Scholarlink

Background <p>Firebrands, small burning particles, are a key driver of wildfire spread. Generated from burning vegetation, they are carried by wind to ignite spot fires ahead of the main fire front. Fire behavior models aim to predict this spread, but limited data on firebrand properties introduces uncertainty. Developing a fast and reliable method for firebrand fuel characterization is essential. Our study uses Terrestrial Laser Scanning (TLS) data to quantify the fine fuels in near-surface, elevated, and canopy forest layers. TLS scans were collected from three different temperate eucalypt forest sites in Victoria, Australia. A set of algorithms was used to extract and build characteristics of firebrand fuel from the TLS point cloud data. We categorized firebrand fuel into three types: twigs (diameter &lt; 5 cm), candlebark and stringybark.</p> Results <p>We were able to estimate variation in firebrand fuel in 1 m<sup>3</sup>. The mean of twig length (diameter &lt; 5.5 cm) was found to be 4.9–7.1 cm in the total plot, 1.5–1.7 cm in the canopy layer, and 15.4–24.8 cm in the elevated layer. The mean of candlebark length was 0.5–4.0 cm in the total plot, 0.2–1.0 cm in the canopy, and 2.9–13.6 cm in the elevated layer. The mean of stringybark surface area was 1218 cm<sup>2</sup> in the elevated layer and 344 cm<sup>2</sup> in the total layer. Fuel from different sites had varying characteristics. Our results indicated the overall trend across the three sites was for a decrease in twig length with increasing diameter, with the exception of candlebark, for which bark length decreased with increasing width. Fuel distribution was found to be highest in the near-surface and elevated layers and lowest in the canopy layer.</p> Conclusions <p>The method we developed can be used as a new approach for quantifying fine fuel and hazard assessment. Collected data is especially valuable for laboratory-scale experiments, where controlled and replicable conditions are critical for studying firebrand generation and behavior. They can also serve as inputs to fire behavior simulation models, particularly for surface and crown fire models. Future research efforts could focus on enhancing the accuracy of algorithms and their implementation to further improve the effectiveness of fuel quantification.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A LiDAR framework to characterise elevated and canopy firebrand fuels for fire prediction models

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月15日

查看全文
  • Ha-Ninh Nguyen,
  • Trent D. Penman,
  • Mika Peace,
  • Nina Hinko-Najera,
  • Sandra Penman,
  • Jean-Baptiste Filippi,
  • Alberto Alonso Pinar,
  • Alexander I. Filkov

摘要

Background

Firebrands, small burning particles, are a key driver of wildfire spread. Generated from burning vegetation, they are carried by wind to ignite spot fires ahead of the main fire front. Fire behavior models aim to predict this spread, but limited data on firebrand properties introduces uncertainty. Developing a fast and reliable method for firebrand fuel characterization is essential. Our study uses Terrestrial Laser Scanning (TLS) data to quantify the fine fuels in near-surface, elevated, and canopy forest layers. TLS scans were collected from three different temperate eucalypt forest sites in Victoria, Australia. A set of algorithms was used to extract and build characteristics of firebrand fuel from the TLS point cloud data. We categorized firebrand fuel into three types: twigs (diameter < 5 cm), candlebark and stringybark.

Results

We were able to estimate variation in firebrand fuel in 1 m3. The mean of twig length (diameter < 5.5 cm) was found to be 4.9–7.1 cm in the total plot, 1.5–1.7 cm in the canopy layer, and 15.4–24.8 cm in the elevated layer. The mean of candlebark length was 0.5–4.0 cm in the total plot, 0.2–1.0 cm in the canopy, and 2.9–13.6 cm in the elevated layer. The mean of stringybark surface area was 1218 cm2 in the elevated layer and 344 cm2 in the total layer. Fuel from different sites had varying characteristics. Our results indicated the overall trend across the three sites was for a decrease in twig length with increasing diameter, with the exception of candlebark, for which bark length decreased with increasing width. Fuel distribution was found to be highest in the near-surface and elevated layers and lowest in the canopy layer.

Conclusions

The method we developed can be used as a new approach for quantifying fine fuel and hazard assessment. Collected data is especially valuable for laboratory-scale experiments, where controlled and replicable conditions are critical for studying firebrand generation and behavior. They can also serve as inputs to fire behavior simulation models, particularly for surface and crown fire models. Future research efforts could focus on enhancing the accuracy of algorithms and their implementation to further improve the effectiveness of fuel quantification.