Scholarlink

Background <p>We evaluated the three-dimensional (3D) structure, composition, and biomass of understory fuels common to prescribed burn programs in pine-dominated forests of the southeastern and western United States. Traditional fuel characterization in these systems has been limited to two-dimensional representations, including biomass, height, and cover estimates per unit area.&#xa0;The objective of this study was to compare close-range photogrammetry with terrestrial lidar scanning (TLS), validated by destructive sampling to inform gridded 3D maps of live and dead understory fuels.</p> Results <p>We developed and analyzed a large dataset of calibrated plots using co-located terrestrial lidar scanning (TLS), close-range photogrammetry, and destructive biomass sampling conducted at 18 field sites across the southeastern and western US. Field sites represented vegetation commonly burned in prescribed fire programs, including southeastern mesic flatwood forests, southeastern loblolly-sweetgum forests, western ponderosa pine and mixed conifer forests, and western grasslands. Scanning methods were compared with metrics derived from fine-scale volumetric fuels sampling to determine the most effective method for mapping fuels in 3D. TLS and SfM-based point cloud metrics were variable in their accuracy dependent on vegetation type and levels of occlusion in understory vegetation, and neither was effective at predicting fuels within the lowest sampled stratum (0–10&#xa0;cm).</p> Conclusions <p>Modeled relationships between photogrammetry and TLS-based metrics and fuel biomass and bulk density can be used to produce unit-scale mapping for prescribed burn programs that are informed by the 3D structure and composition of understory fuels. However, point-cloud metrics may be limited in their utility in dense fuel complexes with high levels of occlusion. This work supports advances in the characterization and mapping of wildland fuel beds for use in physics-based models of fire behavior and effects that rely on gridded, 3D inputs.</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

Quantifying understory fuel structure and biomass using coupled terrestrial lidar scanning, close-range photogrammetry, and field sampling

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月16日

查看全文
  • Deborah G. Nemens,
  • Susan J. Prichard,
  • Gina R. Cova,
  • Brian Drye,
  • Paige Eagle,
  • Eric Rowell,
  • Jessie Thoreson

摘要

Background

We evaluated the three-dimensional (3D) structure, composition, and biomass of understory fuels common to prescribed burn programs in pine-dominated forests of the southeastern and western United States. Traditional fuel characterization in these systems has been limited to two-dimensional representations, including biomass, height, and cover estimates per unit area. The objective of this study was to compare close-range photogrammetry with terrestrial lidar scanning (TLS), validated by destructive sampling to inform gridded 3D maps of live and dead understory fuels.

Results

We developed and analyzed a large dataset of calibrated plots using co-located terrestrial lidar scanning (TLS), close-range photogrammetry, and destructive biomass sampling conducted at 18 field sites across the southeastern and western US. Field sites represented vegetation commonly burned in prescribed fire programs, including southeastern mesic flatwood forests, southeastern loblolly-sweetgum forests, western ponderosa pine and mixed conifer forests, and western grasslands. Scanning methods were compared with metrics derived from fine-scale volumetric fuels sampling to determine the most effective method for mapping fuels in 3D. TLS and SfM-based point cloud metrics were variable in their accuracy dependent on vegetation type and levels of occlusion in understory vegetation, and neither was effective at predicting fuels within the lowest sampled stratum (0–10 cm).

Conclusions

Modeled relationships between photogrammetry and TLS-based metrics and fuel biomass and bulk density can be used to produce unit-scale mapping for prescribed burn programs that are informed by the 3D structure and composition of understory fuels. However, point-cloud metrics may be limited in their utility in dense fuel complexes with high levels of occlusion. This work supports advances in the characterization and mapping of wildland fuel beds for use in physics-based models of fire behavior and effects that rely on gridded, 3D inputs.