The water supply in the Gaza Strip substantially depends on the groundwater resource of the Gaza coastal aquifer. The climate changes and the over-exploiting processes negatively impact the recovery of the groundwater balance. The climate variability is characterized by the decline in the precipitation of −5.2% and an increase in temperature of +1 °C in the timeframe of 2020–2040. The potential evaporation and the sunshine period are expected to increase by about 111 mm and 5 hours, respectively, during the next 20 years. However, the atmosphere is predicted to be drier where the relative humidity will fall by a trend of −8% in 20 years. The groundwater abstraction is predicted to increase by 55% by 2040. The response of the groundwater level to climate change and groundwater pumping was evaluated using a model of a 20-neuron ANN with a performance of the correlation coefficient (r)=0.95–0.99 and the root mean square error (RMSE)=0.09–0.21. Nowadays, the model reveals that the groundwater level ranges between −0.38 and −18.5 m and by 2040 it is expected to reach −1.13 and −28 m below MSL at the northern and southern governorates of the Gaza Strip, respectively.

  • Groundwater in the Gaza Strip faces serious decline of about −18 m below mean sea level.

  • The climate change affects the future groundwater balance by a decline of about −10 m over 2020–2040.

  • Nontraditional water resources should be introduced to cope with the water scarcity in the Gaza Strip.

Groundwater is the dominant water resource for more than half of the domestic and agricultural needs on Earth (Anderson 2017). The Mediterranean arid and semi-arid regions are experiencing serious water supply threats due to the impacts of extreme climate changes which affect the natural recovery of the limited groundwater resources (Gopalakrishnan et al. 2019; Hussain et al. 2019). Moreover, the abuse of the groundwater through high over-pumping processes causes severe and subnational depression in the groundwater table to levels below the mean sea level (MSL) and this, in turn, causes prolonged salinization, irreversible economic losses, and a serious threat to food security (Zekri et al. 2017). In this scope, groundwater modeling-based management becomes crucial to evaluate the groundwater level variability to develop effective mitigation strategies and efficient management policies in order to preserve the groundwater resources sustainably (Gladden & Park 2016; Karimi et al. 2019). The data-driven statistical models are common types of climate and groundwater models that are widely developed to simulate the long-term time series data of groundwater level for future forecasting and decision-making (Yan & Ma 2016; Zhou et al. 2017). In addition, artificial intelligence (AI) techniques are intelligent data-driven methods that can capture efficiently the nonlinear relationships between the groundwater level and other related climatic parameters (Emamgholizadeh et al. 2014). In applicability, the artificial neural networks (ANNs) are the most promising and competitive algorithms among the AI algorithms which are widely utilized in the applications of groundwater modeling (Chang et al. 2016; Ebrahimi & Rajaee 2017). The power of ANN models in groundwater studies refers to their advantages as groundwater management tools for studying the impact of water policies and intervention plans on the sustainability and the recovery of groundwater resources (Krishna et al. 2008; Trichakis et al. 2009; Mohanty et al. 2010). The statistical data-driven methods have been exploited by many researchers to simulate the climate and hydrology of water (Kumbuyo et al. 2014; Al-Najjar et al. 2020). The stochastic ARIMA models are widely used in water resources management applications, especially for modeling hydrological stream flows, groundwater level fluctuations, and drought patterns (Myronidis et al. 2018; Takafuji et al. 2018; Sakizadeh et al. 2019; Al-Najjar et al. 2020). Moreover, the ability of AI in hydrology and water resources management and for groundwater level modeling has been examined by many studies (Rakhshandehroo et al. 2012; Ghose et al. 2018; Kouziokas et al. 2018; Guzman et al. 2019; Lee et al. 2019; Tang et al. 2019). In particular, this study aims to simulate the fluctuations in the groundwater level of the Gaza coastal aquifer in light of the climate change consequences.

The Gaza Strip (Figure 1) is a littoral strip of land that comprises an area of 365 km2 on the southeast coast of the Mediterranean Sea with a stretch of 42 km and a width that ranges between 6 and 12 km. The Gaza Strip is categorized as one of the world's highly populated areas with a population of about two million inhabitants (PCBS 2020).

