It is known that topography is an important factor controlling the variation of road surface temperature (RST). In order to explore possible relationships between RST and topography, the authors used Thermal Mapping data obtained by a vehicle-mounted thermometer in a mountainous area in Nevada, USA, under different weather conditions in December 1994. The data were validated against the measurements of road surface sensors. After that, a step-wise regression technique is employed to find out possible statistic relationships between RST and altitude at different surveying routes. It was found that the relationships exhibited different characteristics in different climate domains and under different weather types. In most cases, the relationships were non-linear. In order to obtain more information about the cause of the residuals and the variation, the error of regression fitting is related to near-ground meteorological parameters (air temperature, dew point and wind, etc.). This study proved that Thermal Mapping is a reliable and effective method to display spatial variation of road surface temperature.