Land Surface Temperature (LST) is a key parameter in the energy balance model. However, the spatial resolution of the retrieved LST from sensors with high temporal resolution is not accurate enough to be used in local-scale studies. To explore the LST–Normalised Difference Vegetation Index relationship potential and obtain thermal images with high spatial resolution, six enhanced image sharpening techniques were assessed: the disaggregation procedure for radiometric surface temperatures (TsHARP), the Dry Edge Quadratic Function, the Difference of Edges (Ts∗DL) and three models supported by the relationship of surface temperature and water stress of vegetation (Normalised Difference Water Index, Normalised Difference Infrared Index and Soil wetness index). Energy Balance Station data and in situ measurements were used to validate the enhanced LST images over a mixed agricultural landscape in the sub-humid Pampean Region of Argentina (PRA), during 2006–2010. Landsat Thematic Mapper (TM) and Moderate Resolution Imaging Spectroradiometer (EOS-MODIS) thermal datasets were assessed for different spatial resolutions (e.g., 960, 720 and 240m) and the performances were compared with global and local TsHARP procedures. Results suggest that the Ts∗DL technique is the most adequate for simulating LST to high spatial resolution over the heterogeneous landscape of a sub-humid region, showing an average root mean square error of less than 1K.