Convergence, divergence or chaos? Consequences of tree trunk decay for pedogenesis and the soil microbiome in a temperate natural forest

The biochemical effects of trees may significantly influence local pedogenesis as well as pedocomplexity, biodiversity and forest dynamics on both stand and landscape scales. One such effect is the decay of tree trunks, which is driven by organisms, and especially by the microbiome. Decomposition modifies soil formation, which due to the existence of many feedbacks affects the composition of the decomposer community. We aimed to uncover general trends in the evolution of Entic Podzols as well as individual trajectories of soil properties below decaying beech trunks in an old-growth mountain forest. In particular, we used mathematical models to distinguish soil convergence, divergence and chaotic behaviour to enhance a general theory of pedogenesis. We further aimed to calculate the depth and time of convergence if this scenario is prevailing in the study plot. Pedogenetic pathways were assessed regarding the changing composition of fungal and bacterial communities in soils to obtain a complex picture of the decaying trunk-soil microbiome system. We sampled the decaying wood layer under 24 lying beech trunks and corresponding organic horizons on adjacent control microsites occupied by decaying beech leaves. At the same time we sampled underlying mineral soil horizons at both microsites (wood vs. leaves), all on Entic Podzols and granite (in total 192 soil samples). Individual trunk fall events were dated using precise dendrochronology, with the resulting chronosequence covering trunks lying for 8–52 years. We analysed decomposition processes (with a wide spectra of organic acids and ions analysed), soil chemistry (28 additional soil properties assessed), and the microbiome composition in both decaying organic matter and soils (relative abundances of the 200 most common bacterial and fungal OTUs analysed). During the first stage of trunk decay, underlying Entic Podzols responded with a significant increase of nutrients, pH, and CEC, and the maximal divergence compared to control sites was reached between 12 and 60 years after the trunk fall. Subsequently, a majority of soil properties slowly converged over a few decades to match the soil properties of control sites. The modelled convergence point occurred at ages between 39 (SO42−) and 229 (Alw) years, with a median convergence time of 53 years. The majority of soil properties converged within 20 cm below the trunk, but mathematical models predicted footprints of some soil properties down to depths of ca 60 cm. In addition, 11 soil properties did not converge at any depth, and for some properties the models even diverged. Differences in bacterial and fungal communities between below-trunk and control positions were relatively minor. Pedochemical drivers of fungal and bacterial communities (nutrients content, Ntot, Cox, Al, Fe, Mn forms) changed significantly in the mineral soil below trunks, and the microbiome partly reflected these depth-related changes. However, we propose that there is a threshold between organic and mineral soil horizons limiting the impacts of trunk decay and pedogenesis in changes to the microbiome.