Figure 1

The geographical location of the Gaza Strip.

Figure 1

The geographical location of the Gaza Strip.

Several agencies identify the current situation in the Gaza Strip as a severe humanitarian crisis where the coastal aquifer that is the only accessible resource for water supply is heavily contaminated and suffers from chronic degradation conditions (UN 2012; PWA 2014). Quantitatively, the total water extracted from the Gaza coastal aquifer is reported by about four times the amount, i.e., 55 million cubic meters per year, that the aquifer can sustainably produce each year (PWA 2013, 2014, 2015). The groundwater of the Gaza coastal aquifer is, in general, neutral with a slight trend to the alkalinity condition, with a pH value that ranges from 6.7 to 8.3, due to the existence of the carbonate mineral dissolution in the form of bicarbonate (HCO3). The electrical conductivity (EC) ranges between 597 and 30,400 μS/cm and demonstrates recorded values ranging between 370 and 18,848 mg/L for total dissolved solids (TDS). The concentrations of the detected ions in the groundwater for chloride (Cl), sodium (Na+), magnesium (Mg+2), calcium (Ca+2), potassium (K+), (SO4−2), (HCO3), and (NO3) were 78–10,318, 41–5,400, 23–665, 25–657, 1.4–155, 8–1,604, 101–1,280, and 18–496 mg/L, respectively (Abu-alnaeem et al. 2018). Climatic drought is a significant occurrence in the Gaza Strip owing to climate change consequences which adversely influence the vulnerability of the coastal aquifer and the sustainability of agricultural activities. Drought investigation studies reveal that the incidence of drought occurrence has increased from about 20% in the 1970s to more than 80% in the last ten years (Al-Najjar et al. 2020). The increase in drought events can be imputed to the decrease in precipitation and the spike of evaporation related to temperature rises. Generally, the total average annual rainfall in the Gaza Strip is typically estimated to be about 370 mm. Climate models indicate that the amount of precipitation is diminishing, and the local downscaling of the drought demonstrates that the southern Gaza Strip governorates are in prolonged drought, while the northern areas are experiencing drought every 9–12 years (Al-Najjar et al. 2020).

The study aims at investigating the climate change traces and modeling the groundwater level of the Gaza coastal aquifer according to the approaches described in Figure 2. The methodology relies on the merit of coupling the stochastic time series models and the artificial neural networks (ANNs) to construct an integrated groundwater management model capable of describing the groundwater dynamic of the Gaza coastal aquifer in terms of the change in climate conditions over the next 20 years to 2040.

Figure 2

The methodology of the study.

Figure 2

The methodology of the study.

The available data for climate and water supply in the Gaza Strip were collected through a field survey of the meteorological stations and the monitoring groundwater wells; however, the lack of possibility and the operational phase of these monitoring stations led to a lack of available data to a period extending only to 2016 for climate parameters and 2018 for groundwater level. The available historical records for the climate parameters of the monthly precipitation (P), minimum temperature (Tmin), average temperature (Tavg), maximum temperature (Tmax), evaporation (Eo), sunshine (So), and humidity (Ho) for the period of 1974–2016 were collected, screened, and statistically analyzed for the meteorological stations distributed over the Gaza Strip. The groundwater table level due to the excessive pumping and the low recharge rate shows significant depression; therefore, the historical water-table level records from ten groundwater wells, shown in Figure 1, which exhibit an influential change in the water level and over-abstraction activity from the Gaza coastal aquifer were collected throughout 1974–2018. In generating the models, 90% of the observed data were utilized for calibration while the other 10% of the data were used for validation and testing the performance of the model in forecasting the future. In terms of models, the stochastic autoregressive integrated moving average (ARIMA) models, mathematically described in Equation (1), were used in this study to forecast the future trend of the time series (Box & Jenkins 1976; Kottegoda 1990; Tong 1990; Polyak 1996; Sharma et al. 2019).
formula
(1)
where, ϕi is the ith autoregressive (AR) parameters, Φis is the ith seasonal autoregressive (AR) parameters, ϴi is the ith moving average (MA) parameters, ϴis is the ith seasonal moving average (MA) parameters, B is the backshift operator, d is the differencing, D is the seasonal differencing, S is the seasonality period, and ɛt is a noise random component.
Moreover, the logistic sigmoid ANN of multi-layer feed-forward perceptron (MLP), shown in Equations (2) and (3), with a single hidden layer was exploited to obtain the relationships between the climate factors and groundwater level (Bishop 1995; Haykin 2009; Sahoo & Jha 2013).
formula
(2)
formula
(3)
where, wjk is the connection weight between jth node of hidden layer and output node k, zj is the output of the jth hidden neuron resulting from the input data, and bj is the connection weight for bias term.

