Wednesday, November 4, 2009
Convention Center, Exhibit Hall BC, Second Floor
Abstract:
Soil classification for forest management, with its associated map units for making site-specific management decisions, has relied on an individual’s observational field skills and a very limited number of soil samples. This approach has led to relatively uniform management prescriptions across a heterogeneous soil landscape. Developing techniques in soil sensing (i.e., near-infrared reflectance spectroscopy and electromagnetic induction) and mapping (i.e., geographic information sciences) are being used to develop continuous-field maps that are based on spatially intensive observational data and are designed to improve site-specific management prescriptions. Here we present results from three locations were spatially explicit soil maps of C, N, P, and K in the upper 0-30 cm have been developed along with maps of electrical conductivity through the upper 3 m to predict pine tree heights. At each location of 1, 3, and 6 ha, respectively, approximate 50% of the spatial variance in tree height could be explained by soil attributes.