Root functional traits are key determinants of the rhizosphere effect on soil organic matter decomposition across 14 temperate hardwood species

Abstract Understanding plant roots induced changes in soil organic matter (SOM) decomposition, the rhizosphere effect, is critical to predict soil carbon and nutrient cycling within and across biomes. Root functional traits are proposed to regulate the rhizosphere effect. We measured the rhizosphere effect on SOM decomposition by incubating rhizosphere and bulk soils sampled from 14 hardwood species with distinct mycorrhizal associations (Arbuscular mycorrhizal, AM vs. Ectomycorrhizal, ECM). We also measured root traits (related to chemistry, morphology and physiology), soil enzymes related to carbon, nitrogen and phosphorous cycling and soil physico-chemical properties, which could explain the mechanisms that regulate interspecific variation of the rhizosphere effect on SOM decomposition. There was large interspecific variation in the rhizosphere effects on SOM decomposition and soil enzymes, which was not influenced by mycorrhizal associations. Root traits were more strongly correlated with the rhizosphere effect on SOM decomposition across the 14 hardwood species than bulk soil physico-chemical properties. Specifically, the rhizosphere effect on SOM decomposition was correlated positively with root nitrogen concentration, and negatively with root carbon/nitrogen ratio and root diameter, respectively. These three root traits were the most important predictors for the interspecific variation of the rhizosphere effect on SOM decomposition amongst all measured variables. Moreover, the rhizosphere effect on SOM decomposition tended to increase with increasing root exudationn rates. This is compelling evidence that the rhizosphere effect on SOM decomposition tends to increase along the root economics spectrum from conservative species to acquisitive species in woody plants. Our findings thus contribute to a better mechanistic link between root traits and rhizosphere effect across species, which could be incorporated into land surface models to predict ecosystem scale consequences of root induced changes in SOM decomposition.