Assessing the Impact of Urban Expansion on Groundwater Recharge Potential Using GIS-based Multi-criteria Analysis: A Case Study of Kahramanmaras¸,Türkiye (2000

Abstract

Urban expansion threatens groundwater sustainability, yet regional-scale assessments can mask localized impacts. This study presents a GIS-based multi-criteria framework to quantify this ‘paradox of scale’, assessing the conflict between urban growth and the potential of groundwater recharge at Kahramanmaras¸, Türkiye, during the period 2000–2025. Using open-source datasets (GHSL, CHIRPS and SoilGrids), natural recharge potential was modelled, and annual loss attributable to urbanization was quantified. Although the regional mean groundwater recharge potential index (GRPI) declined only marginally (0.449→0.447) during this period, spatial analysis revealed a disproportionate, effectively irreversible loss within the most critical recharge zones. In total, 10.7 km² (3.55%) of top-priority ‘hotspot’ areas were converted to urban land, with losses unevenly distributed; the faster-urbanizing district forfeited a larger share of its hotspots. These findings indicate that reliance on regional averages is inadequate for sustainable planning. The resulting hotspot map provides a decision-ready tool to prioritize conservation of strategic recharge zones and strengthen long-term water security in rapidly urbanizing regions.

Keywords

Groundwater recharge Urban expansion GIS Land-use change

Introduction

In recent years, the rapid and often unplanned urbanization witnessed on a global scale is exerting escalating pressure on natural resources and is widely recognized as one of the foremost obstacles to sustainable development (Ampe et al., 2012; Pasquier et al., 2022). Among the factors most severely affected by this pressure are water resources, which are indispensable for both human life and ecosystem integrity (Jasechko et al., 2014). Urban growth fundamentally alters the hydrological cycle by sealing the natural landscape with impervious surfaces such as concrete and asphalt, a transformation that leads to diminished groundwater recharge, increased surface runoff and a heightened risk of flooding (Salem et al., 2023). As groundwater plays a pivotal role in supplying drinking, domestic and agricultural water in numerous regions worldwide (Jasechko et al., 2014), the disruption of these recharge processes threatens not only water quantity but also its quality, due to pollutants carried by urban runoff, thereby jeopardizing long-term water security for millions of people (Nolan et al., 2003).

Although this general relationship between urbanization and groundwater recharge is well-documented in the literature on the subject (Mohan et al., 2017; Pasquier et al., 2022; Salem et al., 2023), a significant research gap persists in the quantitative assessment of this impact, particularly concerning the specific loss of the most valuable recharge zones or ‘hotspots’ (Maqsoom et al., 2022; Mohammed et al., 2025; Waldowski et al., 2023). Many studies conducted at broad administrative scales, such as basins or districts, tend to focus on regional averages (MacDonald et al., 2021; Manna et al., 2019; Waldowski et al., 2023). The minor changes often observed in these averages can create a misleading perception that the urban impact is limited (Ampe et al., 2012; Pasquier et al., 2022). This situation points to a problem that can be described as the ‘paradox of scale’: severe degradation occurring in small but critical, hydrologically active areas can be easily overlooked when averaged across a wide geography (Doble et al., 2012; Manna et al., 2019). Consequently, there is a pressing need for spatial analyses capable of measuring not only the existence of an impact but also the qualitative loss resulting from its disproportionate concentration on the most productive lands.

To address this research gap and the ‘paradox of scale’, the primary objective of this study is to spatially analyse the relationship between urban growth and groundwater recharge in the Kahramanmaraş urban area by developing a GIS-based multi-criteria model. In pursuit of this main objective, the research focuses on three fundamental questions: (a) Where is the region’s natural groundwater recharge potential concentrated? (b) To what extent and in which locations has urban expansion during the 2000–2025 period quantitatively reduced this potential? (c) How much of the critically important high-potential hotspot area has been irreversibly lost during this process?

To achieve these research objectives, a transparent and scientifically reproducible workflow was developed, founded entirely on open-source and globally accessible geospatial datasets (Copernicus Digital Elevation Model (DEM), Global Human Settlement Layer (GHSL), CHIRPS and SoilGrids). The selection of the Kahramanmaraş urban area as the study site was not coincidental. The city serves as a typical example of a rapidly growing medium-sized city in Türkiye, where this growth exerts significant pressure on agricultural lands and the natural landscape. Therefore, the findings and the methodology developed herein have the potential to offer important insights for other regions facing similar urban and hydrological challenges.

