A comparison of Landsat TM and SPOT HRV data for use in the development of forest defoliation models

Publisher Website
Google Scholar
Abstract
Spectral response patterns of the Landsat Thematic Mapper (TM) and Satellite Probatoire d'Observation de la Terre (SPOT) High Resolution Visible (HRV) were compared for a range of defoliation conditions within the boreal montane forests of the Black Mountains of North Carolina, United States of America (U.S.A.). Near-infrared (NIR) data from these sensors were shown to be correlated significantly with field estimates of per cent defoliation (p<0·05). However, neither TM nor HRV data were found to be reliable predictors of defoliation when used as single independent variables in regression equations. An empirical model predicting per cent defoliation from NIR digital numbers and digital elevation model (DEM) data was also developed for each sensor.