Abdelmadjid Boufekane

1468 © IWA Publishing 2020 Journal of Hydroinformatics | 22.
6 | 2020
Delineation of groundwater recharge zones in the Mitidja

plain, north Algeria, using multi-criteria analysis
Abdelmadjid Boufekane, Hind Meddi and Mohamed Meddi
ABSTRACT
This work aims to identify the potential groundwater recharge zones in the Mitidja plain (north Abdelmadjid Boufekane (corresponding author)
Hind Meddi
Algeria) using the multi-criteria approach. The analysis was based on the use of a geographical Mohamed Meddi
GEE Laboratory,
information system (GIS) and remote sensing to establish eight thematic maps, weighted, Ecole Nationale Supérieure d’Hydraulique (ENSH),
Blida,
categorized and inserted, that allowed us to establish the potential zones’ map for groundwater
Algeria
recharge. Three potential groundwater recharge classes were defined corresponding, respectively, E-mail: a.boufekane@ensh.dz
to low (26%), moderate (47%) and high (27%). The best groundwater potential zones are situated in
the piedmont of the Blidean Atlas (the south of the study area), precisely, upstream near to wadis
(wadi El Harrach, wadi Djemaa, wadi Mazafran) and the western aquifer limit, where the
hydrogeological units are formed by the alluvium formation which is characterized by high hydraulic
conductivity, high water flow, low slope, low drainage, low quantity transported sediments and good
water quality. The obtained results, in this work, describe the groundwater recharge potential areas
and supply information for a suitable mapping and the management of aquifer resources in the study
area.
Key words | GIS, groundwater, Mitidja plain, multi-criteria analysis, recharge zones, remote sensing
HIGHLIGHTS
• The study involves choosing the multi-criteria decision analysis approach to delineate
groundwater recharge areas in the Mitidja plain (north Algeria).
• Geographical Information System (GIS) and remote sensing are used to establish eight thematic
maps and the final potential zones map for groundwater recharge.
• The final map shows the presence of three potential groundwater recharge classes, respectively,
low (26%), moderate (47%), and high (27%) to a total surface of 1,450 km2.
doi: 10.2166/hydro.2020.082
Downloaded from http://iwaponline.com/jh/article-pdf/22/6/1468/782111/jh0221468.pdf

by guest
1469 A. Boufekane et al. | Delineation of groundwater recharge zones Journal of Hydroinformatics | 22.6 | 2020
GRAPHICAL ABSTRACT
INTRODUCTION
Groundwater is a vital natural resource; it is considered as a water level (overexploitation) due to human activity (industrial
major factor in the development of urban, industrial and rural and agricultural activities, urbanization) and climate change
regions with more and more increasing demand. Ground- increase the demand on groundwater resources (Tiwari et al.
water presents about 34% of the whole yearly fresh water ). This critical situation is observed in the plain of Mitidja
source worldwide (Boufekane et al. ). In order to identify (north Algeria), which has experienced a reduction in natural
possible new aquifer areas, the combination of multi-criteria recharge that has led to a significant drop in the piezometric
decision analysis (MCDA) and geographical information levels reaching 40 m in some areas (Demmak ). This
system (GIS) techniques were used (Famiglietti ). The reduction is mainly due to the reduction of rainfall by about
groundwater resources have the highest priority of preser- 20% since 1975 (Meddi & Hubert ).
vation. The lithological properties of the aquifer formation Thus, in order to remedy this overexploitation and to
and its porosity have great importance in determining the control the groundwater pollution problem, it is necessary
presence and estimating the amount of groundwater. to pump groundwater in specific areas. Therefore, determin-
In recent years, water demand has intensively increased, ing potential groundwater zones has become very significant
mainly in the arid and semi-arid regions of the world, in several regions of the world.
especially, in the Middle East and North Africa, causing the Recently, several models have been suggested to situate
deficit of these important natural resources. The decreasing potential groundwater recharge zones, such as the artificial

