Fuel Modelling Characterisation Using Low-Density LiDAR in the Mediterranean: An Application to a Natural Protected Area

Aurora Ferrer Palomino,Francisco Rodríguez Y Silva

doi:10.3390/f12081011

Abstract

Fuel structure and characteristics are important to better understand and predict wildfire behaviour. The aim of the present study was to develop a methodology for characterising fuel models using low-density and free LiDAR data that facilitate the work of managers of protected territories. Field inventories were carried out in order to understand the characteristics of the stand and the variables that fuel models must include. This information, together with the use of the intensity and structure provided by LiDAR, was used to perform statistical analyses. The linear regressions obtained to characterise the stand of the mixed Quercus spp.–Pinus ssp.-dominated stand had an R2 value ranging from 0.4393 to 0.66. While working with low-density LiDAR data (which has more difficulties crossing the canopy), in addition to the obtained results, we performed the statistical analysis of the dominant stand to obtain models with R2 values ranging from 0.8201 to 0.8677. The results of this research show that low-density LiDAR data are significant; however, in mixed stands, it is necessary to only use the dominant stratum because other components generate noise, which reduces the predictive capacity of the models. Additionally, by using the decision tree developed in combination, it is possible to update the mapping of fuel models in inaccessible areas, thereby significantly reducing costs.

Highlights

Wildfires have increased in size, frequency, and suppression costs [1]
The estimation of wildfire risks must consider the components that make up the so-called fire triangle [6], which are meteorology, topography, and fuel, for which accurate data collection is necessary
The topography is the most stable feature, as it is not modified in short intervals of time, and we cannot modify it

Summary

Introduction

Wildfires have increased in size, frequency, and suppression costs [1]. In addition to these costs, these fires have had negative effects on properties, air quality, and natural habitats [2]. The estimation of wildfire risks must consider the components that make up the so-called fire triangle [6], which are meteorology, topography, and fuel, for which accurate data collection is necessary. Managers more frequently demand updated vegetation coverage because it directly relates to fuels and where fire spread occurs, as well as because it is susceptible to modifications. This spatial information provides estimations of forest stand conditions and potential fire behaviour; these estimations are essential for guiding the mitigation of damage caused by a potential wildfire through treatments [10,12]

Objectives

Methods

Results

Conclusion