Soil respiration dropped by 20-30% immediately after trenching and then returned to control levels over the following weeks as the cut roots decayed. After eight months, eliminating roots by trenching had decreased soil respiration by around 25% in the control (CT) plots and 45 % in L- plots but only by 13% in the L+ plots, which reflects the reduction in fine root biomass in the soil in this treatment.
Trenching also greatly increased soil water content during the dry season in the CT and L+ plots, but the increase was less pronounced in the L- plots, possibly due to greater evaporation from the exposed soil surface.
Soil respiration was strongly positively correlated with soil temperature in the main litter treatments but the absence of roots in the trenched subplots decreased the sensitivity of soil respiration to changes in soil temperature.
Soil respiration decreased with increasing soil water content in the CT and L- plots but not in the L+ plots; there was no effect of soil water content on soil respiration in the trenched subplots in any of the litter treatments. This suggests that increasing soil water content negatively affects root respiration, rather than heterotrophic respiration.