How Do Plants Spatially Alter Soils Through Litter Fall in Tropical Forests?

Background. Causes of soil heterogeneity are theoretically predicted to shift in importance from abiotic controls to biotic controls at regional scales to plot scales in tropical forests. Tree litter fall is one mechanism that could function as a biotic control at plot scales. The objective of this study was to test how spatial variation in total litter fall relates to soil heterogeneity. The primary objective developed from earlier objectives aimed at characterizing spatial patterns of soil heterogeneity, analysis of abiotic controls related to topography and bedrock, and the finding that soil organic carbon was positively associated with nutrient availability. Methods. I collected a large spatial sample of soil cores from the Barro Colorado Island 50 ha Forest Dynamics Plot (FDP), Republic of Panama. I used results from soil chemical analyses to study the spatial patterns of codependence between nutrients and soil carbon. I also compared the results to 1) a litter manipulation experiment on nearby Gigante Pennisula and 2) spatially explicit data on litter fall and soil chemical heterogeneity in the BCI 50 ha FDP and two other Panamanian forests. Results. Soil macronutrients (P, Mg, Ca, K) were positively associated with soil organic carbon.  However, association strengths varied depending on the nutrient. Ca was most strongly associated with C, Mg and P were intermediate, while K was weakly associated with C. Association strengths varied inversely with reported resorption efficiencies of nutrients from senescent leaves implying that spatial heterogeneity may be resulting from plant nutrient cycling through litter fall.  I compared my results to a litter manipulation experiment on Gigante Pennisula, Panama, and found that litter addition relative to litter removal resulted in the greatest effect on C and Ca, intermediate effects on Mg and P, with the weakest effect on K. Therefore, my spatial findings matched results from the litter experiment.  Lastly, I am in the process of testing how spatial variation in total litter fall relates to soil chemical heterogeneity. Conclusions. My results support the hypothesis that plants contribute to spatial heterogeneity of soil chemical properties, with between 36% and 40% of soil Ca depending on spatial variation in soil carbon.  I am in the process of directly testing how spatial variation in total litter fall relates to soil chemical heterogeneity to better understand the role of plants as causes of variation in soil carbon and nutrient availability in tropical forests.