by guest
neural network (ANN) (Sokeng et al. ), frequency ratio Site description
(FR) (Jothibasu & Anbazhagan ), logistic regression
(LR) (Ozdemir ), random forest (RF) (Zabihi et al. The Mitidja plain covers an area of 1,450 km2, is located in
), weights of evidence (WoE) (Tahmassebipoor et al. the northern part of Algeria and in the northeast is open to
), boosted regression tree (BRT) (Mousavi et al. ), the Mediterranean Sea. It is situated between latitude 36
support vector machine (SVM) (Lee et al. ), evidential 220 30″ and 36 480 58″ north and longitude 2 320 07″
belief function (EBF) (Pourghasemi & Beheshtirad ) and 3 280 14″ east (Figure 1). It has an extent of 100 km
and groundwater recharge mapping (GRM), which we from east to west and between 8 and 18 km from north to
develop and apply in this study. south. This plain is divided into two parts: eastern Mitidja
Potential groundwater recharge mapping is useful to and western Mitidja. The relief of the study area varies
assist water resource managers to better plan water exploi- between 0 m and 600 m above mean sea level. The
tation and management (Saidi et al. ). This time medium rise of the region is 50 m which increases in a regu-
gaining method allows establishment of thematic maps lar way to the piedmont of the Blidean Atlas to the south,
such as geology, hydrology, geomorphology, lineaments and the ripples of the Sahel to the north, allowing the
and slope (Shaban et al. ). Nowadays, computer tech- plain to have a basin form. Administratively, it spans four
nology (programming or software) is used in hydrogeology towns: Algiers, Blida, Boumerdes and Tipaza.
studies. For this study, we used a combination between A Mediterranean-type climate dominates in the Mitidja
MCDA approach and GIS technique to delineate ground- plain. It is characterized by hot and dry summers and
water potential zones. During the last years, several rainy winters. The average temperature varies between
studies have been carried out to assess groundwater poten- 6 C in winter and 33 C in summer, while the mean
tial mapping in different regions of the world with annual rainfall ranges from 500 to 700 mm.
satisfactory results as shown by field surveys (Magesh Its topography and the favourable climate conditions
et al. ); for example, Saidi et al. () in Tunisia, make it an excellent agricultural area (Imache et al. ).
Shaban et al. () in Lebanon, Yeh et al. () in The industry is developed and based in the industrial areas
Taiwan, Fashae et al. () in Nigeria, Abdalla () in located in the central and eastern regions (Blida, Arabaa,
Egypt, Mohebzadeh et al. () in Iran, Şener et al. Boufarik, El Harrach, Rouiba-Reghaia).
() in Turkey, Mahmoud et al. () in Saudi Arabia, The hydrology of the study area is characterized by an
Mandal et al. () in India and Ait El Mekki & Laftouhi intense hydrographic density with four important rivers:
() in Morocco. wadi Reghaia, wadi Hamiz, wadi El Harrach in the eastern
The main purpose of this research is to provide a simple Mitidja and wadi Mazafran in the western Mitidja. The tribu-
methodology to delineate the groundwater potential taries of these rivers are derived from the Blidean Atlas with
recharge zones in the Mitidja plain using a multi-criteria a significant flow in a south–north direction.
analysis that included eight hydrogeological parameters: This coastal aquifer is formed by subsidence and sedi-
lithology, drainage, rainfall, soil, slope, land use, per- mentation. In 1969, geophysical studies showed the
meability and water depth combined using remote sensing presence of two superimposed aquifers in the plain. In
and GIS systems. Remote sensing is used to examine drai- order to understand the lithology variation a hydrogeolo-
nage, slope and land while the GIS is used to manage data gical cross section was established. The AA’ section with
in different thematic layers (lithology, rainfall, soil, per- a NNW–SSE direction correlates six drilling wells
meability and water depth) which are taken into (Figure 2). This correlation shows the presence of: (i)
consideration for assessing groundwater potential zones in the Pliocene (Astian) aquifer is confined, formed by sand-
the Mitidja plain. stone and sandy limestones (Sbargoud et al. ). Its
The results of this research may serve water manage- substratum is trained by the blue marls of the Plaisancien
ment studies by helping water managers to choose the best and its surface is developed by a semi-permeable yellow
potential zones. marl layer named marls of El Harrach. It is located

by guest
Figure 1 | Map of the study area.
between a depth of 250–300 m, from the ground surface; Based on bibliographic data, piezometric measurements
(ii) the quaternary alluvial aquifer is composed of sand, and pumping tests, the direction of groundwater flow is from
gravel and rollers alternating with silts and clays. In the south to north (Freeze & Cherry ; Toubal ;
western part (apart from the Mazafran zone), this aquifer Ait-Ouali ; Zamiche et al. ). In general, the ground-
is unconfined and founded on the yellow marls of El Har- water depth ranges from 5 to more than 130 m and the
rach. Its thickness oscillates between 100 and 200 m. The transmissivity varies from 1.5 × 102 m2/s to 2.1 × 102 m2/s.
piezometric level is encountered between 5 and 35 m. Rainfall is an important source of the aquifer recharge and

by guest
Figure 2 | Hydrogeological section AA’ in the Mitidja plain area (refer to Figure 1 for location).
this is in addition to underground supply that is sourced probable aquifer recharge zones, ArcGIS software (Version
from the Blidean Atlas. In fact, the aquifer is exploited by 10.5) was utilized.
more than 250 boreholes and wells.
Parameters influencing potential recharge
METHODOLOGY
Eight hydrogeological parameters influencing the aquifers’
recharge have been considered: lithology (L), drainage (D),
Data used and potential recharge zones’ identification
rainfall (R), soil (S), slope (Sp), land use (Lu), permeability
(P) and water depth (W).
In the present study, the groundwater prospective zone
The potential groundwater recharge map is a result of
mapping and the thematic maps are prepared from
overlaying these eight parameters depending on their impor-
several sources including (i) satellite images, two scenes of
tance and impact on water infiltration. The potential
Landsat 8, Operational Land Imager (OLI/TIRS C1
groundwater recharge index (PRindex) is calculated by mul-
level-1) were acquired from Landsat’s official website
tiplying the influence of the eight parameters using the
(hearthexplorer.usgs.gov), the first scene covering the cities
following equation:
of Algiers, eastern part of Tipaza and western part of
Boumerdes, dated 7 February 2017 and the second scene PRindex ¼ Lw × Lr þ Dw × Dr þ Rw × Rr þ Sw × Sr þ Spw
covering the city of Blida, eastern part of Ain Defla and × Spr þ Luw × Lur þ Pw × Pr þ Ww × Wr (1)
northern part of Medea, dated 7 February 2017; (ii) the
SRTM image DEM at 90 m spatial resolution of the study where: L, D, R, S, Sp, Lu, P and W are the eight parameters
area which was downloaded from CGIAR-CSI GeoPortal and the subscripts r and w are the corresponding rating and
(srtm.csi.cgiar.org); (iii) the field data and secondary data weights.
were collected from different government agencies: the The partial index of each parameter is then calculated
National Agency of Hydrous Resources (ANRH) and the using Equation (2):
National Institute of Land, Irrigation and Drainage (INSID).
Partial index ¼ weight × rating (2)
The contribution of these sources for each parameter is
presented in the section ‘Thematic maps’ generation’. Fur- This approach is based on a multi-criteria decision-
thermore, in order to make the various study zones’ making technique that was originally developed by Saaty
thematic maps and the final map for the estimation of the ().

