Using aridity as an overarching factor to advance understanding of soil organic carbon storage at the continental scale

Abstract

Efforts to increase soil organic carbon (SOC) storage and predict its responses to climate change demand enhanced understanding of the interrelationships of controls on SOC storage and their dependence on environmental context. To this end, we use structural equation modeling to test a hypothesized structure of controls that includes the mediating influences of plant productivity and soil pH together with the direct effects of climate and soil properties on two contrasting SOC components, particulate (POC) and mineral-associated organic carbon (MAOC), using > 1000 topsoils from across the USA for which POC and MAOC were directly measured or predicted using mid-infrared spectroscopy. We find that separating systems into arid and humid systems by AI (0.65 cutoff) improves understanding controls on POC and MAOC storage, as the relationships between predictors and their effects on POC and MAOC differ between arid and humid systems based on the multigroup structural equation model and random forest models. Net primary productivity is more important for predicting POC and MAOC storage in arid than humid systems, while base cations, pH, and texture are more important in humid than arid systems. Reactive metals (oxalate-extractable Al and Fe) together are the most important predictor of topsoil POC and MAOC storage regardless of climate. We find the negative relationship between MAOC and potential evapotranspiration is stronger than that for POC, suggesting that for the mineral topsoils studied here, MAOC may be more sensitive than POC to increasing aridity. Our results support the concept that SOC storage in arid systems is more constrained by plant inputs than in humid systems, where microbial inhibition via pH and association with minerals and metals are stronger constraints, and point to the sensitivity of MAOC formation to drought. Overall, these results help to clarify the context-dependence of SOC storage and show how representing aridity as an overarching influence over the controls on SOC formation and loss processes can inform its stewardship under climate change.