Biodiversity is a major driver of numerous ecosystem functions. However, consequences of changes in forest biodiversity remain difficult to predict because of limited knowledge about how tree diversity influences ecosystem functions. Litter decomposition is a key process affecting nutrient cycling, productivity and carbon storage, and can be influenced by plant biodiversity. Leaf litter species composition, environmental conditions, and the detritivore community are main components of the decomposition process, but their complex interactions are poorly understood. In this study, we tested the effect of tree functional diversity (FD) on litter decomposition in a field experiment manipulating tree diversity, and partitioned the effects of litter physio-chemical diversity and the detritivore community. We used litterbags with different mesh sizes to separate the effects of microorganisms and microfauna, mesofauna, and macrofauna, and monitored soil fauna using pitfall traps and earthworm extractions. We hypothesized that higher tree litter FD accelerates litter decomposition due to the availability of complementary food components and higher activity of detritivores. Although we did not find direct effects of tree FD on litter decomposition, we identified key litter traits and macro-detritivores that explained part of the process. Litter mass loss was found to decrease with an increase in leaf litter carbon:nitrogen ratio. Moreover, litter mass loss increased with an increasing density of epigeic earthworms, with most pronounced effects in litterbags with a smaller mesh size, indicating indirect effects. Higher litter FD and litter nutrient content were found to increase the density of surface dwelling macrofauna and epigeic earthworm biomass. Based on structural equation modeling, we conclude that tree FD has a weak positive effect on soil surface litter decomposition by increasing the density of epigeic earthworms, and that litter nitrogen-related traits play a central role in tree composition effects on soil fauna and decomposition.

See Methods in Patoine et al. 2020

DATA_Patoineetal_EcolEvol_2020.csv

Data file from the BIOTREE-FD decomposition experiment, for each plot and litterbag mesh size. When data is at the plot level, values are duplicated across rows for each mesh size. The variables include data about litter decomposition, tree community and litter traits, soil fauna and soil chemical and microbial properties.

README_Patoineetal_EcolEvol_2020.csv

Description file of the columns from DATA.