作者: C. Jeganathan , N.A.S. Hamm , S. Mukherjee , P.M. Atkinson , P.L.N. Raju
DOI: 10.1016/J.JAG.2010.11.001
关键词:
摘要: Fine spatial resolution (e.g., <300 m) thermal data are needed regularly to characterise the temporal pattern of surface moisture status, water stress, and forecast agriculture drought famine. However, current optical sensors do not provide frequent at a fine resolution. The TsHARP model provides possibility generate from coarse (≥1 km) on basis an anticipated inverse linear relationship between normalised difference vegetation index (NDVI) land temperature study utilised over mixed agricultural landscape in northern part India. Five variants were analysed, including original model, for their efficiency. Those five global (original); resolution-adjusted model; piecewise regression stratified local model. models first evaluated using Advanced Space-borne Thermal Emission Reflection Radiometer (ASTER) (90 aggregated following resolutions: 180 m, 270 450 630 810 m 990 m. Although sharpening was undertaken resolutions 90 root mean square error (RMSE) <2 K could, average, be achieved only 990–270 ASTER data. RMSE sharpened images data, global, regression, stratification 1.91, 1.89, 1.96, 1.70 K, respectively. yielded higher accuracy, applied sharpen MODIS (1 target resolutions. Aggregated considered as reference respective assess prediction results predicted image 250 3.08, 2.92 1.98 consistently led more accurate predictions by comparison other variants.