Classification of Plot-Level Fire-Caused Tree Mortality in a Redwood Forest Using Digital Orthophotography and LiDAR

Abstract

Aerial and satellite imagery are widely used to assess the severity and impact of wildfires. Light detection and ranging (LiDAR) is a newer remote sensing technology that has demonstrated utility in measuring vegetation structure. Combined use of imagery and LiDAR may improve the assessment of wildfire impacts compared to imagery alone. Estimation of tree mortality at the plot scale could serve for more rapid, broad-scale, and lower cost post-fire assessments than feasible through field assessment. We assessed the accuracy of classifying color-infrared imagery in combination with post-fire LiDAR, and with differenced (pre-and post-fire) LiDAR, in estimating plot percent mortality in a second-growth coast redwood forest near Santa Cruz, CA. Percent mortality of trees greater than 25.4 cm DBH in 47 permanent 0.08 ha plots was categorized as low (<25%), moderate (25%-50%), or high (>50%). The model using Normalized Difference Vegetation Index (NDVI) from National Agricultural Imagery Program (NAIP) was 74% accurate; the model using NDVI and post-fire LiDAR was 85% accurate, while the model using NDVI and differenced LiDAR was 83% accurate. The addition of post-fire LiDAR data provided a modest increase in accuracy compared to imagery alone, which may not OPEN ACCESS Remote Sens. 2014, 6 1955 justify the substantial cost of data acquisition. The method demonstrated could be applied to rapidly estimate tree mortality resulting from wildfires at fine to moderate scale.

A New Pseudoinvariant Near-Infrared Threshold Method for Relative Radiometric Correction of Aerial Imagery

“…It is available at the spatial resolution of 0.6 to 2 meters with very low cloud coverage and consists of repeat images during the growing season with two or three year cycles for more than 15 years [1]. It has been a unique choice for a variety of geospatial mapping applications, such as analysis of land cover and land use change [2][3][4], evaluation of ecosystem services [5][6][7], monitoring of forest health [8][9][10][11], and assessment of urban green infrastructure [12,13]. The NAIP imagery will likely continue to be the one of the best data sources for many research and operational efforts that need high-resolution multispectral imagery for feature extraction, change detection, or collection of ground truth for validate coarse-resolution satellite products.…”

Multi-temporal LiDAR and Landsat quantification of fire-induced changes to forest structure

“…The difficulty in quantifying fire-induced change without pre-fire measurements extends beyond the CBI protocol. There have been a few research opportunities where fire burned through permanent monitoring plots that were subsequently assessed (Bishop et al, 2014;Cocke et al, 2005;Lutz et al, 2016;Wimberly & Reilly, 2007); however, these studies have relied on small numbers of burned plots to represent change over a large area. Long-term monitoring plots associated with the Forest Inventory and Analysis (FIA) project (Gillespie, 1999) have burned with greater frequency in recent years, but the re-measurement intervals between FIA collections (5-10 years) can lead to considerable disconnects between the fire event and the post-fire re-visit, reducing the visibility and magnitude of those effects when data collection does occur (Whittier & Gray, 2016).…”

“…Analyzed in concert with field data, LiDAR returns can also be used to predict forest structure attributes such as basal area, volume, biomass, and leaf area (García et al, 2010;Hudak et al, 2009;Lefsky et al, 2002). LiDAR has been successfully used to quantify the effects of insect outbreaks in forests (Bater et al, 2010;Bright et al, 2012), pre-fire fuel loading (Andersen et al, 2005;García et al, 2011;Riaño et al, 2003Riaño et al, , 2004Seielstad & Queen, 2003), and structural measurements of the post-fire environment (Bishop et al, 2014;Kane et al, 2013Kane et al, , 2014Kwak et al, 2010;Wulder et al, 2009). Structural datasets such as those derived from LiDAR data have been previously highlighted as holding considerable promise for directly quantifying changes in vegetation structure (Smith et al, 2014), but acquisitions of highresolution, comparable pre-and post-fire LiDAR data that provide measure of fire-induced vegetation change have been limited (Bishop et al, 2014;Reddy et al, 2015;Wang & Glenn, 2009;Wulder et al, 2009).…”

“…LiDAR has been successfully used to quantify the effects of insect outbreaks in forests (Bater et al, 2010;Bright et al, 2012), pre-fire fuel loading (Andersen et al, 2005;García et al, 2011;Riaño et al, 2003Riaño et al, , 2004Seielstad & Queen, 2003), and structural measurements of the post-fire environment (Bishop et al, 2014;Kane et al, 2013Kane et al, , 2014Kwak et al, 2010;Wulder et al, 2009). Structural datasets such as those derived from LiDAR data have been previously highlighted as holding considerable promise for directly quantifying changes in vegetation structure (Smith et al, 2014), but acquisitions of highresolution, comparable pre-and post-fire LiDAR data that provide measure of fire-induced vegetation change have been limited (Bishop et al, 2014;Reddy et al, 2015;Wang & Glenn, 2009;Wulder et al, 2009). However, multi-temporal LiDAR is not a novel concept and has been widely applied to quantify other ecosystem properties such as snow volume (Tinkham et al, 2014), forest growth and harvest disturbance (Hudak et al, 2012), boreal forest gap dynamics (Vepakomma et al, 2008), and change in biomass resulting from a Gypsum moth (Lymantria dispar) outbreak (Skowronski et al, 2014), among other applications.…”

Urban tree damage estimation using airborne laser scanner data and geographic information systems: An example from 2007 Oklahoma ice storm

“…Bio-geographers, foresters, and city planners have been developing techniques to measure tree damages due to natural disasters (Bishop et al, 2014;Honkavaara et al, 2013;Vastaranta et al, 2012;Boutet and Weishampel, 2003). Space-borne passive remote sensors have allowed variable level of accuracy in detection and mapping of damaged or changes in single trees and forest stands by capturing and comparing the vegetation reflectance in land use pixels of pre-and post-hazard periods (Franklin et al, 2000;Leckie et al, 1992;Nyström et al, 2013;Olthof et al, 2004;Roberts et al, 1997;Rogan et al, 2002;Wulder et al, 2009).…”