Arab J Geosci

DOI 10.1007/s12517-015-1932-2

ORIGINAL PAPER

Groundwater of Abu Dhabi Emirate: a regional assessment

by means of remote sensing and geographic information system
Samy Ismail Elmahdy 1 & Mohamed Mostafa Mohamed 1,2

Received: 15 January 2015 / Accepted: 13 April 2015

# Saudi Society for Geosciences 2015

Abstract Mapping of geological, topographical, and hydro- timely, cost-effective, and can be used in arid regions for nu-
logical elements is critical for understanding and assessing the merical modeling as well as water balance analysis.
regional hydrological setting in an arid region. In this study, a
synergistic approach has been developed, which uses a com- Keywords Remote sensing . GIS . UAE . Groundwater .
bination of remote sensing data and geographic information Paleochannels . Al Ain
system (GIS) to map factors controlling groundwater re-
charge, discharge, and quality across the Abu Dhabi
Emirate. The Spectral Angel Mapper (SAM) algorithm, which Introduction
uses a n-D angle to match the pixels to reference spectra, was
used to map new water-bearing rocks, and the deterministic The availability of surface water in the United Arab Emirates
eight-node (D8) algorithm, which allows flow to only one of (UAE) is limited because of low rainfall and high evaporation
the eight neighbors based on the direction of steepest descent, rates. Additionally, the UAE has been experiencing rapid
was used to map paleochannels. The terrain category was urbanization and intensive human activities. These activities
applied to simulate seawater intrusion from digital elevation have a negative impact on groundwater quality in the Quaternary
model (DEM). New maps of lithology, normalized difference sand aquifer underlying the city. Mapping of hydrological,
vegetation index (NDVI), and paleochannels were derived environmental, and geological elements is a key component
and interpreted from multi-sources of remote sensing data. to understand the hydrological setting of Abu Dhabi Emirate.
The study indicated that the area was produced by a fluvial In Abu Dhabi Emirate, natural elements such as paleochannels,
and eolian process and recharged by local, intermediate, and lithological units, topographic slope, and anthropological
regional flows. The results showed that the Oman and Hafeet factors affect agricultural activity, aquifer thickness, shallow
Mountains are the natural sources of groundwater recharge as water table, and seawater intrusion resulting in increased
well as HCO3, Ca, Na, and Mg in groundwater. The mapped groundwater salinity and contamination (Mohamed and
factors were spatially correlated with hydrologic anomalies Elmahdy 2014; Elmahdy and Mohamed 2014a, c).
observed in groundwater wells. The integrated approach is Groundwater assessment in Abu Dhabi Emirate provides
the hydrological information to help in groundwater manage-
ment in determining the suitability of various areas in
Abu Dhabi for particular activities such as land development
and urban planning.
* Samy Ismail Elmahdy Several studies, which use remote sensing data, have been
samy@uaeu.ac.ae applied in arid regions to map groundwater recharge and dis-
charge areas over regional scale (Dehaan and Taylor 2002;
1
Civil and Environmental Engineering Department, United Arab Metternicht and Zinck 1997; Mohamed et al. 2010a, b), to
Emirates University, P.O. Box 15551, Al-Ain, United Arab Emirates investigate the influence of lineaments on groundwater
2
Irrigation and Hydraulics Department, Faculty of Engineering, Cairo (Elmahdy and Mohamed 2012a, b; Maged 2014; Mohamed
University, P.O. Box 12211, Giza, Egypt and Elmahdy 2014), and to map the natural and
 Arab J Geosci

