Multispectral LiDAR-Based Estimation of Surface Fuel Load in a Dense Coniferous Forest

Nikos Georgopoulos,Ioannis Z Gitas,Alexandra Stefanidou,Lauri Korhonen,Dimitris Stavrakoudis

doi:10.3390/rs12203333

Abstract

Surface fuel load (SFL) constitutes one of the most significant fuel components and is used as an input variable in most fire behavior prediction systems. The aim of the present study was to investigate the potential of discrete-return multispectral Light Detection and Ranging (LiDAR) data to reliably predict SFL in a coniferous forest characterized by dense overstory and complex terrain. In particular, a linear regression analysis workflow was employed with the separate and combined use of LiDAR-derived structural and pulse intensity information for the load estimation of the total surface fuels and individual surface fuel types. Following a leave-one-out cross-validation (LOOCV) approach, the models developed from the different sets of predictor variables were compared in terms of their estimation accuracy. LOOCV indicated that the predictive models produced by the combined use of structural and intensity metrics significantly outperformed the models constructed with the individual sets of metrics, exhibiting an explained variance (R2) between 0.59 and 0.71 (relative Root Mean Square Error (RMSE) 19.3–37.6%). Overall, the results of this research showcase that both structural and intensity variables provided by multispectral LiDAR data are significant for surface fuel load estimation and can successfully contribute to effective pre-fire management, including fire risk assessment and behavior prediction in case of a fire event.

Highlights

Forest surface fuels are considered the most complex fuel types in fire management and their detailed physical description is required by the majority of fire behavior prediction systems, such as FARSITE, FlamMap, and BEHAVE [1,2,3]
The difference between training and validation relative Root Mean Square Error (rRMSE) increases, which is an indication of possible overfitting
Following the leave-one-out cross-validation (LOOCV) approach, the goodness-of-fit statistics showcased that the incorporation of Light Detection and Ranging (LiDAR) intensity information to the regression analysis workflow leads to the construction of models with higher estimation performance compared to the ones including only structural metrics

Summary

Introduction

Forest surface fuels are considered the most complex fuel types in fire management and their detailed physical description is required by the majority of fire behavior prediction systems, such as FARSITE, FlamMap, and BEHAVE [1,2,3]. The load of individual surface fuel types (SFTs) is critical because each type has a generally different effect on fire behavior; SFTs are incorporated in different surface fuel models. Such models require a numerical description of the physical parameters characterizing each fuel type, including load, which is used in fire danger and behavior prediction systems [14]. A prominent example is litter and fine woody debris (FWD) less than 0.64 cm in diameter, which dry very quickly and can be ignited, whereas woody fuels of larger dimensions are less susceptible to ignition and combustion [6,15]

Objectives

Results

Discussion

Conclusion