This study, through the spatial model developed for Kahramanmaraş, aims to reveal the latent impacts of urban growth on groundwater recharge and to provide a scientific foundation for decision-makers to protect the most critical recharge areas.

Materials and Methods

This section details the methodological framework of the study, encompassing a description of the study area, the datasets employed, data preprocessing procedures and the analytical steps followed for modelling groundwater recharge potential and assessing urban impacts. All geospatial data processing, analysis and visualization were conducted within the open-source QGIS 3.44 software platform, supported by custom Python scripts (version 3.13). The entire analytical workflow, from raw data acquisition to final impact assessment, is schematically summarized in Figure 1. All spatial analyses and resultant outputs were generated within the EPSG: 32,637 (UTM Zone 37N) projection system, utilizing a 100-metre resolution reference grid.

Study Area

This investigation is centred on the urban area of Kahramanmaraş, a province in south-eastern Türkiye, specifically covering the Onikişubat and Dulkadiroğlu districts (Figure 2). The study area presents a diverse landscape, transitioning from high-altitude, rugged mountainous terrain in the west and north-west to low-lying, flat alluvial plains in the south and south-east, which are characterized by intensive agriculture and dense urban settlement. The region’s rapid urban growth poses a significant pressure on its water resources.

Datasets and Preprocessing

The analysis was constructed upon four primary datasets, all of which are open-source and globally available.

DEM: The Copernicus GLO-30 DEM, with a nominal resolution of approximately 30 metres, was acquired via the STAC API from the Microsoft Planetary Computer platform (ESA, 2021). Tiles covering the study area were mosaicked and clipped to the area’s extent. This foundational dataset was then processed using Whitebox Tools (Lindsay, 2016), where a depression-filling procedure based on the Wang and Liu (2006) algorithm was applied, followed by D8 flow direction and flow accumulation analyses, from which the thematic layers of slope (degrees) and the Topographic Wetness Index (TWI) were derived.

Precipitation Data: The CHIRPS v2.0 global monthly precipitation dataset (at ~5.5 km resolution) was obtained in NetCDF format for the 2000–2025 period (Funk et al., 2015). From these monthly records, annual totals were calculated, and a mean annual precipitation map was generated by averaging across the entire period. To mitigate the blocky artefact inherent in the coarse resolution, a NaN-aware Gaussian smoothing filter with a radius of approximately two kilometres was applied to this map.

Urban Area Data: The GHSL GHS-BUILT-S R2023A dataset from the European Commission, providing a 100-metre resolution layer of built-up area fraction, was procured for five-year intervals from 2000 to 2025 (European Commission, 2023). To characterize urban expansion, the difference in built-up fraction between 2000 and 2025 (Δbuilt) was computed, and a binary urban expansion mask was created using a threshold of Δbuilt> 0.05.

Soil Data: Information regarding the sand, silt, and clay content of the topsoil layer (0‒5 cm) was dynamically retrieved from the ISRIC SoilGrids database (Hengl et al., 2017). A permeability proxy was formulated from these components to represent the soil’s infiltration capacity, designed to positively weight sand content while penalizing clay content.

To ensure comparability within the model, all derived continuous layers (slope, TWI, precipitation and permeability) were standardized to a 0–1 range using a robust min–max normalization method, which scales the data based on the 2nd and 98th percentiles to minimize the influence of outliers.

Methodology

The analytical process involved the development of two principal models, followed by an assessment of urban-induced impacts.

