Economic cost of Plateaus
Description
Title: Integrated Python Pipeline for Geospatial Assessment of Groundwater Recharge Loss from Urban Expansion on Laterite Plateaus, Goa, India (2016–2026) Description: This code implements the complete geospatial and economic methodology used in the study "Environmental Cost of Urban Expansion in Goa: A Geospatial Valuation of Laterite Plateaus," quantifying groundwater recharge loss attributable to urban expansion on Goa's laterite plateaus between 2016 and 2026. The pipeline runs as a single self-contained Jupyter notebook (Python 3) and executes four sequential analytical stages in one session: LULC Mapping and Change Detection — processes Dynamic World land-cover composites (Google Earth Engine, 10 m resolution) for 2016 and 2026, reprojects them to a common analysis grid (EPSG:32643, 10 m), masks to the Goa state boundary, and quantifies land-cover class areas, urban gain/loss, source-class transitions, and plateau-specific urbanisation statistics. Recharge Potential Index (RPI), Recharge Coefficient (Rc), and Recharge Volume Estimation — reclassifies six recharge-controlling factors (rainfall, LULC, lineament density, slope, drainage density, and soil hydrological group) to a common 1–5 scale, computes an AHP-weighted RPI and derived Rc for both periods, and estimates annual recharge volume (state-wide, urban-gain-attributable, and plateau-specific zonal statistics). Spatial Statistical Analysis and Sensitivity/Uncertainty Analysis — applies Global and Local (LISA) Moran's I and Getis-Ord Gi* hotspot analysis to the recharge-volume-change surface to identify statistically significant spatial clusters of recharge loss, and performs one-at-a-time sensitivity analysis on the three primary model inputs (Rc, rainfall, AHP LULC weight) to quantify uncertainty in the reported recharge-volume estimates. Economic Valuation of Recharge Loss — converts the estimated recharge-volume loss into monetary terms using three independent valuation methods (replacement cost, water tariff, and avoided cost), triangulates the results, computes 20–30 year net present value under three discount-rate scenarios, and produces a spatially explicit "recharge debt" map. Input data required (not included; user-supplied): Dynamic World LULC composites (2016, 2026; generated via the included Google Earth Engine script), gridded rainfall, lineament density, slope, drainage density, and soil hydrological-group rasters, and Goa state boundary and laterite plateau shapefiles. Outputs: Reprojected and reclassified factor rasters; RPI, Rc, and recharge-volume rasters (2016, 2026, change) for the full study area and boundary-masked to Goa; urban-gain raster; LISA and Gi* cluster rasters; spatial economic "recharge debt" rasters; sensitivity-analysis results (CSV); diagnostic and results figures (PNG); and a complete plain-text run log capturing every printed intermediate and final result for full reproducibility and auditability.