Satellite mapping of air temperature under polar night conditions

Abstract

A novel satellite technique for air temperature mapping with a spatial resolution of ~100 m was proposed for the town of Apatity, the Kola Peninsula (Russia), taken as a case study. The main idea behind this novel technique is to find the statistical relationships between the land surface temperatures in each point of the study area, observed by multiple infrared thermal satellite imagery, and the time series of air temperatures recorded by World Meteorological Organization (WMO) weather station. Fourteen scenes of infrared thermal spectral band of the Landsat satellites for the period 2014–2019 were used, as well as the long-term time series of air temperature from the weather station and the results of air temperature observations carried out by the network of loggers. For calm weather conditions, according to the ground truth, the error of air temperature mapping was σ = 1.5°C, and the precision was estimated as δ = 1.0°C. An analysis of the compiled air temperature map showed that, under polar night conditions, the air temperature on the hilltops was by 10–18°C higher than in the lowlands. It was concluded that, for economic reasons, as well as for the reasons of population health protection in the Arctic, it would be advisable to plan the placement of new cities on the hills. Each of these new areas should be designed in a “semi-isolated” manner in order to minimize the time needed by the local people for crossing the lowlands between the nearby districts. A characteristic feature of modern megalopolises is their internal structure formed by the growing primary settlements that can be considered as nuclei interlinked by transportation routes. Thus, the new Arctic cities can be called “Arctic megalopolises” because of their internal structure that is specific to megalopolises.