Toward reliable model-based soil moisture estimates for forest managers.

The study presents a daily water balance model with a focus on quantifying drought intensity and duration in different forest stands within the Level II monitoring sites in Saxony. The model requires climatic data input and various site and stand parameters, which mostly have physical meaning and can be measured. We estimated the van Genuchten soil parameters of the model by different pedotransfer functions. Water stress is assumed to occur when the relative extractable soil water (REW) or the ratio of real transpiration and potential transpiration or pressure head drops below a certain threshold. The model was not calibrated to test the robustness of physically based parameters but was validated with available soil moisture measurements at forest climate stations (grass vegetation) and inventory locations in various vegetation and climatic conditions with Kling-Gupta efficiencies > 0.5 at most of the sites. The model captures the dynamics of soil water depletion and recharge in a study period well. Day-to-day estimates of water balance components allow us to calculate the duration and intensity of drought events and derive stress indices. A dendro-ecological application is presented: a retrospective analysis of the effects of drought on radial tree growth. Some limitations and potential of applications of the model are discussed. [ABSTRACT FROM AUTHOR]