Assessing Recharge Zones for Groundwater Potential in Dera Ismail Khan (Pakistan): A GIS-Based Analytical Hierarchy Process Approach

Abstract

Groundwater constitutes the primary source of liquid freshwater on Earth and is essential for ecosystems, agriculture, and human consumption. However, rising demand, urbanization, and climate change have intensified groundwater depletion, particularly in semi-arid regions. Therefore, assessing groundwater recharge zones is essential for sustainable water resource management in vulnerable areas such as Dera Ismail Khan, Pakistan. This study aims to delineate groundwater potential zones (GWPZs), using an integrated approach combining the Geographic Information System (GIS), remote sensing (RS), and the analytical hierarchy process (AHP). Twelve factors were identified in a study conducted using GIS-based AHP to determine the groundwater recharge zones in the region. These include land use/land cover (LULC), rainfall, drainage density, soil type, slope, road density, water table depth, and remote sensing indices such as Normalized Difference Vegetation Index (NDVI), Normalized Difference Built-up Index (NDBI), Moisture Stress Index (MSI), Worldview Water Index (WVWI), and Land Surface Temperature (LST). The results show that 17.52% and 2.03% of the area have “good” and “very good” potential for groundwater recharge, respectively, while 48.63% of the area has “moderate” potential. Furthermore, gentle slopes (0–2.471°), high drainage density, shallow water depths (20–94 m), and densely vegetated areas (with a high NDVI) are considered important influencing factors for groundwater recharge. Conversely, areas with steep slopes, high temperatures, and dense built-up areas showed “poor” potential for recharge. This approach demonstrates the effectiveness of integrating advanced remote sensing indices with the AHP model in a semi-arid context, validated through high-accuracy field data (Kappa = 0.93). This methodology offers a cost-effective decision support tool for sustainable groundwater planning in similar environments.

1. Introduction

Groundwater is an essential natural resource, especially in regions where surface water is limited. Although water covers 71% of the Earth’s surface, only 2.5% of it is freshwater suitable for human consumption [1,2]. Aquifers provide around 98.7% of global freshwater, and about a quarter of the world population relies on groundwater for their requirements [3,4]. From 1950 to 2000, global water withdrawals increased by more than three times, rising from 1382 km³ per annum to 3973 km³ per annum. Predictions indicate that this figure could reach 5235 km³ per annum by 2030 [5,6]. In the last decade, over 2 billion people have been residing in water-stressed areas because of over-extraction rates and climate change [7]. Demand for groundwater as a source of freshwater is rising progressively, particularly for domestic and commercial use [8,9,10,11].

Previous studies have shown that groundwater accounts for around one-sixth of freshwater globally [12,13,14], forming a significant source of water not only for humans but also for aquatic ecosystems [15,16]. Furthermore, intensive farming practices and exponential population growth rates have increased demand for groundwater, resulting in over-extraction [17,18,19]. Groundwater is critical for supporting urbanization and sustaining population growth [14,20]. Nonetheless, there is a significant lack of information and research-based findings regarding the likelihood of ensuring an adequate supply of water [21,22]. This shows the importance of groundwater modeling as a method for addressing current water resource challenges [9,20,23]. Groundwater levels have been declining globally in recent years due to climate change and excessive groundwater use for a variety of purposes [24,25]. Additionally, climate change has led to higher evapotranspiration and less rainfall, further undermining the ability of some places to restore their water tables [7,26,27].

About 60–70 percent of the people in Pakistan use groundwater either directly or indirectly for their needs [28,29]. Consequently, the nation’s economic growth, public health, and environmental stability are directly related to better access to fresh water and sustainable groundwater resource management [16,30]. However, the water management system of Pakistan has yet to reach its potential despite its artificial groundwater recharge systems such as the Indus Basin Irrigation System [31]

The excessive use, contamination, and inadequacy of abstraction techniques have a significant impact on the survival of this essential resource [32,33,34]. A balance between efficient use and sustainability is necessary to continue being an adequate source of drinking water and agricultural support in groundwater [17]. However, Pakistan faces numerous hurdles in achieving this, including a weak groundwater policy and a lack of information about groundwater processes [35]. These issues must be resolved to preserve this valuable resource for future generations.

Excessive pumping and the effects of climate change have depleted groundwater supplies, a critical problem particularly in Dera Ismail Khan, one of the districts of Khyber Pakhtunkhwa (KP) province, Pakistan [29,36]. The semi-arid climate of this area has left it mostly dependent on groundwater for crop irrigation and other purposes [22]. The amount of groundwater available continues to be pressured by increased demand for available water supplies created by population density, as well as urban and agricultural expansion. In such areas, groundwater quality plays an important role in meeting the needs of the people for population as well as industrial and agricultural needs and for drinking water purposes [16,30,36]. The growing demand for water in agriculture, domestic use, and industrial activities poses a challenge that must be addressed; we chose to map the groundwater potential zone in the study area as a main step in sustainable water resource management.

The parameters used for groundwater potential zone mapping in this study include rainfall, elevation, drainage density, soil, slope, road density, LULC, NDVI, NDBI, MSI, WVWI, and LST. These parameters have been commonly used in many research studies due to their important roles in defining the factors affecting the groundwater recharge and potential zone mapping [37,38,39]. This study integrated the AHP to draw weightages for parameters directly into the GIS environment to develop the evaluation of the potential zones. This approach seeks to offer a thorough comprehension of groundwater resources so that well-informed decisions may be made for sustainable water management [4,14,40]. Furthermore, this research holds significant value in enhancing water supply and ensuring the efficient use of sustainable water resources.

In fact, conventional methods of groundwater assessment such as field surveys, manual well sampling, and hydrogeological mapping are often constrained by high implementation costs, sampling errors, labor intensity, and limited spatial coverage, which can lead to oversimplified interpretations of groundwater dynamics [41,42]. As this leads to limitations, researchers have sought innovative ways to explore these methods to improve efficiency and spatial accuracy in analyzing groundwater resources. RS and GIS have proven to be powerful tools for delineating groundwater potential zones over large areas with greater precision and temporal consistency [28,43,44]. Building upon these capabilities, recent studies have integrated RS and GIS with multi-criteria decision-making and machine learning techniques such as the AHP [37,45], Weighted Overlay Method [38], Frequency Ratio [19,46], Principal Component Analysis (PCA) [15], Decision Tree Models [1], and ensemble classifiers like Random Forest, Quick Unbiased Efficient Statistical Tree, and Support Vector Machines [47,48,49] to further enhance accuracy and predictive robustness. These integrative methods not only overcome the inherent drawbacks of traditional techniques but also provide the ability to incorporate diverse hydrogeological, environmental, and socio-economic variables, thereby offering more comprehensive solutions for groundwater resource assessment and sustainable management.

