Mapping Progression and Severity of a Southern Colorado Spruce Beetle Outbreak Using Calibrated Image Composites

Brian Woodward,Anthony Vorster,Paul Evangelista

doi:10.3390/f9060336

Abstract

An ongoing spruce beetle (Dendroctonus rufipennis Kirby.) epidemic in southern Colorado has resulted in the death of thousands of acres of forests primarily dominated by Engelmann spruce (Picea engelmannii Parry.). To evaluate the ecological and economic impacts of this massive mortality event, researchers and land managers need to efficiently track its progression, spread, and severity across large spatial extents. In this study, mortality severity (0–100% dead) was successfully mapped at the Landsat pixel scale (30 × 30 m) across a large (5000 km2), persistently cloud-covered study area using multi-sensor (Landsat 5 Thematic Mapper (TM), Landsat 7 Enhanced Thematic Mapper Plus (ETM+) and Landsat 8 Operational Land Imager (OLI)) harmonized tasseled cap image composites as spectral predictors of gray stage spruce beetle mortality. Our maps display the distribution and severity of this landscape-scale mortality event in 2011 (R2 = 0.48, root mean squared error (RMSE) = 7.7) and 2015 (R2 = 0.55, RMSE = 11.6). Potential applications of this study include efficient landscape-scale forest health monitoring, targeted forest and timber management, and assessment of ecological impacts of bark beetle outbreaks.

Highlights

Monitoring of forest mortality across large spatial extents is a traditionally complex and resource-intensive endeavor [1,2,3]
The primary objective of this study was to map the severity and progression (2011–2015) of the spruce beetle outbreak occurring in southern Colorado spruce/fir forests
This study displayed an effective forbased detecting gray stage spruce The mortality in spruce mortality when applied within amethodology regression tree modelling framework

Summary

Introduction

Monitoring of forest mortality across large spatial extents is a traditionally complex and resource-intensive endeavor [1,2,3]. In the wake of a changing climate, multi-year drought conditions, and severe insect and disease activity, the complexity of monitoring and understanding change across vast expanses of public and privately owned forestlands in the western United States has only increased [4]. Colorado has killed Engelmann spruce (Picea engelmannii Parry.) trees across thousands of acres of spruce and fir dominated forests [5]. With spruce/fir forest types representing approximately 20%. Of statewide forest cover [6] and encompassing the largest number of forested acres under public ownership within the state [7], monitoring and management of these forests are top priorities for both public and private land managers and owners.

Objectives

Methods

Results

Conclusion