The Groundwater Recharge Potential Index (GRPI) Model

A two-stage approach was adopted to develop the model. First, a natural recharge potential index was calculated, deliberately excluding any urban influence. This was achieved by combining the four fundamental standardized layers with equal weighting (w = 0.25) (Ahmed et al., 2021; Bhunia, 2020; Saraf et al., 2004):

\begin{array}{l} N a t u r a l p o t e n t i a l = 0.25 \times \Pr e c i p i t a t i o n_s t d + 0.25 \times \\ T W I_s t d + 0.25 \times S l o p e S u i t a b i l i t y_s t d + 0.25 \times \\ P e r m e a b i l i t y_s t d \end{array}

In this formulation, SlopeSuitability_std is defined as 1−Slope_std to reward areas with low gradients. In the second stage, the annual GRPI was computed by integrating the urban fabric of the corresponding year as a suppression factor (Ampe et al., 2012; Pasquier et al., 2022):

G R P I_y e a r = N a t u r a l p o t e n t i a l \times (1 - B u i l t - u p F r a c t i o n_y e a r)

This two-step methodology first establishes the landscape’s intrinsic potential and subsequently models how that potential is diminished by the presence and growth of urban development.

Hotspot Analysis and Urban Impact Assessment

Hotspots, defined as areas possessing the most favourable conditions for groundwater recharge (Bhunia, 2020; Maqsoom et al., 2022; Navane & Sahoo, 2020), were identified using a set of flexible criteria: Natural potential ≥ 0.5, slope ≤ 10° and permeability ≥ 0.4. The extent to which these critical areas were lost to urban expansion between 2000 and 2025 was quantified by spatially intersecting the hotspot layer with the urban expansion mask. Furthermore, the magnitude of urban impact was analysed through a counterfactual scenario, which modelled 2025 conditions under the assumption of a 2000-level urban footprint, resulting in an ‘urban suppression’ map. The robustness of the study’s findings was further interrogated through a sensitivity analysis that tested alternative weighting schemes and threshold values.

Map Production and Visualization

All thematic maps and final figures presented in this study were generated using Python’s Matplotlib library. Standard cartographic elements, including a legend, north arrow and scale bar, were incorporated into the map design to ensure clear communication of spatial patterns. The accuracy of intermediate layers and analytical results was visually verified at each stage within the QGIS environment.

Results

This section presents a quantitative and spatial analysis of the groundwater recharge potential across the study area, which encompasses the Onikişubat and Dulkadiroğlu districts of Kahramanmaraş. The investigation quantifies the impact of urban expansion between 2000 and 2025 on this potential and assesses the resulting loss of high-potential hotspots.

Spatial Distribution of Natural Recharge Potential and Urban Expansion

The investigation commenced by modelling the area’s inherent groundwater recharge potential under purely natural conditions, a scenario constructed to deliberately exclude the influence of the urban fabric. An index for ‘natural groundwater recharge potential’ was derived from the synthesis of four principal hydrological determinants: precipitation, the TWI, slope and soil permeability. The outcome of this process—a map depicting the region’s natural capital—is presented in Figure 3a. Within this map, shades of yellow and bright green signify zones of high recharge potential (index > 0.7), which are prominently located in areas characterized by alluvial deposits within flat, low-elevation river valleys. In stark contrast, the potential noticeably diminishes (indicated by blue and purple hues) across the elevated and rugged terrain of the western and north-western sectors, a decline directly associated with increasing slope gradients.

Urban expansion, identified as the foremost anthropogenic pressure on this natural potential, was analysed for the 2000–2025 period utilizing data from the GHSL. Figure 3b illustrates the increase in the built-up area fraction over this 25-year interval. The bright purple pixels highlight territories where new residential, industrial or transportation infrastructure has been established. It is clearly observable that this expansion has concentrated along the peripheries of the existing urban core, particularly on the flatlands situated in the southern and south-eastern portions of the study area. A concurrent examination of these two maps introduces a critical question that forms the basis for subsequent analyses: whether urban growth has encroached upon the very lands that hold the highest natural potential for groundwater recharge.

An examination at the district level reveals noteworthy distinctions in the distribution of both natural potential and anthropogenic pressure. The Dulkadiroğlu district, with a mean natural recharge potential of 0.499, presents a more advantageous baseline compared to the Onikişubat district’s average of 0.422. Conversely, the pressure from urban expansion during the 2000–2025 period was more pronounced in Onikişubat. This district experienced an urbanization of 45.03 km² (representing 1.85% of its total area), whereas Dulkadiroğlu saw a smaller expansion of 17.60 km² (1.34% of its area). These preliminary figures suggest that the ultimate impact of urban development on groundwater potential may vary in magnitude between the two districts.