To address the increasing problems with groundwater usage in semi-arid areas such as Dera Ismail Khan, this work presents a novel way to integrate the AHP with GIS and remote sensing methods to identify possible groundwater zones. In contrast to traditional techniques, our method uses a hierarchical, expert-validated framework that incorporates twelve important parameters covering climatic, topographic, hydrological, land cover, and human aspects. The originality of this study is found in its robust multi-criteria assessment framework, which is improved by spatial modeling and ratio consistency checks. This framework increases the precision of defining both surface and subsurface recharge potential zones. This integrated methodology offers a scalable, data-driven approach to sustainable groundwater management, with practical benefits for policymakers, water resource managers, and planners in water-stressed areas.

2. Research Methodology

2.1. Study Area

Dera Ismail Khan is in the southern part of the KP province of Pakistan. It lies between 31°15′ N and 32°32′ N and 70°11′ E and 71°20′ E (Figure 1). The total area of Dera Ismail Khan is 7325 km². Recent census records have shown that the district has a population of nearly 2 million [50]. Dera Ismail Khan experiences a semi-arid climate characterized by low and erratic rainfall, high evapotranspiration, and sharp seasonal temperature fluctuations. The region’s average summer temperatures peak around 44–46 °C in June, while winter temperatures can drop to 4–6 °C in January [51]. The mean annual rainfall varies between 150 and 250 mm, mostly received during the monsoon season (July–September), with minor winter precipitation [51]. Such climatic conditions significantly impact groundwater recharge potential by reducing infiltration rates and increasing evaporation losses, especially in uncovered or barren areas [22,29,52].

2.2. Data and Their Sources

Various data sources were used to achieve the objectives of the study. Annual rainfall data for the study area were obtained from the regional meteorological department in KP province, Pakistan. The SRTM Digital Elevation Model (DEM) with a 30 m resolution, obtained from the United States Geological Survey (USGS) database (

Table 1. Used data sources.

Dataset	Description	Spatial Resolution	Temporal Resolution	Source
DEM	Used for Slope and Drainage Density data	30 m	Multi-day	USGS Earth Explorer,
Landsat	Used for making maps of land cover, MSI, WVWI, NDVI, LST, and NDBI	30 m	8 days	Copernicus Data Space Ecosystem (Landsat 9)
Road Network	Used to calculate distance from roads	Variable	Single instance	Local Transportation Departments
Rainfall	Used to create rainfall maps	1000 m	Annually	Regional Meteorological Department
Soil	Essential for making maps of soil types	Scale 1:2,000,000	-	Soil Survey of Pakistan
Water Depth	Utilized for making the water depth map	-	-	Pakistan Council of Research in Water Resources

), was used to calculate slope and drainage density. A soil type toposheet from the Soil Survey of Pakistan was used for delineating the study area. For land use and land cover classification, a Landsat 9 image, captured on 22 July with 30 m resolution, was obtained from the USGS website (Table 1). This image was also utilized to generate maps of MSI, WVWI, NDVI, LST, and NDBI. Road network data were obtained from the local transportation department, while the water depth data were acquired from the Pakistan Council of Research in Water Resources (PCRWR), Islamabad.

2.3. Preparation of Thematic Maps

Thematic maps representing variables influencing groundwater recharge were generated to visualize their spatial distribution within a GIS environment. As shown in

Table 2. Classification of groundwater recharge parameters with percentage coverage, ratings, rankings, and weights.

Parameter	Class	Percentage Coverage (%)	Class Ratings	Class Rankings	Parameter Weight (%)
Water depth	20–94	7.08	Very High	5	10
	94–144	20.29	High	4
	144–174	28.19	Moderate	3
	174–203	33.37	Low	2
	203–318	11.07	Very Low	1
NDVI	−0.195–0.18	3.32	Very Low	1	07
	0.019–0.09	56.47	Low	2
	0.091–0.164	22.07	Moderate	3
	0.165–0.26	10.14	High	4
	0.261–0.505	8.00	Very High	5
MSI	0.533–0.783	11.43	Very High	5	07
	0.784–0.910	10.57	High	4
	0.911–1.025	12.78	Moderate	3
	1.026–1.104	27.10	Low	2
	1.105–1.653	38.12	Very Low	1
NDBI	−0.304–(−0.141)	8.29	Very High	5	10
	−0.14–(−0.072)	10.34	High	4
	−0.071–(−0.007)	10.69	Moderate	3
	−0.006–(−0.041)	22.47	Low	2
	0.042–(−0.246)	48.21	Very Low	1
Slope	0–2.471	90.14	Very High	5	07
	2.472–7.413	4.70	High	4
	7.414–13.796	2.89	Moderate	3
	13.797–22.032	1.69	Low	2
	22.033–52.507	0.58	Very Low	1
Rainfall	356.918–387.328	17.56	Very Low	1	05
	387.329–410.548	19.09	Low	2
	410.549–440.959	22.06	Moderate	3
	440.960–480.79	16.63	High	4
	480.791–532.958	16.66	Very High	5
WVWI	−0.431–(−0.166)	48.48	Very Low	1	07
	−0.165–(−0.117)	37.71	Low	2
	−0116–(−0.04)	7.48	Moderate	3
	−0.039–0.088	3.83	High	4
	−0.089–0.267	2.50	Very High	5
Land cover	Vegetation	14.88		4	13
	Barren	59.47	Very Low	1
	Buildup	14.49	Low	2
	Water Bodies	11.16	Very High	5
LST	20–24	02	Very High	5	07
	25–28	19.09	High	4
	29–32	21.13	Moderate	3
	33–36	29.76	Low	2
	37–39	28.02	Very Low	1
Drainage density	0.001–10.416	59.25	Very High	5	13
	10.417–20.833	26.11	High	4
	20.834–31.249	9.88	Moderate	3
	31.25–41.665	4.3	Low	2
	41.666–52.082	0.92	Very Low	1
Road distance	0–0.5	53.14	Very Low	1	10
	0.5–1	7.23	Low	2
	1–2	13.96	Moderate	3
	2–3	7.23	High	4
	3–5	2.23	Very High	5
Soil type	Haplic	80.43	Very High	5	05
	Calcaric	17.61	Moderate	3
	Lithosol	0.92	Very Low	1

, which includes percentage coverage, class ratings, ranks, and weights, each parameter was categorized based on its potential contribution to recharge. The workflow of the methodology, including data sources and processing steps, is illustrated in Figure 2.