anthropological factors affecting groundwater quality. A toward the west, southwest, and northwest, passing through
DRASTIC approach has been applied to map the relative re- alluvial plain and sand dune and ends in the Arabian Gulf.
charge rate (Shirazi et al. 2012), and to evaluate the aquifer The hydrological framework in Abu Dhabi comprises four
vulnerability and contamination (Shirazi et al. 2013). Salama hydrological units (UAE National Atlas United Arab Emirates
et al. (1994) integrated remote sensing data and field National Atlas 1993): (1) the carbonate aquifer in the east, (2)
measurements to map recharge and discharge areas over a the sand dune aquifer in the middle, (3) the western gravel
large scale. Leblanc et al. (2003b) also highlighted the use of aquifer in the east, and (4) the coastal aquifer in the west.
various hydrological data and land use maps to identify Each aquifer has its own hydrologic and physical characteris-
groundwater recharge and discharge areas. tics (Table 1). They are characterized by heterogeneity in po-
Although geophysical survey and field observation studies rosity, thickness, and groundwater quality (Mohamed and
(Woodward and Al Sharhan 1994; Mohamed et al. 2009; Hatfield 2005; Mohamed and Hatfield 2011). For example,
Elmahdy and Mohamed 2012b; Elmahdy and Mohamed the thickness aquifers vary from 5 m near the shoreline to
2013a, b; Elmahdy and Mohamed 2014a, b) can assist with 38 m at Liwa Oasis under the influence of the fault displace-
hydrological interpretations, the most useful can be obtained ments and folding (Brook et al. 2006).
from remote sensing and geographic information system According to Brook and Dawoud (2005), groundwater for-
(GIS) technique. This technique enables assessment of mations in the UAE can be categorized into two types: (1) the
groundwater quantity and quality using geological, hydrolog- Tertiary sediments which consist of loosely cemented, calcar-
ical, topographical, and environmental elements. The main eous sandstones, sandy limestone, silty chalk, and gypsum
objective of this study is to precisely map elements such as layers, and (2) the Quaternary unconsolidated clastic sedi-
the m ain a quifers within the surfacial deposits, ments which cover most of the desert parts throughout the
palaeochannels, and areas of agricultural activity from remote country.
sensing data using automated algorithms. This study also aims The results of geophysical survey and drilling
to use remote sensing and hydrological data to identify boreholes by HydroConsult (1987) (Fig. 2a), Mohamed
sources of groundwater recharge and discharge and factors et al. (2010), and US Geological Survey (2001) also re-
controlling groundwater accumulation and contamination. vealed the presence of three hydrological aquifers and
The regional groundwater assessment can be used in develop- four resistivity layers (Table 1 and Fig. 3a). The first layer
ing a conceptual model of hydrological setting within (shallow aquifer) consists of unconsolidated sandstone,
Abu Dhabi, UAE. gravel, and silt with thickness of 20 m. Its storage coeffi-
cient is 4.0 ×10−3 and specific capacity of 3 m2/h. The
third layer (deep aquifer) consists of dolomite carbonate
Study area with serpentine, gravel, marl, shale, and dolomite with a
thickness of 300 m, storage coefficient of 2.8×10−3, spe-
The study area extends between the longitudes E 51° 64′ 41″ cific capacity of 2 m2/h, and porosity percentage of 30.
and E 55° 55′ 00″ and the latitudes N 23° 20′ 20″ and N 24° The average depth to water table in the shallow aquifer
21′ 15″. The Emirate is bounded on the east by the Sultanate ranges predominantly between 4 and 324 m and the
of Oman and on the south by Saudi Arabia. It extends from the highest values are mostly recorded in the gravel plains
Oman Mountains in the east and the Arabian Gulf in the west, (Brook and Dawoud 2005). Average groundwater salinity
occupying an area of about 60,330 km2 (Fig. 1a). Rainfall is and electrical conductivity generally increase from the
highly variable in time and space (Fig. 1b). Most of the rainfall east to the west (Alsharhan et al. 2001; Mohamed et al.
in winter occurs as a result of convergence zones caused by an 2009).
upper level trough to the west of the Gulf area. Heavy, short-
duration rainfall produces the best opportunities for aquifer
recharge. The total annual rainfall in the UAE ranges from Datasets and methods
40 to 160 mm (Fig. 1b) and the total surface water is estimated
to be about 39.6 Mm3 and annual recharge ranges from 21.8 to Datasets
32.7 Mm3 (Al-Rashed and Sherif 2000; Mohamed et al.
2010a). Two remotely sensed datasets were used in this study: (i) the
Surface water occurs in the non-vegetated Oman Shuttle Radar Topography Mission (SRTM) of spatial resolu-
Mountains and collects in wadis which drain into the UAE, tion of (∼90 m), and (ii) Landsat images of 30-m spatial res-
eventually recharging into the shallow alluvial gravel aquifers olution. The Shuttle Radar Topography Mission (SRTM) ob-
(Al-Rashed and Sherif 2000; Brook and Dawoud 2005; tained digital elevation model (DEM) on a near-global scale is
Elmahdy and Mohamed 2012c; Elmahdy and Mohamed used to generate the most complete high-resolution digital
2014a, d). Water flows from the east (Oman Mountains) topographic database of the Earth. SRTM consisted of a
Arab J Geosci

Fig. 1 Maps of the UAE, including Abu Dhabi Emirate (a), annual rainfall (b), and groundwater aquifers (c) (modified after UAE National Atlas United
Arab Emirates National Atlas 1993). Red points highlight the observed groundwater well locations

specially modified radar system that flew onboard the Space in DEMs and Landsat images to decide if patterns expressed
Shuttle Endeavor during an 11-day mission in February of by high-resolution optical data are different.
2000 (www2.jpl.nasa.gov/srtm).
The SRTM DEM is of spatial resolution of 90 m. The Methods
SRTM DEM indicates slope gradient, terrain configuration,
and drainage pattern. The Landsat 8 image of spatial resolu- To map elements which control the regional groundwater re-
tion 30 m is available as a geographic (long/lat) projection, charge and discharge in regional and remote access areas, it is
with the WGS84 horizontal datum currently available from important to employ algorithms and remote sensing data
US Geological Survey USGS browser (http:// (Marghany et al. 2009). The advantage of using remote sens-
glovis.usgs.gov/). These various datasets were used to com- ing data is in mapping the spatial extent, multi-temporal, in a
pare visually the textural features (drainage network) evident cost-effective manner (Hoffmann 2005; Elmahdy et al. 2011;
 Arab J Geosci

Table 1 The hydraulic properties of the hydrogeologic units in the United Arab Emirates compiled from geophysical survey (HydroConsult 1987; US
Geological Survey 2001)

Aquifer Lithology Thickness Resistivity Transmissivity Storage Specific capacity Porosity

(m) (Ω-m) (m2/day) coefficient (m2/h) (%)

Shallow aquifer Sand and gravel, with 20 75 85 4.0×10−3 3 35

clay intercalations
Aquiclude Semi-consolidated gravel 200 19 ???? 10−5 to 10−7 ?? 5
with minor sand shale
and marl
Deep aquifer Dolomite carbonate with 300 29 51 2.8×10−3 2 30
serpentine, gravel, marl,
shale, and dolomite

