On the role of soil water storage capacity and soil nutrients on tree growth of selected tree species in Central Europe.

While numerous correlational studies on the impact of climatic variation on tree ring formation consider plant functional traits, masting cycles and stochastic disturbances, the role of soil properties is frequently neglected due to insufficient data availability. Using a homogenous dataset of increment cores collected in mature stands at 1562 different forest sites with detailed plot specific climate, stand and soil information in the province of Styria in Austria we are focusing on the role of soil water storage (plant available water capacity – AWC) and soil nutrient status on tree ring formation. The study area covers a wide altitudinal and climate gradient with mean annual temperatures ranging from 2.1°C to 10.2°C and mean annual precipitation ranging from 695 mm to 2024 mm. Generalized Additive Mixed Models (GAMM) for annual tree ring width over 38 years (1980 to 2018) were fitted for six tree species (Abies alba, Fagus Sylvatica, Larix decidua, Picea abies, Pinus sylvestris, Quercus robur/petrea). Individual tree characteristics, stand attributes, general site characteristics (terrain information calculated from high resolution ALS), downscaled climate information in high temporal resolution, and soil information were used as independent variables. Soil data was derived from a morphological description of 80 cm soil pits dug on each corresponding forest site. Via pedo-transfer-functions (using functions available from the literature as well as derived from laboratory analyses of approximately 25% of the pits), soil characteristics such as AWC and soil nutrient status were calculated.

In a two-step procedure, we first developed a general tree growth model, including solely tree and stand attributes (e.g. age, competition) and general site specific information available in high spatial resolution (e.g. slope, aspect, irradiance, mean annual temperature, precipitation). Subsequently, we added soil attributes to the model and checked for their effect on model parameters. AWC and soil nutrient status do have significant influence on tree ring formation when added to the tree ring model. However, this effect varies amongst tree species. The results are consistent with tree species specific traits as available from literature: E.g. deep rooting species like Quercus benefit more from high water storage capacity than shallow rooting Picea abies; the effect of soil nutrient status is most pronounced for Fagus sylvatica, which has high nutritional requirements and more negligible for Pinus sylvestris, with low nutritional requirements. As soil formation itself depends upon geological substrate, landform and climate, the improvement of model quality when adding additional soil information is moderate.