The Impact of Urban Expansion on Groundwater Recharge Potential

To quantitatively assess the relationship between natural potential and the spatial patterns of urban expansion, a GRPI was calculated for each analysis year (2000, 2005, 2010, 2015, 2020 and 2025). This index was formulated by multiplying the natural potential map by the corresponding year’s non-built-up fraction, calculated as 1 minus the built-up fraction. Such an approach integrates the escalating effect of urbanization, primarily through increased impervious surfaces, as a suppression factor on the recharge potential.

Figure 4 visualizes this impact by presenting the GRPI maps for the years 2000 and 2025, along with a delta map illustrating the change between these two periods. As depicted in Figure 4a, the more limited extent of the urban fabric in the year 2000 allowed for a broader distribution of high-potential zones (indicated by yellow hues). By 2025, however, as seen in Figure 4b, there was a marked contraction of these yellow areas, which have largely been supplanted by zones of lower potential in shades of green. This transition is particularly evident in the southern and south-eastern sectors where urban expansion was most concentrated.

The magnitude and location of this transformation are explicitly detailed in the delta map shown in Figure 4c. The blue tones in this map pinpoint the exact locations where the potential has diminished most significantly over the 25-year period, with values approaching—1 representing the greatest loss. The direct correspondence of these areas with the newly built-up zones identified in Figure 3b provides compelling evidence that urban expansion is the primary driver of the decline in GRPI. Across the entire study area, the mean GRPI value decreased from 0.449 in 2000 to 0.447 in 2025. Although modest when averaged across the region, this decline serves as a quantitative measure of the cumulative and persistent pressure that urban development exerts on groundwater resources.

The Loss of High-potential Areas (Hotspots)

In this phase of the analysis, the focus shifts beyond regional averages to examine the direct impact of urban expansion on the most critical and favourable lands for groundwater recharge. For this purpose, areas with high natural potential (≥0.5), low slope (≤10°) and high soil permeability (≥0.4) were defined as hotspots. These criteria delineate the territories that provide the most ideal conditions for water infiltration into the ground.

As illustrated in Figure 5a, these hotspots constitute a significant portion of the study area, predominantly covering valley floors and plains. Calculations reveal that a total area of 301.3 km² within the study region meets these favourable characteristics. However, the extent to which these valuable lands are under anthropogenic pressure is starkly revealed in Figure 5b. The red zones on this map indicate where high-potential hotspots have been directly and irreversibly eliminated by urban expansion during the 2000–2025 period. The quantitative data show that a total of 10.7 km² of hotspot area was lost to construction and development over the 25-year span. This figure accounts for 3.55% of the entire hotspot territory. Contrary to the modest decline in regional averages, this finding demonstrates that urban growth exerts a disproportionately large pressure on the most productive and essential lands that warrant protection. Furthermore, Figure 5c, which maps the spatial intensity of potential loss due to urban suppression, confirms that this impact is most severe in hotspots located in close proximity to the urban core.

District-level Comparison of Findings

In the final phase of the result analysis, the differences in groundwater recharge potential and urban pressures between the Onikişubat and Dulkadiroğlu districts were investigated. The key findings, summarized in Table 1, reveal that urban development has impacted each district in a distinct manner.

Table 1.

Comparison of Key Findings for Onikis¸ubat and Dulkadirog˘lu Districts.

Metric	Onikis¸ubat	Dulkadirog˘lu	Unit
Mean natural potential	0.422	0.499	Index
Urban expansion area (2000‒2025)	45.03	17.60	km²
Total hotspot area	184.64	116.67	km²
Lost hotspot area	9.00	1.71	km²
Hotspot loss rate	4.87%	1.46%	%

In terms of natural endowment, the Dulkadiroğlu district possesses more favourable baseline conditions, with a mean natural potential index of 0.499 compared to 0.422 in Onikişubat. However, the pressure of urban development has been felt far more intensely in the Onikişubat district. During the 2000–2025 period, Onikişubat experienced 45.03 km² of new urbanization, whereas this figure was limited to 17.60 km² in Dulkadiroğlu.

The most striking consequence of this disparity is observed in the loss of hotspot areas, which are of critical importance for groundwater recharge. Onikişubat, having faced more intensive urban expansion, lost 9.0 km² of its most productive land. In contrast, the loss in Dulkadiroğlu was 1.71 km². Proportionally, Onikişubat lost 4.87% of its total hotspot area, while Dulkadiroğlu lost 1.46%. This finding demonstrates that the absolute magnitude of urban pressure has a direct and disproportionate impact on the loss of the most valuable natural resources.

