Water Resour Manage

DOI 10.1007/s11269-016-1421-8

Delineation of Groundwater Potential Zones of Coastal

Groundwater Basin Using Multi-Criteria Decision
Making Technique

Uday Mandal 1 & Satiprasad Sahoo 2 &

Selva Balaji Munusamy 3 & Anirban Dhar 3 &
Sudhindra N. Panda 4 & Amlanjyoti Kar 5 &
Prasanta K. Mishra 6

Received: 25 February 2016 / Accepted: 24 June 2016

# Springer Science+Business Media Dordrecht 2016

Abstract Delineation of groundwater potential zones (GWPZ) has been performed for a
coastal groundwater basin of eastern India. The groundwater potential zone index (GWPZI)
map is generated by using Analytic Hierarchy Process (AHP) from different influencing
features, e.g., Land Use/Land Cover (LU/LC), soil (S), geomorphology (GM), hydrogeology

Electronic supplementary material The online version of this article (doi:10.1007/s11269-016-1421-8)

contains supplementary material, which is available to authorized users.

* Anirban Dhar
[email protected]; [email protected]

Uday Mandal
[email protected]

Satiprasad Sahoo
[email protected]

Selva Balaji Munusamy

[email protected]

Sudhindra N. Panda
[email protected]

Amlanjyoti Kar
[email protected]
Prasanta K. Mishra
[email protected]

1
Hydrology and Engineering Division, ICAR-Indian Institute of Soil & Water Conservation,
Dehradun 248195, India
2
School of Water Resources, Indian Institute of Technology Kharagpur, Kharagpur 721302, India
 U. Mandal et al.

(HG), surface geology (SG), recharge rate (RR), drainage density (DD), rainfall (RF), slope
(Sl), surface water bodies (SW), lineament density (LD), and Normalized Difference
Vegetative Index (NDVI). Recharge rate values are estimated from hydrological water balance
model. Overlay weighted sum method is used to integrate all thematic feature maps to generate
GWPZ map of the study area. Four zones have been identified for the coastal groundwater
basin [very good: 36.39 % (273.53 km2, good: 43.57 % (327.47 km2), moderate: 18.27 %
(137.30 km2), and poor: 1.77 % (13.27 km2)]. Areas in north to south-west and south-east
direction show very good GWPZ due to the presence of low drainage density. GWPZ map and
well yield values show good agreement. Sensitivity analysis reveals that exclusion/absence of
rainfall and lineament density increases the poor groundwater potential zones. Omission of
hydrogeology, soils, surface geology, and NDVI show maximum increase in good GWPZ.
Obtained GWPZ map can be utilized effectively for planning of sustainable agriculture. This
analysis demonstrates the potential applicability of the methodology for a general coastal
groundwater basin.

Keywords Coastal groundwater basin . Groundwater potential zone . Analytical hierarchy

process (AHP) . Remote sensing (RS) . Geographic information system (GIS)

1 Introduction

Coastal groundwater basins are important for human settlement and development. Half of the
world’s population lives within 60 km of the shoreline, and this number could increase up to
three quarters over a decade (UNCED 1992). Increase in unplanned groundwater use disturbs
the natural recharge-discharge equilibrium. On the whole, over exploitation of coastal aquifers
may have serious consequences on environment, ecology and economy of the region
(Moreaux and Reynaud 2006). Planning of groundwater management framework requires
information about GWPZs. GWPZ is defined for an area based on relative likelihood of getting
groundwater resources. GWPZs provide an imprecise assessment of groundwater resources
based on remote sensing and conventional secondary data. Thus, a multicriteria decision
making tool along with hydrological model is applied for groundwater potential zoning of a
coastal groundwater basin in Eastern India.
GWPZ can be precisely identified by using integrated hydrogeological and geophysical
surveys coupled with borehole logging. However, these methods are time consuming and
expensive. Thus large amount of spatio-temporal RS images/ data along with limited field
observations can be utilized to get a reasonable guess of the groundwater potential for an area.
Multicriteria decision making and geoinformatic tools can provide a qualitative estimate of
groundwater resources (Al-Adamat et al. 2003; Srivastava and Bhattacharya 2006; Prasad et al.
2007; Madrucci et al. 2008; Chowdhury et al. 2009; Nagarajan and Singh 2009; Avtar et al. 2010;

3
Department of Civil Engineering, Indian Institute of Technology Kharagpur, Kharagpur, WB 721302,
India
4
Department of Agricultural and Food Engineering, Indian Institute of Technology Kharagpur,
Kharagpur 721302, India
5
CGWB, Bhujal Bhawan, Faridabad, Haryana, India
6
ICAR-Indian Institute of Soil & Water Conservation Dehradun, Dehradun 248195, India
Delineation of Groundwater Potential Zones of Coastal Groundwater Basin