by guest
Determination of weighting Table 3 | Assigned rating for potential groundwater recharge parameters in the study
area
To delineate the groundwater potential zones’ map of the Parameters Classes Rating
study area, the eight thematic layers’ maps were reclassified Lithology Very permeable: modern formations 10
and overlaid with weighted overlay in ArcGIS 10.5 software. (Quaternary), gravel, sand
The weight of each parameter was calculated on the basis of Permeable: silt, conglomerate, sand, gravel 8
Low permeable: sandstone, sandy limestone 4
normalized weight using pair-wise correlation matrix Semi-permeable: yellow marl, gravel, silts and 2
according to Igwe et al. (). The PRindex weights are clay
assigned from 1 to 5 according to the importance of the the- Impermeable: blue marls 1
matic layers. The normalized and assigned weights are Drainage 0.0 –0.5 10
(km/km2) 0.5–1.0 8
presented in Table 1. 1.0–1.5 6
1.5–2.0 4
2.0 2
Evaluation of ratings
Rainfall 800–900 9
(mm) 700–800 8
A rating scale was determined based on the range of vari- 600–700 7
550–600 6
ation of the contribution factors to determine homogenous
500–550 5
ratings for all the parameters (Ait El Mekki & Laftouhi
Soil Sandy gravelly 9
). The rating of each parameter is assigned on a scale Sandy 8
from 1 to 10 (Table 2). Moreover, the factors were classified Sandy loamy 5
Loamy 3
according to this scale to obtain independent ratings for
Clayey loamy 1
each factor (Table 3).
Slope (%) 0–2 9
2–4 7
Table 1 | Weight settings of the potential groundwater recharge 4–7 5
7–10 3
Symbol Parameter Weight 10–12 1
Land use Water bodies (dams, wadis, etc.) 9
L Lithology 5
Irrigated areas 7
D Drainage 4 Forest, etc. 5
R Rainfall 3 Hills, rocky outcrops, bare soil, etc. 3
Urban areas 1
S Soil 3
Permeability 0.5 × 104 8
Sp Slope 2
(m/s) 1.0 × 104 7
Lu Land use 2 1.5 × 104 6
P Permeability 2 2.0 × 104 4
2.5 × 104 2
W Water depth 1
Water depth 0–2 9
(m) 2–5 8
5–7 7
7–10 6
Table 2 | Rating scale intensity parameter 10 3
Class Rating
Processing parameters
Very high 10
High 7
Several processing parameters are considered in this work:
Moderate 5
•
Low 3
The lithology (L): the lithology of geological formations is
Very low 1
the important parameter for the delineation of potential

by guest
groundwater recharge zones. However, the hydraulic • The water depth (W): the depth to the groundwater table
conductivity of these geological formations directly influ- depends on geological and hydrological characteristics of
ences the infiltration. In addition, the aquifer recharge the area. In general, shallow aquifers display a high
was influenced by lithology and that is done by ruling groundwater potentiality. However, the deeper aquifers
the water flow purification (El-Baz & Himida ). present less groundwater potential.
• The drainage (D): the drainage density is considered as
one of the important parameters to the aquifer recharge
evaluation as it is attached directly to the runoff and infil- RESULTS AND DISCUSSION
tration (Boughariou et al. ). It is the overall length of
the whole rivers existing in a discharge basin split by the Mapping potential groundwater recharge zone
whole zone of the discharge basin. The thematic analysis
of a drainage network aids value to the features of a Thematic maps’ generation
groundwater recharge area.
• The rainfall (R): rainfall is the most important influencing
• The lithology (L): the classification of this area was based
parameter in the groundwater potentiality of any zone on 180 drilling logs for each well, and on the Mitidja geo-
because the groundwater recharge increases with the logical report and synthesis map provided by the
water quantity of precipitation (Ibrahim-Bathis & National Agency of Hydrous Resources (ANRH). It was
Ahmed ). The zones with low precipitation were classified into five categories:
assigned a low rating value while the highest rating
- Very permeable: gravel and sand
value was appointed to zones receiving the highest
- Permeable: conglomerate, sand, gravel and silt
precipitation.
- Low permeable: sandstone and sandy limestone
• The soil (S): the type of soil is an important factor for deli-
- Semi-permeable: yellow marl, gravel, silt and clay
neating the potential groundwater recharge zones via the
- Impermeable: blue marls.
study of pedological data (soil resource inventory, soil
map, soil profiles). Fine-grained soils limit infiltration Figure 3 shows the distribution of lithology in the study area.
due to the low permeability, conversely, coarse-grained • The drainage (D): the drainage of the study area was
soil materials favour infiltration (Fashae et al. ). checked with satellite images and calculated after digitiz-
• The slope (Sp): the slope is the parameter that affects ation of all drainage existing in the study area. The high
runoff and infiltration. A steeper slope leads to less infil- drainage density area indicates low-infiltration rate
tration but increases the runoff and the rate of erosion whereas the low-density areas are favourable to high-
(sediment transport). On the other side, in the low infiltration rate. The values for the drainage density
slope area, the surface runoff is low allowing more time ranged from 0 to 2.2 km/km2 and were grouped into
for infiltration of rainwater. five classes (Figure 4).
• The land use (Lu): the land use of a region performs an • The rainfall (R): observation data from ten regional rain-
important role in runoff, infiltration and recharge of aqui- fall stations (ANRH stations) over a period of 33 years
fers. Nowadays, the high spatial resolution of satellite (1971–2004) indicated an annual average rainfall of
images indicates very small details in urban or farmland about 616 mm. The average for the station with the
zones ( Jin & Davis ). lowest rainfall (Hadjout) was 502 mm/year and that for
• The permeability (P): the permeability presents the ability the station with the highest rainfall (Attatba) was
of the aquifer to transmit water, enabling the rate at 812 mm/year. To prepare a rainfall map of the study
which groundwater flow is controlled under a specified area, inverse distance-weighted (IDW) interpolation
hydraulic gradient (Sinan & Razack ). In fact, a method of ArcGIS Spatial Analyst was applied using
higher permeability favours increased groundwater the annual rainfall. The resulting map was divided into
yields by infiltration. five rainfall classes (Figure 5).