Discussion

Interpretation of Principal Findings

The findings from this investigation present a multi-layered narrative of the impacts that urban expansion has exerted on the groundwater recharge potential within the Kahramanmaraş urban area between 2000 and 2025. A primary observation is that the spatial distribution of urban growth was not random. The concentration of expansion within the flat, low-lying terrains along river valleys—areas possessing the highest natural recharge potential—strongly suggests that socio-economic drivers, such as the ease of construction and the re-zoning of agricultural lands, are imposing pressure on the most hydrologically sensitive zones.

One of the study’s most striking revelations is what can be termed ‘paradox of scale’. While a cursory glance at the regional average GRPI, which decreased only marginally from 0.449 to 0.447 over 25 years, might suggest a negligible impact, this interpretation is misleading. This minimal shift indicates that the effects of urbanization were not diffuse across the landscape but were instead intensely focused within a relatively small fraction of the total study area, thereby failing to significantly alter the regional average. The true story lies concealed behind this average, in the qualitative nature of the losses.

The most compelling evidence of this qualitative loss emerges from the hotspot analysis. The fact that 10.7 km² (representing 3.55%) of the 301.3 km² of prime rechargeable land was directly lost to development is the study’s most critical finding. This is not merely a quantitative reduction in area but signifies the irreversible destruction of the most valuable natural capital that serves as a crucial buffer for the region’s long-term water security.

Finally, the results expose a significant imbalance between the two districts. Despite possessing a less favourable natural potential, the Onikişubat district was subjected to far more intensive urban pressure. Consequently, Onikişubat lost 4.87% of its total hotspot area, a figure starkly contrasting with the 1.46% loss observed in Dulkadiroğlu. This disparity serves as a tangible indicator of how localized urban planning and land-use decisions can produce widely divergent environmental outcomes.

Comparison of Findings with Existing Literature

The findings of this study are consistent with the broader conflict between urbanization and groundwater resources observed in other rapidly urbanizing regions globally. However, this study’s analysis, particularly through the lens of hotspot loss, introduces a novel quantitative dimension to this relationship.

The tension between urbanization and recharge documented by Ávila-Carrasco et al. (2023) for the southern basin of Mexico City strongly corroborates this study’s results. Their study reported a decline in groundwater recharge (as a percentage of precipitation) corresponding with an increase in urbanized areas between 1970 and 2019. This causal link aligns perfectly with the two-stage framework of this study, which first maps natural potential and then models the suppressive effect of annual urban growth; the expansion of impervious surfaces is the primary, spatially structured driver of recharge reduction. This study’s analysis, however, further demonstrates that regional averages, whether at the basin or district scale, can effectively mask the disproportionate and permanent loss of the most permeable flatlands.

Similarly, a study by Tiwari et al. (2024) employing an AHP–MCDA framework for Ajmer, India, revealed ‘very poor’ to ‘poor’ recharge potential in core urban areas, contrasted with ‘moderate’ to ‘very good’ potential in the urban periphery. While this work also confirms that impervious surface expansion spatially suppresses recharge, this study’s temporal model (2000–2025) allows for a quantitative assessment of this impact over time. A unique contribution of this study is the demonstration that a marginal decline in the regional mean GRPI (from 0.449 to 0.447) can conceal a significant, disproportionate loss of critical hotspot areas (3.55%). Whereas the static classification from the Ajmer study answers the question ‘where to protect’, this study’s approach adds the crucial dimensions of ‘how much have we lost, and where are we losing it most rapidly’?

At the process level, isotope-based studies, such as those in the Shiyang River basin (Li et al., 2023) that show urbanization altering surface groundwater hydraulic connectivity and reducing residence times, validate the underlying mechanism of this study’s findings. While such studies prove how urbanization alters hydrological processes, this study’s GHSL-derived suppression and GRPI-hotspot framework quantifies where and to what extent this impact manifests in metrics relevant for planning.