Chenini et al. 2010; Jha et al. 2010; Machiwal et al. 2010; Oh et al. 2011; Adiat et al. 2012;
Bagyaraj et al. 2013; Magesh et al. 2012; Mukherjee et al. 2012; Nag and Ghosh 2012; Agarwal
et al. 2013a; Agarwal et al. 2013b; Deepika et al. 2013; Gumma and Pavelic 2013; Mahmoud
et al. 2014; Dhar et al. 2015a, 2015b; Zaidi et al. 2015; Agarwal and Garg 2016; Sahoo et al. 2016;
Senanayake et al. 2016). Moreover, artificial groundwater recharge zones can also be identified by
using similar approach (Ravi Shankar and Mohan 2005; Chowdhury et al. 2010).
Groundwater potential of an area varies spatially depending on the variability of
influencing features. In this study, analytic hierarchy process (AHP; Saaty 1980) along
with geoinformatic tools are utilized for delineation of GWPZs. Land use/land cover
(LU/LC), topography (slope), geology, hydro-geology, geomorphology, soil map,
drainage density, rainfall, lineament density, normalized difference vegetation index
(NDVI), surface water body, and recharge rate are considered as influencing features
during the analysis. In the present study an attempt has been made to incorporate the
recharge rate considering overall hydrologic balance. Practical applicability of the
resulting groundwater potential zone map is validated by the spatially distributed
actual well yield values. Sensitivity analysis is performed to see the degree of
influence of the individual features on groundwater occurrence. The coastal area
between the Subarnarekha and the Budhabalanga rivers covering the Balasore
District, Odisha State (India) is selected as the study area.

2 Materials and Methodology

2.1 Study Area

The study area is a coastal groundwater basin situated in the Balasore district of Odisha State in
Eastern India. Area is bounded by the Subarnarekha River in the east, the Budhabalanga River
in the west, hills in the north, and the Bay of Bengal in the south (Fig. 1). It consists of three
administrative blocks (Balasore, Basta, and Baliapal). The basin lies between latitude 21° 27′

Fig. 1 Study area location of coastal groundwater basin, Balasore (Odisha), India
 U. Mandal et al.

0″ to 21° 45′ 45″ N and longitude 86° 56′ 15″ to 87° 20′ 30″ E covering an area of about
751.58 km2. In the northern part, the hilly region is the extension of the Eastern Ghats
(Mountains). The surface elevation of the area ranges from 0 to 40 m above the mean
sea level. The basin mainly comprises of two rivers, the Subarnarekha and the
Budhabalanga, which drain into the Bay of Bengal. There is one lined coastal
navigational canal carrying saline water running from west to east direction. The
study area comprises of alluvial deposits of recent to older in origin. Fine and coarse
loamy are the predominant soil character. The climate of the district is humid sub-tropical and
characterized by three distinct seasons viz. winter, summer and rainy seasons. The winter
seasons extends from the late November to the end of February, followed by the summer season
which continues up to the middle of June.
The maximum temperature rises up to maximum of 45 °C in summer and falls to a
minimum of 7 °C in winter season. The average temperature ranges between 25 to
32 °C. The relative humidity is high and varies from 40 to 85 % during the year.
During the monsoon season, the humidity is high in the coastal area. The potential
evapotranspiration ranges from 50 mm in January to 280 mm in May. Southwest
monsoon is the principal source of rainfall in the district. The rainfall is recorded at
all the block headquarters. The 20 years (1991–2010) average annual rainfall of the
study area is 1760 mm.
Agriculture is the main source of income for the rural population. Paddy is the mjor crop
(97 % during monsoon season) and 47 % area covered under winter crop. Rainfall and
groundwater are the only sources of irrigation water. Due to erratic behavior of southwest
monsoon, supplementary irrigation is also required during Kharif (monsoon) season (Rejani
et al. 2003). Moreover, frequent extreame events are observed this part of the coastal Odisha.
Socio-economic status can be improved by optimum crop planning, irrigation scheduleing,
locating tube wells, selecting sustainable pumping strategy along with a strong groundwater
legislation.

2.2 Groundwater Potential Zones

The methodology utilized will be useful to determine GWPZs using integrated spatial science
tool (RS & GIS). Identification of potential groundwater zones depend on many features as
well as their individual sub-features (Table S1). In this study, 12-thematic layers of LU/LC, S,
GM, HG, SG, RR, DD, RF, Sl, SW, LD, and NDVI are considered for delineation of
GWPZs. Standard index approach is applied for delineation of groundwater potential
zones. Individual feature layers are reclassified into sub-features and ranks are
assigned accordingly. Goundwater potential index map is generated in GIS platform
by using weighted linear combination of individual feature layers. Groundwater
potential index (GWPI) is calculated as

!
X X   
v 
GWPI i; j ¼ Wk wkl χAk C i; j  ð1Þ
l k
k∈ F l∈S k

where index (i, j) denotes row column location of a pixel; F denotes the set of all features, k denotes
element of feature set; Sk denotes set of sub-features for kth feature; l denotes element of sub-feature
Delineation of Groundwater Potential Zones of Coastal Groundwater Basin

set; Wk normalized weight of kth feature; wkl normalized weight of lth sub-feature for kth feature;


C vi; j  denotes the class value of the cell (i,j) for kth feature; Akl denotes the lth sub-feature interval
k
for kth feature; χAk denotes the indicator function for lth sub-feature of kth feature and defined as
l

8 
  < 1 if C vi; j  ∈Akl ;
χAk C vi; j ¼ k ð2Þ
l : 0 if C v  ∉Ak :
i; j l
k

Analytic Hierarchy Process (AHP) can be applied for estimation of Wk and wkl . In AHP
(Saaty 1980), 1–9 scale (i.e., extremely unimportant, strongly unimportant, unimportant,
moderately unimportant, equally important, moderately important, more important, strongly
important, extremely important) is adopted for constructing judgment matrices. The following
steps are adopted for calculation of weights and consistency ratio (CR):

Step I. Assignment of ranks based on experts’ opinion and site specific conditions.
Development of judgment matrices (A) by pair wise comparison.
Step II. Calculation of relative weight Wk:
.X
W k ¼ GM k GM m ð3Þ
m∈ F

Where, the geometric mean of the kth row of judgment matrix is calculated as
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
GM k ¼ NF ak1 ak2 :::akN F , NFis the total number of features.

