Land surface temperature (LST) is one variable in the earth-atmosphere interactive system. Temperatures are naturally high in tropical environments under continental influence. In Brazil, the problem of high temperature is intensified by the urbanization process, which is characterized by vegetation removal, concentration of buildings, and the use of building materials unfavorable to the climate in the tropics. Therefore, in this article, we analyze the spatial variability and the frequency of the surface temperature during the dry and rainy seasons in Penápolis, a small population city in the countryside of São Paulo State. The analysis was performed by establishing the relationship between the normalized vegetation index (NDVI) and the surface temperature in order to contribute to urban planning. The adopted procedure used 10 images from satellite Landsat 8, Operational Land Imager (OLI) and Thermal Infrared Sensor (TIRS) for the year 2018. The results showed that the surface urban heat island has significant seasonal variations that are directly associated with the rainfall regime of the region. The intensity was not elevated during the year, and the highest magnitudes occurred in the summer, which is a hot and rainy season. In the winter, being the dry season, the magnitudes were lower. The frequency of heat island intensity taken from 10 points of the images is related to construction density, building materials, and the vegetation cover index. These results are important to understanding the generation of heat in the city and proposing planning measures for the use and coverage of urban land.