Tree and stand growth differ among soil classes in semi-natural forests in central Europe

We determined the size of differences in stand and tree growth in semi-natural forests with respect to 16 reference soil groups. The forest area of Slovenia (11.8 thousand km2) was used as the study area, and reference soil units were derived from the national soil map at a 1:25,000 scale consisting of 10,781 polygons with an average size of 117.95 ha. Stand growth was defined as periodic stand basal area increment, while the growth of Norway spruce, silver fir, Scots pine, European beech and sessile oak trees was estimated by the periodic diameter increment of 238,349 dominant trees on 67,061 permanent sampling plots. A linear fixed-effects model and linear mixed-effect models were used for studying stand and tree growth in different site, stand and tree conditions. The soil unit was the dummy variable with Dystric Cambisols set as the reference category. Soil contributed 4.3 % to the explained variability of basal area increment and 4–27 % to the explained variability of the diameter increment of the five tree species. Soil was a stronger driver of stand and tree growth than climate or topography. Stand and tree species production rate on soil units was in the interval of −28 % to +5 % and −47 % to +14 % of that on the reference soil unit, respectively. Stand growth was the highest on Eutric Gleysols and the lowest on Histosols, and tree species generally exhibited the highest and the lowest growth rates on different soil units. We suggest that soil should be considered in growth models and studied interrelatedly with climatic, site and stand variables