Maged Marghany 2012; Elmahdy and Mohamed 2012; Geological and topographical element mapping
Elmahdy and Mohamed 2013a). Thus, automated algorithms and characterization
use the multi-sourced remote sensing data which can effec-
tively contribute to the mapping of factors that control The Spectral Angle Mapper (SAM) algorithm of Kruse et al.
groundwater recharge/discharge and groundwater quality. (1993) implemented in ENVI v.4.5 software was used here.
These factors are particularly useful, especially when The algorithm is a physically based spectral classification that
regional-scale areas have to be mapped and investigated uses a n-D angle to match the pixels to reference spectra. The
and field measurements are difficult to conduct and access algorithm determines the spectral similarity between two
(Mohamed et al. 2007). spectra by calculating the angle between the spectra and

Fig. 2 Photos of recharge and (a) (b)

discharge areas such as sand
dune, agricultural areas and sand
(b, c), and paleochannels (d)

(c) (d)

Paleochannels
Arab J Geosci

Fig. 3 Lithological succession of groundwater-bearing rocks (a), and E–W topographic profile across the Abu Dhabi Emirate, UAE (b), showing the
groundwater flow directions in the UAE (modified after (Alsharhan et al. 2001))

treating them as vectors in a space with dimensionality equal areas. The topographic map was derived from DEM
to the number of bands. The classification of lithological units using ArcGIS v. 9.2 software.
was performed using 200 endmember spectra which were
extracted from Landsat images (ftp.glcf.umd.edu/glcf/ Hydrological element mapping and characterization
Landsat) and calibrated by SAM spectral libraries, and then
the algorithm compares the angle between the endmember The hydrological indicators include palaeochannels and maps
spectrum vector and each pixel vector in n-D space. The pro- interpolated from data collected from boreholes. The
duced map was compared visually against the geological map palaeochannels were derived from DEM using the D8 flow
of the UAE (UAE National Atlas United Arab Emirates route based on the eight-cell neighborhood approach (Jenson
National Atlas 1993). and Domingue 1988) implemented in the ArcGIS v. 9.2 soft-
To map subsurface geological structures such as faults and ware. The algorithm derives drainage network in the follow-
folds, the geological maps of the UAE released by the environ- ing steps:
mental agency of Abu Dhabi, UAE (http://enviroportal.ead.ae/
mapviewer/Default.aspx), and UAE ATLAS (UAE National 1. DEM is filled to generate a seamless surface and to repli-
Atlas United Arab Emirates National Atlas 1993) were cate the conditions, facilitating an uninterrupted move-
screened digitized visually. The surface lineaments were visu- ment of water across the surface.
alized and mapped from the filtered Landsat images and veri- 2. Generating the flow direction grid.
fied using terrain parameter maps pertaining to slope, aspect, 3. Calculation of the flow accumulation grid function.
and shaded relief generated from SRTM DEM. Slope breaks 4. Delineate the palaeochannel network by counting all the
and tone changes in the shaded relief maps could be used to cells upstream of a given cell.
easily identify linear geological features. The obtained map was 5. Selecting a suitable threshold value for the number of
validated by comparing it with geological map produced by the cells.
UAEU (1993). 6. Delineating drainage basins by locating the pour points at
Topographic factors, such as slope, affect water flow the edges of the analysis window. In this study, a threshold
and infiltration rate over the landscape (Subba 2006; value of 100 cells was selected for palaeochannels as it
Sherif et al. 2010). Thus, the slope may be useful to gave the best result in the extraction of major
identify potential groundwater recharge and discharge palaeochannel network from DEM.
 Arab J Geosci