2.3.1. LULC

Land cover plays a critical role in determining groundwater potential by influencing hydrological processes such as water infiltration, retention, and recharge. Vegetation cover, particularly forests, enhances groundwater recharge by intercepting rainfall, reducing surface runoff, and conserving soil moisture [4,11]. In contrast, urbanized areas with impervious surfaces increase runoff, limit infiltration, and reduce recharge potential [6]. Integrating LULC data refines the delineation of recharge zones based on favorable groundwater conditions. The LULC of this study was prepared using a supervised maximum likelihood classification algorithm in an ArcGIS Pro environment.

2.3.2. Rainfall

Rainfall in Pakistan, particularly in semi-arid regions, significantly influences groundwater recharge by controlling deep percolation rates and aquifer replenishment. Precipitation patterns, especially monsoon rains, play a crucial role in replenishing aquifers, as intense rainfall can exceed soil infiltration capacity, leading to enhanced vertical percolation in permeable areas [25,53]. Poor rainfall and low recharge rates typically occur in areas with low infiltration, while permeable regions like floodplain enhance recharge. Incorporating rainfall into the AHP model improves the assessment of recharge potential by accounting for spatial and temporal distribution as well as its impact on vegetation and land use.

The data of annual rainfall in 2022 were obtained from the Regional Meteorological Center of KP, Pakistan. The meteorological observatory data was obtained in points format and interpolated using the Inverse Distance Weighting (IDW) spatial method to obtain the spatial distribution of rainfall in the study area.

2.3.3. Drainage Density

Higher surface water drainage may indicate higher drainage density because it indicates better surface water drainage, resulting in increased infiltration and better groundwater recharge [22]. An area with low drainage density, however, may allow for surface water retention for an extended time, promoting infiltration within the favorable geological conditions (high permeability and porosity) [27]. However, areas with impermeable soil, despite low drainage density, still exhibit low recharge potential. The interaction of these multiple factors that influence groundwater recharge is highlighted. Using ArcGIS Pro, drainage density was then calculated using the total length of drainage lines derived from a high-resolution DEM via Flow Direction and Flow Accumulation tools, as well as a set threshold for delineating streams; then, the length of the total drainage was measured.

2.3.4. Water Depth

The interaction between surface water and groundwater is influenced by distance to the water table, and this is one of the key factors in determining aquifer recharge capacity. Shallow water bodies have higher infiltration rates, therefore providing an opportunity for groundwater to recharge, while deep water bodies have less potential for recharge since the distance between the surface water and aquifers is much greater. High recharge rates are present in saturated areas, such as wells or rivers, which are also potential replenishment zones [2,42]. In addition, the identification of recharge potential is furthermore improved when water depth data are combined with other datasets, even in regions with topography change and groundwater flow [43]. Also, such mapping aids in making effective use of water resources since it enables estimation of rechargeable quantities based on water depth [2].

Water depth data, collected from PCRWR, were processed in ArcGIS Pro. The point data, including geographic coordinates and water table depth values, were interpolated using the IDW spatial analysis tool to generate a continuous water depth surface, which accurately estimated a comprehensive water depth map.

2.3.5. Soil

Groundwater recharge is strongly influenced by soil characteristics, in particular infiltration and drainage properties and water retention. Sandy and loamy are the two major types of soil, and both are suitable for recharging groundwater. Contrary to this, infiltration in clayey or compacted soils will be poor, since these soils are not very permeable [1]. Therefore, the AHP incorporated soil type as a parameter, as groundwater dynamics are highly dependent on soil type. Such a mapping results in identifying possible groundwater recharge areas based on the interaction of soil characteristics with physical factors [54]. The soil data of locations and soils were processed in ArcGIS Pro, and interpolation of data by the IDW tool was used to create continuous surface maps. The natural neighbor tool was used to make soil zones and to display the spatial soil variety in the study area.

2.3.6. Slope

Slope significantly impacts groundwater recharge by controlling runoff and infiltration. Gentle slopes promote water infiltration, whereas steep slopes increase runoff, limiting recharge [55]. In the AHP for groundwater potential zoning, slope data help to identify areas that are conducive to recharge and those prone to high runoff [43].

The Slope tool was used to process high-resolution DEM in ArcGIS Pro to calculate the elevation changes; this allowed the Reclassify tool to be applied to categorize slopes for evaluating recharge potential and for other analyses.

2.3.7. Road Density

Road density, defined as the total length of roads per unit area, influences surface runoff and infiltration characteristics. In urban areas with high road density and extensive impervious surfaces like concrete, surface runoff increases while groundwater recharge decreases. In contrast, areas with low road density, often featuring permeable surfaces, facilitate enhanced recharge potential. As a result, road density is utilized as a parameter in the AHP for groundwater potential zoning, reflecting its impact on infiltration. Mapping road density helps identify areas with limited recharge due to impermeability, enabling a more accurate assessment of recharge potential in AHP models [56].

The Euclidean Distance tool was used in ArcGIS Pro to examine the road network layer in order to obtain a raster map of distance from roads. This subsequently was reclassified into distance zones for analysis and visualization.

