Spatial and temporal patterns of near‐surface ground temperature in the Arctic mountain catchment

• Published in: Land Degradation & Development
• Main Authors: Kasprzak, Marek, Szymanowski, Mariusz
• Other Authors: Narodowe Centrum Nauki
• Format: Article in Journal/Newspaper
• Language: English
• Published: Wiley 2023
• Subjects: Arctic; Spitsbergen
• Online Access: http://dx.doi.org/10.1002/ldr.4841; https://onlinelibrary.wiley.com/doi/pdf/10.1002/ldr.4841

Abstract We consider the case of a mountain catchment representative of southwest Spitsbergen to investigate the spatial and temporal variation in temperature of the near‐surface ground. We set up 20 thermistor strings distributed in different parts of the catchment, combine the measured data with land surface temperature (LST) retrieved from Landsat imagery, and verify the depth of ground thawing using electrical resistivity tomography. Under current climatic conditions, the thickness of the active layer (ALT) is at least 1.5 m, and in the lower parts of the catchment, between 3.5 and 4 m. The maximum ground surface temperature in the study area can be 27°C, and the temperature at a depth of 1.5 m, 5.4°C. We identified unfrozen ground (taliks) to a depth of 10–30 m in zones of water concentration at foot slopes and under river channels. The persistence of year‐round taliks at lake shores is possible. We verify the hypothesis that topographic parameters significantly determine the spatial variation of near‐surface ground temperature. For each level of temperature aggregation (from 6 h to 12 months data) and each examined the depth of the active layer, it is possible to identify environmental variables—related to energy conditions, heights, terrain relief, and ground surface moisture—significantly correlated with ground temperature, and consequently to specify multiple regression models. Topographic exposure and LST (used only in daily models), altitude, and various approximations of relative height are crucial in their construction. The ground temperature variance, unexplained in the regression analysis, prompts recognition of the critical role of factors not included in the modelling (groundwater, soil structure).