Step III. Strength assessment of judgment matrix based consistency ratio (CR)
.
C:R: ¼ C:I: R:C:I: ð4Þ

Where CI is the consistency index; and RCI is random consistency index. Consistency
index (C.I.) is evaluated as
λmax −N F
C:I: ¼ ð5Þ
N F −1
where the latent root of judgment matrix is calculated as

X ðAWÞ
λmax ¼ m
ð6Þ
m∈ F
N F W m

where Wis the weight vector (column). Random consistency index (RCI) can be obtained from
standard tables (Alonso and Lamata 2006). CR value less than 0.1 is acceptable for a specific
judgment matrix. However, revision in judgment matrix is needed for CR ≥ 0.1.
Same procedure should be followed for wkl calculation. Finally, potential zone map can be
generated from the above mentioned procedure. The results were validated by using statially
distributed actual well yield data within the study area. The methodology adopted in this study
is given in the schematic flowchart (Fig. 2).
 U. Mandal et al.

Fig. 2 Schematic flowchart for groundwater potential zone mapping

2.3 Sensitivity Analysis

Sensitivity analysis can be peformed by omitting individual features used in AHP. Major
changes due to omission of a particular thematic feature indicates the influence of that feature
in the final GWPZ map. Sensitivity can be expressed as

S −ij −S jF
DS ij ¼  100 ð%Þ ð7Þ
S jF

where i represents parameter number and j represents type of zone (very good, good, moderate,
poor). DS ij is the percentage change (+/−) in jth type of GWPZ area due to the absence of ith
feature. S −ij is the jth type of GWPZ area due to omission of ith feature, and S jF is the jth type of
GWPZ area using all features.
Delineation of Groundwater Potential Zones of Coastal Groundwater Basin

3 Results and Discussions

Twelve thematic maps are identified for the delineation of GWPZs. Details of all the thematic
maps relevant to groundwater prospect for this study are discussed in the following sub-
sections:

3.1 Land Use/Land Cover Classes

For the identification of LU/LC patterns of the study area, Enhanced Thematic
Mapper Plus (ETM+) image is used. Total eight numbers of different classes are
identified based on unsupervised classification. The details of the land use classes are
given in Table 1 and shown in Fig. 3(a). Vegetation retards the surface runoff,
increases water detention time and enhances infiltration. Higher weightage values
are assigned for the river/surface water body, mixed forest, sand, moist land, and
crop land. Whereas agricultural fallow, open forest, dry fallow, and built-up areas are
given lower weightage. Built-up and settlement (urbanization) units retard rainfall
infiltration and thus prevent recharged water to reach groundwater reservoirs.

3.2 Soil

Soil also plays a significant role in mapping GWPZs, e.g., coarse textured soil is
generally permeable while fine textured soils exibit less permeability. Highly perme-
able soils permit relatively rapid rate of infiltration wherein much of the rainwater can
reach groundwater table at a faster rate. Soil map of the area is collected from the
ICAR-National Bureau of Soil Survey and Land Use Planning, Nagpur, India. Soil of
the study area is classified as six types namely: sandy loamy, sandy clay, clay, coarse
loamy, fine loamy and sandy (Fig. 3(b)). Types of soil and percentage area covered
are given in Table 1.
Maximum portion of the study area are covered by the fine loamy soil (481 km2) and
course loamy soil (247.20 km2). Coarse loamy soil is concentrated in the central part and the
eastern part of the study area. Sandy loam soil is occupying less geographical area (24.22 km2)
and could be seen near the Subarnarekha River. This type of soil possesses high specific yield
and porosity value (32.38 %), which indicates comparatively higher recharge from precipita-
tion than it happens in other areas.

3.3 Geomorphology

For delineation of GWPZs, geomorphology is one of the controlling features.

Geomorphology map helps to identify various geomorphic units. Geomorphology
and associated features as identified through visual interpretation of satellite image
and existing maps from Groundwater Survey and Investigation (GSI) Division,
Baripada, Govt. of Odisha are shown in Fig. 3(c). Mainly three types of geo-
morphological feature are shown: hard greyish green clay occupies 577.88 km2; old
sand dune, marine deposit, fluvial silt, clay, and deltaic deposit cover 191.41 km2; and
flood plain, recent sand dune and marine deposit cover 95.16 km2. The flood plain
area comprises mainly of sand and slits with minor intercalations of clay and they act
as good aquifers. The groundwater potential of flood plain is very high.
 U. Mandal et al.

Table 1 Details of feature layers and their features ranks and normalized weights

Sl. Feature Rank Normalized Sub-feature Area Groundwater Rank Normalized

no. assigned weight cover prospect weight
(%)