2.3.8. Normalized Difference Vegetation Index (NDVI)

The vegetation index is used to quantify and monitor the greenness and density of vegetation. It is calculated using the red band and near-infrared (NIR) band, as shown in Equation (1):

$$\mathrm{DVI}={\displaystyle \frac{\mathrm{N}\mathrm{I}\mathrm{R}-\mathrm{R}\mathrm{E}\mathrm{D}}{\mathrm{N}\mathrm{I}\mathrm{R}+\mathrm{R}\mathrm{E}\mathrm{D}}}$$

(1)

This is the difference in the red band (which absorbs vegetation) versus the NIR band (reflected vegetation). NDVI values depend on how healthy and green the vegetation is, with values closer to 1 being healthier. It is an important parameter in controlling surface runoff and infiltration and thus groundwater recharge [57]. Areas that have high NDVI values are those that contain an abundance of vegetation and good water retention, which facilitate infiltration and replenishment of groundwater [58]. On the contrary, values of NDVI lower than 0.33 (below 0.33) indicate the sparseness of vegetation or arid areas and consequently high runoff and low recharge [59].

2.3.9. Normalized Difference Built-Up Index (NDBI)

The NDBI index quantifies built-up surfaces by analyzing reflectance from the short-wave infrared (SWIR) and NIR bands, as shown in Equation (2):

$$\mathrm{NDBI}={\displaystyle \frac{\mathrm{S}\mathrm{W}\mathrm{I}\mathrm{R}-\mathrm{N}\mathrm{I}\mathrm{R}}{\mathrm{S}\mathrm{W}\mathrm{I}\mathrm{R}+\mathrm{N}\mathrm{I}\mathrm{R}}}$$

(2)

Urbanized, impervious surfaces (concrete, asphalt, etc.) have high NDBI values as they restrict water infiltration and increase the potential for surface runoff and a loss of water for groundwater recharge potential. Higher NDBI values indicate reduced recharge capacity, while lower values are more common in rural areas, giving a higher infiltration [24]. In terms of AHP models for groundwater potential zoning, NDBI helps to establish the hydrological impacts of urbanization to assist sustainable water management [38].

2.3.10. Moisture Stress Index (MSI)

MSI index is used to quantify vegetation stress due to moisture demands, calculated using the SWIR and NIR reflectance values, as shown in Equation (3):

$$\mathrm{MSI}={\displaystyle \frac{\mathrm{S}\mathrm{W}\mathrm{I}\mathrm{R}}{\mathrm{N}\mathrm{I}\mathrm{R}}}$$

(3)

MSI can therefore serve as a proxy for the vegetation’s water content and its potential to hold and transmit water; therefore, it is important in groundwater potential zoning. The higher MSI values indicate vegetation stress, less water retention, and reduced groundwater recharge. On the other hand, low MSI values indicate healthy vegetation and good water-holding capacity, leading to recharge [41]. By adding the vegetation health and moisture variabilities into the AHP model, this study enhances the accuracy of groundwater recharge mapping by incorporating MSI [11].

2.3.11. World View Water Index (WVWI)

The WVWI index, derived from World View satellite imagery, is used to assess surface water extent and vegetation water relationships. It is calculated using green band and short-wave infrared (SWIR) reflectance values, as shown in Equation (4):

$$\mathrm{WVWI}={\displaystyle \frac{\mathrm{G}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{n}-\mathrm{S}\mathrm{W}\mathrm{I}\mathrm{R}}{\mathrm{G}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{n}+\mathrm{S}\mathrm{W}\mathrm{I}\mathrm{R}}}$$

(4)

This formula accentuates the difference between water and non-water surfaces using reflectance WVWI values of green and SWIR bands, and when water bodies are present, the WVWI values are increased. Accurate water feature mapping can be carried out using this method, and this helps with water resource management.

For groundwater potential zoning, the combined factors that affect groundwater recharge are useful for WVWI. An area with high WVWI values indicates a region with lots of surface water, wetlands, or vegetation [19]. Low WVWI values typically represent arid or impervious areas. This study incorporates WVWI into the AHP model to facilitate high-resolution mapping of hydrological processes by improving identification of recharge suitable zones, which is instrumental for sustainable groundwater resource management.

2.3.12. Land Surface Temperature (LST)

LST is one of the key parameters in groundwater potential zoning. LST directly affects evapotranspiration, soil moisture, and vegetation cover. Therefore, LST indicates the thermal characteristics of the surface, i.e., its ability to absorb and emit heat. Groundwater zoning is only possible through LST, whereby the surface thermal properties are linked to hydrological processes. If verified, high LST values will indicate areas with high evaporation and low recharge potential, and cooler areas are suitable for conservation and recharge. LST is incorporated into AHP groundwater potential assessments to improve accuracy [39,60].

2.4. Analytical Hierarchy Process (AHP)

The AHP was developed by American theorist Thomas L. Saaty [61]. AHP is a systematic and widely used method for prioritizing factors in decision-making problems involving multiple criteria [62]. In this study, we applied the AHP method to assess the groundwater potential zoning of the study area using normalized weights (Figure 3). Following this approach, we selected the recharge potential-influencing factors, assigned relative scores (

Table 3. Relative scale values.

Score	Importance Intensity	Definition
1	Equal Importance	Both elements contribute equally.
3	Moderate Importance	One element is moderately preferred over the other.
5	Strong Importance	One element has strong importance over the other.
7	Very Strong Importance	One element is very strongly preferred over the other.
9	Extreme Importance	One element is extremely preferred over the other.
2, 4, 6, 8	Intermediate Values	Values when importance lies between two intensities.

), conducted a pairwise comparison matrix (

Table 4. Pairwise comparison matrix.

Factor	LULC	Rainfall	D. Density	W. Depth	Soil	Slope	R. Density	NDVI	NDBI	MSI	WVWI	LST
LULC	1	2.0	1.5	1.5	3.0	2.0	1.5	2.0	1.5	1.5	1.5	1.5
Rainfall	0.5	1	0.50	0.50	1.5	1.0	0.5	1.0	0.5	0.5	0.5	0.5
D. Density	0.67	2	1	1.0	2	1.5	1	1.5	1	1.5	1.5	1.5
W. Depth	0.67	2	1	1	2	1.5	1	1.5	1	1.5	1.5	1.5
Soil	0.33	0.67	0.5	0.5	1	0.5	0.5	0.67	0.5	0.67	0.67	0.67
Slope	0.5	1	0.67	0.67	2	1	1	1	0.67	0.67	0.67	0.67
R. Density	0.67	2	1	1	2	1	1	1.5	1	1.5	1.5	1.5
NDVI	0.5	1	0.67	0.67	1.5	1	0.67	1	1	1	1	1
NDBI	0.67	2	1	1	2	1.5	1	1	1	1.5	1.5	1.5
MSI	0.67	2	0.67	0.67	1.5	1	0.67	1	0.67	1	1	1
WVWI	0.67	2	0.67	0.67	1.5	1	0.67	1	0.67	1	1	1
LST	0.67	2	0.67	0.67	1.5	1	0.67	1	0.67	1	1	1

), and computed a normalized vector (

Table 5. Normalized vector for key flood parameters.