Ultimately, when read in conjunction with this body of literature, the ‘paradox of scale’ identified in this study becomes clear: basin-scale averages may obscure the disproportionate loss of high-potential zones that are often the most hydrologically active. This underscores the critical need for targeted policy interventions, such as the delineation of hotspot protection zones, impervious surface limits and the implementation of nature-based infiltration solutions in urban planning.

Methodological Considerations and Strengths

The validity of this study’s findings rests on its methodological robustness, the transparency of the adopted approaches and the stability of its results. A critical strength of the investigation is the sensitivity analysis performed to test the model’s reliability.

First, the influence of the thematic layer weights on the final GRPI was examined. In this study’s baseline scenario (S0_baseline), where all layers were assigned equal weights (25%), the mean GRPI for 2025 was calculated as 0.561. When the weights were altered to favour factors related to wetness (S1_wetness_favoured), the resulting mean value was 0.559. The negligible difference between these scenarios demonstrates that the model’s core findings are highly robust and insensitive to the initial weighting assumptions.

Second, the impact of the threshold values used in the hotspot definition was tested. Even when a stricter, more conventional scenario was applied (natural potential ≥ 0.7, Slope ≤ 5°, permeability ≥ 0.6), a loss of 0.261 km² of hotspot area due to urban expansion was still identified. Although this figure is lower than this study’s main finding of 10.7 km², it does not alter the fundamental conclusion that an irreversible loss of the most valuable lands is occurring. This consistency shows that the study’s primary conclusion—that urban growth exerts a disproportionate pressure on the most productive lands—remains valid even when the defining thresholds are changed.

Finally, the construction of the entire study upon open-source software (QGIS and Python) and globally accessible, free datasets (Copernicus, GHSL, CHIRPS and SoilGrids) maximizes the methodology’s transparency and scientific reproducibility. The automation of the entire analytical process via scripts not only minimized human error but also enables the straightforward adaptation and verification of this work by researchers in other parts of the world. These features enhance the value and credibility of the study for the scientific community.

Limitations and Future Research

Although the methodology employed in this study is robust and reproducible, several limitations should be considered when interpreting the results. First, some of the foundational datasets utilized in the model, such as CHIRPS precipitation and ISRIC SoilGrids soil data, are global in scale and possess a relatively coarse spatial resolution that may not fully capture local-scale dynamics. Future investigations could enhance the model’s precision and accuracy by incorporating higher-resolution precipitation data from local meteorological stations or through detailed field-based soil surveys.

Second, this study is predicated on the average climatic conditions of the 2000–2025 period. However, future climate change scenarios, such as prolonged droughts or an increased frequency of extreme precipitation events, could significantly alter the groundwater recharge regime. Therefore, a valuable next step for future research would be to run the model developed herein under various climate projections (e.g., IPCC scenarios) to understand the future vulnerability of the region’s water resources.

Finally, while this study spatially analyses where and how much urban expansion has occurred, it does not model the underlying socio-economic dynamics—such as land value, zoning regulations or proximity to transportation networks—that explain why this expansion has proceeded in specific directions. Interdisciplinary research that integrates these driving forces of urban growth could not only diagnose the current situation but also provide more accurate forecasts of future urban development scenarios.

Conclusion

This study has presented a spatially explicit and quantitative methodology to reveal the pressure that rapid urban growth exerts on one of the most critical natural resources: groundwater recharge potential. In the case of the Kahramanmaraş urban area, it has been demonstrated that urban expansion between 2000 and 2025, while causing only a marginal decline in regional averages, has led to a disproportionate and irreversible loss of the most valuable and hydrologically active hotspot areas. The ‘paradox of scale’ revealed by this research is significant, as it shows that regional-scale environmental indicators can mask severe degradation occurring in the most ecologically critical locations.

The results confirm that urban growth is not random but is preferentially directed towards the flat and productive lands most favourable for groundwater recharge. This trend signifies more than just a loss of area; it represents the systematic degradation of the natural infrastructure that ensures the region’s long-term water security and resilience against drought. The high rate of hotspot loss observed, particularly in the Onikişubat district, underscores how variable the environmental costs of urban planning decisions can be.

