In ecological niche models (ENM), the soil regularly plays a subordinate role compared to the climate. The German Forest Soil Inventory (BZE II) offers an ideal database for the realistic assessment of the contribution of soil properties to the habitat specificity of tree species. The factors of soil chemistry were the C/N ratio and the base saturation (BS), the available water capacity (AWC) as soil physical parameters, the average temperature in the vegetation period (Tmp) and the climatic water balance (CWB) as climate and water budget parameters. Among the possible modeling techniques, Maxent proved to be particularly suitable, since this machine learning method requires only a few observations and the aim was to analyze the habitat specificity of as many tree species as possible. Only for eight out of 40 tree species too few observations were available for modeling. In 20 cases, the overall model quality was low, especially because of a weak limitation of tree species due to the climate (period 1970–2000) and the prevailing soil conditions within Germany (soil indifferent tree species). According to the models, there is a strong dependency on soil chemistry for 17 tree species. The soil was hardly relevant for three tree species compared to the climate. The impact of AWC was generally small. For most tree species, the occurrences were more determined by the C/N ratio than by BS. For the tree species most dependent on the soil, however, the variable importance of the BS was similarly large. The strong influence of the C/N ratio is also due to the recycling of nutrients to the soil humus via the litter and is therefore not a purely soil-based effect. The results are fundamentally in line with previous assessments of tree species requirements. However, deviating findings as well as diverging expert opinions on habitat specificity underscore the need for data-based methods to objectify and readjust our knowledge. With the help of databased ecograms, the site-specific preferences and tolerances of rare and interesting alternative tree species in climate change could be estimated, for which there are hardly any relevant references available in Central Europe.
展开▼