1. Land 7 0.1029 Agriculture, Crop 3.09 Very good 9 0.1795

use/land land
cover Forest land 3.73 Very good 8 0.1596
Water body 4.43 Good 7 0.1396
Sandy area 0.75 Good 7 0.1396
Wet land 14.12 Moderate 6 0.1224
Agriculture, 6.33 Moderate 5 0.0997
Plantation
Built-up, Rural 15.54 Moderate 5 0.0798
agriculture, fellow 52.01 Poor 4 0.0798
land
2. Soil 6 0.0882 Course sandy loam 2.80 Very good 9 0.2368
Coarse loamy 28.60 Very good 8 0.2105
Sandy 5.26 Good 7 0.1842
Fine loamy 55.66 Moderate 6 0.1579
Sandy clay 2.77 Moderate 5 0.1316
Clay 4.91 Poor 3 0.0789
3. 9 0.1324 Flood plain, Recent 11.01 Very good 9 0.4737
Geomor- dune sand,
phology Marine deposit
Old dune sand and 22.14 Good 7 0.3684
marine deposit,
Fluvial silt, Clay
and deltaic
deposit
Hard grayish green 66.85 Poor 3 0.1579
clays
4. Surface 7 0.1029 Younger alluvium 24.60 Very good 8 0.5333
geology Older alluvium 75.40 Good 7 0.4667
5. 7 0.1029 Younger and older 76.03 Good 7 0.5833
Hydroge- alluvium
ology Younger aeolian 23.97 Moderate 5 0.4167
sand and sand
dunes
6. Recharge rate 8 0.1176 69–90 5.83 Very good 9 0.3462
(cm year−1) 48–68 5.39 Good 7 0.2692
33–47 26.12 Moderate 5 0.1923
24–32 23.55 Poor 3 0.1154
13–23 39.10 Poor 2 0.0769
7. Drainage 7 0.1029 0–1 30.26 Good 7 0.4088
density 2–4 7.75 Moderate 5 0.3284
(km km−2)
5–7 24.14 Poor 3 0.1971
8–12 37.83 Poor 1 0.0657
8. Rainfall 5 0.0735 1903–2008 14.63 Very good 8 0.1333
(mm year−1) 1799–1902 14.97 Good 7 0.1667
1688–1798 40.53 Moderate 6 0.2000
Delineation of Groundwater Potential Zones of Coastal Groundwater Basin

Table 1 (continued)

Sl. Feature Rank Normalized Sub-feature Area Groundwater Rank Normalized

no. assigned weight cover prospect weight
(%)

1574–1687 17.56 Moderate 5 0.2333

1441–1573 12.31 Poor 4 0.2667
9. Slope 5 0.0735 0–2.10 45.44 Very good 9 0.360
2.10–4.80 36.20 Good 7 0.280
4.80–26.98 10.94 Moderate 5 0.20
26.98–61.77 5.32 Poor 3 0.12
61.77–76.47 2.09 Poor 1 0.04
10. Surface water 2 0.0294 - 1.61 - - 1
body
11. Lineament 3 0.0441 2.40–3.70 2.52 Good 6 0.46
density 1.20–2.30 4.00 Moderate 4 0.31
(km/km2)
0.30–1.10 11.10 Poor 2 0.15
0–0.29 82.36 Very Poor 1 0.08
12. Normalized 2 0.0294 0.09–0.50 55.25 Good 5 0.42
Difference −0.03–0.08 31.54 Moderate 3 0.33
Vegetation
Index −0.27–−0.04 8.83 Poor 2 0.17
−0.80–−0.28 4.37 Very Poor 1 0.08

3.4 Surface Geology

The age of the geologic formations of the basin ranges from Archean to Quaternary. Tertiary
and Quaternary formations are predominant over the basin, while the Archeans formations are
prevalent near the northern boundary. In the area the clastics comprising clay, fine and coarse
sand with gravel are deposited by the Subarnarekha and Budhabalanga rivers and their
tributaries. Mainly two types of geological formations:. Younger alluvium (212.58 km2);
and Older alluvium (651.87 km2) could be seen in the surface geology map of the study area
(Fig. 3(d)).

3.5 Hydrogeology

The occurrence and movement of groundwater is controlled by the porosity, permeability,

disposition of the aquifers, their horizontal distribution, recharge area, and exploitation of the
aquifers. In the study area, both the Quaternary and Tertiary aquifers possess primary
porosities which facilitates the formation of potential occurrence of groundwater in the
coastal aquifers. Mainly two types of hydrogeological features (Fig. 3(e)) are present:
Younger aeolian sand and sand dunes (Saline tract) (207.24 km2); and Younger and
older alluvium (657.21 km2). The deposits of Tertiaries and Quaternaries are mainly
represented by gravel, sand, silt, and clay. Exploration data shows predominance of
finer clastics, i.e. silt and clay in the sub-surface below the average depth of 100 m.
The potential deeper aquifers are generally restricted between 25 and 100 m depth.
These deeper aquifers are tapped by tube wells operated by hand pumps for drinking
purpose and energized pumps for irrigation.
 U. Mandal et al.

Fig. 3 a Land use/ land cover; b Soil texture; c Geomorphology; d Surface geology e Hydrogeology and f
Recharge rate classification map of the study area

3.6 Recharge Rate

Recharge rate is calculated by using Soil and Water Assessment Tool (SWAT). Recharge in
SWAT is calculated as (Dhar et al. 2014):

    
wrchrg;i ¼ 1−exp −1=δgw wseep þ exp −1=δgw wrchrg;i−1 ð8Þ
Delineation of Groundwater Potential Zones of Coastal Groundwater Basin

where wrchrg , i is the amount of recharge entering the aquifer on day i(mm); δgw is the delay
time or drainage time of the overlying geologic formations (days); wseep is the total amount of
water exiting the bottom of the soil profile on day i (mm); and wrchrg , i − 1 is the amount of
recharge entering the aquifer on day i − 1(mm). A distributed parameter model, SWAT, has
been used for estimating the average groundwater recharge rate for Balasore district in Odisha,
India. Hydrological and meteorological data [Data source: NASA Climatology Resource for
Agroclimatology Daily Averaged Data (http://power.larc.nasa.gov)] for 12 years (1997–2008)
has been used. Unsupervised classification (Fig. 3(a)) method is performed for land use
classification from satellite image (LANDSAT 7, ETM+) and used in SWAT model. Soil map is
classified into six different classes (Fig. 3(b)) and used in SWAT model. Details of the groundwater
recharge rate of the study area and percentage of area covered is given in Table 1. Recharge rate,
13–23 and 24–32 cm year−1 covered an area of 39.10 % (82 sub-basin) and 23.55 % (46 sub-basin)
of the total study area, respectively. High recharge rate, 48–68 and 69–90 cm year−1, covered less
area of 5.39 % (82 sub-basin) and 5.83 %, (46 sub-basin) of the entire study area, respectively.
Subwatershed wise groundwater recharge rate map has been produced and shown in Fig. 3(f).