by guest
Figure 3 | Thematic layer of lithology parameter.
Figure 4 | Thematic layer of drainage parameter.

by guest
Figure 5 | Thematic layer of rainfall parameter.
• The soil (S): the soil parameter was obtained by digitizing • The permeability (P): based on the literature (Castany
the existing soil maps (1:50,000) of ANRH, which ), the permeability classification is based on the
cover the entire region, established by Ecrement & hydraulic conductivity. In this study, the permeability
Seghir (). The analysis highlighted five soil types values are obtained from pumping test results of 81
(Figure 6): sandy gravelly, sandy, sandy loamy, loamy wells (Toubal ). These values were interpolated
and clay loam. using ordinary kriging method via the Surfer software
• The slope (Sp): the slope map of the study area was pre- (Version 13.4) to plot the permeability map (Figure 9).
pared based on the SRTM 90 m data following standard It was classified into five classes with values ranging
GIS routines using ArcGIS software. The Mitidja plain from 0.5 to 2.5 × 104 m/s.
was divided into five slope classes. For this study, it has • The water depth (W): the groundwater depth data were col-
been proposed that the better class indicates the areas lected from 43 observation wells provided by ANRH in
having 0–2% slope (nearly flat) which give a high infiltra- May 2017. These data were exploited to establish the
tion rate. Figure 7 illustrates the slope map of the study groundwater level contour map. Generally, in this region,
area with classes ranging from 0 to 12%. shallow groundwater levels vary between 10 m and 20 m,
• The land use (Lu): the land use of the study area is indicating medium groundwater potentiality (Figure 10).
assessed in accordance with the map established by the
National Institute of Land, Irrigation and Drainage
(INSID). This map (Figure 8) focuses on the definition Sensitivity analysis
of five areas containing: water bodies, irrigated areas,
forest rocky outcrops and bare soil, and urban areas. In this study, the sensitivity analysis is carried out to verify
Around 45% of the total zone is under cultivation and the coherence of the thematic maps’ results and to establish
40% is an urban area. a proper potential groundwater recharge map. According to

by guest
Figure 6 | Thematic layer of soil parameter.
Figure 7 | Thematic layer of slope.

by guest
Figure 8 | Thematic layer of land use parameter.
Figure 9 | Thematic layer of permeability parameter.

by guest
Figure 10 | Thematic layer of water depth parameter.
Lodwick et al. (), the map-removal sensitivity measure Table 4 | Modified weight for standard parameters influencing potential recharge based
on effective weights
represents the sensitivity associated with removing one or
more maps. This measure can be expressed as: Standard Standard Effective Effective
Parameters weight weight (%) weight (%) weight
S ¼ ((PrPw)=V) × 100 (3) L 5 22.73 24.15 5.31

D 4 18.18 16.19 3.56
R 3 13.64 11.75 2.58
where: W is the effective weight of each parameter, Pr is the
S 3 13.64 11.56 2.54
rating value and Pw is the weight value of each parameter,
Sp 2 9.09 10.44 2.30
and V is the potential groundwater index, as calculated
Lu 2 9.09 9.57 2.10
from Equation (1).
P 2 9.09 10.24 2.25
New effective weighting factors have been obtained by
W 1 4.54 6.10 1.34
the use of the potential groundwater recharge map which
are used for sensitivity analysis. The calculated mean effec-
tive weights based on the previously explained formula are • The lithology is considered as the most important par-
shown in Table 3. Obviously, we notice that there are ameter in the potential groundwater recharge zone.
some significant differences compared to the theoretical • The lithology, the slope, the land use, the permeability and
values proposed by Ait El Mekki & Laftouhi () and the water depth parameters’ effective contribution weights
Igwe et al. () as the weighting value of parameters chan- are respectively, 5.31, 2.30, 2.10, 2.25 and 1.34 instead of
ged, because the calculated values using weighting are based the standard weights of, respectively, 5, 2, 2 and 1.
on the eight hydrogeological parameters influencing aqui- • Unlikely, the drainage, the rainfall and the soil par-
fers’ recharge. The results in Table 4 show that: ameters are over-estimated. They show, respectively, the