Factor	LULC	Rainfall	D. Density	W. Depth	Soil	Slope	R. Density	NDVI	NDBI	MSI	WVWI	LST	AVRG	%
LULC	0.13	0.10	0.15	0.15	0.14	0.14	0.15	0.14	0.15	0.11	0.11	0.11	0.13	13
Rainfall	0.07	0.05	0.05	0.05	0.07	0.07	0.05	0.07	0.05	0.04	0.04	0.04	0.05	5
D. Density	0.09	0.10	0.10	0.10	0.09	0.11	0.10	0.11	0.10	0.11	0.11	0.11	0.10	10
W. Depth	0.09	0.10	0.10	0.10	0.09	0.11	0.10	0.11	0.10	0.11	0.11	0.11	0.10	10
Soil	0.04	0.03	0.05	0.05	0.05	0.04	0.05	0.05	0.05	0.05	0.05	0.05	0.05	5
Slope	0.07	0.05	0.07	0.07	0.09	0.07	0.10	0.07	0.07	0.05	0.05	0.05	0.07	7
R. Density	0.09	0.10	0.10	0.10	0.09	0.07	0.10	0.11	0.10	0.11	0.11	0.11	0.10	10
NDVI	0.07	0.05	0.07	0.07	0.07	0.07	0.07	0.07	0.10	0.07	0.07	0.07	0.07	7
NDBI	0.09	0.10	0.10	0.10	0.09	0.11	0.10	0.07	0.10	0.11	0.11	0.11	0.10	10
MSI	0.09	0.10	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	7
WVWI	0.09	0.10	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	7
LST	0.09	0.10	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	7
SUM	1	1	1	1	1	1	1	1	1	1	1	1	1

) to evaluate the relative importance of the twelve parameters concerning the model’s objectives.

2.4.1. Pairwise Comparison Matrix

All factors selected for groundwater potential mapping were analyzed using the weighted overlay analysis tool within a GIS environment. The AHP was employed to identify the key water recharging-influencing factors, assign relative scores, conduct pairwise comparisons, and assess matrix consistency. In the pairwise comparison process, an expert compares each factor (e.g., land cover) with the others, one at a time, using Saaty’s 1–9 scale to evaluate their relative importance in identifying water recharge. An Excel sheet automatically generates a 12 × 12 pairwise comparison matrix, which is used to compute the normalized principal eigenvector for the AHP-based flood-prone area mapping (Table 4).

2.4.2. Assessing Matrix Consistency

According to [63], the matrix consistency was assessed by calculating the consistency ratio (CR), which compares the degree of consistency in the pairwise comparison matrix. The CR helps determine if the judgments made in the matrix are consistent enough, based on the eigenvalue (λ_max). If the CR exceeds a certain threshold (typically 0.1 or 10%), it indicates inconsistency, and the matrix may need to be revised. The specific formula used to calculate the CR is provided in Equation (6), which typically involves the ratio of the matrix (λ_max) to the consistency index (CI) (Equation (5) and compares it to a random index (RI) for the corresponding number of factors.

$$CI={\mathsf{\lambda }}_{\mathrm{m}\mathrm{a}\mathrm{x}}-{\displaystyle \frac{\mathrm{n}}{\mathrm{n}-1}}$$

(5)

$$CI=12.03-{\displaystyle \frac{12}{12-1}}$$

$$CI=0.01$$

where CI is the consistency index, and RI is the random index. The values for RI and CI are provided in

Table 6. Priority vector matrix consistency index (C1).

Factor	LULC	Rainfall	Drainage Density	Water Depth	Soil	Slope	Road Density	NDVI	NDBI	MSI	WVWI	LST	SUM	Consistancy Vectore
LULC	0.13	0.11	0.15	0.15	0.14	0.13	0.15	0.14	0.15	0.11	0.11	0.11	1.60	12.03
Rainfall	0.07	0.05	0.05	0.05	0.07	0.07	0.05	0.07	0.05	0.04	0.04	0.04	0.64	12.01
Drainage Density	0.09	0.11	0.10	0.10	0.09	0.10	0.10	0.11	0.10	0.11	0.11	0.11	1.24	12.03
Water Depth	0.09	0.11	0.10	0.10	0.09	0.10	0.10	0.11	0.10	0.11	0.11	0.11	1.24	12.03
Soil	0.04	0.04	0.05	0.05	0.05	0.03	0.05	0.05	0.05	0.05	0.05	0.05	0.56	12.04
Slope	0.07	0.05	0.07	0.07	0.09	0.07	0.10	0.07	0.07	0.05	0.05	0.05	0.81	12.03
Road Density	0.09	0.11	0.10	0.10	0.09	0.07	0.10	0.11	0.10	0.11	0.11	0.11	1.20	12.05
NDVI	0.07	0.05	0.07	0.07	0.07	0.07	0.07	0.07	0.10	0.07	0.07	0.07	0.86	12.02
NDBI	0.09	0.11	0.10	0.10	0.09	0.10	0.10	0.07	0.10	0.11	0.11	0.11	1.20	12.03
MSI	0.09	0.11	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.90	12.05
WVWI	0.09	0.11	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.90	12.05
LST	0.09	0.11	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.07	0.90	12.05
Average														12.03

and

Table 7. Random index (RI).

n	1	2	3	4	5	6	7	8	9	10	11	12
RI	0	0	0.58	0.9	1.12	1.24	1.32	1.41	1.45	1.49	1.51	1.54

, respectively. CI is calculated using Equation (5).

