006). Also this group points to the areas which are affected highly by human activities (corrosion of water system pipes, agricultural practice, fertilizers, and pesticides).

The third cluster (III) is characterized by its relatively low concentration values of pH, EC, TDS, K, Na, Mg, Ca, Cl, SO4, PO4, V, Cr, Ni, Cu, As, Se, Pb, and U and high values of depth, Eh, and HCO3 (Table 3). The groundwater of this class is largely influenced by mineral dissolution. This group is distributed in the southwestern part of the studied area (Fig. 13), it reflects the short groundwater path from the recharge area (recharged by rainfall, runoff infiltration, and groundwater lateral flow from south to north), and therefore leaching of some minerals takes place. This is confirmed with the high negative saturation indices of several minerals within the area (El Alfy and Merkel 2006); also this cluster points to the moderate human activities in this area.

Factor analysis

Factor analysis describes processes occurring in the investigated environment, identification of the factors allows genetic interpretation of the environment. Therefore, it is a useful tool to define the factors that impact on the groundwater quality and its hydrochemical processes (Dalton and Upchurch 1978; Al Yamany et al. 1994; El Alfy and Merkel 2006). Prior knowledge of hydrogeologic processes affecting the studied environment is required for an effective application of factor analyses (Lawrence and Upchurch 1983). Prior to analysis, the initial data set was standardized with criteria presented by Davis (2002). R-mode factor analysis was carried out for the entire set of data, where it used 21 cases and 21 variables, and then the factor loadings matrix was rotated to an orthogonal simple structure according to Varimax rotation technique (Davis 2002). The results of this operation are high factor loadings (close to 1 or −1) obtained for the variables correlated in the factor and low factor loadings (close to 0) obtained for the remaining variables. The number of factors which best describe the variance of the analyzed data have Eigenvalue > 1 and therefore can be reasonably interpreted. The sum of squared factor loadings shows how the obtained factors describe variance of particular variables and then the factor scores were calculated for each sample.

There are six factors representing a cumulative total variance of 82.5% (Table 4). The first factor obtained explains the biggest part of variance, it accounts 31% of the total variance and 6.5 of the Eigenvalue. High factor loadings indicate strong relationship between the variable and the factor describing this variable. This factor has high loadings with TDS, Cl, Mg, Na, and Ca (0.95, 0.94, 0.93, 0.90, and 0.80, respectively) and moderate loadings with SO4 (0.67). The factor loadings reflect two processes affecting the water chemistry; the first one is the evaporation of the groundwater (depth to the water table is <10 m in the northern part) and the circulation of dissolved salts in the irrigation water itself (Fig. 4). The second process is the role of dissolution of several minerals occurred in aquifer sediments such as gypsum (CaSO4·2H2O), anhydrite (CaSO4), and halite (NaCl), due to the high loading of Na and Cl as well as Ca and SO4 (El Alfy and Merkel 2006).

Table 4 Factor analysis loadings and Eigenvalues of Quaternary aquifer in Al Arish area
Full size table

The second factor accounts 19.7% of the total variance and 4.1 of the Eigenvalue, it has high loadings with K and U (0.89 and 0.86), moderate loading with NO3, SO4, depth, and PO4 (0.59, 0.45, −0.38, and 0.33). This indicates that the influence of human activity is one of the most important factors controlling groundwater chemical composition in the study area. This factor corresponds to the role of unwise use of potassium, nitrate, sulfate, and phosphate fertilizers, where U is often found as a by-product (Figs. 5, 6, and 12). However, the nature and spatial distribution of contamination observed in the aquifer indicate that the main contamination sources came from the agricultural activities and unsewered rural and urban areas.

The third factor accounts for 10.9% of the total variance and 2.3 of the Eigenvalue, there are high loadings for Cr, V, and Se (0.83, 0.81, and 0.67) and moderate loading with the depth (−0.48). This is referred to high human activites; waste disposal, petroleum and carbon consumption, and agricultural practices especially in the urbanized northern part of the study area (Figs. 7, 8, and 10).

The fourth factor accounts 8.5% of the total variance and 1.8 of the Eigenvalue, it has high loading with Pb, Cu, and Ni (0.91, 0.88, and 0.77) and moderate loading with NO3 (0.59). This factor points to the contamination of water as a result of these elements dissolution and corrosion from household plumbing systems (Fig. 11).

The fifth factor shows the role of rainfall recharge through the soil and this can be evidenced by the high loading of HCO3 and depth (−0.79 and −0.64) and moderate loading with pH (0.48) which occurs under alkaline conditions. It accounts 7.2% of the total variance and 1.5 of the Eigenvalue. On the other hand it reflects the the dissolution of HCO3 (aqueous phase) and its deposition as CO3 (solid phase) with the increase of percolation depth in the aquifer pore spaces forming calcareous sandstone (El Alfy and Merkel 2006).

The sixth factor accounts 5.3% of the total variance and 1.1 of the Eigenvalue, it has high loading with As (0.89) and moderate loading with PO4 and pH (0.44 and 0.31). This factor reflects human impact on the aquifer, where it shows the role of unwise use of arsine pesticides and phosphate fertilizers in the agricultural practice (Fig. 9).

Conclusions

The area of study is located at Al Arish area (North Sinai Peninsula, Egypt), it is a promising area for agricultural and industrial projects, but sustainable development of the area is retarded by pollution of the limited groundwater resources. The fast growth of human activities in the study area is accompanied by the over-exploitation of the water resources. This area is characterized by a Quaternary aquifer at relatively shallow depths, which is most vulnerable to pollution. Hydrochemistry of the aquifer was studied involving the description of groundwater constituents and their relationships with each other. Numerous areal distribution maps were constructed for correlating water quality with possible contributing factors such as location, land-use, and aquifer depth. These maps identified possible anthropogenic and natural processes affecting groundwater quality of the Quaternary aquifer in the study area. The nature and spatial distribution of contamination observed in the aquifer indicate that the main contamination sources came from agricultural activities, unsewered rural and urban areas, and upward leakage of highly saline water. Cluster analysis was used to classify groundwater chemistry and determine the hydrochemical groups, Q-mode dendogram is interpreted and there are three main clusters. Factor analyses identify six factors that have impacted on the groundwater quality and its hydrochemical processes; dissolution of gypsum and halite, groundwater evaporation, unwise use of fertilizers and pesticides, waste disposal, and rainfall recharge.