by guest
following effective weights of 3.56, 2.58 and 2.30 instead corresponding to steep slopes, with low rainfall and low per-
of standard weights of, respectively, 4, 3 and 3. meability soil. The classes of moderate to high capacity of
groundwater correspond to relatively flat areas associated
with important rainfall and infiltration potentialities. The
Potential groundwater recharge zones mapping best groundwater potential zone is located in the alluvium
formations.
The potential recharge map (Figure 11) was prepared by
compiling and integrating the eight thematic maps for the
various individual factors using Equation (1). It is defined Results validation with borehole data
as a map estimating the probability that groundwater will
occur in a study area (Kim et al. ). To valid the potential groundwater recharge map, we used
According to Ibrahim-Bathis & Ahmed’s () work the results of groundwater productivity data (transmissivity)
based on the relationship of area statistics, weightage obtained from the pumping test results carried out at 102
values and their corresponding rankings for potential boreholes in the study area by Toubal (). The transmis-
groundwater recharge zone, the groundwater potential sivity values range from 1.40 × 103 to 63.60 × 103 m2/s.
zone of this study area can be divided into three classes: Furthermore, according to the classification proposed by
16–38 (low), 38.1–60 (moderate) and 60.1–73 (high). various authors (Lasm ; Dibi et al. ; Yao et al.
About 26% of the total area is included in the low class, ): low (T < 105 m2/s), moderate (105 < T < 104 m2/s)
47% belongs to the moderate class and 27% of the study and high (T > 104 m2/s), the transmissivity values are classi-
area is located in the high groundwater potential zone fied in the high class. The boreholes’ locations are shown in
2
with 1,450 km (Table 5). The low class occupies the areas Figure 11.
Figure 11 | Potential groundwater recharge map of the Mitidja plain.

by guest
Table 5 | Classification of groundwater potential zones Quantitatively (high potential groundwater recharge),
the most favourable zones to recharge are those of the
Potential zones Area (km2) Area (%)
extreme south between Bouinane and Arabaa cities and
Low 377.00 26
the southwest one (Chiffa). These zones, with a high
Moderate 681.50 47
recharge capacity, are eligible for recovery practices such
High 391.50 27
as artificial recharge from the hydrographic network, for
Total area 1,450.00 100
example, using the water of wadi El Harach in the region
of Bouinane, the water of wadi Djemaa (principal affluent
The comparison of transmissivity values (boreholes) of wadi El Harach) in the region of Arabaa and the water
with potential groundwater recharge map (Figure 11 and of wadi Mazafran in the region of Chiffa.
Table 6) shows that 55 values (boreholes) correspond to high To control the quality and quantity of water used for arti-
recharge area, 38 values (boreholes) coincide with moderate ficial recharge in this region, a follow-up was carried out in
recharge area and 9 values (boreholes) with low recharge zone. the upstream part of wadi El Harrach, precisely at the
These validation results demonstrate that around 91% of Rocher Pigeons station, for the period 2009–2013.
the boreholes accurately match with the zonation of moder- Table 7 shows that the water quantity from wadi El Har-
ate–high groundwater potential map. The high groundwater rach, in its upstream part, is largely sufficient to feed the
potential zones coincide with groundwater productivity Mitidja aquifer artificially, especially during flood periods.
areas (west, south, south-west and east central parts of the Generally, groundwater is profoundly linked to the presence
study area) due to favourable conditions such as alluvial of nearby water sources where the properties of the wadi
plain, shallow groundwater depth, low slope, flat topogra- linked to the groundwater are the width and depth of the
phy, optimum rainfall, favourable soil porosity condition river, slope and the flow velocity (Taormina & Chau ).
and permanent wadis. Thus, considering the results of this However, since runoff values and some flood values are
study, it can be concluded that this model for groundwater high, which could be affected by high sediment transport,
potential zone mapping gives very good results in the pre- we proposed the achievement of infiltration basins in the
sent research. upstream part of wadi El Harrach.
The analysis results of the superficial water (Table 8)
from wadi El Harrach, in its upstream part, show that this
DISCUSSION water has good quality that can be used as a source for the
Mitidja groundwater recharge.
The results obtained in this study show that the best ground-
water recharge potential area is situated in the southern
region (piedmont of the Blidean Atlas) due to the presence
CONCLUSIONS
of the alluvium formations, the zone being nearly flat and
the high rainfall with a high infiltration capacity. Addition-
In this study, a mapping approach has been proposed to identify
ally, in this zone, the important flow of wadis (presence of
the potential areas of groundwater recharge using remote sen-
continuous springs) also helps the stream flow recharge
sing and GIS techniques in the Mitidja plain, northern Algeria.
towards the groundwater system.
Table 7 | Volume of runoff and flood at the Rocher Pigeons station (period: 2009–2013)
Table 6 | Transmissivity values of boreholes into potential groundwater recharge classes
Year Runoff (Mm3) Flood (Mm3)

Potential groundwater recharge Number of the transmissivity values
zone (boreholes) 2009–2010 114.23 49.55
Low 9 2010–2011 154.65 71.50
Moderate 38 2011–2012 159.62 59.17
High 55 2012–2013 101.01 36.17

by guest
Table 8 | Analysis of the superficial water (wadi El Harrach) at the Rocher Pigeons station (period: 2010–2012)
Ca Mg Na K Cl SO4 HCO3 NO3 DR pH EC

