Environmental and pedological factors influencing organic carbon storage in Italian forest soils

Le résumé fourni par la source

As a large and persistent carbon sink, forest soils have an essential role in the carbon cycle, thus performing valuable services to society. This paper aims to investigate the role of several environmental factors in driving soil organic carbon (SOC) storage variability in forest soils. The Italian ICP-Forests (International Co-operative Programme on Assessment and Monitoring of Air Pollution Effects on Forests) dataset provides an excellent opportunity to analyse an extensive range of variables in a land spanning different ecological regions and climatic gradients, thus being a case of broad interest. We estimated SOC storage for mineral forest soils, considering the topsoil (0–20 cm; SOCM02) and the whole soil profile (0–80 cm; SOCM08). Boosted regression trees (BRTs) were applied to explore the relative influence of environmental predictors and to model SOC storage responses. The reference soil groups (RSG-WRB) and the total subsoil phosphorus (P) content were the factors with the highest performance in explaining SOC storage for both models (SOCM02-SOCM08). Parameters related to vegetation, such as tree species, biodiversity and plant traits, clearly influenced topsoil SOC storage, while their impact was reduced with depth. Climate directly controls SOC storage, but indirect influences via plant or soil characteristics were also identified. Interactions between different environmental factors were discussed to elucidate and analyse SOC responses evidenced by the models. Our results highlighted how soil types (RSGs) can be an effective environmental factor in explaining SOC storage variability, which would likely improve SOC models on national and global scales. Interactions between different environmental factors were seen to be most important, pointing out the strong links between SOC storage and the general ecological context.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.