For quality control and quality assurance, both R-statistical analysis language and the Statistical Package for Social Sciences (SPSS) were used in this research to evaluate the nature of the rainfall time series. The R-statistical analysis language is highly recommended for climatic studies because of the vast availability of case studies. The SPSS is a familiar and established tool to confirm the consistency of results. The MATLAB software was used to establish the ANN where the MATLAB has a high and fast ability to manipulate the long and complex networks better than R. In addition, the use of MATLAB gives some indication about the quality control of the data integrity and the model workability.

Forecasting of climate parameters

The forecasted rainfall time series data up to 2040 was obtained from Al-Najjar et al. (2020). The rainfall model reveals that the rainfall declines by a yearly average trend of about −0.26%, hence the average yearly rainfall for the Gaza Strip over the next 20 years is predicted as 370 mm. The stochastic time series model (Tables 1 and 2) of the structure (3,1,2) (2,1,1)12 was recommended to simulate the manner of minimum temperature, average temperature, and maximum temperature. Moreover, the stochastic models of (2,1,5) (2,1,1)12, (5,1,3) (2,1,3)12, (4,1,2) (5,1,2)12 were structured to demonstrate the time series of evaporation, sunshine and humidity, respectively. In terms of climate change tracking, as shown in Figure 3, the effect of climate change is tangible in the Gaza Strip where, according to expectations of the stochastic models, there is a significant increasing trend in the temperature of approximately +0.03 to +0.09 °C each year, which is averagely compatible with the IPCC assessment scenarios of climate change where the worst climate scenario of RCP 8.5 indicates that the temperature will increase by 1.5 °C above the normal by 2040. However, in the most optimistic scenario of RCP 2.8 the increase in temperature will be about 1 °C by 2040. Therefore, it is expected that the average temperature in the Gaza Strip be between 21 °C in the winter seasons and 25 °C in the summer seasons with an overall average temperature of about 23 °C by the year 2040.

Table 1

The model non-seasonal autoregressive (ar) and moving average (ma) parameters for the climate parameters

ParameterModelΦ1Φ2Φ3Φ4Φ5ϴ1ϴ2ϴ3ϴ4ϴ5
Minimum temperature (3,1,2) (2,1,1)12 0.8972 −0.273 0.0476 – – −1.5116 0.5117 – – – 
Average temperature (3,1,2) (2,1,1)12 0.9692 −0.2214 0.0184 – – −1.6557 0.6560 – – – 
Maximum temperature (3,1,2) (2,1,1)12 0.9377 −0.1725 0.0154 – – −1.6136 0.6203 – – – 
Evaporation (2,1,5) (2,1,1)12 −0.1266 0.6175 – – – −0.5577 −0.7155 0.4327 −0.1447 −0.0148 
Sunshine (5,1,3) (2,1,3)12 −1.0384 −0.8514 0.0600 −0.0779 −0.0676 0.2856 −0.0353 −0.8846 – – 
Humidity (4,1,2) (5,1,2)12 −0.8294 0.0325 0.008 0.0839 − −0.0415 −0.8815 – – – 
ParameterModelΦ1Φ2Φ3Φ4Φ5ϴ1ϴ2ϴ3ϴ4ϴ5
Minimum temperature (3,1,2) (2,1,1)12 0.8972 −0.273 0.0476 – – −1.5116 0.5117 – – – 
Average temperature (3,1,2) (2,1,1)12 0.9692 −0.2214 0.0184 – – −1.6557 0.6560 – – – 
Maximum temperature (3,1,2) (2,1,1)12 0.9377 −0.1725 0.0154 – – −1.6136 0.6203 – – – 
Evaporation (2,1,5) (2,1,1)12 −0.1266 0.6175 – – – −0.5577 −0.7155 0.4327 −0.1447 −0.0148 
Sunshine (5,1,3) (2,1,3)12 −1.0384 −0.8514 0.0600 −0.0779 −0.0676 0.2856 −0.0353 −0.8846 – – 
Humidity (4,1,2) (5,1,2)12 −0.8294 0.0325 0.008 0.0839 − −0.0415 −0.8815 – – – 
Table 2

