Skip to Main Content
Characterizing fire effects on conifers at tree level from airborne laser scanning and high-resolution, multispectral satellite dataAuthor(s): Carine Klauberg; Andrew T. Hudak; Carlos Alberto Silva; Sarah A. Lewis; Peter R. Robichaud; Terrie B. Jain
Source: Ecological Modelling. 412: 108820.
Publication Series: Scientific Journal (JRNL)
Station: Rocky Mountain Research Station
Download Publication (4.0 MB)
DescriptionPost-fire assessment is made after a wildfire incident to provide details about damage level and its distribution over burned areas. Such assessments inform restoration plans and future monitoring of ecosystem recovery. Due to the high cost and time to conduct fieldwork, remote sensing is an appealing alternative to assess post-fire condition over larger areas than can be surveyed practically in the field. The aim of this study is to use remote sensing data to characterize post-fire severity at tree level in a mixed conifer forest following the Cascade and East Zone megafires of 2007 in central Idaho, USA. We used remote sensing metrics derived from Airborne Laser Scanning (ALS) data (2008) and high-resolution QuickBird (QB) multispectral satellite imagery (2007-2009) for calibrating and validating predictive models with field data (2008). We compared fire effects on trees in open canopies within recent fuel treatments to similar trees in closed canopies on adjacent, untreated sites. We observed more trees with charred crowns in high fire severity sites, mostly untreated, whereas we observed more trees with live crowns in low fire severity sites, independent of the treatment. Individual trees were more accurately detected from ALS data in treated sites with open canopies than untreated sites with closed canopies. For detected trees, the response variables predicted from ALS and QB metrics were total height (Ht), crown base height (CBH), total basal area (BAT), live basal area (BAL), scorched basal area (BAS), charred basal area (BAC) and crown severity (CS). None of the selected QB metrics were strongly correlated with the selected ALS metrics, which justified combining both data types into the predictive models. Random Forest regression models combining ALS+QB metrics or using ALS metrics alone performed similarly but clearly better than models using only QB metrics. This study shows the superiority of ALS data to high resolution, multispectral QB imagery for mapping fire severity at tree level. Managers with limited resources to plan for restoration of fire affected forests are advised to prioritize spending for data collection on ALS data and a modest number of field inventory plots, rather than QB or other broadband satellite imagery.
- 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.
CitationKlauberg, Carine; Hudak, Andrew T.; Silva, Carlos Alberto; Lewis, Sarah A.; Robichaud, Peter R.; Jain, Terrie B. 2019. Characterizing fire effects on conifers at tree level from airborne laser scanning and high-resolution, multispectral satellite data. Ecological Modelling. 412: 108820.
Keywordscrown fire severity, fire effects, fuel treatment effectiveness, individual tree attributes, Random Forest
- Evaluation of linear spectral unmixing and deltaNBR for predicting post-fire recovery in a North American ponderosa pine forest
- Indicators of burn severity at extended temporal scales: a decade of ecosystem response in mixed-conifer forests of western Montana
- Using hyperspectral imagery to estimate forest floor consumption from wildfire in boreal forests of Alaska, USA
XML: View XML