EFFECTS OF GAP SIZE ON LITTER DECOMPOSITION AND MICROBIAL ACTIVITY IN A SUBTROPICAL FOREST

We examined the effects of gap size on rates of mass loss of leaf, branch and bark litter, and microbial activity in a Castanopsis kawakamii forest over a 360-d period. Fiver forest gap size classes that comprise a gap size gradient were chosen: (1) closed canopy of pure Castanopsis kawakamii vegetation; (2) small gaps with a diameter of < 5 m; (3) small-to-intermediate gaps with a diameter of 5-15 m; (4) intermediate-to-large gaps with a diameter of 15-30 m; (5) large gaps with a diameter of > 30 m. After 360 d, plant litter decomposing under closed canopy or within the small gaps lost mass more rapidly than those in the large gaps. Among the litter quality parameters measured, initial id concentrations were most strongly positively correlated with the annual decay rates, and among the microclimatic factors, soil moisture content was the best predictor. Total substrate-induced respiration (SIR) was highest under closed canopy, intermediate in small gaps, and lowest in large gap size classes. Within gap size classes, total microbial respiration was highest for leaves, intermediate for bark, and lowest for branches. Across all sample dates, soil moisture content was the best predictor of microbial activity among the measured microclimatic factors. Plant litter C/N ratios were best predictors of microbial activity among the measured litter quality indices. Decomposition rate constants were linearly correlated with overall mean values of microbial activities for all three types of decomposing litter within five different gap size classes (R(2) = 0.984). The SIR rates of decomposing plant fitter as a measure of potential active microbial biomass reflected the microclimate and litter quality. Our results suggest that large gaps significantly reduce microbial activity and decomposition rates by changing environmental conditions that consequently should reduce nutrient cycling rates in this system.