3.7 Drainage Density

Drainage density (Dd = L/A) is defined as the length of drainage (L) per unit area (A). SWAT is
used to delineate watershed and subwatershed. It generates 221 total subwatersheds from
Cartosat-1 DEM [Bhuvan 2014 (http://bhuvan-noeda.nrsc.gov.in)]. Drainage density is
calculated from the watershed delineation (Table 1). Drainage basins with high drainage
densities indicate that a large proportion of the precipitation goes as runs off. On the other
hand, low drainage densities indicate high rainfall infiltration. Low drainage density value is
more favourable for high groundwater potential and assigned higher weight. A large number of
sub-basin (97) covers an area of 30.26 % of the total study area which has low drainage
densities (0–1 kmkm−2) indicating high infiltration rate (Fig. 4(a)).

3.8 Rainfall

Rainfall is the major source of recharge to groundwater. Five rain gauge stations are situated in
and around the study area. 20-years (1991–2010) rainfall data has been used as collected from
the GSI Division, Baripada, Govt. of Odisha. The average annual rainfall of the study area is
1760 mm. Maximum rainfalls received in Baliapal (2792 mm in 1999) and minimum in Basta
(1012 mm in 1996 & 2004) blocks. South-west monsoon is the principal source of rainfall in
these districts. The rainfall map (Fig. 4(b)) has been grouped into five classes which are 1441–
1573, 1574–1687, 1688–1798, 1799–1902, and 1903–2008 mm year−1. Slope directly affects
the runoff and consequently affect groundwater recharge. Spatial distribution of rainfall along
with less slope helps creation of groundwater potential zones. High amount of rainfall is more
favourable for high groundwater potential; hence weight has been assigned higher.

3.9 Slope

For creation of groundwater potential, infiltration from precipitation is an important hydrolog-

ical parameter. Infiltration depends on many features like, soil type, vegetative cover, and land
slope. Due to high slope, water cannot be ponded on the land surface for a prolonged duration
to infiltrate. Slope is calculated from Cartosat-1 DEM. Result shows that 45.44 % of the study
 U. Mandal et al.

Fig. 4 a Drainage density; b Rainfall; c; Slope d Surface water body e Lineament density and f NDVI
classification maps of the study area

area (384.52 km2) is under 0–2.10 % slope, and 36.20 % of the area (306.35 km2) is under
2.10–4.80 % slope. Almost plain land indicates there is high chance of occurrence of good
groundwater potential. Figure 4(c) shows that the slope map of the study area and details are
given in Table 1.

3.10 Surface Water Body

Surface water system represent the main recharge/discharge route of groundwater. It is the
pathway of convergence of groundwater from the shallow aquifers. However, identification of
the discrete surface water bodies in the study area is performed from google earth image. Large
water bodies such as rivers/streams are not considered in the surface water bodies. Figure 4(d)
shows spatially distributed discrete surface water bodies and it covered 12.01 km2 area.
Delineation of Groundwater Potential Zones of Coastal Groundwater Basin

3.11 Lineament Density

Lineament is the extensive linear surface feature on a planet, as a fault/fracture line, which
increases secondary porosity and permeability (Abdullah 2009; Bagyaraj et al. 2013).
Lineaments are the manifestation of linear features that can play a major role in
identifying suitable sites for groundwater recharge (Chowdhury et al. 2010; Avtar
et al. 2010). Lineament facilitates groundwater movement in the subsurface. Cartosat-1
DEM is used to identify the linear topographic features. Lineament density (Ld) is defined
as ratio between total lengths of all delineated lineament to total area (A) under consideration
(Avtar et al. 2010) and it express as:

X
n
Ld ¼ Li =A ð9Þ
i¼1

where, Li is lineament length, and i indicate the number of lineament. The lineament
density map of the study area is shown in Fig. 4(e) and classified into four classes
(Table 1). High lineament density is observed in the center, north-east, and south-east
parts of the study area with range from 0 to 3.70 km km−2. However, most of the
study area (82.36 %) covered low lineament density value (0–0.29 km km−2). A high
lineament-length density is an indicative of good groundwater prospect as presence of
lineament increases secondary porosity and permeability of the area and accordingly
assigned higher weight.

3.12 Normalized Difference Vegetation Index (NDVI)

Complex interactions of weather, soil moisture, soil, and crop types influence the crop
condition during its growth period. A NDVI takes into account the amount of infrared reflected
by plants (Eq. 10).