In light of these findings, it is recommended that the hotspot map produced in this study be utilized not as a static result, but as a dynamic planning tool. For local authorities and urban planners in Kahramanmaraş, this map should define ‘strategic water reserve zones’ that must be protected in future development plans. The implementation of urban design guidelines that either restrict development in these hotspot areas or promote water infiltration through nature-based solutions (e.g., rain gardens and permeable surfaces) would be one of the most tangible steps towards securing the city’s water future. This study provides a replicable and adaptable framework for other cities with similar geographical and socio-economic characteristics to navigate the delicate balance between urban sustainability and water resource management.

Footnotes

Author’s Contribution

Erdi Ekren: Conceptualization; methodology; software; validation; formal analysis; investigation; data curation; visualization; writing: original draft, review and editing.

Data Availability Statement

All underlying datasets used in this study are open-access and are cited in the manuscript (e.g., GHSL, CHIRPS, global DEM and SoilGrids). Processed layers, code used to generate the GRPI, and the resulting hotspot maps may be made available in a public repository (e.g., Zenodo/OSF).

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Ethical Approval

This study did not involve human participants, animal subjects, clinical interventions or personally identifiable data. It relies exclusively on open-access, aggregated environmental and geospatial datasets. Accordingly, institutional ethics approval was not required.

Funding

The author received no financial support for the research, authorship and/or publication of this article.

References

Ahmed

A. A.

, Alrajhi

, & Alquwaizany

(2021). Identification of groundwater potential recharge zones in the Flinders Ranges, South Australia using remote sensing, GIS, and MIF techniques. Water, 13(18), 2571. https://doi.org/10.3390/w13182571

Ampe

, Vanhamel

, Salvadore

, Dams

, Bashir

, Demarchi

, Chan

, Sahli

, Canters

, & Batelaan

(2012). Impact of urban land-cover classification on groundwater recharge uncertainty. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 5, 1859–1867. https://doi.org/10.1109/JSTARS.2012.2206573

Ávila-Carrasco

J. R.

, Hernández-Hernández

M. A.

, Herrera

G. S.

, & Hernández-García

G. D. J.

(2023). Urbanization effects on the groundwater potential recharge of the aquifers in the southern part of the Basin of Mexico. Hydrology Research, 54(5), 663–685. https://doi.org/10.2166/nh.2023.103

Bhunia

(2020). An approach to demarcate groundwater recharge potential zone using geospatial technology. Applied Water Science, 10, 1–12. https://doi.org/10.1007/s13201-020-01231-1

Doble

, Crosbie

, Smerdon

, Peeters

, & Cook

(2012). Groundwater recharge from overbank floods. Water Resources Research, 48(9). https://doi.org/10.1029/2011WR011441

ESA (European Space Agency). (2021). Copernicus DEM GLO-30. OpenTopography. https://doi.org/10.5069/G9028PQB

European Commission. (2023). GHSL data package. Publications Office of the European Union. https://doi.org/10.2760/098587

Funk

, Peterson

, Landsfeld

, Pedreros

, Verdin

, Shukla

, Husak

, Rowland

, Harrison

, Hoell

, & Michaelsen

(2015). The climate hazards infrared precipitation with stations: A new environmental record for monitoring extremes. Scientific Data, 2(1), 1–21.

Hengl

, De Jesus

J. M.

, Heuvelink

G. B. M.

, Gonzalez

M. R.

, Kilibarda

, Blagotić

, Shangguan

, Wright

M. N.

, Geng

, Bauer-Marschallinger

, Guevara

M. A.

, Vargas

, MacMillan

R. A.

, Batjes

N. H.

, Leenaars

J. G. B.

, Ribeiro

, Wheeler

, Mantel

, & Kempen

(2017). SoilGrids250m: Global gridded soil information based on machine learning. PLoS One, 12(2), e0169748.

10.

Jasechko

, Birks

S. J.

, Gleeson

, Wada

, Fawcett

, Sharp

, McDonnell

, & Welker

(2014). The pronounced seasonality of global groundwater recharge. Water Resources Research, 50, 8845–8867. https://doi.org/10.1002/2014WR015809

11.

, Zhu

, Lu

, Sang

, Meng

, Chen

, Jiao

, & Wang

(2023). Effects of urbanization on the water cycle in the Shiyang River basin: Based on a stable isotope method. Hydrology and Earth System Sciences, 27, 4437–4452. https://doi.org/10.5194/hess-27-4437-2023

12.

