Experimental litterfall manipulation drives large and rapid changes in soil carbon cycling in a wet tropical forest

Global changes such as variations in plant net primary production are likely to drive shifts in leaf litterfall inputs to forest soils, but the effects of such changes on soil carbon (C) cycling and storage remain largely unknown, especially in C-rich tropical forest ecosystems. We initiated a leaf litterfall manipulation experiment in a tropical rain forest in Costa Rica to test the sensitivity of surface soil C pools and fluxes to different litter inputs. After only 2 similar to years of treatment, doubling litterfall inputs increased surface soil C concentrations by 31%, removing litter from the forest floor drove a 26% reduction over the same time period, and these changes in soil C concentrations were associated with variations in dissolved organic matter fluxes, fine root biomass, microbial biomass, soil moisture, and nutrient fluxes. However, the litter manipulations had only small effects on soil organic C (SOC) chemistry, suggesting that changes in C cycling, nutrient cycling, and microbial processes in response to litter manipulation reflect shifts in the quantity rather than quality of SOC. The manipulation also affected soil CO 2 fluxes; the relative decline in CO 2 production was greater in the litter removal plots (-22%) than the increase in the litter addition plots (+15%). Our analysis showed that variations in CO 2 fluxes were strongly correlated with microbial biomass pools, soil C and nitrogen (N) pools, soil inorganic P fluxes, dissolved organic C fluxes, and fine root biomass. Together, our data suggest that shifts in leaf litter inputs in response to localized human disturbances and global environmental change could have rapid and important consequences for belowground C storage and fluxes in tropical rain forests, and highlight differences between tropical and temperate ecosystems, where belowground C cycling responses to changes in litterfall are generally slower and more subtle.