ðNIR−RÞ
NDVI ¼ ð10Þ
ðNIR þ RÞ

The analysis of NDVI is regarded as the rough estimation of vegetation amount

present and groundwater prospect over the space. The NDVI map is prepared from the
Landsat7 ETM+ images for the year 2000. Figure 4(f) shows the NDVI map of the
study area and it indicates the grassland and scattered agricultural land. NDVI values
ranges in the study area is −0.8 to 0.50 and it classified into four classes. NDVI
values ranges from 0.09 to 0.50 covered maximum area (55.25 %). Higher NDVI
values increase possibility of groundwater prospect, therefore, assigned more weight
(Table 1). Negative values of NDVI (values approaching −1) correspond to deep
water. Values close to zero (−0.1 to 0) generally correspond to barren areas with
rock, sand, or snow exposure. Low, positive values represent shrub and grassland
(approximately 0.2 to 0.4), while high values indicate temperate and tropical
rainforests (values approaching 1). The typical range is between about −0.1 (for a
not very green area) to 0.6 (for a very green area).
 U. Mandal et al.

3.13 Delineation of Groundwater Potential Zones

Detailed description of weight (Table S2), rank assigned for all thematic maps and
their corresponding sub-features are shown in Table 1. AHP based weighted sum
method is used for gnerating the GWPZ map. Figure 5 shows GWPZ map with four
type of zones namely: very good, good, moderately, and poor. The map shows
36.39 % area (273.53 km −2 ), 43.57 % area (327.47 km −2 ), 18.27 % area
(137.30 km−2), 1.77 % area (13.27 km−2) covered under very good, good, moderate,
and poor, respectively. GWPZ map shows that the very good and good potential
zones are coming together. It is also showing a patch of very good GWPZ area from
north to south-west and south-east direction (with low drainage density, high recharge
rate, and less slope value).

3.14 Validation

To validate the GWPZ map, yield data of the pumping wells for the entire area are
utilized. Figure 5 represents the GWPZ map along with pumping well locations.
Average well yield is of about 1958 m3 day−1. It is observed that almost all the
existing pumping wells for irrigation purpose are coming under good and very good
category of GWPZ area. GWPZ map also clearly shows the classified irrigation
potential of the study area for future planning. Table 2 shows well-wise validation.
24 Well ID’s agree with the GWPZ map (very good /good potential zone), 5 Well
ID’s disagree with GWPZ map (moderate/poor category area,at present defunct con-
dition) while 6 Well ID’s show partial agreement with the GWPZ map. Overall
accuracy level is about 85 %.

Fig. 5 Groundwater potential zone map with well location

Delineation of Groundwater Potential Zones of Coastal Groundwater Basin

Table 2 Accuracy assessment of GWPZ map with actual pumping well yield

Sl. No. Well ID Latitude Longitude Actual yield Well location on Validation
(m3/day) GWP maps remarks

1 103 21.5313 86.9546 1635.30 Moderate/poor Disagree

2 117 21.5251 86.9920 1471.77 Moderate Partially agree
3 118 21.5274 86.9847 1985.07 Moderate Partially agree
4 177 21.5435 86.9722 1817.00 Good Agree
5 96 21.5478 86.9641 1271.90 Good Agree
6 288 21.5389 87.1133 635.95 Very good/good Agree
7 224 21.6295 87.0267 1635.30 Moderate Partially agree
8 201 21.5789 87.0117 1817.00 Moderate Partially agree
9 278 21.6110 87.1007 1862.42 Very good Agree
10 38bas 21.6310 87.2562 2180.40 Very good Agree
11 132 21.6563 87.2674 1817.00 Good Agree
12 89 21.6550 87.2951 1635.30 Good Agree
13 44 21.6641 87.2890 2725.50 Very good/good Agree
14 5 21.6700 87.2952 1998.70 Very good Agree
15 138 21.6700 87.2311 2725.50 Moderate Defunct/Agree
16 64 21.6715 87.2393 1726.15 Moderate Partially agree
17 126 21.6747 87.2525 2089.55 Very good Agree
18 52 21.6665 87.2600 2543.80 Very good Agree
19 25 21.6997 87.1529 1817.00 Moderate Partially agree
20 185 21.6585 87.1570 2362.10 Good Agree
21 157 21.6473 87.1652 1817.00 Good/moderate Agree
22 73 21.6104 87.1327 1670.37 Good Agree
23 216 21.7021 87.2004 1453.60 Very good Agree
24 18 21.6372 87.0464 1499.02 Very good Defunct/Disagree
25 27 21.6571 87.2023 1894.22 Very good/good Agree
26 38 21.6361 87.1780 2089.55 Very good Agree
27 7 21.6670 87.1163 1817.00 Very good Agree
28 22 21.6764 87.1458 1907.85 Very good Agree
29 223 21.7453 87.1600 2066.83 Moderate/poor Disagree
30 77bas 21.7570 87.1547 1544.45 Moderate/good Agree
31 66 21.7030 87.0420 1371.83 Very good/good Agree
32 26 21.6239 87.1789 1907.85 Good Agree
33 18bas 21.7027 87.2299 2271.25 Very good Agree
34 130 21.6245 87.2856 5996.09 Good Defunct/Disagree
35 33 21.6363 87.2717 1181.05 Good Agree

Source: Lift Irrigation Corporation Ltd. (OLIC), Balasore, India

3.15 Sensitivity Analysis

Sensitivity analysis has been carried out using Eq. (7) by excluding each feature to identify the
sensitivity/importance of each feature relevant to GWPZ map for the present study. The effect
of each features is then evaluated. Table 3 shows, how the percentage area changes with the
 U. Mandal et al.

