Monitoring land degradation risk using ASTER data:: The non-evaporative fraction as an indicator of ecosystem function

There is a need to develop operational land degradation indicators for large regions to prevent losses of biological and economic productivity. Disturbance events press ecosystems beyond resilience and modify the associated hydrological and Surface energy balance. Therefore, new indicators for water-limited ecosystems can be based on the partition of the Surface energy into latent (lambda E) and sensible heat flux (H). In this Study, a new methodology for monitoring land degradation risk for regional scale application is evaluated in a semiarid area of SE Spain. Input data include ASTER surface temperature and reflectance products, and other ancillary data. The methodology employs two land degradation indicators, one related to ecosystem water use derived from the non-evaporative fraction (NEF=H/(lambda E+H)), and another related to vegetation greenness derived front the NDVI. The surface energy modeling approach used to estimate the NEF showed errors within the range of similar Studies (R-2=0.88; RMSE=0.18 (22%)). To create quantitative indicators suitable for regional analysis, the NEF and NDVI were standardized between two possible extremes of ecosystem status: extremely disturbed and undisturbed in each climatic region to define the NEFS (NEF Standardized) and NDVIS (NDVI Standardized). The procedure Was Successful, as it statistically identified ecosystem status extremes for both indicators without supervision. Evaluation of the indicators at disturbed and undisturbed (control) sites, and intermediate Surface variables such as albedo or surface temperature, provided insights on the main Surface energy Status controls following disturbance events. These results suggest that ecosystem functional indicators, such as the NEFS, can provide information related to the surface water deficit, including the role of soil properties. (C) 2008 Elsevier Inc. All rights reserved.