Latitudinal patterns of particulate and mineral-associated organic matter down the soil profile in drylands

Understanding the controls on particulate organic matter (POM) and mineral-associated organic matter (MAOM) accumulation is essential in order to accurately predict carbon-climate feedbacks and for ecosystem manage-ment. However, how POM and MAOM fractions vary across latitudinal gradients in dryland ecosystems, and what drives this variation remains unclear. In this study, we sampled soils from 100 natural shrubland sites across a latitudinal gradient (23?N to 32?N) in the dry valleys of southwestern China to identify latitudinal patterns of POM and MAOM fractions at four soil depths (0-10 cm, 10-20 cm, 20-30 cm, and 30-50 cm). We found that both C and N fractions in POM and MAOM exhibited a binomial trend with latitude at all depths, whereas the N fractions in POM and MAOM exhibited a pronounced increase at higher latitudes in subsurface soils (20-30 cm and 30-50 cm). Variation in C fractions (i.e., particulate organic carbon (POC) and mineral -associated organic carbon (MAOC)) were mainly explained by the independent effect of soil properties, with the most important explanatory factor being exchangeable calcium. In contrast, variation in N fractions (i.e., particulate organic nitrogen (PON) and mineral-associated organic nitrogen (MAON)) were largely explained by interactions between soil , climate and vegetation properties, with the most important driver being mean annual temperature and mean annual precipitation for PON, exchangeable calcium for MAON at 0-10 cm and 10-20 cm depths, and mean annual temperature for MAON at 20-30 cm and 30-50 cm soil depths. Interestingly, climate was more important in explaining variation in C and N fractions in the subsurface soils, and climate and vegetation properties had higher predictive power for soil N fractions than C fractions in the subsurface soils. These results advance our understanding of the role of soil properties on SOM accumulation in dryland eco-systems. Our study also highlights the important impact of climate and vegetation properties in explaining the spatial variation of SOM in subsurface soils, which should be incorporated into large-scale assessments to better predict SOM dynamics.