Estimating ground cover in the mixed prairie grassland of southern alberta using vegetation indices related to physiological function

Native grasslands are an important forage resource for the cattle industry and play a vital role in hydrological, carbon, and nutrient cycles; energy flow; faunal and floral biodiversity; and recreational services. Despite their importance, information on the health of Canada’s native grasslands is extremely limited. This study investigated the use of functional relationships, namely, differences in the Normalized Difference Vegetation Index (NDVI) and a shortwave infrared index (SWIR75 or CAI) related to cellulose and lignin content, along with spectral mixture analysis (SMA), to derive estimates of broad categories of grassland ground cover at 4 test sites established in the mixed prairie grassland in southern Alberta, Canada. Field campaigns were carried out in 2009 and 2010, at peak grass production, to collect fractional ground cover of photosynthetic vegetation (fPV), nonphotosynthetic vegetation (fNPV), and background (f B), as well as ground spectra of various grassland components. The ability to separate PV, NPV, and B using NDVI and the SWIR75 or CAI indices derived from field spectroradiometer data and Landsat-5 TM data was investigated. Although reasonable estimates of fPV were derived using the NDVI and SWIR75 indices with SMA (r = 0.82) from Landsat imagery, the similarity in the spectral characteristics of soil, lichens, litter, and standing senescent vegetation confounded the ability to estimate fNPV and fB. In the absence of a hyperspectral satellite system offering the ability to derive CAI, the ability to evaluate grassland health in the mixed grass prairie is limited.