Estimation of Gross Primary Production of Sorghum Using Landsat Imagery and Eddy Covariance Data.

Accurate estimates of Growth Primary Production (GPP) provide a context for assessing carbon budget and evaluating the impact of climate change on crop productivity. Satellite Remote Sensing is convenient and expedient technique that provides the spatiotemporal distribution of GPP at regional and global scales. Here, we develop a simple model for estimating crop GPP that relies on products of chlorophyll-related Vegetation Index (VIs) and Ground Cover (GC), retrieved from Landsat data, and incident Photosynthesis Active Radiation (PAR_in). We calibrate and validate the model by using tower-based CO₂ flux observations over a 3-year period (2013-2015) for an irrigated sorghum field in the Texas High Plains. The accuracy and uncertainties of the model are compared using different chlorophyll-based vegetation indices derived from Landsat data. The results show that determination coefficient (R²) is improved when GPP is estimated from VI×GC rather than a single VI. The model is also able to accurately estimate daily GPP values and its temporal daily variations patterns.