Here, λ_max is the highest eigenvalue of the matrix, and n is the number of factors or criteria being evaluated.

$$CR={\displaystyle \frac{CI}{RI}}$$

(6)

$$CR={\displaystyle \frac{0.01}{1.54}}$$

$$CR=0.006$$

2.4.3. Validation of Groundwater Potential Zones

To ensure the reliability of the groundwater potential zoning, an accurate assessment was carried out using both field observation data and water table depth data. Hydrogeological studies frequently use the validation method outlined for groundwater potential zoning, which compares predicted zones with field data and uses statistical measures, including overall accuracy and the Kappa coefficient. For example, Arulbalaji et al. [64] validated groundwater potential maps in India’s Vamana Puram River Basin, achieving an overall accuracy of 85%. Similarly, a recent study conducted in Jinan, China, combined receiver operating characteristic (ROC) analysis with well-observed groundwater levels, showing strong performance in heterogeneous karst environments with an AUC of 0.736 and 74% classification accuracy [65]. In Nigeria, Ajayi et al. [66] reported a 68% agreement between hydro-geophysical vertical electrical sounding (VES) data and GIS-derived groundwater potential zones, emphasizing the value of multi-method validation in hard-rock terrains. In this study, actual field data collected by PCRWR were compared with estimated groundwater potential zones as part of the validation process. A total of 83 water table depth measurements were obtained across the study area, encompassing regions of varying topography and hydrogeological conditions. The recorded water depths ranged from 20 m to over 320 m, reflecting the significant variability in groundwater availability within the region. Based on these measurements, the observation was categorized into the following five potential classes:

• Very Good Potential: 0–60 m
• Good Potential: 61–120 m
• Moderate Potential: 121–180 m
• Low Potential: 181–240 m
• Very Low Potential: greater than 240 m

The validation methodology involved overlaying the observed data on the predicted groundwater potential map and assessing the alignment of field observations with the modeled results.

2.4.4. Kappa (K) Analysis

Kappa (K) analysis is a multivariate method used to assess accuracy by measuring the agreement between predicted and observed classifications. The Khat statistic, derived from Kappa, is widely used in evaluating inter-rater reliability (IRR) in environmental modeling. Weighted Kappa, which accounts for the degree of disagreement, is preferred for ordered categorical variables, while Cohen’s Kappa is suited for nominal data [67]. These techniques are essential in validating predictive models and evaluating the precision of geographical classifications. The Kappa coefficient (K) is calculated using the formula given in Equation (7):

$$\mathrm{K}={\displaystyle \frac{\mathrm{O}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{l}\mathrm{l} \mathrm{A}\mathrm{c}\mathrm{c}\mathrm{u}\mathrm{r}\mathrm{a}\mathrm{c}\mathrm{y}-\mathrm{E}\mathrm{x}\mathrm{p}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{e}\mathrm{d} \mathrm{A}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t}}{1-\mathrm{E}\mathrm{x}\mathrm{p}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{e}\mathrm{d} \mathrm{A}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t}}}$$

(7)

3. Results

Groundwater potential zoning was assessed using twelve factors, including LULC, rainfall, drainage density, soil type, slope, road density, NDVI, NDBI, MSI, WVWI, and LST. These factors were analyzed with GIS-based AHP to determine their relative significance for groundwater recharge potential (Figure 4).

The analysis revealed the following trends in recharge potential based on various factors.

LST showed that very high recharge potential was found in cooler areas (2.00%, 20–24 °C), while higher temperatures were associated with progressively lower recharge potential, ranging from high (19.09%, 25-28 °C) to very low (28.02%, 37–39 °C) potential (Figure 4A). In terms of water depth, the study found that shallow depths (7.08%, 20–94 m) exhibited the highest recharge potential, with moderate to very low potential increasing at greater depths (up to 33.37% at 203–318 m) (Figure 4B). The MSI analysis revealed that high-soil-moisture zones (11.43%, 0.533–0.783 MSI) provided very high recharge potential, but as moisture stress increased, recharge potential decreased, with very-low-moisture zones showing 38.12% at 1.105–1.653 MSI (Figure 4C). For NDBI, highly built-up areas (48.21%, 0.042–0.246 NDBI) showed very low groundwater recharge potential, while less built-up zones exhibited varying degrees of recharge potential, from low (22.47%) to very high (8.29%, −0.304 to −0.141) (Figure 4D). Drainage density analysis revealed that areas with dense drainage (0.001–10.416) had very high recharge potential (59.25%), with less dense areas showing moderate to low recharge potential (4.3% at >41.666) (Figure 4E). The land cover analysis demonstrated that water bodies (11.16%) had very high recharge potential, whereas barren land (59.47%) had very low recharge potential. Also, the built-up areas (14.49%) showed low recharge due to impervious surfaces, but vegetated areas (14.88%) supported high recharge potential (Figure 4F).

Figure 4G shows the spatial and temporal distribution of NDVI in the study area. The result revealed that decreasing vegetation cover corresponds to reduced recharge potential, with moderate (22.07%), high (10.14%), low (56.47%), and very low vegetation cover (3.32%, −0.195–0.018) exhibiting progressively lower recharge potential. However, areas with dense vegetation (8.00%, 0.261–0.505 NDVI) were found to have very high recharge potential. In terms of slope, flat terrain (90.14%, 0–2.471°) had the highest recharge potential, followed by moderate (2.89%), low (1.69%), and extremely low (0.58%, >22°) slope zones (Figure 4H).

The spatial pattern of Rainfall showed that very rainy areas (16.63%) had the greatest recharge potential, with progressively lower recharge potential in areas with moderate (22.06%), low (19.09%), and very low (17.56%, 356.918–387.328 mm) rainfall. Areas with very high rainfall (16.66%, 480.791–532.958 mm) provided optimal conditions for groundwater recharge (Figure 4I). The WVWI analysis revealed that areas with very high surface wetness (2.50%, −0.089 to 0.267 WVWI) exhibited high recharge potential, with decreasing potential in areas of high (3.83%), moderate (7.48%), low (37.71%), and very low (48.48%, −0.431 to −0.166) wetness (Figure 4J). Regarding Road distance, areas closer to roads (53.14%, 0–0.5 km) had low recharge potential, which improved with increasing distance from roads, reaching very high recharge potential at 3–5 km (2.23%) (Figure 4K). Lastly, as shown in Figure 4L, Soil type analysis revealed that Haplic soil dominated the area, covering 80.43%, while Lithosol and Calcaric soils were present in smaller proportions (0.92% and 17.61%, respectively). These findings thus demonstrate the various ways in which environmental interactions impact the groundwater resource recharge process. Therefore, at various study scales, it is shown that important characteristics such as drainage density, slope, vegetation cover, and water depth relief have a substantial impact on recharging chances.