Hydrological data, such as hydraulic head and groundwater affected by faults and folds (Fig. 4a). Their trends are
salinity, were interpolated using Kriging interpolator tool im- in the NW–SE, NE–SW, and WNW–ESE directions and
plemented in the ArcGIS v. 9.2 software. The produced maps share similar trends as those of Ham, Hatta, and Dibba
were used to indicate factors controlling groundwater recharge Fault zones (Alsharhan et al. 2001). Their displacements
and discharge and quality. control topographic slope and surface flow over land-
scape (Elmahdy and Mohamed 2012b; Emahdy and
Environmental mapping and characterization Mohamed 2012). The map also shows that the region
comprises four major lithological units. Every unit has
Seawater intrusion and agricultural activity are also en- its own hydrological, chemical, and physical properties.
vironmental factors affecting groundwater quality These properties control groundwater recharge, dis-
(Mohamed et al. 2009). Seawater intrusion is partially charge, and quality in Abu Dhabi (Maged 2014).
controlled by water table, fresh–saline transition zone, Based on their physical characteristics and type locality, the
and groundwater extraction. In this regard, we proposed water-bearing rocks in Abu Dhabi Emirate can be categorized
a method for the simulation of Arabian Gulf water in- into the following three groups: (i) the clastics and granular
trusion into the coastal basin based on the change in water-bearing rocks, (ii) the non-clastic and carbonate fissured
water table and fresh–saline transition zone. water-bearing rocks, and (iii) clastic–non-clastic coastal
The terrain category tool, which is implemented in the open aquifer.
source Microdem v. 12 software (GuthPL 2009), was used to The clastic water-bearing rocks include the western gravel
see the terrain through the seawater intrusion when the water and sand dune aquifers. The western gravel aquifer consists of
table depleted below sea level due to overpumping in the coarse clastic fragments brought from the Oman Mountain via
coastal areas of the UAE. The model highlights the areas that a fluvial process that becomes gradually finer in grain size
may have experienced seawater intrusion, especially when further west. The aquifer occupies an area of 542 km2
water table and fresh–saline transition zone are depleted be- (0.89 %) and carries fresh water at the foot of Oman
low the sea level. Mountain and becomes gradually saline as the distance from
Agricultural activity is a natural source of nitrate in Hafeet and the Oman Mountains increases (Table 2 and
groundwater and represents the factor controlling Fig. 4a).
groundwater discharge (Hounslow 1995). To map the The sand dune aquifer is located between the western grav-
spatial distribution of vegetation in Abu Dhabi, the nor- el aquifer in the east and coastal aquifer in the west. Based on
malized difference vegetation index (NDVI) was calcu- the shape, the aquifer comprises two types of sand dunes. The
lated from Landsat 8 images using transform tool im- first type is linear dunes while the second one is barchan-like
plemented in ENVI v. 4.5 software using Eq. (1): in shape. The linear dune aquifer receives its water from the
. western side of Hafeet and the western gravel aquifer. The
NDVI ¼ ðN IR − RÞ ðN I R þ RÞ ð1Þ barchan dune aquifer is located in the southern part of
Abu Dhabi and stretched between Liwa and Madinate
IR and Red are the near-infrared (NIR) and red bands, Zayed. The aquifer receives its groundwater from Oman and
which are from 0.7 to 1 lm and 0.6 to 0.7 lm of the electro- Hegaz Mountains in Saudi Arabia (Fig. 2).
magnetic spectrum. The NDVI value, which indicates the The non-clastic water-bearing rocks include the carbonate
presence and intensity of vegetation in the study area, was aquifer, which is mainly composed of highly fractured carbon-
applied to Al Ain (east) and Liwa (south) areas. ate rocks at an elevation of 1000 m (above m.s.l.). This area
has a lower flux of water infiltration and recharging, has limb
areas at lower elevation, and is composed carbonate frag-
Results and discussion ments, providing effective conduits for groundwater recharge.
The aquifer includes symmetrical anticline folds (Hafeet
A set of modified maps of the factors controlling groundwater Mountain) and occupies an area of about 186 km2 (Table 2).
recharge and discharge and quality was derived from remote The flow in carbonate aquifer is local in scale and
sensing data using automated algorithms. These maps include approximately in the east and west directions. The water
geology, hydrology, terrain parameters, and environment. recharges to the aquifer depend on the leakage from the
deep aquifer and then discharges its water into the west-
Geological elements ern gravel aquifer in the east and the sand dune aquifer
in the west (Fig. 3). The chemical dissolution of car-
The modified map of geological elements controlling ground- bonate rocks of Hafeet is the origin of Ca, Mg, Na, and
water recharge and discharge and quality is shown in Fig. 4a. HCO3, in groundwater. The hydrology of the fissured
It is clear from the geological map that the region is carbonate rocks is not well understood.
Arab J Geosci

(a)

Oman

Saudi Arabia

(b)

(c)

Saudi Arabia

Hafeet M.

Fig. 4 Geological (a) and topographic slope (b) maps of Abu Dhabi Emirate show the main geological elements auto-mapped from Landsat images
using SAM algorithm and slope classes derived from DEM (modified after (UAEU 1993))

Table 2 The hydrologic properties of the main groundwater aquifers in the UAE (Brook and Dawoud 2005)

Aquifer Total area (km2) Hydraulic head (m) Productivity Salinity (ppm) Flow Nitrate (mg/L)

Eastern gravel 280 200–250 High 14,000–28,000 Local 50–200

Western gravel 15,177 120–350 High 800–5700 Intermediate 50–200
Ophiolite 10,195 700 Low <800 Local 50–100
Linear sand dune 68,376 50–120 Moderate 5000–15,000 Regional 100–1000
Star sand dune 8,194 80–105 High 800–2000 Intermediate 250–550
Coastal 22,760 0–50 High >15,000 Local 0–50
Carbonate 4,956 0–25 High ??? Local 100–200
 Arab J Geosci

