Soil microbial dynamics and biogeochemical cycling in moist tropical forests

The overall objective of my dissertation research was to elucidate some of the links between microbial community structure and function and biogeochemistry. I used natural gradients in phosphorus (P) fertility created by soils of widely varying ages (oxisols and mollisols) to investigate the factors that regulate the structure and function of the soil microbial community, and the possible linkages between microbial community structure and function and biogeochemical cycling in both natural (forest) and managed (cattle pasture) systems in the humid tropics.; In Chapter 1, I used the natural P-fertility gradient, combined with direct manipulations of carbon (C) and P supply, to test the effects of P availability on the decomposition of multiple forms of C, including dissolved organic carbon (DOC) and soil organic C. Results from a combination of laboratory and field experiments suggested that C decomposition in old, highly weathered oxisol soils is strongly constrained by P availability, with potentially profound impacts for C and P cycling in this system. In Chapter 2, I explored the seasonal variability in the structure and function of the microbial community. I found strong seasonal patterns in both the magnitude and activity of the soil microbial community, and results suggested that in contrast to many temperate ecosystems, where N appears to regulate microbial processes, P availability may ultimately control microbial dynamics in tropical ecosystems. Finally, in Chapter 3, I again used forest and pasture sites on the natural soil gradient to investigate how conversion of tropical rain forest to cattle pasture affects the size and function of the microbial community, and to explore possible relationships between microbial dynamics and biogeochemistry following conversion. I found that land conversion led to fundamental changes in the magnitude and activity of the soil microbial community. Microbial biomass was consistently higher in forests than in pastures, particularly in the oxisol sites, where it was more than twice the pasture value. Forest sites were also characterized by a microbial community that was more active and responded more rapidly to carbon substrate additions, and was functionally different from pasture sites in important ways. My results provide evidence that changes in biogeochemical cycling following land conversion observed here and elsewhere may be directly related to changes in microbial community structure and function. (Abstract shortened by UMI.)...