The model seasonal autoregressive (ar) and moving average (ma) parameters for the climate parameters

ParameterModelΦ1sΦ2sΦ3sΦ4sΦ5sϴ1sϴ2sϴ3s
Minimum temperature (3,1,2) (2,1,1)12 −0.0482 0.0605 – – – −1.0000 − − 
Average temperature (3,1,2) (2,1,1)12 −0.0707 0.0927 – – – −0.9723 − −- 
Maximum temperature (3,1,2) (2,1,1)12 −0.0961 0.0244 – – – −0.9992 – – 
Evaporation (2,1,5) (2,1,1)12 0.1487 −0.1042 – – – −0.9210 – – 
Sunshine (5,1,3) (2,1,3)12 −0.576 −0.8670 – – – −0.4308 0.5696 −0.9059 
Humidity (4,1,2) (5,1,2)12 −1.0158 −0.3057 −0.1495 −0.1191 −0.1211 0.0613 −0.6850 – 
ParameterModelΦ1sΦ2sΦ3sΦ4sΦ5sϴ1sϴ2sϴ3s
Minimum temperature (3,1,2) (2,1,1)12 −0.0482 0.0605 – – – −1.0000 − − 
Average temperature (3,1,2) (2,1,1)12 −0.0707 0.0927 – – – −0.9723 − −- 
Maximum temperature (3,1,2) (2,1,1)12 −0.0961 0.0244 – – – −0.9992 – – 
Evaporation (2,1,5) (2,1,1)12 0.1487 −0.1042 – – – −0.9210 – – 
Sunshine (5,1,3) (2,1,3)12 −0.576 −0.8670 – – – −0.4308 0.5696 −0.9059 
Humidity (4,1,2) (5,1,2)12 −1.0158 −0.3057 −0.1495 −0.1191 −0.1211 0.0613 −0.6850 – 
Figure 3

Time series modeling for: (a) minimum temperature, (b) average temperature, (c) maximum temperature, (d) sunshine, (e) evaporation, and (f) humidity.

Figure 3

Time series modeling for: (a) minimum temperature, (b) average temperature, (c) maximum temperature, (d) sunshine, (e) evaporation, and (f) humidity.

The climatic parameters of evaporation, sunshine, and humidity also reflect tangible traces regarding the tendency of climate changes in the Gaza Strip. The annual period of sunshine shows a rising behavior by +1 hour where the simulation manner indicates that the sunshine is expected to reach 2,891 hours by the year 2040. In response, the evaporation reveals an increasing trend of about 7 mm per year. However, the humidity demonstrates a declining yearly trend of −0.3%.

The performance testing of the models that were examined using the correlation coefficient (r) and the root mean square error (RMSE) shows that the stochastic models introduce proper simulations for the climate data and gives a good indication in forecasting the future. The data fitted for temperature by the suggested stochastic models indicate a robust simulation manner where the models give a correlation coefficient of r=98–99% and an RMSE of 0.69–0.94. The performance of the sunshine and evaporation models reveal high simulating quality with a correlation coefficient of r=0.94–0.96, RMSE=21.68–23.49 and r=0.97, RMSE=10.30–11.03 for the sunshine and evaporation, respectively. However, the stochastic model of humidity shows less performance in fitting the data and in forecasting the future, where the correlation coefficient was r=0.74–0.81, RMSE=3.07–2.88.

