Mapping and assessing ecosystem service hotspots and bundles in rapidly urbanizing, data-scarce regions: a case study of Kabul, Afghanistan

Context Urban expansion in the Global South challenges ecological sustainability. Understanding ecosystem service (ES) distribution and bundling in these rapidly changing landscapes is vital for informed urban planning and resilience. However, there is a research gap in applying integrated, high-resolution ES assessments to data-scarce, informally developing urban contexts. Objectives This study investigates ES patterns and trade-offs in the Kabul metropolitan region (KMR) in Afghanistan, a fast-growing, data-scarce city. Objectives are to: (1) map spatial patterns, including hotspots and coldspots, of five key ESs (water yield, heat mitigation, stormwater runoff retention, nature access, habitat quality), and (2) identify distinct ES bundles and their associated synergies and trade-offs. Methods The study focuses on KMR, as a representative case of a rapidly urbanizing, data-scarce region facing significant environmental pressures. Using multi-source data, a high-resolution land-use/land-cover map is created for ES modeling (InVEST), hotspot analysis (Getis-Ord Gi*), ES bundle identification (k-means clustering), and conservation prioritization. Results The assessment reveals significant spatial heterogeneity across five key ESs. ES hotspots corresponded to green/blue infrastructure, while coldspots are prevalent in dense urban areas. Distinct ES bundles are mapped, revealing context-dependent synergies (e.g., heat mitigation, stormwater runoff retention) and trade-offs (e.g., water yield, stormwater runoff retention). Finally, conservation priority areas are mapped based on multi-service supply. Conclusions This study provides spatially explicit information for KMR's sustainable urban planning, using a procedure transferable to other cities in the Global South. Findings enable targeted interventions to enhance urban resilience by identifying priority zones for ES protection. Future research should prioritize model validation with local data.