Skip to Main Content
Utilizing random forests imputation of forest plot data for landscape-level wildfire analysesAuthor(s): Karin L. Riley; Isaac C. Grenfell; Mark A. Finney; Nicholas L. Crookston
Source: In: Viegas, Domingos Xavier, ed. Advances in Forest Fire Research. Coimbra, Portugal: Imprensa da Universidade de Coimbra. p. 583-590.
Publication Series: Paper (invited, offered, keynote)
Station: Rocky Mountain Research Station
Download Publication (726.0 KB)
Related Research Highlights
A Tree Level Model of Forests in the Western United States
DescriptionMaps of the number, size, and species of trees in forests across the United States are desirable for a number of applications. For landscape-level fire and forest simulations that use the Forest Vegetation Simulator (FVS), a spatial tree-level dataset, or “tree list”, is a necessity. FVS is widely used at the stand level for simulating fire effects on tree mortality, carbon, and biomass, but uses at the landscape level are limited by lack of availability of forest inventory data for large contiguous areas. Detailed mapping of trees across large areas is not feasible with current technologies, but statistical methods for matching forest plot data with biophysical characteristics of the landscape offer a practical means to populate landscapes with a limited set of forest plot inventory data. We used a modified random forests approach, with Landfire vegetation and biophysical predictors at 30m grid resolution. In essence, the random forests method creates a “forest” of decision trees in order to choose the forest plot with the best statistical match for each grid cell in the landscape. Landfire data was used in this project because is publicly available, offers seamless coverage of variables needed for fire models, and is consistent with other datasets, including burn probabilities and flame length probabilities generated for the continental US by Fire Program Analysis (FPA). We used the imputed forest plot data to generate a map of forest cover and height as well as existing vegetation group for a study area in eastern Oregon, and examined correlations with Landfire data. The results showed good correspondence between the two data sets (84-97% within-class agreement, depending on the variable). In future research, the new imputed grid of inventory data will be used for landscape simulation studies to determine risk to terrestrial carbon resources from wildfire as well as to investigate the effect of fuel treatments on burn probability and fire sizes.
- You may send email to email@example.com to request a hard copy of this publication.
- (Please specify exactly which publication you are requesting and your mailing address.)
- We recommend that you also print this page and attach it to the printout of the article, to retain the full citation information.
- This article was written and prepared by U.S. Government employees on official time, and is therefore in the public domain.
CitationRiley, Karin L.; Grenfell, Isaac C.; Finney, Mark A.; Crookston, Nicholas L. 2014. Utilizing random forests imputation of forest plot data for landscape-level wildfire analyses. In: Viegas, Domingos Xavier, ed. Advances in Forest Fire Research. Coimbra, Portugal: Imprensa da Universidade de Coimbra. p. 583-590.
Keywordsimputation, forest plot data, Landfire, random forests
- Seeing the forest for the trees: utilizing modified random forests imputation of forest plot data for landscape-level analyses
- Mapping forest vegetation for the western United States using modified random forests imputation of FIA forest plots
- TreeMap, a tree-level model of conterminous US forests circa 2014 produced by imputation of FIA plot data
XML: View XML