Modeling and forecasting of groundwater

The groundwater table was evaluated and simulated through a 20-neuron ANN which demonstrates the linkage processes for the relationship between the inputs and the output within a reasonable period. The monthly data of the P(t), Tmin(t), Tavg(t), Tmax(t), Eo(t), So(t), and Ho(t) were chosen as a combination of inputs to represent the comprehensive influence of the climatic and hydrological factors on the aquifer water level. The combination of the input parameters within the ANN enhances the interpretation of the groundwater time series and it shows better presentations for the outlier points of the observed data. The network was developed using MATLAB by training the network on 540 combinations of the data for each time series of the ten groundwater wells. In terms of performance testing, the generated stochastic-ANN model shows a valid presentation of the observed groundwater level. Overall, as shown in Figure 4, the stochastic models describe the relationship between the observed and the simulated data by a correlation coefficient (r) of 94–99% and RMSE of 0.1–0.22.

Figure 4

Performance testing of ANN.

Figure 4

Performance testing of ANN.

The fluctuation of groundwater levels was represented by an ANN of 20 neurons, and the model was generalized (Table 3) to simulate the groundwater level at all of the ten water wells. Generally, the groundwater levels demonstrate a declining trend over time due to the groundwater over-pumping activities and the negative effects of climate changes.

Table 3

ANN parameters for groundwater

Water wellANN parameters
P(t)Tmin(t)Tavg(t)Tmax(t)Eo(t)So(t)Ho(t)
C/48 0.3144 0.2179 −0.2246 −0.2746 0.0283 −0.0411 −0.1023 
E/45 0.4368 0.207 −0.2662 −0.4434 −0.0791 0.0106 −0.0032 
G/24B −0.9390 −0.3240 0.7909 0.0644 −0.1924 −0.1409 0.0095 
F/68B 0.1147 0.4551 0.1237 −0.544 −0.2337 −0.0305 −0.0105 
S/15 −0.0248 0.2567 0.1335 −0.3582 −0.1767 −0.0672 −0.0879 
L/86 0.4634 0.2497 −0.5946 −0.2145 −0.1113 0.0214 0.1273 
L/66 −0.1522 0.1769 0.3225 −0.4422 −0.3210 0.0373 −0.1208 
N/12 −0.0313 0.4298 0.1935 −0.5718 −0.3219 −0.1102 0.0795 
N/16 −0.3282 0.6212 0.4375 −0.8122 −0.3422 0.0127 0.1050 
P/48A 0.0778 −0.8018 0.1100 0.7464 −0.0357 −0.1446 −0.1755 
Water wellANN parameters
P(t)Tmin(t)Tavg(t)Tmax(t)Eo(t)So(t)Ho(t)
C/48 0.3144 0.2179 −0.2246 −0.2746 0.0283 −0.0411 −0.1023 
E/45 0.4368 0.207 −0.2662 −0.4434 −0.0791 0.0106 −0.0032 
G/24B −0.9390 −0.3240 0.7909 0.0644 −0.1924 −0.1409 0.0095 
F/68B 0.1147 0.4551 0.1237 −0.544 −0.2337 −0.0305 −0.0105 
S/15 −0.0248 0.2567 0.1335 −0.3582 −0.1767 −0.0672 −0.0879 
L/86 0.4634 0.2497 −0.5946 −0.2145 −0.1113 0.0214 0.1273 
L/66 −0.1522 0.1769 0.3225 −0.4422 −0.3210 0.0373 −0.1208 
N/12 −0.0313 0.4298 0.1935 −0.5718 −0.3219 −0.1102 0.0795 
N/16 −0.3282 0.6212 0.4375 −0.8122 −0.3422 0.0127 0.1050 
P/48A 0.0778 −0.8018 0.1100 0.7464 −0.0357 −0.1446 −0.1755 

The data of the groundwater level and the findings of the model, shown in Figure 5, reveal that the groundwater resource faces real threats in terms of water balance. Historically, the groundwater level was in an abundant state in the 1970s. However, the groundwater is declining and it is significantly overexploited where the groundwater level shows a drop to less than −15 m below the MSL.

