Effects of two key plant trait spectra on litter layer properties and habitat provision functions

Plant litter harbours a wide range of organisms, such as soil animals, microbes, vascular plant seeds and seedlings, mosses and lichens. While it is well-recognized that litter drives carbon and nutrient cycling through its traits related to the plant resource economics spectrum (PES), its explanatory power for predicting the litter-dwelling biological community has been very low. In this era of biodiversity change, litter habitat provisioning functions should receive more research attention. A conceptual trait space defined by a physical trait axis related to the litter size and shape spectrum (SSS) and a biochemical trait axis (PES) has been proposed, to explain habitat as well as nutrition functions for soil organisms. We aim to improve the explanatory power of litter traits for the litter-dwelling animal communities by testing this concept, and by connecting fundamental traits related to the size and shape of single litter particles with litter layer properties more directly related to habitat quality (e.g. moisture regime, habitat space). We tested the concept by using 16 woody species, and examined how different traits or trait axes determine the litter and litter layer properties directly related to habitat quality for soil animals. In addition, we assessed in a field experiment with litter layers in mesocosms how this new trait framework explains the soil invertebrate community. The two plant trait spectra together accounted for >60% of the variance in litter quality in a principal component analysis. SSS-related traits strongly explained litter layer habitat-related properties, such as water-holding capacity and litter bulk density. The SSS axis and these SSS-associated properties more strongly determined soil animal communities assembled over 7 weeks in the field experiment, surpassing the effects of the PES axis. Synthesis. This study has provided the first empirical evidence for the litter trait space defined by both SSS and PES spectra occupied by wide-ranging plant species, and its utility to explain litter layer properties related to the physical environment and associated habitat provisioning function. We suggest that this trait-based SSS-PES framework will help us to disentangle mechanisms underpinning real-world interactions between above-ground vegetation and below-ground communities.