High tree diversity and abiotic heterogeneity contribute to complex patterns of nutrient availability across the tropics that often include strong nitrogen (N) availability gradients. We used airborne imaging spectroscopy from intact forest on the Osa Peninsula in southwestern Costa Rica to identify 0.25 ha plots with relatively high or low mean canopy N content (±1 SD of the landscape mean; n = 5 each). Plots were located within 1 km2 on a gently sloping terrace underlain by similar soils. Previous work showed that soil inorganic N concentrations were >10 times higher and nitrification rates were >2 times faster under high N canopies. Here, we hypothesized that high canopy N plot soils would support larger and more diverse microbial communities with lower N demands. We collected ten 0-10 cm soil cores from each plot, and 1) analyzed phospholipid fatty acid (PLFA) content to assess coarse differences in microbial community structure; 2) performed extracellular enzyme assays as a proxy to assess nutrient demand; and 3) conducted a laboratory fertilization (N x P) incubation experiment using native soils and leached dissolved organic matter (DOM) as a C substrate to assess relative microbial nutrient limitation in the high versus low N plots.
Results/Conclusions
Fungal:bacterial ratios were similar (P=0.29) in the plots, but live microbial biomass (P=0.086) and microbial functional group diversity (P=0.009), as measured by PLFA analysis, were both higher in high canopy N plots. However, gram-positive biomarkers (P=0.006) and the activity of C-, N-, and P-degrading enzymes (P=0.024; 0.034; 0.038) as well as polyphenol oxidase (P=0.033) were higher in the low canopy N plots. The dominance of gram-positive bacteria and relatively high rates of hydrolytic and oxidative enzyme activity suggest that low N soil communities were more nutrient limited. In the incubation experiment, DOM+N enhanced soil respiration, but responses were similar between high N soils amended with DOM leached from high N litter and low N soils amended with DOM leached from low N litter, suggesting that DOM decomposition was N limited in both settings. DOM+P stimulated decomposition in the low N plots, but not in the high N plots. Taken together, these results indicate that remotely sensed canopy N can be mapped to compositionally and functionally distinct microbial communities at the sub-hectare scale.