The clastic–non-clastic water-bearing rocks include the from east to west. In general, high quantity of freshwater oc-
coastal aquifer, which is composed of sand and sabkha curs in the upper stream at the foot of the Oman Mountain in
(evaporites) along the Arabian Gulf region. It occupies an area the east, and groundwater quantity and quality trends increase
of 3861 km2 and receives its water from seawater intrusion down-gradient in the west (Mohamed and Elmahdy and
and the sand dune aquifer. Mohamed 2014c).
The major recharge areas mapped are the fissured carbon- Water quality changes from Mg-HCO3 to Na-Mg-Cl-
ate aquifer and western gravel aquifer comprising of HCO3 type and groundwater salinity increases in the flow
compromising low clay and high boulders while the discharge direction from the east to the west (Tang et al. 2001;
areas mapped are the sand dunes and coastal aquifers. Elmahdy and Mohamed 2012). This is because the water
has the sufficient time to dissolve the salts in the main wadis.
Hydrological elements A hydraulic map presented in Fig. 5b highlights the rela-
tively high hydraulic head within the western gravel aquifer in
The new map of palaeochannels delineated from DEM using the east. This map showed a gradual decrease in water table
D8 algorithm and are shown in Figs. 5a and 3. As demonstrat- from 325 m at the foot of Hafet and Oman Mountains and
ed from Figs. 3 and 5a, the general groundwater flow direction Liwa area (recharge areas) to 4 m near the coastal areas in the
within the area is to the northwest, west, and southwest toward west (discharge area). The paleochannels and hydraulic head
the Arabian Gulf. The spatial pattern of palaeochannels illus- map provide an overlay of downward and lateral components
trates the presence of groundwater divide near Hafeet and of groundwater and surface flow within the shallow aquifer
Oman Mountains and south of Liwa City. The map also shows (Fig. 5). Up-streams and highlands in the east are areas where
that the water flows from the Oman Mountain in the east and recharge most likely goes into local-scale flow system and
Hegaz in the southwest to the Arabian Gulf. some discharges into coastal areas in the west. Close to these
The nature of the revealed palaeochannels identified in areas, the water table is shallow and thus indicates groundwa-
Abu Dhabi is shallow, narrow, and short compared with those ter and potential for groundwater discharge. The low hydraulic
identified in the Eastern Sahara (Paillou et al. 2009; Elmahdy head (<5 m) is found in saline groundwater and coastal areas
and Mohamed 2013b; Elmahdy and Mohamed 2014c), indi- (Fig. 5b).
cating a long period of wet seasons during the late Quaternary The spatial analysis indicates that the syncline folds at Liwa
age in the Eastern Sahara than this part of the Arabian and Baynouna and the middle part of Abu Dhabi Emirate
Peninsula. Similar to Hutchinson (1996), this study also pro- (Fig. 4a) show a low total dissolved solids (TDS) value of
poses a possible regional flow (link) from the Oman Mountain 1000 mg/L, compared with 35,000 mg/L in the coastal
to the Arabian Gulf (west) via palaeochannels underneath Sabkha (west), implying that the lithological contacts (gaps)
sand dunes and sand dune corridors. Most of palaeochannels and the syncline folds played the role of freshwater reservoirs
are longitudinal in shape, reflecting a terrain of very gentle (Fig. 6a).
slope (Fig. 4b), and therefore high precipitation and ground-
water recharge. Environmental elements
The flow system in Abu Dhabi Emirate can be categorized
into three types under the influence of slope and fault dis- The most important problems related to groundwater in
places (Figs. 3 and 4) (Alsharhan et al. 2001; Wood and Abu Dhabi are the contamination of shallow aquifer due to
Sanford 2002): (1) local-scale flow, (2) intermediate-scale the dissolution of different sources (JICA 1996). These
flow, and (3) regional-scale flow (Figs. 3 and 5a). The local sources include natural and anthropological factors such as
flow is spread from Mountainous areas to the western gravel host rock, agricultural activity, and seawater intrusion which
aquifer and contains freshwater and reflects steep slope. The have direct and indirect effects on groundwater quality. The
intermediate flow is spread from western gravel to sand dune direct effects include dissolution and transport of excess quan-
and contains brackish water. The regional flow extends from tities of fertilizers and associated materials, while the indirect
sand dune areas to coastal areas and contains saline water. At effects include host rock dissolution (Hounslow 1995).
the regional-scale flow of the study area (coastal aquifer), the The excessive pumping of groundwater and scarcity of
flow lines indicate regional groundwater discharge into the rainfall are factors which have caused serious depletion of
Arabian Gulf (Fig. 5a). At local-scale flow, the hydraulic head potentiometric water level along the coastline of the UAE
increases, where NDVI indicates groundwater discharge. (Hutchinson 1996). As a result, a landward-directed pressure
Some of these palaeochannels are filled with alluvial de- gradient caused seawater intrusion toward the coastal aquifer
posits and represent an excellent groundwater reservoir, espe- and extends to the Al Ain area in the east as indicated by the
cially those distrusted in the alluvial plain. Within these dominance of MgCl2 and NaCl salts (Hounslow 1995).
palaeochannels, groundwater quantity and quality are highly The map of seawater intrusion simulated from DEM illus-
variable and may spatially change along the flow direction trated in Fig. 6b highlights the areas that may have
Arab J Geosci

(a)

Oman

Saudi Arabia

400 (b)

0m

Oman

Saudi Arabia

Fig. 5 Paleochannel map delineated from DEM using D8 algorithm (a), and hydraulic head map (b) indicating groundwater discharge to the Arabian
Gulf and Ruh A Khali in Saudi Arabia. Arrows highlight flow directions from Oman and Hegaz Mountains (modified after (Hutchinson 1996))

experienced seawater intrusion when the water table depletes dominance of NaCl salts. The Na/CI ratio, which indi-
10 m below sea level. The map shows that the coastal aquifer, cates seawater intrusion, is less than unity (0.85) and
which covers an area of about 8600 km2 may have experi- greater than unity (Hounslow 1995; Elmahdy and
enced seawater intrusion, whenever the water table depth and/ Mohamed 2012c).
or fresh–saline transition zone drops 10 m below the sea level. The vegetation density, which reflects the agricultural ac-
Of the 72,820 km2, 46,554 km2 of the west coast of the UAE tivity, was mapped for the entire UAE, and it ranges from very
may experience seawater intrusion, if water table drops down high to a very low value (Fig. 7a). An area where the vegeta-
to 50 m under sea level (Fig. 6b). tion density is very high (>50) indicates regions of groundwa-
The result also shows that the shoreline from Abu Dhabi ter discharge, and high evapo-transpiration which is also an
Emirate to Ras Al Khaimah Emirate might get affected by indication of the potential groundwater discharge. These areas
seawater intrusion if the water table drops to 10 m below sea are spatially distributed across in Liwa and Baynouna, sand
level. Saltwater intrusion in this area is indicated by the dune corridors, and Wadi Al Ain covering an area of 19,
 Arab J Geosci

37000
(a)

0 mg/L

Oman

Saudi Arabia

30-50
(b)
20-30

0- 20 m

Oman

Saudi Arabia

Fig. 6 Maps of groundwater elevation (a), and thickness of shallow aquifer showing the spatial variation of groundwater level and aquifer thickness
under the influence of topographic slope and geological structures

