Utilizing Geostatistics to Increase the Efficacy of Sampling Protocols Used to Measure Soil Organic Carbon.

Several studies have suggested that geostatistics could increase the efficacy of sampling protocols used to measure soil organic carbon, allowing for more accurate quantification of soil carbon (C) for application in a modeling environment.  This paper illustrates how information about the range of spatial autocorrelation of soil organic carbon can help determine the sampling density required for estimating mean carbon stock at the scale of a watershed with a given accuracy. .Accounting for spatial autocorrelation reduces the standard errors and narrow the confidence intervals associated with sample estimates of carbon per unit area. Utilizing secondary data such as soil nitrogen and soil moisture improves spatial accuracy. This information may be useful for modelers who work with land-use managers in terms of future challenges in implementing land-use practices that reduce risks related to global climate changes.