Sampling date mg/L / μ/cm
14/10/10 76 38 35 0 120 116 214 4.0 707 8.0 1,093

16/12/10 82 73 165 2 240 200 214 0.0 744 8.0 1,438
18/01/11 82 47 200 2 251 216 244 6.0 1,014 7.8 1,572
22/05/11 68 37 200 2 190 149 238 6.0 604 8.2 1,019
13/02/12 81 54 64 2 112 158 275 2.4 634 7.8 1,100
20/05/12 91 43 45 2 93 243 214 1.2 715 8.2 1,080
20/09/12 106 77 298 3 580 332 159 0.0 1,875 8.2 1,690
DR, dry residual.
This approach consists of the delimitation of groundwater poten- approach can be used by water sector managers in selecting
tial recharge zones in order to protect the water resources from suitable sites for groundwater recharge and even for ground-
the point of view of quantity and quality in this region. water resources management, e.g., drilling new boreholes
Eight hydrogeological parameters, including lithology, and wells.
drainage, rainfall, soil, slope, land use, permeability and The obtained results will especially serve water manage-
water depth were taken into consideration and evaluated ment works in Mitidja plain as this reservoir is the largest
to determine potential recharge zones. freshwater aquifer in Algeria and forms part of the socio-
These hydrogeological factors are created and super- economic development of Algiers and surrounding cities
posed (thematic layers) to develop the best model in order (Blida, Boumerdes and Tipaza).
to establish a thematic map of potential recharge. The
obtained results show that the final groundwater potential
zone map was divided into three classes: high, moderate ACKNOWLEDGEMENTS
and low. Also, the results indicate that 47% of the study
area has moderate groundwater potential; about 27% of the This work was carried out as part of the project
total area was identified as high potential zone for ground- ERANETMED_WATER-13-166, Groundwater Resilience
water recharge corresponding to the southern part of the to Climate Change and High Pressure within an IWRM
study area, exactly in the piedmont of the Blidean Atlas, Approach. We are grateful to Directorate General for
near to wadi El Harach in the region of Bouinane, wadi Scientific Research and Technology Development staff
Djemaa in the region of Arabaa and wadi Mazafran in the (DGRSDT-MESRS) for funding, support and help with this
region of Chiffa. These areas are recommended to create arti- project.
ficial groundwater recharge structures to store rainwater and
surface runoff, especially during flood periods.
The low potential groundwater recharge zones which DATA AVAILABILITY STATEMENT
are considered unsuitable areas for groundwater recharge
processes are located especially in the northern and eastern All relevant data are included in the paper or its Supplemen-
areas, corresponding to a low permeability soil. tary Information.
This approach showed the best suitable artificial rechar-
ging sites in the Mitidja plain, in order to improve the REFERENCES
groundwater level and to confront the seawater intrusion
problem in the coastal zone situated in the northeastern Abdalla, F.  Mapping of groundwater prospective zones using
part of the study area. The final map obtained by this remote sensing and GIS techniques: a case study from the

by guest
Central Eastern Desert, Egypt. Journal of African Earth Enugu State, Southeastern, Nigeria. Sustainable Water
Sciences 70, 8–17. Resources Management 6, 39.
Ait El Mekki, O. & Laftouhi, N. E.  Combination of a Imache, A., Le Goulven, P., Bouarfa, S. & Chabaca, M. 
geographical information system and remote sensing data to Evolution de la demande en eau agricole dans la plaine
map groundwater recharge potential in arid to semi-arid irriguée de la Mitidja, Algérie (Evolution in the Demand for
areas: the Haouz Plain, Morocco. Earth Science Informatics Agricultural Water in the Mitidja irrigated plain, Algeria).
9, 465–479. Actes du troisième atelier régional SIRMA (Nabeul, Tunis),
Ait-Ouali, A.  Synthèse hydrogéologique et vulnérabilité à la CD-Rom, Édit. CIRAD, Montpellier, France.
pollution du système aquifère quaternaire de la Mitidja Jin, X. & Davis, C. H.  An integrated system for automatic
(Hydrogeological Synthesis and Vulnerability to Pollution of road mapping from high-resolution multispectral satellite
the Mitidja Quaternary Aquifer System). Master Thesis, imagery by information fusion. International Journal of
USTHB, Algiers, Algeria. Information Fusion 6, 257–273.
Boufekane, A., Saibi, H., Benlaoukli, B. & Saighi, O.  Scenario Jothibasu, A. & Anbazhagan, S.  Spatial mapping of
modeling of the groundwater in a coastal aquifer (Jijel plain groundwater potential in Ponnaiyar River basin using
area, Algeria). Arabian Journal of Geosciences 12, 799. probabilistic-based frequency ratio model. Modeling Earth
Boughariou, E., Saidi, S., Barkaoui, A. E., Khanfir, H., Zarehloul, Systems and Environment 3 (1), 33.
Y. & Bouri, S.  Mapping recharge potential zones and Kim, J. C., Jung, H. S. & Lee, S.  Groundwater productivity
natural recharge calculation: study case in Sfax region. potential mapping using frequency ratio and evidential belief
Arabian Journal of Geosciences 8, 5203–5221. function and artificial neural network models: focus on
Castany, G.  Principes et méthodes de l’hydrogéologie topographic factors. Journal of Hydroinformatics 20 (6),
(Principles and Methods of Hydrogeology). Pierre et Marie 1436–1451.
Crue University (Paris VI), Paris, France. Lasm, T.  Hydrogéologie des réservoirs fracturés de socle:
Demmak, A.  Impact sur les ressources en eau et stratégie de analyses statistique et géostatistique de la fracturation et des
lutte des années 1975–2000 (Impact on water resources and propriétés hydrauliques; application à la région des
control strategy drought in Algeria 1975–2000). In 2nd MEDA montagnes de Côte d’Ivoire domaine archéen) (Hydrogeology
Water Regional Event, 28–30 April 2008, Marrakech, Morocco. of Fractured Basement Reservoirs: Statistical and
Dibi, B., Inza, D., Goula, B. T., Savané, I. & Biémi, J.  Analyse Geostatistical Analyzes of Fracturing and Hydraulic
statistique des paramètres influençant la productivité des Properties; Application to the Mountains Region of Ivory
forages d’eau en milieu cristallin et cristallophyllien dans la Coast (Archean Domain). Doctorate Thesis, Poitiers
région d’Aboisso (Sud-Est de la Côte d’Ivoire) (Statistical University, France, p. 272.
analysis of parameters influencing drilling water productivity Lee, S., Hong, S. M. & Jung, H. S.  GIS-based groundwater
in crystalline and crystalliferous environment region of potential mapping using artificial neural network and
Aboissso (South-East of Ivory Coast)). Sud Sciences & support vector machine models: the case of Boryeong city in
Technologies 13, 22–31. Korea. Geocarto International 33 (8), 847–861.
Ecrement, Y. & Seghir, B.  Etude agro pédologique de la Lodwick, W. A., Monson, W. & Svoboda, L.  Attribute error
Mitidja (Mitidja Agro Pedological Study). ANRH, Algiers, and sensitivity analysis of map operations in geographical
Algeria. information systems: suitability analysis. International
El-Baz, F. & Himida, I.  Groundwater Potential of the Sinai Journal of Geographical Information Systems 4 (4), 413–428.
Peninsula, Egypt, US Agency for International Development, Magesh, N. S., Chandrasekar, N. & Soundranayagam, J. P. 
Cairo, Egypt. Delineation of groundwater potential zones in Theni district,
Famiglietti, J. S.  The global groundwater crisis. Nature Tamil Nadu, using remote sensing, GIS and MIF techniques.
Climate Change 4 (11), 945–948. Geoscience Frontiers 3, 189–196.
Fashae, O. A., Tijani, M. N., Talabi, A. O. & Adeleji, O.  Mahmoud, S. H., Alazba, A. A. & Amin, M. T.  Identification
Delineation of groundwater potential zones in the crystalline of potential sites for groundwater recharge using a GIS-based
basement terrain of SW Nigeria: an integrated GIS and decision support system in Jazan region-Saudi Arabia. Water
remote sensing approach. Applied Water Science 4, 19–38. Resources Management 28, 3319–3340.
Freeze, R. A. & Cherry, J. A.  Groundwater. Prentice-Hall, Mandal, U., Sahoo, S., Munusamy, S. B., Dhar, A., Panda, S. N.,
Englewood Cliffs, NJ, USA. Kar, A. & Mishra, P. K.  Delineation of groundwater
Ibrahim-Bathis, K. & Ahmed, S. A.  Geospatial technology for potential zones of coastal groundwater basin using multi-
delineating groundwater potential zones in Doddahalla criteria decision making technique. Water Resources
watershed of Chitradurga district, India. The Egyptian Management 30, 4293–4310.
Journal of Remote Sensing and Space Sciences 19, 223–234. Meddi, M. & Hubert, P.  Impact of the modification of the
Igwe, O., Ifediegwu, S. I. & Onwuka, O. S.  Determining the rainfall regime on the water resources of North-West Algeria.
occurrence of potential groundwater zones using integrated International Association of Hydrological Sciences 278, 1–7.
hydro-geomorphic parameters, GIS and remote sensing in (in French).