Table 3 Sensitivity analysis results

Zone Types (j)

1 2 3 4
Features (i) Poor (%) Moderate (%) Good (%) Very good (%)

1 LU/LC +6.36 +15.48 +14.38 −36.22

2 Soil +1.33 +5.76 +29.01 −36.10
3 Geomorphology +3.32 +15.30 +16.54 −35.17
4 Surface geology +0.81 +4.23 +31.27 −36.31
5 Hydrogeology +0.76 +1.68 +33.89 −36.34
6 Recharge rate +2.10 +14.81 +18.03 −34.95
7 Drainage density +2.40 +19.88 +14.09 −36.37
8 Rainfall +17.42 +14.08 −7.92 −23.57
9 Slope (%) +4.79 +16.18 +15.21 −36.17
10 Surface water body +8.50 +17.12 +10.71 −36.32
11 Lineament density +10.17 +7.19 −7.88 −9.49
12 NDVI +0.78 +3.42 +32.13 −36.33
SF All parameter 1.77 18.27 43.57 36.39

Bold values indicate significant results

exclusion of each feature. It is observed that exclusion of rainfall and lineament density
increases 17.42 % and 10.17 % area of poor category of GWPZ, respectively. Maximum area
increased in moderate type of GWPZ map for the exclusion of drainage density map. It is also
observed that the maximum area increased in good type of GWPZ map by 32.13 % and
31.27 % with the exclusion of drainage density and NDVI thematic map, respectively.
Exclusion of hydro-geology, NDVI, and surface geology increases good GWPZ area by
33.89 %, 32.13 % and 31.27 %, respectively. For the very good type GWPZ map, minimum
9.49 % area reduction is observed with the exclusion of only lineament density followed by the
rainfall (23.57 % area reduction). Sensitivity analysis can quantify the influence of each feature
and corresponding sub-features. Details regarding individual weights for sensitivity analysis
are provided in supplementary material (Table S3 to S14).

4 Conclusions

In this study, spatial science tools are utilized to identify the most plausible groundwater
potential zones for a coastal groundwater basin. Multi-criteria decision technique (AHP) has
been successfully applied to determine the normalized weight and ranking for each features
and sub-features. SWAT model is used to prepare thematic map for groundwater recharge rate
on sub-watershed basis. Total twelve feature layers are used for integration. The integrated
thematic map is classified into four category (very good, good, moderate, and poor). Area
coverages under different category i.e. very good, good, moderate, and poor, respectively, were
273.53 km2 (36.39 %), 327.47 km2 (43.57 %), 137.30 km2 (18.27 %), and 13.27 km2
(1.77 %). Validation is performed by field data along with the help of google earth image.
Yield performance of the study area and their location are correlated with GWPZ map and
show a good agreement. Sensitivity of each feature has been conducted. This groundwater
Delineation of Groundwater Potential Zones of Coastal Groundwater Basin

potential zone map is useful for future irrigation planning and for framing sustainable
groundwater development plans. Spatial redistribution of cropping patten can be suggested
based on the groundwater potential zone map. The major findings of the present study is the
possible location of the tubewells for groundwater extraction on a long-term sustainable basis
in the area that can be planned very efficiently. Moreover, GWPZ map act as primary
information for detailed field survey. This approach can be suitably applied to other coastal
basin with/ without modification.

Acknowledgments The authors are thankful to GSI Division, Baripada and Department of Water Resources,
Goverenment of Odisha; and the Central Ground Water Board (CGWB), Bhubeneswar and the Chairman,
CGWB, Faridabad for providing necessary input data.

References

Abdullah A (2009) A comparison of Landsat TM and SPOT data for lineament mapping in Hulu Lepar area,
Pahang, Malaysia. Eur J Sci Res 34(3):406–415
Adiat KAN, Nawawi MNM, Abdullah K (2012) Assessing the accuracy of GIS-based elementary multi criteria
decision analysis as a spatial prediction tool – a case of predicting potential zones of sustainable groundwater
resources. J Hydrol 440-441:75–89
Agarwal R, Garg PK (2016) Remote sensing and GIS based groundwater potential & recharge zones mapping
using multi-criteria decision making technique. Water Resour Manag 30:243–260
Agarwal E, Agarwal R, Garg RD, Garg PK (2013a) Delineation of groundwater potential zone: an AHP/ANP
approach. J Earth Syst Sci 122(3):887–898
Agarwal R, Garg PK, Garg RD (2013b) Remote sensing and GIS based approach for identification of artificial
recharge sites. Water Resour Manag 27:2671–2689
Al-Adamat RAN, Foster IDL, Baban SMJ (2003) Groundwater vulnerability and risk mapping for the basaltic
aquifer of the Azraq basin of Jordan using GIS, remote sensing and DRASTIC. Appl Geogr 23(4):303–324
Alonso J, Lamata T (2006) Consistency in the analytic hierarchy process: a new approach. Int J Uncertainty
Fuzziness Knowledge Based Syst 14:445–459
Avtar R, Singh CK, Shashtri S, Singh A, Mukherjee S (2010) Identification and analysis of groundwater potential
zones in Ken–Betwa river linking area using remote sensing and geographic information system. Geocarto
Int 25(5):379–396
Bagyaraj M, Ramkumar T, Venkatramanan S, Gurugnanam B (2013) Application of remote sensing and GIS
analysis for identifying groundwater potential zone in parts of Kodaikanal Taluk, South India. Front Earth
Sci 7(1):65–75
Bhuvan (2014): URL:http://bhuvan-noeda.nrsc.gov.in/download/download/download.php
Chenini I, Mammou AB, El May M (2010) Groundwater recharge zone mapping using GIS-based multi-criteria
analysis: a case study in Central Tunisia (Maknassy Basin). Water Resour Manag 24(5):921–939
Chowdhury A, Jha MK, Chowdary VM, Mal BC (2009) Integrated remote sensing and GIS-based approach for
assessing groundwater potential in West Medinipur district, West Bengal, India. Int J Remote Sens 30(1):
231–250
Chowdhury A, Jha MK, Chowdary VM (2010) Delineation of groundwater recharge zones and identification of
artificial recharge sites in West Medinipur district, West Bengal, using RS, GIS and MCDM techniques.
Environ Earth Sci 59:1209–1222
Deepika B, Avinash K, Jayappa KS (2013) Integration of hydrological factors and demarcation of groundwater
prospect zones: insights from remote sensing and GIS techniques. Environ Earth Sci 70(3):1319–1338
Dhar A, Sahoo S, Dey S, Sahoo M (2014) Evaluation of recharge and groundwater dynamics of a shallow
alluvial aquifer in central Ganga Basin, Kanpur (India). Nat Resour Res 23(4):409–422
Dhar A, Sahoo S, Mandal U, Dey S, Bishi N, Kar A (2015a) Hydro-environmental assessment of a regional
groundwater aquifer: Hirakud command area (India). Environ Earth Sci 73(8):4165–4178
Dhar A, Sahoo S, Sahoo M (2015b) Identification of groundwater potential zones considering water quality
aspect. Environ Earth Sci 74(7):5663–5675
Gumma MK, Pavelic P (2013) Mapping of groundwater potential zones across Ghana using remote sensing,
geographic information systems, and spatial modeling. Environ Monit Assess 185(4):3561–3579
http://power.larc.nasa.gov/cgi-bin/cgiwrap/solar/[email protected]
 U. Mandal et al.

