This study assessed the relationship of Normalized Difference Vegetation Index (NDVI) and Normalized Difference Water Index (NDWI) with land surface temperature (LST), using Landsat images of La Vieja river watershed, in the Andean region of Colombia. We evaluated land covered by Andean forest, forest plantation, coffee, pasture and urban area. Negative correlations between indices and LST were identified, and significant differences (p<0.01) between land cover types were found. Pasture presented mean LST greater than urban area, and coffee lower than forests and plantations. Transitions from pasture to Andean forest, plantation and coffee showed decreases in LST up to 9 °C, from urban area to coffee up to 8 °C, while from urban area to pasture increased up to 1.28 °C. We concluded that pasture can have higher capacity than urban area to promote heating of land surface, and that coffee is similar to forests and plantations in its cooling capacity. It was suggested that there is to pay attention not only to the urban heat islands but to the rural heat islands associated to livestock activities that have not been deeply studied.
Normalized Difference Vegetation Index (NDVI); Normalized Difference Water Index (NDWI); Remote Sensing; Heat Island; Landsat
