Estimating leaf area index of winter oilseed rape using high spatial resolution satellite data

The leaf area index (LAI) provides critical information about vegetation density and vitality, and the temporal feature of field crops' green LAI could provide valuable information about the growing status, including environmental stresses, of crops. Quantitatively mapping the spatial and temporal variations of crop growth conditions has been enabled by the emerging of a large number of remote sensing data at various spatial and temporal resolutions recently. The vegetation indices (VIs) composed by different bands of satellite data generally show close relationships with the biophysical or physiological properties of plants. The aim of this paper was to estimate oilseed rape green LAI using high spatial resolution satellite data. The field experiment was carried out in northern Zhejiang province, China. The VIs we evaluated including the normalized difference vegetation index (NDVI), optimized soil adjusted vegetation index (OSAVI), ratio vegetation index (RVI), enhanced vegetation index (EVI), two band enhanced vegetation index (EVI2), green normalized difference vegetation index (GNDVI), and the modified triangular vegetation index (MTVI2). High spatial resolution satellite data, i.e., Pleiades-1 A, Worldview-2, Worldview-3, and SPOT-6, collected from 2014 to 2015 were used to compare with the ground observed LAI. Total 82 field LAI samples were collected at different growth stages using a portable handheld leaf area meter. We evaluated the effectiveness of exponential model built with the remote sensing VIs and observed LAI data, with the coefficients of determination being 0.708 for GNDVI, 0.702 for NDVI, 0.701 for OSAVI, 0.690 for EVI2, 0.689 for EVI, 0.677 for MTVI2, and 0.641 for RVI, respectively. The corresponding RMSE for LAI were 0.679, 0.690, 0.703, 0.729, 0.728, 0.750, and 0.803, respectively. The results showed that, GNDVI and NDVI had higher correlations with ground observed LAI than the other VIs. We concluded that the high spatial resolution remote sensing data could be applied to map the within-field LAI variability at a relative regional scale and serve the purpose of precision agriculture.