Beyond Forest Classification: Continuous Mapping of Forest Species Composition, Mortality, and Sub-Canopy Using Landsat Imagery and Agnostic Analysis
Abstract
Past forest mapping and monitoring efforts generally focused on classification based on dominant species or total percent cover/crown closure. These data, however, are inadequate for many purposes, for example when key species are not the dominant species or sub-canopy composition must be evaluated following non-stand replacing disturbances. Previous mortality studies, such as those of beetle infestations, similarly generally focused on presence or absence instead of measures of intensity, which have greater value in application. Modern moderate-resolution sensors and analysis techniques might enable more detailed information regarding forest structure. We evaluated the ability of Landsat OLI imagery and agnostic image analyses at three different study sites in the Rocky Mountains to map (1) percent conifer composition by species in mixed stands, (2) percent mortality from beetle infestation, and (3) post disturbance sub-canopy composition. Average differences between predictions and observations for validation data for (1) percent cover by species ranged from 0.5-1.8% (RMSE 11-19%), depending on species, (2) percent mortality from beetle infestation was < 0.1% (RMSE 10%), and (3) counts of surviving post disturbance sub-canopy composition by species ranged from 0.2-4.7 stems per pixel (RMSE 3-13), depending on species. These results substantiate the capacity of OLI imagery to map forest composition over large areas in much greater detail than traditionally practiced. Applications of this technique include studies evaluating Canada lynx habitat, where knowledge of spruce and fir forest cover at levels well below canopy dominance is needed for habitat evaluation, and studies of economic evaluation of harvesting following bark beetle mortality, where intensity, rather than presence absence, is critical.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2016
- Bibcode:
- 2016AGUFM.B43A0544L
- Keywords:
-
- 0426 Biosphere/atmosphere interactions;
- BIOGEOSCIENCESDE: 0439 Ecosystems;
- structure and dynamics;
- BIOGEOSCIENCESDE: 0476 Plant ecology;
- BIOGEOSCIENCESDE: 0480 Remote sensing;
- BIOGEOSCIENCES