by guest
Mohebzadeh, H., Fathi, A., Navabi, S., Parisouj, P. & Nazari, A. International Journal of Innovation and Applied Studies
 A GIS-based modelling for identifying high priority areas 15 (4), 747–759.
for groundwater recharge. Journal of Earth Science and Tahmassebipoor, N., Rahmati, O., Noormohamadi, F. & Lee, S.
Engineering 8, 93–102.  Spatial analysis of groundwater potential using weights
Mousavi, S. M., Golkarian, A., Naghibi, S. A., Kalantar, B. & of-evidence and evidential belief function models and remote
Pradhan, B.  GIS-based groundwater spring potential sensing. Arabian Journal of Geosciences 9, 79.
mapping using data mining boosted regression tree and Taormina, R. & Chau, K. W.  Neural net work river
probabilistic frequency ratio models in Iran. AIMS forecasting with multi-objective fully informed particle
Geosciences 3 (1), 91–115. swarm optimization. Journal of Hydroinformatics 17 (1),
Ozdemir, A.  Using a binary logistic regression method and 99–113.
GIS for evaluating and mapping the groundwater spring Tiwari, A. K., Lavy, M., Amanzio, G., De Maio, M., Singh, P. K. &
potential in the Sultan Mountains (Aksehir, Turkey). Journal Mahato, M. K.  Identification of artificial groundwater
of Hydrology 405 (1–2), 123–136. recharging zone using a GIS-based fuzzy logic approach: a
Pourghasemi, H. R. & Beheshtirad, M.  Assessment of a data case study in a coal mine area of the Damodar Valley, India.
driven evidential belief function model and GIS for Applied Water Science 7, 4513–4524.
groundwater potential mapping in the Koohrang Watershed, Toubal, A. C.  Apport de la géophysique à l’étude des
Iran. Geocarto International 30 (6), 662–685. problèmes hydrodynamique et d’invasion marine en milieu
Saaty, R. W.  The analytic hierarchy process – what it is and souterrains. Exemple des plaines d’Annaba, de la Mitidja et
how it is used. Mathematical Modelling 9 (3), 161–176. de la baie d’Alger (Contribution of Geophysics to
Saidi, S., Hosni, S., Mannai, H., Jelassi, F., Bouri, S. & Anselme, B. Hydrodynamic Study and Marine Invasion Problems in
 GIS-based multi-criteria analysis and vulnerability Groundwater. Example of Annaba and Mitidja Plains, and
method for the potential groundwater recharge delineation, the bay of Algiers). Doctorate Thesis, USTHB, Algiers,
case study of Manouba phreatic aquifer, NE Tunisia. Algeria.
Environmental Earth Sciences 76, 511. Yao, K. A., Kouassi, A. M., Koffi, Y. B. & Biemi, J. 
Sbargoud, S., Hartani, T., Aidaoui, A., Herda, F. & Bachir, H.  Caractérisation hydrodynamique et hydrogéochimique des
Assessment of groundwater vulnerability to nitrate based on aquifères fissurés de la région de Toumodi (centre de la Côte
the optimised DRASTIC models in the GIS environment d’Ivoire) (Hydrodynamical and hydrogeochemical
(case of Sidi Rached Basin, Algeria). Geosciences 7, 1–23. characterization of the fractured aquifers of Toumodi area
Ş ener, E., Ş ener, Ş . & Davraz, A.  Groundwater potential (central Ivory Coast)). Journal of Environmental Hydrology
mapping by combining fuzzy-analytic hierarchy process and 18, 26.
GIS in Beyşehir Lake Basin, Turkey. Arabian Journal of Yeh, H. F., Cheng, Y. S., Lin, H. I. & Lee, C. H.  Mapping
Geosciences 11, 187. groundwater recharge potential zone using a GIS approach
Shaban, A., Khawlie, M. & Abdallah, C.  Use of remote sensing in Hualian River, Taiwan. Sustainable Environment Research
and GIS to determine recharge potential zones: the case of 26, 33–43.
Occidental Lebanon. Hydrogeology Journal 14, 433–443. Zabihi, M., Pourghasemi, H. R., Pourtaghi, Z. S. & Behzadfar, M.
Sinan, M. & Razack, M.  An extension to the DRASTIC  GIS-based multivariate adaptive regression spline and
model to assess groundwater vulnerability to pollution: random forest models for groundwater potential mapping in
application to the Haouz aquifer of Marrakech (Morocco). Iran. Environmental Earth Sciences 75, 665.
Environmental Geology 57, 349–363. Zamiche, S., Hamaidi-Chergui, F., Demiai, A. & Belaidi, M. 
Sokeng, V. J., Kouamé, F., Ngatcha, B. N., N’da, H. D., You Akpa, Identification of factors controlling the quality of
L. & Rirabe, D.  Delineating groundwater potential groundwater in Mitidja plain (North-Algeria) using indexing
zones in Western Cameroon Highlands using GIS based method and statistical analysis. Journal of Fundamental and
Artificial Neural Networks model and remote sensing data. Applied Sciences 10, 248–267.
First received 8 May 2020; accepted in revised form 29 June 2020. Available online 3 September 2020