Lindsay

J. B.

(2016). Whitebox GAT: A case study in geomorphometric analysis. Computers & Geosciences, 95, 75-84.

13.

MacDonald

, Lark

, Taylor

, Abiye

, Fallas

, Favreau

, Goni

, Kebede

, Scanlon

, Sorensen

, Tijani

, Upton

, & West

(2021). Mapping groundwater recharge in Africa from ground observations and implications for water security. Environmental Research Letters, 16(3), 034012. https://doi.org/10.1088/1748-9326/abd661

14.

Manna

, Murray

, Abbey

, Martin

, Cherry

, & Parker

(2019). Spatial and temporal variability of groundwater recharge in a sandstone aquifer in a semiarid region. Hydrology and Earth System Sciences, 23(4), 2187–2205. https://doi.org/10.5194/HESS-23-2187-2019

15.

Maqsoom

, Aslam

, Khalid

, Ullah

, Anysz

, Almaliki

A. A.

, & Hussein

E. E.

(2022). Delineating groundwater recharge potential through remote sensing and geographical information systems. Water, 14(11), 1824. https://doi.org/10.3390/w14111824

16.

Mohammed

S. H.

, Mohammed

M. A. A.

, Karim

, Al-Manmi

, Aziz

B. Q.

, Mustafa

A. I.

, & Szűcs

(2025). Integrating geospatial, hydrogeological, and geophysical data to identify groundwater recharge potential zones in the Sulaymaniyah basin, NE of Iraq. Scientific Reports, 15(1), 9920. https://doi.org/10.1038/s41598-025-94603-z

17.

Mohan

, Western

, Wei

, & Saft

(2017). Predicting groundwater recharge for varying land cover and climate conditions: A global meta-study. Hydrology and Earth System Sciences, 22, 2689–2703. https://doi.org/10.5194/HESS-22-2689-2018

18.

Navane

V. S.

, & Sahoo

(2020). Identification of groundwater recharge sites in Latur district of Maharashtra in India based on remote sensing, GIS and multi‐criteria decision tools. Water and Environment Journal, 35(2), 544–559. https://doi.org/10.1111/wej.12650

19.

Nolan

, Baehr

, & Kauffman

L. J.

(2003). Spatial variability of groundwater recharge and its effect on shallow groundwater quality in southern New Jersey. Vadose Zone Journal, 2, 677–691. https://doi.org/10.2136/vzj2003.6770

20.

Pasquier

, Vahmani

, & Jones

(2022). Quantifying the city-scale impacts of impervious surfaces on groundwater recharge potential: An urban application of WRF–Hydro. Water, 14(19), 3143. https://doi.org/10.3390/w14193143

21.

Salem

, Abduljaleel

, Dezsȍ, J., & Dénes

(2023). Integrated assessment of the impact of land use changes on groundwater recharge and groundwater level in the Drava floodplain, Hungary. Scientific Reports, 13(1), 5061. https://doi.org/10.1038/s41598-022-21259-4

22.

Saraf

, Choudhury

, Roy

, Sarma

, Vijay

, & Choudhury

(2004). GIS-based surface hydrological modelling in identification of groundwater recharge zones. International Journal of Remote Sensing, 25, 5759–5770. https://doi.org/10.1080/0143116042000274096

23.

Tiwari

, Kumar

, & Mukherjee

(2024). GIS and AHP-based groundwater recharge potential zones in urban region: A study of Ajmer City, Rajasthan, India. Urban Climate, 54, 101840. https://doi.org/10.1016/j.uclim.2024.101840

24.

Waldowski

, Sánchez-León

, Cirpka

, Brandhorst

, Franssen

H. H.

, & Neuweiler

(2023). Estimating groundwater recharge in fully integrated PDE‐based hydrological models. Water Resources Research, 59(3). https://doi.org/10.1029/2022WR032430

25.

Wang

, & Liu

(2006). An efficient method for identifying and filling surface depressions in digital elevation models for hydrologic analysis and modelling. International Journal of Geographical Information Science, 20(2), 193–213. https://doi.org/10.1080/13658810500433453

Assessing the Impact of Urban Expansion on Groundwater Recharge Potential Using GIS-based Multi-criteria Analysis: A Case Study of Kahramanmaras¸,Türkiye (2000–2025)