Mapping Defoliation Severity
Can multispectral imagery be used to quantify the severity and extent of cumulative spruce budworm defoliation?
There have been several studies that have used vegetation indices such as NDVI and variations to ISR derived from satellite imagery to identify defoliated areas 6,7,8,9,10. The extent and degree of accuracy in which severity can be discriminated and/or modelled needs to be addressed before operational procedures for national-scale mapping can be developed.
The purpose of this study is to use multivariate techniques (i.e. discriminate analysis, MANOVA, multiple regression) to distinguish logical classes of defoliation severity and determine which spectral bands and/or indices drive these classes (Figure 2). Pending results are as expected, the extent of each defoliation class can be determined and resulting model applied to the imagery. This study will also investigate the ability to detect defoliation using multidate imagery between conifer and mixed stands.
The success of this analysis can prove beneficial for regional and national reporting requirements, carbon accounting commitments, and industry. Using satellite imagery to map defoliation severity provides a cost-effective alternative to aerial-sketch mapping, and a better potential to map consistently across regional boundaries and temporal scales. Regional and temporal defoliation mapping exercises can provide further insight into distribution and abundance patterns, enable the prediction of stand vulnerability, and identify potential stands for future monitoring and/or for implementing management programs.