Modeling the spatial variability of soil organic matter in a deeply dissected landscape - Bisley Watershed, Puerto Rico.
Kristofer Johnson and Fred Scatena. University of Pennsylvania, 4519 Locust St., Philadelphia, PA 19139
Soil organic matter is a key indicator of soil quality and important for understanding soil carbon storage. However, SOM is highly variable and therefore a challenge to quantify and map. Simplified techniques that can predict the variation of SOM over a landscape are needed to identify areas for protection or potential accumulation. Spatial parameters that are easily derived from digital elevation models (e.g. curvature, slope, flow accumulation and Euclidean distance to ridge) may explain a substantial proportion of the variation in SOM. To illustrate this, preliminary analysis is being carried out with the soils in the Bisley Watershed, Puerto Rico. Soils were sampled at three soil depths (0-10cm, 10-35cm and 35-60 cm) every 40 m. In this deeply dissected landscape the SOM at a 60 cm depth was significantly greater in the ridge soils than in the valley and slope soils. Empirical values correlated minimally with topographic factors, but this could be overcome in the future with algorithms and techniques that simulate the natural environment. This exercise illustrates the potential of increasingly accessible topographic data to be used for modeling the spatial variability of SOM.