Soil Moisture Variations From Boreal Forests to the Tundra

Abstract

Soil moisture has a profound influence on life on Earth, and this vital water resource varies across space and time. Here, we explored soil moisture variations in boreal forest and tundra environments, where comprehensive soil moisture datasets are scarce. We installed soil moisture sensors up to 14 cm depth at 503 measurement sites within seven study areas across northern Europe. We recorded 6,138,528 measurements to capture soil moisture variations of the snowless season from April to September 2020. We described the spatio‐temporal patterns of soil moisture, and test how these patterns are linked to topography and how these links vary in space and time. We found large spatial variation and often less pronounced temporal variation in soil moisture across the measurement sites within all study areas. We found that throughout the measurement periods both univariate and multivariate models with topographic predictors showed great temporal variation in their performance and in the relative influence of the predictors within and across the areas. We found that the soil moisture‐topography relationships are site‐specific, as the topography‐based models often performed poorly when transferred from one area to another. There was no general solution that would work well in all the study areas when modeling soil moisture variation with topography. This should be carefully considered before applying topographic proxies for soil moisture. Overall, these data and results highlight the strong spatio‐temporal heterogeneity of soil moisture conditions in boreal forest and tundra environments.

A Gridded Microclimate Dataset from a Sub-Arctic Biodiversity Hotspot in Finland

“…In general, the cross-validation statistics indicated that the models performed well, but also that there was large variability across the response variables ( Table 3 ). On average, the cross-validated R 2 value was 0.61 (median 0.64), which can be considered good when compared to previous studies in high-latitude regions with relatively similar study settings (Aalto et al, 2022; Greiser et al, 2018; Kemppinen et al, 2023). The best performing models had response variables which described summer and early autumn conditions.The worst-performing models were related to the freeze-thaw frequency in the soil and to temporal variability in soil moisture.…”

“…All logger data were quality-filtered and preprocessed (see detailed descriptions of the data cleaning methods in Aalto et al 2022 andKemppinen et al 2023). In summary, 1) we removed all data outside the period during which the logger was in field; 2) we removed all unrealistically high or low temperatures and moisture values; 3) we cross-correlated the time series across loggers aiming to identify unrealistically high sudden peaks (or reversed drops) in the data and removed them; 4) we plotted all individual time series, one site and month at the time, and checked visually whether there were any clearly erroneous data and removed them; 5) if there were data for a logger from a time when the logger was under stable 'office' conditions outside of the field, we used that data to check and correct for potential systematic differences across the three temperature sensors within the logger.…”

“…Microclimate is a multifaceted phenomenon: different seasons are often characterised by distinct microclimate patterns (Bokhorst et al, 2009; Niittynen et al, 2020); soil and air temperatures can be locally decoupled (Kropp et al, 2016); and mean and extreme temperatures may exert contrasting effects on biotic communities (Bokhorst et al, 2015; Parmesan et al, 2000). Another crucial aspect of microclimate is soil moisture, which also displays great local-scale variability due to the strong topographic control over top-soil moisture (Burt and Butcher, 1985; Kemppinen et al, 2023; le Roux et al, 2013). Arctic and alpine regions, characterised by overall cold temperatures, strong seasonality, and complex terrains, may differ from other ecosystems in regard to the main drivers of microclimate variability (Aalto et al, 2022; Opedal et al, 2015).…”

Modelling spatio‐temporal soil moisture dynamics in mountain tundra

“…They also have important local ecosystem impacts. Spatio‐temporal variations of soil moisture are an important driver of greenhouse gas fluxes (Lohila et al, 2016; Virkkala, 2020) as well as fine‐scale patterns of vegetation properties (Kemppinen, 2020; le Roux et al, 2013). Therefore, through various feedback mechanisms, soil moisture in the tundra plays an important role in global change and its accurate predictions are fundamental to our ability to understand tundra ecosystems now and in the future.…”

Summer litter decomposition is moderated by scale‐dependent microenvironmental variation in tundra ecosystems

“…Local‐scale experimental warming was found to increase decomposition (Björnsdóttir et al 2021), yet microclimate and topography were not found to explain the variation decomposition in a tundra landscape in Western Greenland (von Oppen et al 2022), highlighting remaining questions about the controls of tundra litter decomposition across scales. Tundra ecosystems exhibit heterogeneity in both in vegetation patterning and in local environmental conditions – for example, unique topographical and soil conditions driven by permafrost dynamics and seasonal snowmelt dynamics (Elmendorf et al 2012a, b2012a, Myers‐Smith et al 2015, Davidson et al 2016, Ackerman et al 2017, Bjorkman et al 2018, Kemppinen et al 2023). Therefore, we may expect to see variance in decomposition explained by regional macroclimate, and some explained by local‐scale conditions.…”