Figure 5

Groundwater simulation for: (a) C/48, (b) E/45, (c) G/24B, (d) F/68B, (e) S/15, (f) L/86, (g) L/66, (h) N/12, (i) N/16, and (j) P/48H.

Figure 5

Groundwater simulation for: (a) C/48, (b) E/45, (c) G/24B, (d) F/68B, (e) S/15, (f) L/86, (g) L/66, (h) N/12, (i) N/16, and (j) P/48H.

The general indication of the groundwater model reveals that the level is decreasing continuously except for some parts in the eastern region of the study area that form semi-separated small basins. The groundwater simulation findings, shown in Figure 6, illustrate that the groundwater level drop is between −0.38 and −18.49 m below MSL in 2020 and between −1.13 and −27.77 m below MSL in 2040. Geographically, the southern governorates of the Gaza Strip, especially in Rafah, show more deficit in the groundwater balance than other locations where the decline in the groundwater will reach −27.77 m below MSL in 2040.

Figure 6

Groundwater level of the Gaza coastal aquifer.

Figure 6

Groundwater level of the Gaza coastal aquifer.

The southern part of the Gaza Strip is the most populated area. The municipal wells pump the groundwater at an extensive rate of more than 100 m3 per hour which adversely affects the quality and quantity of the Gaza coastal aquifer in this part of the area. Regionally, the groundwater depression cone started to form in 1992 and followed an expansion pattern towards the north of the Gaza Strip. The diameter of the cone was less than 1 km in 1992 and it is expected to reach 4–5 km in 2040. In consequence, the zero-lateral flow recharging of the groundwater was supposed as deep groundwater wells were excavated along the eastern border of the Gaza Strip to catch the water before passing to the Gaza Strip and this, in turn, causes seawater intrusion which is the most dominant phenomenon impacting the quality of the groundwater by lifting the chloride concentration to an acceptable level. In comparison with the Mediterranean Sea countries, the groundwater level investigations show significant depression below sea water level where in Cairo the drop could reach −27.81 m (Mohamed Ibrahem 2020). In Jordan, the groundwater is being rapidly depleted with observed groundwater level declines of 0.9–3.5 m per year (Yoon et al. 2021).

The environment of the Gaza Strip is changing dramatically as a result of global warming, putting a strain on the Gaza coastal aquifer, which is the only viable water source. Furthermore, the rapid increase in groundwater pumping operations in tandem with the rapid increase in population has a direct impact on the coastal aquifer's long-term productivity. The lack of a reliable simulator to research groundwater behavior for the Gaza coastal aquifer inhibits proper knowledge of groundwater dynamics through time and space. The stochastic and ANN models are both capable of simulating data and identifying abnormal data values. The modeling outputs reveal the following points:

  • The period 2020–2040 is critical for climate and water security in the Gaza Strip, as monthly average precipitation will be assigned to about 21–33 mm by 2040, and the temperature is expected to increase by +1 °C by 2040.

  • The low recharge rate due to the decrease in rainfall and the high temperature and evaporation rate are causing a depression in the groundwater level to reach a low level of about −28 m in 2040.

  • The variation in the groundwater distribution pattern of the Gaza coastal aquifer will alter by about 51%, indicating greater deterioration.

  • At all areas in the Gaza Strip, the groundwater table will be below the MSL. In this context, rapid water intervention plans and optimal management strategies are strongly required to support the coastal aquifer's long-term sustainability and to boost the Gaza Strip's economic activity. The management measures should primarily improve the manner of groundwater use, while also encouraging the use of non-conventional resources like seawater desalination and wastewater reclamation.

