Decomposition of coarse woody debris in a long-term litter manipulation experiment: A focus on nutrient availability

The majority of above-ground carbon in tropical forests is stored in wood, which is returned to the atmosphere during decomposition of coarse woody debris. However, the factors controlling wood decomposition have not been experimentally manipulated over time scales comparable to the length of this process. We hypothesized that wood decomposition is limited by nutrient availability and tested this hypothesis in a long-term litter addition and removal experiment in a lowland tropical forest in Panama. Specifically, we quantified decomposition using a 15-year chronosequence of decaying boles, and measured respiration rates and nutrient limitation of wood decomposer communities. The long-term probability that a dead tree completely decomposed was decreased in plots where litter was removed, but did not differ between litter addition and control treatments. Similarly, respiration rates of wood decomposer communities were greater in control treatments relative to litter removal plots; litter addition treatments did not differ from either of the other treatments. Respiration rates increased in response to nutrient addition (nitrogen, phosphorus, and potassium) in the litter removal and addition treatments, but not in the controls. Established decreases in concentrations of soil nutrients in litter removal plots and increased respiration rates in response to nutrient addition suggest that reduced rates of wood decomposition after litter removal were caused by decreased nutrient availability. The effects of litter manipulations differed directionally from a previous short-term decomposition study in the same plots, and reduced rates of bole decomposition in litter removal plots did not emerge until after more than 6 years of decomposition. These differences suggest that litter-mediated effects on nutrient dynamics have complex interactions with decomposition over time.