Underlying surface temperature is an important parameter of underlying surface thermal radiation and can be used in monitoring forest fires, coal fires, urban thermal radiation and developing climate models. Ground-based observations provide temperature information for small areas around weather stations and in fact cannot provide a high density of surface temperature data. Remote sensing technologies are promising in this respect. However, due to the low spatial resolution in the infrared channel, the surface temperature calculated from Landsat and Aster images does not always have the required detail needed when studying small areas. The results of images from Sentinel 2A and Landsat 8 satellites combination (joint digital processing) made in order to increase spatial resolution of underlying surface temperature are presented. Comparison of surface temperature extreme values shows that in spite of small difference in extreme values of temperature, the spatial field of temperature in case of combined images was more detailed and variable. This is evidenced by a significant increase in the variability of the temperature standard deviation. Direct visual observations of image fragments also confirm that combining Sentinel 2A and Landsat 8 images increases the spatial resolution of the surface temperature when compared to the Landsat 8 image