Jha MK, Chowdary VM, Chowdhury A (2010) Groundwater assessment in Salboni block, West Bengal (India)
using remote sensing, geographical information system and multi-criteria decision analysis techniques.
Hydrogeol J 18(7):1713–1728
Machiwal D, Jha MK, Mal BC (2010) Assessment of groundwater potential in a semi-arid region of India using
remote sensing, GIS and MCDM techniques. Water Resour Manag 25(5):1359–1386
Madrucci V, Taioli F, de Araújo CC (2008) Groundwater favorability map using GIS multicriteria data analysis
on crystalline terrain, São Paulo State, Brazil. J Hydrol 357(3–4):153–173
Magesh NS, Chandrasekar N, Soundranayagam JP (2012) Delineation of groundwater potential zones in Theni
district, Tamil Nadu, using remote sensing, GIS and MIF techniques. Geosci Front 3(2):189–196
Mahmoud SH, Alazba AA, Amin MT (2014) Identification of potential sites for groundwater recharge using a
GIS-based decision support system in Jazan region-Saudi Arabia. Water Resour Manag 28(10):3319–3340
Moreaux M, Reynaud A (2006) Urban freshwater needs and spatial cost externalities for coastal aquifers: a
theoretical approach. Reg Sci Urban Econ 36(2):163–186
Mukherjee P, Singh CK, Mukherjee S (2012) Delineation of groundwater potential zones in arid region of
India—a remote sensing and GIS approach. Water Resour Manag 26(9):2643–2672
Nag SK, Ghosh P (2012) Delineation of groundwater potential zone in Chhatna block, Bankura District, West
Bengal, India using remote sensing and GIS techniques. Environ Earth Sci 70(5):2115–2127
Nagarajan M, Singh S (2009) Assessment of groundwater potential zones using GIS technique. J Indian Soc
Remote Sens 37:69–77
Oh H-J, Kim Y-S, Choi J-K, Park E, Lee S (2011) GIS mapping of regional probabilistic groundwater potential in
the area of Pohang City, Korea. J Hydrol 399(3–4):158–172
Prasad RK, Mondal NC, Banerjee P, Nandakumar MV, Singh VS (2007) Deciphering potential groundwater
zone in hard rock through the application of GIS. Environ Geol 55(3):467–475
Ravi Shankar MN, Mohan G (2005) A GIS based hydrogeomorphic approach for identification of site-specific
artificial-recharge techniques in the Deccan Volcanic Province. J Earth Syst Sci 114(5):505–514
Rejani R, Jha MK, Panda SN, Mull R (2003) Hydrologic and hydrogeologic analyses in a coastal groundwater
Basin, Orissa, India. Appl Eng Agric 19(2):177–186
Saaty TL (1980) The analytic hierarchy process: planning, priority setting, resource allocation. McGraw-Hill,
New York
Sahoo, S., Dhar, A., Kar, A., & Ram, P. (2016). Grey analytic hierarchy process applied to effectiveness
evaluation for groundwater potential zone delineation. Geocarto Int, 1–18. doi:10.1080/10106049.2016.
1195888
Senanayake IP, Dissanayake DMDOK, Mayadunna BB, Weerasekera WL (2016) An approach to delineate
groundwater recharge potential sites in Ambalantota, Sri Lanka using GIS techniques. Geosci Front 7(1):
115–124
Srivastava PK, Bhattacharya AK (2006) Groundwater assessment through an integrated approach using remote
sensing, GIS and resistivity techniques: a case study from a hard rock terrain. Int J Remote Sens 27(20):
4599–4620
UNCED (1992) BAgenda 21: Chapter 17.3^, URL: http://www.un.org/depts/los/consultative_process/
documents/A21-Ch17.htm [June, 2015].
Zaidi FK, Nazzal Y, Ahmed I, Naeem M, Jafri MK (2015) Identification of potential artificial groundwater
recharge zones in Northwestern Saudi Arabia using GIS and Boolean logic. J Afr Earth Sci 111:156–169