3.1. Development of Groundwater Potential Zoning

The distribution of groundwater potential zones (GWPZs) in the Dera Ismail Khan district is determined by climatic and geographic factors (

Table 8. Groundwater potential zones.

GWPZ	Area (km2)	Percentage (%)
Very Poor	42.36	1
Poor	2500.86	33
Moderate	3551.81	47
Good	1280.00	17
Very Good	148.00	2

). The zone of very good potential, which covers only 2% of the district, is found in areas with high soil permeability and drainage density, typically along the eastern edge of the Indus River. The zone of good potential, making up 17% of the district, is in regions with dense vegetation, shallow water tables, and favorable recharge conditions, such as parts of Kulachi Tehsil and the Indus River floodplain. The moderate-potential zone, covering 47% of the area, is concentrated in the central region, including some built-up and agricultural areas with gentle slopes. The poor-potential zone, making up 33% of the district, is characterized by arid, barren land with scarce vegetation, particularly in the southwest region of the study area. The very-low-potential zone, which makes up 1% of the district, is confined to isolated, hilly regions in the west, characterized by steep slopes and deep-water tables (Figure 5). These findings highlight the district’s varied groundwater recharge potential, with limited capacity in arid and hilly regions and better recharge suitability in areas near the Indus River and vegetated zones.

3.2. Validation of Groundwater Potential Zones and Accuracy Assessment

Out of the 83 observations of water table depths, 79 were accurately classified according to their respective groundwater potential zones, while 4 exhibited discrepancies (Figure 6). The overall accuracy of the groundwater potential zoning map was calculated as follows:

$$\mathrm{Overall} \mathrm{Accuracy}={\displaystyle \frac{79}{83}}\times 100=95.18\%$$

The overall accuracy of the groundwater potential zoning was 95.18%, indicating strong agreement between predicted and observed values. The observed water depths varied greatly in terms of groundwater availability, ranging from 20 m to more than 320 m. Sample accuracy numbers are presented in

Table 9. Sample accuracy check.

Class	Very Good	Good	Moderate	Poor	Very Poor	Total	Correct Samples
Very Good	8	0	0	0	0	8	8
Good	0	12	0	0	0	12	12
Moderate	0	0	29	2	0	31	29
Poor	0	0	2	23	0	25	23
Very Poor	0	0	0	0	7	7	7
Total	8	12	31	25	7	83	79

.

In Kappa analysis, the Percentage Correct Agreement with Observed Values is the proportion of correct samples for each class. Pe can be calculated as the sum of the expected proportion of each class being correctly classified by chance (

Table 10. Expected agreement calculation (Pe).

Class	Total	Proportion	Correct Sample	Expected Contribution
Very Good	8	8/83 = 0.0964	8	0.0964 × 0.0952
Good	12	12/83 = 0.1446	12	0.1446 × 0.1446
Moderate	31	31/83 = 0.3735	29	0.3735 × 0.3735
Poor	25	25/83 = 0.3012	23	0.3012 × 0.3012
Very Poor	7	7/83 = 0.0843	7	0.0843 × 0.0843

).

$$\begin{array}{c}\mathrm{P}\mathrm{e}=(0.0964\times 0.0964)+(0.1446\times 0.1446)+(0.3735\times 0.3735)+(0.3012\times 0.3012)\\ \hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}+(0.0843\times 0.0843)\hfill \end{array}$$

$$\mathrm{P}\mathrm{e}=0.0093+0.0209+0.1395+0.0907+0.0071$$

$$\mathrm{P}\mathrm{e}=0.2675$$

The Kappa coefficient (K) is then calculated as follows:

$$\mathrm{K}={\displaystyle \frac{0.95-0.2675}{1-0.2675}}$$

$$\mathrm{K}={\displaystyle \frac{0.6825}{0.7325}}$$

$$\mathrm{K}=0.932$$

The Kappa value is 0.93, which indicates that agreement is almost perfect, meaning groundwater potential zoning classification is highly accurate and reliable. The agreement between predicted and observed values is almost perfect.

4. Discussion

This study utilized GIS, remote sensing, and AHP techniques to assess groundwater recharge potential in Dera Ismail Khan, Pakistan. Previous research in Dera Ismail Khan as well as other arid to semi-arid regions [22,29,36] has centered around the analysis and management of groundwater quality. Most of these studies used standard GIS-based weighted overlay methods. Integrated assessments of groundwater potential lack effective multi-criteria decision-making methodologies. Our research fills this methodological gap through the application of a detailed AHP which works together with geospatial techniques. The hierarchical multi-criteria evaluation framework structures the model to achieve internal consistency through ratio consistency checks and expert judgment validation. The methodological design enhances both reliability and scientific rigor throughout the analysis. Our model stands apart from earlier studies because it includes twelve parameters that have been thoroughly justified and combines remote sensing indices like NDVI, MSI, and WVWI with topographic characteristics and land cover alongside hydro-climatic variables. The combination of different indicators for surface and subsurface recharge improves groundwater potential mapping accuracy in semi-arid regions. Our method shows better methodological progress than similar GIS–AHP-based research carried out in areas with similar climatic conditions, including Meki Catchment in Ethiopia [4], Vaigai Upper Basin in Tamil Nadu [10], the Semi-Arid Lower Ravi River Basin in Pakistan [28], Kohat District in KP Pakistan [35], Habawnah Basin in Saudi Arabia [37], the northern Nile region in Egypt [38], and the Jinan Karst Spring Basin in China [65]. The classification accuracy of our model reached 95.18% with a Kappa coefficient of 0.93, which exceeds the typical accuracy range of 74–85% found in previously reported studies. The results highlight our combined AHP–GIS approach as a precise method for mapping groundwater recharge areas.

