Heterogeneity of CASI-Estimated Leaf Chlorophyll in Corn: Assessment and Comparison with Ground Truth from L'Acadie GEOIDE Experimental Site

摘要

As apart of GEOIDE project for precision farming, collaboration between York University (Toronto) and Agriculture and Agri-Food Canada (St-Jean-sur-Richelieu) has focussed on the evaluation of the potential of CASI hyperspectral data to characterize the nitrogen status of corn canopies, and to assess the consequences of adjusting the nitrogen dose at top-dressing using the crop chlorophyll status as the surrogate measure. In this context, a canopy modelling inversion methodology has been developed using CASI airborne hyperspectral and multi-spectral canopy reflectance to generate maps of estimated leaf chlorophyll content in corn plots with a wide range of nitrogen treatments. Quantitative leaf chlorophyll estimates at the pixel level, derived from CASI, have demonstrated the existence of patterns in spatial variability both between different plots (different nitrogen treatments) and within each plot (same nitrogen treatment). Spatial heterogeneity has also been revealed by leaf chlorophyll content measurements in the laboratory from plot field sampling and with SPAD meter. Based on these observations, and given the correlation between the chlorophyll and nitrogen concentration in green vegetation, a careful analysis has been undertaken to assess chlorophyll heterogeneity over the field plots and to evaluate the correspondence between airborne and field measures of pigment, and by inference, nitrogen. This airborne information may prove an indicator of spatial variability of nitrogen both in the crop and the underlying soil.