SPATIAL VARIATION OF ROOT ZONE SOIL TEMPERATURE IN THE NORTHERN TAIGA OF WEST SIBERIA

Studies of soil heat regime are commonly confined to one vertical soil profile. Some studies on the basis of short-term temperature data series reported significant horizontal variation of topsoil horizons due to spatial variation in tree stand densities, plant cover and soil cover heterogeneity. Introduction of programmed automated temperature data loggers facilitated studies of horizontal heterogeneity in soil temperature, i.e. so called temperature fields with long temporal data series. However horizontal variation of soil temperature has not practically been studied in most Russian ecosystems, including those in the West Siberian northern taiga zone. The study of soil temperature regime was carried out in the environs of Noyabrsk (Yamalo-Nenetsky Autonomous Region, Russia) in the four key study sites: climax larch (Larix sibirica Ledeb.) and Scots pine (Pinus sylvestris L.) forests, and flat palsa mire (separately palsa and adjacent hollow). On each study site automated temperature data loggers were used in 5 replicated subsites to measure soil temperature with 0.5°С and record it every 4 hours, i.e. at 1, 5, 9, 13, 17 and 21 o'clock, daily. The loggers were placed at 10 cm soil depth in the root zone. The data from the meteorological station in Noyabrsk were also used in the study. The standard deviation of temperature values, recorded in a study site at one time point, was used as a measure of temperature variation. In all study sites the maximal temperature variation was observed at the beginning of the warm season. In both forest sites soil temperature began varying already after snow thawed, which was most likely due to the heterogeneity of vegetation cover and forest litter; in the pine site soil the heterogeneity of direct solar radiation as a result of the crown shadowing also contributed to temperature variation. The biggest soil temperature variation during this period was observed at the pine forest site, with difference between simultaneous replicated measurements reaching at certain time points as much as 9.6ºС. In the larch study site the corresponding difference in June did not exceed 7-8ºС. The minimal spatial variation of soil temperature in the root zone was observed in hollows of the flat palsa mire site, most likely due to the absence of the microrelief influence and high water saturation. The highest standard deviation values were observed during snow thawing at the beginning of May, when temperature difference between replicated measurements could be as high as 9.5ºС. In June soil temperature variation decreased, staying low during the rest of the year. The palsa soil temperature at the flat palsa mire site and its variation in winter time showed strong response to changing air temperature. This winter variability was most likely due to unevenness of the snow cover thickness, resulting from the pronounced surface microrelief. During show thawing period the root zone temperature varied less, apparently because of even penetration and distribution of snowmelt water throughout the studied soil layer. Rapid soil warming resulted in increased temperature variability, which remained high during entire summer and the beginning of autumn, with differences between replicates being as high as 10.1ºС. Summer variability of soil temperature at this site could be attributed to the permafrost layer, which in different microrelief positions is located at different depth as related to the root zone, thus exerting different influence on the temperature of the soil layers above. The most pronounced effect of diurnal fluctuations of root-zone temperature on its spatial variability was observed in pine forest during entire summer period, whereas on the palsa mire ecosystem it was observed at the beginning of summer during intensive thawing of the seasonally frozen layer.