Extent and Contributing Factors of Delayed Post-Fire Mortality in Glacier National Park
Naficy CE and Veblen TT. 2017. Extent and Contributing Factors of Delayed Post-Fire Mortality in Glacier National Park. Boulder, CO
Field observations of areas burned within Glacier National Park (GNP) over the last twenty years suggest that substantial delayed post-fire tree mortality may be occurring in some areas classified as low-moderate burn severity by initial assessments. To quantify the spatio-temporal patterns and potential causes of delayed tree mortality, we combined remote sensing and field survey data for five fires burned between 1999-2003 in the western portion of GNP. We developed a remote sensing monitoring tool based on spectral and textural features derived from 1 m2 color imagery from NAIP that produces validated time series maps of tree- to stand-level forest condition including: mature green trees, regenerating trees, dead red-phase trees, dead grey-phase trees, snags, non forest vegetation, shadows, water, and snow/ice. Based on these maps, we document that delayed tree mortality occurred throughout a substantial portion of the fires studied. Geographical variation in the magnitude and timing of delayed tree mortality was observed—in some cases delayed mortality completely transformed landscape conditions since the initial post-fire assessments, while in others it had minimal effects. Lagged detection of initial fire effects (e.g. girdling), primary and secondary bark beetles, and potentially climate all influenced the spatio-temporal patterns of delayed tree mortality that we documented. Forest cover composition was a key determinant of the likelihood of the magnitude of delayed mortality and its likely causal mechanisms. Thus, cover type may be a useful guide for prioritization of future post-fire monitoring efforts.
- Type
- Unpublished Report
- Authors
- Naficy, Cameron; Veblen, Thomas
- Date of Issue
- 2017-12-08
- Units
- CRCO , GLAC
- Keywords
- burn severity, delayed mortality, fire, fire effects, remote sensing, vegetation