A Satellite-based Approach to Evaluate the Impact of Land Use Change on Recharge Rates in the Southern High Plains

A conceptual satellite-based ecohydrological modelling approach was used to evaluate the impact of land use changes on recharge rates in the Southern High Plains. Spatially-distributed recharge rates were estimated as the residual between mean annual values of precipitation (MAP) and evapotranspiration (ET). Evapotranspiration is computed pixel-by-pixel according to the positive/negative anomalies observed between the actual vegetation productivity and an expected rainfall-dependent productivity value. Both anomalies, positive and negative, define a dual-domain of ET-excess and recharge conditions, respectively, in which annual ET is assumed to be linearly correlated with local precipitation and the vegetation productivity anomaly. The Enhanced Vegetation Index (EVI) from MODIS/Terra was used as a surrogate for vegetation productivity and a regional EVI-MAP relationship was built using topographic and land cover criteria to obtain the expected rainfall-dependent productivity values. A time-series of the EVI product (2000-2009), a 60 m digital elevation model and mean annual precipitation maps from the PRISM database were integrated into a Geographic Information System and used as inputs to the model. The ability of the satellite-based approach to estimate ET and recharge rates was tested against independent estimates of recharge and ET rates derived from unsaturated chloride soil profiles and Large Aperture Scintillometer measurements, respectively. Results were spatially combined and differences in recharge rates among representative land cover types (drylands, natural rangelands and, rain-fed and irrigated areas) were highlighted and discussed.