| Summary: | The spatial and temporal variations in soil moisture can cause natural hazards, such as droughts and floods. Soil moisture and its mediating environmental factors are fairly unknown from the point of view of climate change, even though climate is a major controlling factor. A significant feedback loop between climate and soil moisture affects various ecosystem processes, especially in the Arctic. Soil moisture is fairly fast and simple to measure in the field, but even so it has not been the subject of comprehensive research. Especially the temporal changes in soil moisture in the Arctic are poorly understood. In low-energy systems, such as Arctic-alpine regions, soil moisture is an important limiting factor for vegetation. This study aims to understand which factors mediate soil moistures spatial variation and temporal change. In addition, one of the objectives of this study is to determine the importance of soil moisture for vascular plant distribution at the landscape scale. The data comprehends 945 individual square meter plots from a 130 km² research area in northern Norway, where relative elevation reaches almost 1000 meters. The variable relief covers a vast range of extreme environmental gradients over short distances, such as a moisture gradient from dry, sun-baked slopes to water-logged peatlands and a vegetation gradient from rugged mountain tops to flood-meadows rich in species. The data consists of in situ field measurements and observations as well as data based on topography, remote sensing and climate models. Factors controlling the spatial variation and temporal change of soil moisture are studied with multivariate models (GLM: Generalized linear model and GAM: Generalizes additive model) at different scales. The importance of soil moisture for vegetation is examined by using four modelling techniques (GLM, GAM, GBM: Generalized boosted regression model, and RF: Random forest) within BIOMOD, an ensemble platform for species distribution modelling. Soil moisture is mediated by various factors, ...
|
|---|