  • To stop the groundwater decline in the southern governorates a quantity of water equaling 60 and 90 million cubic meters per year is needed by 2020 and 2040, respectively. However, at present, the whole area of the Gaza Strip needs a quantity of water equal to 123 million cubic meters, and by 2040 the quantity should be 193 million cubic meters.

All relevant data are included in the paper or its Supplementary Information.

Al-Najjar
H.
,
Ceribasi
G.
,
Dogan
E.
,
Abualtayef
M.
,
Qahman
K.
&
Shaqfa
A.
2020
Stochastic time-series models for drought assessment in the Gaza Strip (Palestine)
.
Journal of Water and Climate Change
11
(
S1
),
85
114
.
Anderson
D. J.
2017
Coastal Groundwater and Climate Change. WRL Technical Report 2017/4
.
Water Research laboratory, School of Civil Engineering, University of New South Wales
,
Sydney
,
Australia
.
Bishop
C. M.
1995
Neural Networks for Pattern Recognition
.
Oxford University Press
,
New York, NY
,
USA
.
Box
G. E. P.
&
Jenkins
G. M.
1976
Time Series Analysis: Forecasting and Control
, 2nd edn.
Holden-Day
,
San Francisco, CA
,
USA
.
Chang
F. J.
,
Chang
L. C.
,
Huang
C. W.
&
Kao
I. F.
2016
Prediction of monthly regional groundwater levels through hybrid soft-computing techniques
.
Journal of Hydrology
541
,
965
976
.
Gladden
L. A.
&
Park
N. S.
2016
Coastal Groundwater Development: Challenges and Opportunities
, 1st edn.
CRC Press
,
Boca Raton, FL
,
USA
.
Gopalakrishnan
T.
,
Hasan
M. K.
,
Haque
A. S.
,
Jayasinghe
S. L.
&
Kumar
L.
2019
Sustainability of coastal agriculture under climate change
.
Sustainability
11
(
24
),
7200
.
Guzman
S. M.
,
Paz
J. O.
,
Tagert
M. L. M.
&
Mercer
A. E.
2019
Evaluation of seasonally classified inputs for the prediction of daily groundwater levels: NARX networks vs support vector machines
.
Environmental Modelling and Assessment
24
,
223
234
.
Haykin
S.
2009
Neural Networks and Learning Machines
, 3rd edn.
Pearson Education, Inc.
,
Upper Saddle River, NJ
,
USA
.
Karimi
L.
,
Motagh
M.
&
Entezam
I.
2019
Modeling groundwater level fluctuations in Tehran aquifer: results from a 3D unconfined aquifer model
.
Groundwater for Sustainable Development
8
,
439
449
.
Kottegoda
N. T.
1990
Stochastic Water Resources Technology
, 1st edn.
Palgrave Macmillan
,
London
,
UK
, p.
384
.
Kouziokas
G. N.
,
Chatzigeorgiou
A.
&
Perakis
K.
2018
Multilayer feed forward models in groundwater level forecasting using meteorological data in public management
.
Water Resources Management
32
,
5041
5052
.
Krishna
B.
,
Satyaji Rao
Y. R.
&
Vijaya
T.
2008
Modelling groundwater levels in an urban coastal aquifer using artificial neural networks
.
Hydrological Processes
22
(
8
),
1180
1188
.
Kumbuyo
C. P.
,
Yasuda
H.
,
Kitamura
Y.
&
Shimizu
K.
2014
Fluctuation of rainfall time series in Malawi: an analysis of selected areas
.
GEOFIZIKA
31
(
1
), 13–28.
Mohamed Ibrahem
S. M.
2020
Groundwater hydrology and characteristics of the tertiary aquifers, Northwest Cairo, Egypt
.
NRIAG Journal of Astronomy and Geophysics
9
(
1
),
420
432
.
doi:10.1080/20909977.2020.1751924
.
Mohanty
S.
,
Jha
M. K.
,
Kumar
A.
&
Sudheer
K. P.
2010
Artificial neural network modeling for groundwater level forecasting in a river island of eastern India
.
Water Resources Management
