A methodological approach for integration of spatial data from remote sensing with time series of local characteristics of near surface boundary layer of atmosphere is proposed. The approach is implemented by the example of satellite mapping of risks (probabilities) of urban surface overheating. Knowledge of these risks is necessary for assessing the economic losses from traffic disruption as a result of road cover softening. The urgency of the problem is caused by the global warming. St. Petersburg was chosen as the object of study. Materials for the study were Landsat scenes and the time series of standard observations of few meteorological stations. The theoretical foundations of satellite mapping of the risk of urban surface overheating are given. The analysis of the sustainability of proposed algorithm of risk assessments was performed depending on the choice of the reference meteorological station and on the number of satellite images. The result was a map of the risk of urban surface overheating. It was shown that industrial zones were characterized by highest risk of overheating, while recreational zones, as well as the territories built up with five-story buildings had minimal risk. The method can be used in supporting management decisions aimed to parry threats of overheating as a result of global warming. It is concluded that the proposed approach is highly effective and cost efficient.