Spatial variations in soil moisture in temperate forest independent of topographic moisture indices, yet ERA5-Land retrievals accurately reflect their temporal variations

Soil moisture is crucial for ecosystem functioning, as it influences biological and biogeochemical processes. It regulates water, energy, and carbon cycles, playing a key role in ecosystem organization, biodiversity, and vegetation resilience. However, soil moisture dynamics are increasingly impacted by climate change. Shifts in precipitation patterns, rising temperatures, and intensifying droughts are amplifying spatio-temporal variability. These challenges highlight the need for reliable representations of soil moisture, to enable adaptive forestry practices, and strategies to mitigate ecosystem vulnerabilities. In general, topographic models are considered as reliable sources for representing soil moisture.Extensive research has validated topographic modelling of soil moisture, but most studies have focused on northern regions, leaving a scarcity of empirical data for Central Europe. This study investigated five sites in temperate forests of Germany, dominated by cambisols, with over 2,000 measurement locations. The objectives were to (1) analyse the spatio-temporal variability of soil moisture, (2) examine correlations with topographic indices under varying seasonal conditions, and (3) validate soil moisture estimates provided by the ERA5-Land dataset.The results indicated that temporal variability in soil moisture was approximately 3.6 times greater than spatial variability. Flow-accumulation-based indices were poor predictors of spatial moisture patterns. The variability explained (R²) by indices such as the depth-to-water index ranged between 1% and 4% only and did not align with expected seasonal trends. Mesorelief, represented by the topographic position index, showed weak but consistent correlations at selected sites. Temporal variations in soil moisture were effectively captured by ERA5-Land reanalysis data, with site-specific adaptations yielding R² values of up to 98%.These findings reveal the limitations and potential applications of soil moisture modelling. Moreover, they can contribute to improving soil–plant–atmosphere models and inform sustainable forest management strategies in the context of a changing climate.