685 km2, which is 16 % of the UAE land area. This situation The integrated map of groundwater recharge, discharge,
has led to the generation and accumulation of all types and recharge–discharge areas is shown in Fig. 8. The map
of poisons and pollution (Elmahdy and Mohamed shows a gradual of spatial distribution of groundwater re-
2012a). Nitrate (NO3) is leached easily and poses health charge, discharge, and recharge–discharge areas in the study
risks for infants and alike. It is the most commonly area. The recharge areas are the smallest (1658 km2) and com-
recognized contaminants affecting groundwater (Freeze prise the carbonate aquifer (Hafeet Mountain). The aquifer
and Cherry 1979). NO3 value of 1000 mg/L is encoun- provides runoff from rainfall to actively recharge a number
tered in Baynouna and Liwa in the south and western of significant channels (Figs. 3 and 5a). The discharge areas
gravel areas, while the lower value (20 mg/L) for nitrate are the largest (45,54 km2) and comprise linear and barchan
concentration is encountered in the sand dune and coast- dunes and affected by syncline folds containing large amounts
al areas (Fig. 7b). This positive relationship is an evi- of fossil waters. The areas are dotted with small depressions of
dence for effectiveness of leaching fertilization. sabkha and have surface water present (Brook et al. 2006).
Arab J Geosci

-50 - 0 (a)

0 - 50

Saudi Arabia

760
(b)

0 mg/L

Oman

Saudi Arabia

Fig. 7 NDVI map calculated from Landsat 8 images (a), and NO3 concentration map (b) showing the positive correlation of agricultural activity and
concentration of NO3 in groundwater

The recharge–discharge areas occupy an area of 11,618 km2 map is less successful in identifying potential discharge areas.
and comprise the coastal sabkha due to high rate of evapora- The mapping of factors controlling groundwater recharge, dis-
tion and sea water intrusion (Fig. 8). charge, and quality is integral to managing water resources
Graphs of elevation versus topographic slope and elevation and understanding hydrological setting of remote and inacces-
versus hydraulic head and groundwater salinity are shown in sible areas.
Fig. 9. A graph of elevation versus the slope suggests in- In semi-arid regions, shallow groundwater table, scarce
creases in observed slope and water level near mountainous rainfall, and overextraction of groundwater can result in the
areas. In the same context, graph of elevation/slope versus degradation of water and resources (Cartwright et al. 2004;
groundwater salinity suggests that salinity increases as the Mohamed and Hatfield 2005). The derived maps present hy-
slope and elevation decrease from Hafeet and the Oman drological information on shallow aquifer over a regional
Mountain in the east to the coastal aquifer in the west. In scale. The recharge areas are mountainous areas with steep
comparison to the groundwater salinity and paleochannel slopes and have a relatively low infiltration capacity. These
maps, the results shown in Fig. 8 suggest that the NDVI results have permitted a better understanding of the
 Arab J Geosci

Recharge area

Discharge area

Recharge –discharge area

Fig. 8 Map of groundwater discharge, preferential recharge, and aquifer. Coastal aquifer is recharge and discharge areas. Arrows
recharge–discharge areas. The map shows that the sand dunes are highlight flow direction and seawater intrusion
potential discharge areas and western gravel is potential recharge

hydrological setting of Abu Dhabi. For future work, it would Conclusion

be interesting to model water balance by using polarimetric
covers dual HH/VV and HH/H of the SAR data and thermal The current study presented an approach based on a set of
band of Landsat images, and compare the results of that study automated algorithms, which use multiple sources of remote
with results of this study. sensing data to map factors controlling groundwater recharge,

Fig. 9 Graphs of elevation via (a)

topographic slope (a), and altitude
via hydraulic head and
groundwater salinity
Groundwater recharge

