| Support for sustainable agriculture by farmers and consumers is increasing as environmental and socio-economic issues rise due to more intensive farm practices. Site-specific crop management is an important component of sustainable agriculture, within which remote sensing can play an integral role. Field and image data were acquired over a farm in Saskatchewan as part of a national research project to demonstrate the advantages of site-specific agriculture for farmers. This research involved the estimation of crop biophysical parameters from airborne hyperspectral imagery using Spectral Mixture Analysis (SMA), a relatively new sub-pixel scale image processing method that derives the fraction of sunlit canopy, soil and shadow that is contributing to a pixel's reflectance. SMA of three crop types (peas, wheat and canola) performed slightly better than conventional vegetation indices in predicting leaf area index (LAI) and biomass using Probe-1 imagery acquired early in the growing season. Other potential advantages for SMA were also identified, and it was concluded that future research is warranted to assess the full potential of SMA in a multi-temporal sense throughout the growing season. |
