9416819 Coleman Studies of decomposition and nutrient cycling in ecosystems have been very important processes for the interpretation of ecosystem function. They have become "keystone" parameters in the LTER program. There has been an increasing interest in the functional roles of hundreds of species of litter- and microbial- feeding fauna which occur in virtually all terrestrial sites. How do these species operate and interact, and are they "necessary' for optimal functioning of the system? This project will explore the interaction between biodiversity and decomposition processes, by studying factors causing marked differences which have been found between three moist, warm, broad-leaved forest sites, all on similar soil types. Two tropical sites, Luquillo (LUQ) in Puerto Rico, and La Selva (LAS Rica, have low and high species richnesses of soil microarthropods, respectively. Cowetta (CWT), in the Southern Appalachian mountains, has a species richness of microarthropods approaching that of LAS. Both CWT and LUQ are in the general substrate decomposition network study currently being conducted by Harmon et al. The same or similar litter substrates to those of the Harmon et al. network, will be contained in litter bags identical to the previous study. The network will be expanded to include LAS, to test the hypothesis: forests with litters containing the most diverse litter faunas will have the most rapid decomposition rates. Conversely, forests with relatively low species richness (LUQ) will have significantly lower rates of decomposition, when corrected for differences in moisture/temperature regimes. CWT is expected to have a decomposition rate intermediate to that of the two tropical sites, because of its seasonal cycles of warm wet weather. This project uses simple, proven techniques to compare decomposition rates across sites and should be the groundwork for more widespread studies of biotic interactions and processes across the LTER network.
