Spatially Explicit Modeling of Switchgrass Carbon Sequestration in Oklahoma.

Switchgrass (Panicum virgatum L.) acreage is expanding for cellulosic biofuels production due to its perennial habitat and sustainable carbon sequestration potential. Recent literature review has identified the need for regional ecosystem-level switchgrass carbon sequestration potential estimates. The objective of this study was to conduct seasonal (April to October) spatial modeling of net ecosystem exchange (NEE) of C across potential switchgrass growing areas of Oklahoma. Approximation of potential switchgrass production areas in Oklahoma was done by reclassifying the USDA-NASS Crop Data Layer (CDL) in ArcGIS. Weather data (2008 to 2011) obtained from the Oklahoma Mesonet for 120 weather stations across Oklahoma was processed to derive values for 30-minute intervals. Using eddy covariance measurements of 2011, empirical models, a) modified rectangular hyperbolic light-response curve to include limitation of vapor pressure deficit on photosynthesis and b) temperature response function were developed and fitted to estimate gross ecosystem productivity (GEP) and ecosystem respiration (ER) on a seasonal scale across Oklahoma. A total of 9.74 M ha were identified, by reclassifying NASS Crop Data layer, as suitable for switchgrass production in Oklahoma. Average seasonal net ecosystem production (NEP) across Oklahoma ranged from 2.3 to 6.9 t C ha^-1. No correlation was observed between rainfall and NEP. However, results based on a simple linear model analysis suggested that there were significant differences in NEP among years. Spatial interpolation of NEP across potential switchgrass production areas indicated that an average of 5.7 Tt of carbon could be sequestered annually in Oklahoma.