by guest

Abdelmadjid Boufekane

Uploaded by

Copyright:

Available Formats

Abdelmadjid Boufekane

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Abdelmadjid Boufekane

Uploaded by

Copyright:

Available Formats

1468 © IWA Publishing 2020 Journal of Hydroinformatics | 22.

Delineation of groundwater recharge zones in the Mitidja

Downloaded from http://iwaponline.com/jh/article-pdf/22/6/1468/782111/jh0221468.pdf

Downloaded from http://iwaponline.com/jh/article-pdf/22/6/1468/782111/jh0221468.pdf

Downloaded from http://iwaponline.com/jh/article-pdf/22/6/1468/782111/jh0221468.pdf

Figure 1 | Map of the study area.

Downloaded from http://iwaponline.com/jh/article-pdf/22/6/1468/782111/jh0221468.pdf

Parameters inﬂuencing potential recharge

Downloaded from http://iwaponline.com/jh/article-pdf/22/6/1468/782111/jh0221468.pdf

Downloaded from http://iwaponline.com/jh/article-pdf/22/6/1468/782111/jh0221468.pdf

Downloaded from http://iwaponline.com/jh/article-pdf/22/6/1468/782111/jh0221468.pdf

Figure 3 | Thematic layer of lithology parameter.

Figure 4 | Thematic layer of drainage parameter.

Downloaded from http://iwaponline.com/jh/article-pdf/22/6/1468/782111/jh0221468.pdf

Figure 5 | Thematic layer of rainfall parameter.

Downloaded from http://iwaponline.com/jh/article-pdf/22/6/1468/782111/jh0221468.pdf

Figure 6 | Thematic layer of soil parameter.

Figure 7 | Thematic layer of slope.

Downloaded from http://iwaponline.com/jh/article-pdf/22/6/1468/782111/jh0221468.pdf

Figure 8 | Thematic layer of land use parameter.

Figure 9 | Thematic layer of permeability parameter.

Downloaded from http://iwaponline.com/jh/article-pdf/22/6/1468/782111/jh0221468.pdf

Figure 10 | Thematic layer of water depth parameter.

S ¼ ((PrPw)=V) × 100 (3) L 5 22.73 24.15 5.31

Downloaded from http://iwaponline.com/jh/article-pdf/22/6/1468/782111/jh0221468.pdf

Figure 11 | Potential groundwater recharge map of the Mitidja plain.

Downloaded from http://iwaponline.com/jh/article-pdf/22/6/1468/782111/jh0221468.pdf

Year Runoff (Mm3) Flood (Mm3)

Downloaded from http://iwaponline.com/jh/article-pdf/22/6/1468/782111/jh0221468.pdf

Ca Mg Na K Cl SO4 HCO3 NO3 DR pH EC

14/10/10 76 38 35 0 120 116 214 4.0 707 8.0 1,093

DR, dry residual.

Downloaded from http://iwaponline.com/jh/article-pdf/22/6/1468/782111/jh0221468.pdf

Downloaded from http://iwaponline.com/jh/article-pdf/22/6/1468/782111/jh0221468.pdf

Downloaded from http://iwaponline.com/jh/article-pdf/22/6/1468/782111/jh0221468.pdf

You might also like