Location: Soil and Water Management ResearchTitle: Remote estimation of crop gross primary production with Landsat data) Author
Submitted to: Remote Sensing of Environment
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/22/2012
Publication Date: 3/21/2012
Publication URL: http://handle.nal.usda.gov/10113/59553
Citation: Gitelson, A., Peng, Y., Masek, J., Rundquist, D., Verma, S., Suyker, A., Baker, J.M., Hatfield, J.L., Meyers, T. 2012. Remote estimation of crop gross primary production with Landsat data. Remote Sensing of Environment. 121:404-414. Interpretive Summary: Assessment of the amounts of crop production is critical for the projections of expected food and feed production for the world. We currently lack the tools to accurately assess crop productivity and carbon budgets across large areas of the United States or the World. To address this problem we developed a method which combined reflectance data obtained from satellite images using the green color of the leaves as a measure of the amount and health of the vegetation. This information was combined with the measure of the total amount of solar radiation available for plant photosynthesis. Combining these two components we assembled a model to estimate the amount of productivity in corn and soybean canopies. This approach was capable of accurately estimated the growth of corn and soybean across Nebraska, Iowa, Minnesota, and Illinois. Methods developed using these sources of information will be valuable in conducting regional assessments of crop productivity and provide scientists who are attempting to develop large scale carbon budgets a more accurate estimate of the role of agricultural systems in the global carbon budget.
Technical Abstract: An accurate and synoptic quantification of gross primary productivity (GPP) in crops is essential for studies of carbon budgets at regional and global scales. In this study, we developed a model relating crop GPP to a product of total canopy chlorophyll (Chl) content and potential incident photosynthetically active radiation (PARpotential). The approach is based on remotely sensed data; specifically, vegetation indices (VI) that are proxies for total Chl content and PARpotential, which is incident PAR under a condition of minimal atmospheric aerosol loading, calculated from astronomical data. Using VIs retrieved from surface reflectance Landsat data, we found that the model is capable of accurately estimating GPP in maize, with coefficient of variation (CV) below 23%, and in soybean with CV below 30%. The algorithms established and calibrated over three Nebraska AmeriFlux sites showed a great potential for estimating maize and soybean GPP in AmeriFlux study areas in Minnesota, Iowa and Illinois with acceptable accuracy.