Groundwater discharge

(b)
Salinity (mg/l)
Elev. (m)
Arab J Geosci

discharge, and quality across remote regions. Factors control- UPAR-7-2013-31N166). The authors acknowledge the assistance provid-
ed by both institutes.
ling groundwater recharge, groundwater discharge, and
groundwater quality were effectively identified by mapping
lithology, palaeochannels, slope, hydraulic head, NDVI, and
References
seawater intrusion.
The results showed that the carbonate rocks of Hafeet and
Al-Rashed MF, Sherif MM (2000) Water resources in the GCC countries:
the Oman Mountains (sources of recharge) are preferable re-
an overview. Water Resour Manag 14(1):59–75
charge areas and, at the same time, sources of HCO3, Ca, Mg, Alsharhan AS, Rizk ZA, Nairn AE, Bakhit DW, Alhajari SA (2001)
Cl, and Na in the groundwater. The palaeochannels mapped Hydrogeology of an arid region: the Arabian gulf and adjoining
from DEM and hydraulic head mapped from groundwater areas. Elsevier B.V, The Netherlands, p P.331
wells showed that the recharge area in the east, a discharge Brook MC and Dawoud M (2005) Coastal Water Resources Management
in the Coastal Zones of United Arab Emirates. Symposium on
area in the west, and the groundwater salinity increase in the Integrated Coastal Zone Management, Abu Dhabi, UAE
flow directions. The results also showed a positive correlation Brook M, Al Houqani C, Darawsha T, Al Alawneh M, Achary S (2006)
between the spatial distribution of NDVI and groundwater Groundwater Resources; Development and Management in
discharge areas and NO3 concentration in the groundwater. Abu Dhabi Emirate, UAE. In Proceedings of 3rd Joint UAE-Japan
Symposium, Sustainable GCC Environment and Water Resources,
The slope also helped in identifying surface water infiltration EWR, 2006
rate, while seawater intrusion map indicates that the coastal Cartwright I, Weaver TR, Fulton S, Nichol C, Reid M, Cheng X (2004)
aquifer is a hazardous area with high salinity. Hydrogeochemical and isotopic constraints on the origins of dryland
The terrain category map suggests a significant amount of salinity, Murray basin, Victoria, Australia. Appl Geochem 19:1233–
1254
seawater intrusion into the shallow aquifer, especially the
Dehaan RL, Taylor GR (2002) Field-derived spectra of salinized soils and
coastal aquifer. This is due to overpumping and withdrawal vegetation as indicators of irrigation-induced soil salinization.
of groundwater in excess of recharge in Abu Dhabi. The inte- Remote Sens Environ 80:406–417
grated map showed that the region comprises groundwater Elmahdy SI and Mohamed MM (2012a) Influence of geological struc-
recharge, discharge, and recharge–discharge areas. The most tures on groundwater accumulation and groundwater salinity in
Masandam Peninsula, UAE and Oman, Geocarto International,
important elements affecting groundwater in Abu Dhabi can DOI:10.1080/10106049.2012.724455
be summarized as follows: Elmahdy SI, Mohamed MM (2012b) Topographic attributes control
groundwater flow and groundwater salinity of Al Ain UAE: a pre-
(i) Regional fault displacements that control slope and flow diction method using remote sensing and GIS. J Environ Earth Sci
2(8):1–13
direction and palaeochannels
Elmahdy SI, Mohamed MM (2012c) Geological lineament detection,
(ii) The properties of water-bearing rocks such as porosity characterization and association with groundwater contamination
and chemical properties in carbonate rocks of musandam peninsular using digital elevation
(iii) The palaeochannels and flow direction that represent an model (DEM). Open Hydrol J 6(Suppl 1-M3):45–51
excellent groundwater reservoir and groundwater salin- Elmahdy SI, Mohamed MM (2013a) Natural hazards susceptibility map-
ping in Kuala Lumpur, Malaysia: an assessment using remote sens-
ity indicator. ing and geographic information system (GIS). Geomatics Nat
(iv) Seawater intrusion and agricultural activity and fertiliza- Hazards Risk 4(1):71
tion that are sources of NO3, Cl, Na, HCO3, and Ca in Elmahdy SI, Mohamed MM (2013b) Remote sensing and GIS applica-
groundwater in sand dune and coastal aquifers tions of surface and near-surface hydromorphological features in
Darfur region, Sudan. Int J Remote Sens 34(13):4715–4735
Elmahdy SI, Mohamed MM (2014a) Relationship between geological
By mapping factors controlling groundwater recharge structures and groundwater flow and groundwater salinity in Al
and discharge and quality using remote sensing and Jaaw Plain, United Arab Emirates; mapping and analysis by means
GIS, new maps were constructed and hydrological in- of remote sensing and GIS. Arab J Geosci 7(3):1249–1259
Elmahdy SI, Mohamed MM (2014b) Probabilistic frequency ratio model
formation is obtained for numerical modeling and cost-
for groundwater potential mapping in Al Jaww plain UAE. Arab J
effective water balance analysis. Finally, it would be Geosci 15:1–12
interesting to map factors controlling groundwater re- Elmahdy SI, & Mohamed MM (2014c) Automatic detection of near
charge, discharge, and quality from high-resolution re- surface geological and hydrological features and investigating their
mote sensing data. These data will reduce the uncer- influence on groundwater accumulation and salinity in southwest
Egypt using remote sensing and GIS. Geocarto International,
tainties in mapping the factors controlling groundwater (ahead-of-print), 1-13
recharge, discharge, and quality. Elmahdy SI, Mohamed M (2014d) Groundwater potential modelling
using remote sensing and GIS: a case study of the Al Dhaid area
United Arab Emirates. Geocarto Int 29:433–450
Elmahdy SI, Mansor S, Rodsi A, Bujang BK (2011) Geotechnical model-
Acknowledgments This research was funded by the UAE National ing of fractures and cavities that are associated with geotechnical
Water Center (NWC) and the UAE University Program for Advanced engineering problems in Kuala Lumpur limestone, Malaysia.
Research (UPAR) (projects’ numbers NWC -3-2013- 31R008 and Environ Earth Sci 62(1):61–68
 Arab J Geosci

