Download (PDF 437 KB): https://research.fs.usda.gov/download/treesearch/50366.pdf
Abstract
This study examined the utility of dense time series of Landsat imagery for small area estimation and mapping of change in forest conditions over time. The study area was a region in north central Wisconsin for which Landsat 7 ETM+ imagery and field measurements from the Forest Inventory and Analysis program are available for the decade of 2003 to 2012. For the periods 2003-2007 and 2008-2012, the monthly imagery was processed using harmonic analysis in order to capture seasonal trends in reflectance across spectral bands. A nonparametric modeling approach was used with predictor variables and field measurements at two points in time to predict change in live tree basal area. Predictions for individual plots poorly matched observations of change, however the resultant maps of change compared favorably to a purposive sample of locations of high predicted relative change, based on multi-date, high-resolution aerial photography. This suggests the need for a larger sample of plots or further tuning of the model.
Parent Publication
Citation
Wilson, Barry Tyler. 2015. Harmonic analysis of dense time series of landsat imagery for modeling change in forest conditions. In: Stanton, Sharon M.; Christensen, Glenn A., comps. 2015. Pushing boundaries: new directions in inventory techniques and applications: Forest Inventory and Analysis (FIA) symposium 2015. 2015 December 8–10; Portland, Oregon. Gen. Tech. Rep. PNW-GTR-931. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station: 200-205.
