Satellite remote sensing of cumulative spruce budworm defoliation in Prince Albert National Park

Abstract

This thesis is a review and study of cumulative defoliation caused by spruce budworm in Prince Albert National Park as assessed using satellite remote sensing technology. The work presented is organized into three parts: 1) Chapter 1 and 2 is a review of satellite remote sensing technology as applied to defoliation monitoring; 2) Chapter 3 is an introduction to the case study methods, and chapter 4 is the case study submitted in January of 2006 to the Canadian Journal of Remote Sensing outlining the success obtained in using different satellite image techniques to detect spruce budworm defoliation; 3) Lessons learned and synthesis of findings from the literature review in part 1 as they pertain to the research results in part 2 is presented in chapter 5. Thirty five (n=35) sites in mature closed canopy conifer stands of black spruce (Picea mariana), white spruce (Picea glauca) and balsam fir (Abies balsamea) were analyzed using multi-temporal Landsat 5 TM images (June 15, 1992 and July 15, 2004) and a single date SPOT4 HRVIR image (August 19, 2004). Correlation analysis was performed to assess spectral relationships of cumulative eastern spruce budworm (Choristoneura fumiferana Clemens) defoliation to three vegetation indices and their Landsat TM differences. Field estimates of defoliation were collected using digital hemispherical photography and field estimates of cumulative defoliation in three severity classes and a non effected class categorized by nil, light, moderate and severe. Results indicated that weak empirical relationships exist between Landsat TM and SPOT4 HRVIR vegetation indices to cumulative defoliation. Multi-temporal Landsat data performed better than single date Landsat data. The SPOT4 HRVIR NDMI produced the best correlation to defoliation with r2 = 0.78. These results may be helpful in designing and implementing an automated satellite remote sensing defoliation monitoring system required for effective management and reporting of forest conditions in Canada.