24
,
1845
1865
.
Myronidis
D.
,
Ioannou
K.
,
Fotakis
D.
&
Dörflinger
G.
2018
Streamflow and hydrological drought trend analysis and forecasting in Cyprus
.
Water Resources Management
32
(
5
),
1759
1776
.
PCBS
2020
Estimated Population in the Palestinian Territory mid-Year by Governorate, 1997–2021
.
IOP Publishing Physics Web
.
Available from: http://www.pcbs.gov.ps (accessed 25 December 2020)
.
Polyak
I.
1996
Computational Statistics in Climatology
.
Oxford University Press
,
Oxford
,
UK
.
PWA
2013
National Water and Wastewater Policy and Strategy for Palestine: Toward Building A Palestinian State From Water Perspective
.
Palestinian Water Authority
.
PWA
2014
Gaza Strip: No Clean Drinking Water, Not Enough Energy, and Threatened Future
.
Palestinian Water Authority
.
PWA
.
2015
Gaza Strip: Desalination Facility Project: Necessity, Politics and Energy
.
Palestinian Water Authority
.
Rakhshandehroo
G. R.
,
Vaghefi
M.
&
Asadi Aghbolaghi
M.
2012
Forecasting groundwater level in Shiraz plain using artificial neural networks
.
Arabian Journal for Science and Engineering
37
,
1871
1883
.
Sharma
P.
,
Machiwal
D.
,
Jha
M. K.
2019
Overview, current status, and future prospect of stochastic time series modeling in subsurface hydrology
. In:
GIS and Geostatistical Techniques for Groundwater Science
(
Venkatramanan
S.
,
Viswanathan
P. M.
&
Chung
S. Y.
, eds).
Elsevier
,
Amsterdam
,
The Netherlands
, pp.
133
151
.
Takafuji
E. H. D. M.
,
Rocha
M. M. D.
&
Manzione
R. L.
2018
Groundwater level prediction/Forecasting and assessment of uncertainty using SGS and ARIMA models: a case study in the Bauru Aquifer System (Brazil)
.
Natural Resources Research
28
,
407
453
.
Tang
Y.
,
Zang
C.
,
Wei
Y.
&
Jiang
M.
2019
Data-driven modeling of groundwater level with least-square support vector machine and spatial–temporal analysis
.
Geotechnical Geological Engineering
37
,
1661
1670
.
Tong
H.
1990
Nonlinear Time Series: A Dynamical System Approach
.
Oxford University Press
,
Oxford
,
UK
.
Trichakis
I. C.
,
Nikolos
I. K.
&
Karatzas
G. P.
2009
Optimal selection of artificial neural network parameters for the prediction of a karstic aquifer's response
.
Hydrological Processes
23
(
20
),
2956
2969
.
UN
2012
Gaza 2020: A Livable Place? A Report by the United Nations Country Team in the Occupied Palestinian Territory
.
The United Nations
,
New York
,
USA
.
Yoon
J.
,
Klassert
C.
,
Selby
P.
,
Lachaut
T.
,
Knox
S.
,
Avisse
N.
,
Harou
J.
,
Tilmant
A.
,
Klauer
B.
,
Mustafa
D.
,
Sigel
K.
,
Talozi
S.
,
Gawel
E.
,
Medellin-Azuara
J.
,
Bataineh
B.
,
Zhang
H.
&
Gorelick
S. M.
2021
A coupled human-natural system analysis of freshwater security under climate and population change
.
Proceedings of the National Academy of Sciences of the United States of America
118
(
14
),
e2020431118
.
https://doi.org/10.1073/pnas.2020431118
.
Zekri
S.
,
Madani
K.
,
Bazargan-Lari
M. R.
,
Kotagama
H.
&
Kalbus
E.
2017
Feasibility of adopting smart water meters in aquifer management: an integrated hydro-economic analysis
.
Agricultural Water Management
181
,
85
93
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY 4.0), which permits copying, adaptation and redistribution, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/).