Is neglecting soil variables in tree growth studies justified?Trifković, Vasilije (Avtor) Bončina, Andrej (Avtor) reference soil groupsbasal area incrementtree-growth modellingsessile oakpedunculate oakTurkey oakWe analysed differences in growth among Quercus petraea, Q. robur, and Q. cerris in natural forests across 16 reference soil groups. The study encompassed the forested area of Slovenia (11.8 thousand km2), with soil information derived from the national soil map at a scale of 1:25,000. Oak growth was evaluated using periodic basal area increment data from 52,333 trees measured on 17,733 permanent sample plots. The effects of site, stand, and individual tree characteristics on growth were examined using linear mixed-effects models. Soil unit was included as a dummy variable. Soil contributed 16.3, 11.2, and 14.0% to the explained variability in basal area increment of Q. petraea, Q. robur, and Q. cerris, respectively. The difference between the highest and the lowest mean basal area increment across soil units for Q. petraea, Q. robur, and Q. cerris amounted to 70, 39, and 29% of their mean basal area increment, respectively. The three oak species reached the highest and lowest growth rates on different soil units. Q. petraea performed best on Dystric and Eutric Planosols, Q. robur on Eutric and Dystric Gleysols, and Q. cerris on Haplic Luvisols. This study demonstrates that soil-related factors represent an important component of growth modelling and should be evaluated alongside tree, stand, site and climatic variables.20262026-06-19 04:13:52Članek v reviji183780sl