Elmahdy SI, Mansor S, Huat BB, Mahmod AR (2012) Structural geolog- Mohamed M, Hatfield K (2011) Dimensionless Monod parameters to
ic control with the limestone bedrock associated with piling prob- summarize the influence of microbial inhibition on the attenuation
lems using remote sensing and GIS: a modified geomorphological of groundwater contaminants. Biodegradation 22:877–896
method. Environ Earth Sci 66(8):2185–2195 Mohamed M, Hatfield K, Perminova IV (2007) Evaluation of Monod
Freeze RA, Cherry JA (1979) Groundwater. Prentice-Hall, Englewood kinetic parameters in the subsurface using moment analysis: theory
Cliffs, 604p and numerical testing. Adv Water Resour 30(9):2034–2050
GuthPL (2009) Geomorphometry in MICRODEM. In: T. Hengl and H.I. Mohamed M, Hatfield K, Hassan AE, Klammler H (2009) Stochastic
Reuter, eds. Geomorphometry: concepts, software, applications [on- evaluation of subsurface contaminant discharges under physical,
line]. Developments in soil science series. USA Navy Academy, chemical, and biological heterogeneities. Adv Water Resour 33(7):
351–366. Available from: http://www.usna.edu/Users/ oceano/ 801–2010
pguth/website/microdemdown [Accessed 13 September 2012] Mohamed MMA, Saleh NE, Sherif MM (2010a) Modeling in situ ben-
Hoffmann J (2005) The future of satellite remote sensing in hydrogeolo- zene bioremediation in the contaminated Liwa aquifer (UAE) using
gy. Hydrogeol J 13:247–250 the slow-release oxygen source technique. Environ Earth Sci 61(7):
Hounslow AW (1995) Water quality data analysis and interpretation. 1385–1399
CRC Press Inc Lewis Publishers, New York, 397p Mohamed M, Saleh N, Sherif M (2010b) Sensitivity of natural attenua-
Hutchinson CB (Ed.) (1996) Ground-water resources of Abu Dhabi tion to variations in kinetic and transport parameters. Bull Environ
Emirate, U.S. Geological Survey Administrative Report prepared Contam Toxicol 84(4):443–449
for the National Drilling Company of Abu Dhabi, P.135 Paillou PH, Schuster M, Tooth S, Farr T, Rosenqvist A, Lopez S,
HydroConsult (1987) Reconnaissance Report and Development Malezieux MJ (2009) Mapping of a major paleodrainage system
Proposals, Abu Dhabi, (UAE) Eastern Region Water Resources,^ in eastern Libya using orbital imaging radar: The Kufrah River.
Ministry of Petroleum and Mineral Resources Report, Government Earth Planet Sci Lett 277(2009):327–333
of Abu Dhabi, UAE, 126 p Salama R, Tapley I, Ishii T, Hawkes G (1994) Identification of areas of
JICA (Japanese International Cooperation Agency) (1996) The master recharge and discharge using Landsat-TM satellite imagery and ae-
plan study on the groundwater resources development for agricul- rial photography mapping techniques. J Hydrol 162:119–141
ture in the vicinity of Al Dhaid in the UAE, JICA International Sherif MM, Mohamed MM, Shetty A, Almulla M (2010) Rainfall-runoff
Cooperation Agency, (1996), Final Report modeling of three wadis in the northern area of UAE. J Hydrol Eng
Jenson SK, Domingue JO (1988) Extracting topographic structure from 16(1):10–20
digital elevation data for geographic information system analysis. Shirazi SM, Imran HM, Akib S (2012) GIS-based DRASTIC method for
Photogramm Eng Remote Sens 54(11):1593–1600 groundwater vulnerability assessment: a review. J Risk Res 15(8):
Kruse FA, Lefkoff AB, Boardman JB, Heidebrecht K, BShapiro AT, 991–1011
Barloon PJ, Goetz AFH (1993) The spectral image processing sys- Shirazi SM, Imran HM, Akib S, Yusop Z, Harun ZB (2013) Groundwater
tem (SIPS) - interactive visualization and analysis of imaging spec- vulnerability assessment in the Melaka state of Malaysia using
trometer data. Remote Sens Environ 44:145–163 DRASTIC and GIS techniques. Environ Earth Sci 70(5):2293–2304
Leblanc M, Razack M, Dagorne D, Mofor L, Jones C (2003) Application Subba RN (2006) Groundwater potential index in a crystalline terrain
of meteosat thermal data to map soil infiltrability in the central part using remote sensing data. Environ Geol 50:1067–1076. doi:10.
of the lake Chad basin, africa. Geophys Res Lett 30(19):1998. doi: 1007/s00254-006-0280-7
10.1029/2003GL018094 Tang C, Machid I, Shindo S, Kondoh A, Sakura Y (2001) Chemical and
Maged MM (2014) Particle Swarm Optimization for Geological Feature isotopic methods for confirming the roles of wadis in regional
Detection from Palsar Data.Proceeding of 35th Asian Conference on groundwater recharge in a regional arid environment: a case study
Remote Sensing will be held at Nay Pyi Taw, Myanmar from 27- 31, in Al Ain, UAE hydrological processes hydrology. Process 15:
October 2014 2195–2202
Marghany M (2012) Fuzzy B-spline algorithm for 3-D UAE National Atlas (United Arab Emirates National Atlas) (1993)
lineamentreconstruction. Int J Phys Sci 7(15):2294–2301 Remote sensing center: UAE University, p. 188
Marghany M, Mansor S, Hashim M (2009) Geologic mapping of united US Geological Survey (2001) Bibliography of National Drilling
arab emirates using multispectral remotely sensed data. Am J Eng Company – United States Geological Survey Reports on the Water
Appl Sci 2:476–480 Resources of Abu Dhabi Emirate. U.S Geological Survey Open File
Metternicht G, Zinck JA (1997) Spatial discrimination of salt-and sodi- Report 2001-XXX
um-affected soil surfaces. Int J Remote Sens 18(12):2571–2586 Wood WW, Sanford WE (2002) Hydrogeology and solute chemistry of
Mohamed M, Almualla A (2010) Water demand forecasting in Umm Al- the coastal – sabkha aquifer in the emirate of Abu Dhabi. In: Joerg
Quwain using the constant rate model. Desalination 259:161–186 Barth H, Benno B (eds) Sabkha ecosystem, Vol 1- the sabkhas of the
Mohamed MM, Elmahdy SI (2014) Natural and anthropogenic factors Arabian peninsula and adjacent countries. Kluwer Academic
affecting groundwater quality in the eastern region of the United Publishers, Dordrecht
Arab Emirates. Arab J Geosci 12:1–15 Woodward GL, Al Sharhan AS (1994) Domlomitization and
Mohamed MM, Hatfield K (2005) Modeling microbial-mediated reduc- chertification of the early Eocene Rus formation in Abu Dhabi
tion in batch reactors. Chemosphere 59(8):1207–1217 UAE. Sediment Geol 91:273–285