Effects of microtopography and moisture on peatland soil temperature regime (a case study of the ridge-hollow complex Mukhrino bog)

Northern peatlands represent one of the largest terrestrial carbon reservoirs, playing a crucial yet potentially vulnerable role in the global carbon cycle under ongoing climate change. The stability of this vast carbon stock is intrinsically linked to the thermal regime of the peat soil, which controls key biogeochemical processes such as microbial decomposition, methane production, and plant productivity. However, peatlands are not thermally uniform; they are characterized by a pronounced microtopography, typically featuring a mosaic of elevated, drier features (ridges, hummocks) and water-saturated hollows. Understanding the spatiotemporal dynamics of temperature within this micro-landscape is therefore fundamental for accurate prediction of peatland response to warming. While the general influence of microrelief on temperature is recognized, there is a significant lack of detailed, high-frequency, and multi-year datasets that simultaneously capture the thermal behavior of all key microform elements in continental boreal peatlands, particularly in the vast and critically important region of Western Siberia. This study aims to fill this gap by providing a comprehensive, quantitative analysis of the soil and surface temperature regime in a typical ridge-hollow complex. The research was conducted at the Mukhrino Bog, a large oligotrophic mire in the Middle Taiga zone of Western Siberia. The study site features a classic ridge-hollow complex (RHC) with well-defined shrub-Sphagnum-dominated ridges (1-3 m wide, up to 60 cm high) and water-saturated sedge-Sphagnum hollows. To investigate the temperature regime, a dedicated monitoring system was deployed in July 2020. The setup included an array of automatic temperature sensors (DS18B20) installed along a 27-meter transect crossing a sequence of a northern hollow, a ridge, and a southern hollow. A total of 11 soil temperature profilers were placed on characteristic microfeatures (hummocks, depressions, slopes), each measuring temperature at depths of 0, 2, 5, 10, 15, 20, 40, and 60 cm. One additional deep probe monitored temperatures down to 320 cm at a ridge location. Air temperature was measured at 2 m and 15 cm above the surface. Data were logged hourly from July 2020 to November 2022. Complementary field measurements included precise leveling of the microrelief and, in April 2023, sampling of frozen peat monoliths from both a ridge and a hollow for subsequent laboratory determination of natural moisture content and absolute dry peat density. The three-year monitoring period captured a significant range of meteorological conditions, including an extremely cold and low snow accumulation winter in 2020/2021 and a milder with high snow accumulation winter in 2021/2022, as well as contrasting summer conditions. The results reveal a persistent and clear spatial pattern: the saturated hollows were consistently warmer than the elevated ridges throughout the annual cycle. However, the underlying physical drivers of this thermal contrast were seasonally distinct. During the warm season (April-October), the primary mechanism is the difference in the thermophysical properties of the peat. Water-saturated hollow peat has high volumetric heat capacity and thermal conductivity. This leads to efficient absorption, deeper penetration, and slower release of heat. Consequently, diurnal temperature amplitudes are strongly dampened with depth in the hollows. In contrast, the aerated, drier peat of the ridges has a high pore air content, which acts as an effective insulator. This results in extreme surface heating (up to +34.4°C) and cooling, but a very sharp attenuation of these fluctuations with depth. The ridge peat essentially functions as a "reverse thermos," inhibiting heat transfer into the deeper layers. Data from July 2020, the warmest month, quantitatively illustrate this: the mean monthly surface temperature was +21.2°C in the northern hollow versus +19.6°C on the ridge. More importantly, the temperature difference increased with depth, reaching +2.2°C at 60 cm. The southern hollow was consistently 1-2°C warmer than the northern hollow, likely due to higher water saturation. In winter (November-March), the dominant factor shifts to the snow cover distribution. The microtopography dictates snow accumulation: deeper snowpacks form in the hollows, while wind exposure keeps the ridges relatively snow-free. Snow is an excellent insulator. Therefore, the thick snow layer over the hollows effectively decouples the soil from the extreme cold air temperatures, maintaining surface temperatures close to 0°C. Over the ridges, the thin snow cover provides minimal insulation, leading to intense soil cooling. In December 2020, the mean surface temperature on the ridge was -9.8°C, while in the hollow it was only -0.9°C—a difference of 9°C. This snow-mediated effect directly controls the depth of seasonal frost. During the harsh winter of 2020/2021, the frost depth reached 60 cm on the ridge but only about 30 cm in the hollow. In the following, milder winter, maximum frost depths were ~55 cm and ~12 cm, respectively. The latent heat released during freezing of the water-saturated hollow peat further moderates cooling. High-resolution data from a 13-day period in June 2022 further elucidated the diurnal dynamics. Under cloudy, rainy periods, temperatures were uniform across the microrelief. With the onset of clear, anticyclonic conditions, strong diurnal contrasts emerged. During the day, the northern hollow heated most strongly (up to +33.5°C), while shaded areas on the ridge remained cooler. At night, the hollows (especially the southern one) cooled more slowly than the ridge, maintaining a higher temperature. These near-surface patterns persisted to a depth 10 cm. At 20 cm depth, diurnal cycles were almost absent in the hollow but remained pronounced (5-7°C amplitude) on the ridge. Below 40 cm, diurnal variations vanished everywhere, revealing the persistent background spatial pattern: warmer hollows and cooler ridges. Analysis of peat properties confirmed the foundational differences: the hollow peat maintained a very high natural moisture content (94-98%), while the ridge peat showed lower and more variable moisture (85-95%) and slightly higher dry density in the surface layer. In conclusion, this study demonstrates that the thermal regime of a boreal peatland is governed by a dynamic interplay between microtopography, moisture content, and snow cover, with seasonally switching dominant mechanisms. The water-saturated hollows act as thermally buffered, energy-accumulating elements, while the aerated ridges experience thermal extremes and function as insulators. The spatial pattern of temperature—warmer hollows, cooler ridges—is a robust feature sustained year-round. The quantified relationships and the extensive dataset presented here are essential for improving process-based models of heat and water transfer in peatlands. This, in turn, enhances our ability to forecast the fate of the massive carbon stored in these ecosystems under changing climatic conditions, particularly for the extensive and vulnerable peatlands of Western Siberia. The observed thermal heterogeneity underscores the necessity of representing microtopographic diversity in landscape-scale models to prevent significant biases in forecasting carbon cycle feedbacks.