The results indicate that factors such as topography, hydrology, vegetation, and built-up environment play a complex role in groundwater recharge dynamics. Areas with mild slopes (0–2.471°), which make up 90.14% of the study region, were identified as highly conducive to infiltration and aquifer recharge. These findings align with global trends, while steeper slopes (>22°), covering only 0.58% of the area, predominantly promote runoff rather than recharge [43]. Furthermore, the aquifer depth also plays a critical role in groundwater recharge, with zones at depths of 20–94 m, representing 7.08% of the study area, showing active recharge. In contrast, deeper zones (>203 m) show limited recharge potential due to some geological factors [32,36]. In addition, vegetation density emphasizes the impact of land use on groundwater recharge. Dense vegetation areas, covering 8% of the region, correspond to high recharge potential. Conversely, scrub and arid zones (56.47%) demonstrate low recharge potential, highlighting the need for afforestation to enhance water retention.

The challenges of groundwater recharge are exacerbated by urbanization, with built-up areas covering 48.21% of the study region and exhibiting a low recharge index due to the inaccessibility of infiltrative surfaces. This finding aligns with Sharp’s [68] research, which links urban sprawl to negative impacts on groundwater quality and quantity. Additionally, the distribution of annual rainfall influences recharge potential; areas with precipitation between 480 and 533 mm show favorable recharge conditions. In contrast, areas receiving less rainfall require alternative management strategies, such as rainwater harvesting and moisture retention, to support recharge [54,56].

Moreover, the Land Surface Temperature (LST) dataset reveals significant thermal effects on groundwater recharge. Only 2% of areas with low LST (20–24 °C) demonstrate a higher recharge potential, whereas regions with high LST (37–39 °C), which cover an area of 28.02%, negatively impact recharge due to increased evapotranspiration. These findings align with global studies indicating that elevated temperatures can substantially reduce infiltration and soil water content [60].

Our analysis revealed that approximately 1428 km² (comprising 148 km² of very-good-potential and 1280 km² of good-potential zones) has been identified as a groundwater recharge area, accounting for 19% of the total 7325 km² area of Dera Ismail Khan district. These zones are predominantly located along the Indus River floodplain, primarily within Dera Ismail Khan and Paharpur Tehsils, where conditions such as shallow water tables, gentle slopes, high drainage density, and vegetated land cover favor recharge. Based on the spatial distribution of population and recharge zones, an estimated 347,664 individuals—approximately 19% of the district’s total population of 1,829,811—reside in or near these recharge areas [50]. These populations are expected to directly benefit through improved groundwater availability for domestic, agricultural, and municipal water supply. Additionally, the delineation of these zones provides a scientifically grounded foundation for sustainable groundwater management and targeted recharge interventions across the district.

Currently, the region exhibits moderate to poor potential for groundwater recharge, but there are several ways to enhance this potential. First, high recharge areas with low slopes, shallow water tables, and high vegetation should be considered for protection first. It is possible to further utilize these areas for recharging basins and rehabilitation of infiltration rates through afforestation. Furthermore, targeted measures are needed to mitigate natural recharge processes in the urban areas, where the substantial impermeable surfaces substantially hinder recharge. The implementation of permeable pavements, green roofs, and increased urban green spaces can significantly enhance groundwater recharge by allowing rainwater to percolate into the ground. Furthermore, policies promoting rainwater harvesting and the construction of check dams in the surrounding areas could also support the replenishment of the aquifer. Future development plans for study area should integrate these strategies into urban planning, ensuring that both urban growth and groundwater recharge are balanced for long-term sustainability.

The use of GIS, remote sensing, and AHP in this study offers valuable insights and could be further refined by incorporating additional variables, such as aquifer transmissivity and sustainable hydrological studies. By promoting these methodologies, this research provides a strong foundation for addressing water scarcity in arid and semi-arid regions through enhanced groundwater resource management. The study provides valuable insights into groundwater potential zones in Dera Ismail Khan using recent geospatial techniques; however, certain limitations must be acknowledged.

The use of moderate-resolution satellite data (30 m) may overlook finer-scale variability, and static environmental parameters fail to account for temporal changes, such as seasonal variabilities or climate change impacts. Key hydrological and geological factors, such as aquifer transmissivity, were not included due to data limitations, and reliance on expert judgments in AHP may introduce biases. The limited ground validation and reliance on a single-year dataset (e.g., 2022) constrain the robustness of the results. Additionally, anthropogenic influences like groundwater extraction were not explicitly modeled. Finally, while the methodology applies well to semi-arid regions, there may be certain components of the methodology that may require adaptation for a wider application. To increase the accuracy and generalizability of the results, future studies could fill in these gaps.

5. Conclusions

The present study effectively identified groundwater potential zones (GWPZs) in Dera Ismail Khan, Pakistan, utilizing GIS, RS, and the AHP model. A comprehensive evaluation of groundwater recharge capacity was conducted, considering key factors such as land surface temperature, drainage density, LULC, water depth, NDVI, slope, soil, road density, NDBI, WVWI, MSI, and rainfall. The findings reveal that 47% of the area falls within the moderate-recharge-potential zone, while 17% and 2% are classified in the good-and very-good-potential zones, respectively. However, 33% of the region is categorized as poor and 1% as very poor, highlighting significant challenges in water resource management.

This study makes a significant contribution to advancing scientific understanding by demonstrating how GIS, RS, and AHP can be integrated to create a robust framework for mapping groundwater recharge zones in semi-arid regions. The inclusion of innovative remote sensing indices (NDVI, NDBI, MSI, and WVWI), alongside traditional factors, enhances the methodological rigor and provides a reproducible model for similar studies in other regions. By highlighting the role of meteorological conditions and human activities, such as road density and LST, the analysis is further strengthened. The findings have critical implications for sustainable groundwater management, addressing issues like urban sprawl and over-extraction in low-potential areas and providing a scientific basis for prioritizing high-potential zones for conservation, afforestation, and recharge basin development.

The sustainable management of water resources to ensure long-term availability is emphasized in Sustainable Development Goal 6 (Clean Water and Sanitation), which is in line with this study. This study supports Target 6.4, which focuses on boosting sustainable withdrawals and improving water use efficiency to prevent water scarcity, by identifying groundwater potential zones using GIS and AHP. The results herein will give policy makers a solid scientific foundation on which to build groundwater conservation plans, maximize agricultural water use, and protect supplies of drinking water. In semi-arid areas such as Dera Ismail Khan, where groundwater serves as the primary source of water for domestic and agricultural use, the implementation of this strategy is particularly critical. Future research may enhance this methodology by integrating aquifer transmissivity and temporal datasets to better account for evolving hydrological systems.