Abstract Tree mortality from major disturbances can greatly increase dead wood in forested areas, affecting fire intensity and behavior, wildlife habitat, and carbon dynamics. Accurately quantifying regional standing dead tree (SDT) pools, as conducted by the U.S. Forest Service Forest Inventory and Analysis (FIA) program, remains a prominent challenge. Little empirical work has been done accounting for structural changes in SDT volume across decay classes due to measurement and sampling challenges associated with SDT. Light-detection-and-ranging (LiDAR) represents a remote sensing technology with the potential to improve sampling efficacy and provide volume estimates of SDT via non-destructive sampling. Following this, the goal of this study was to explore the feasibility of empirically quantifying and assessing structural volume in southern pine SDT by decay class using terrestrial LiDAR. To meet this goal, we addressed three objectives, 1) construct empirical volume estimates of SDT by decay class using terrestrial LiDAR and a voxel-based, volume calculation algorithm capable of accounting for occlusion and point cloud quality, 2) develop allometric relationships of aboveground SDT component volumes by decay class and assess error in models and predictions, and 3) quantify proportion-remaining volume of SDT components from terrestrial LiDAR-derived volumes relative to predicted intact tree volumes. This study represents the first to develop empirically-based, terrestrial LiDAR-derived allometric volume relationships and proportion-remaining volume of SDT by decay class. Results indicate that terrestrial LiDAR-derived volumes of SDT produced robust allometric equations by decay class for total above-stump and stem-plus-bark components (adjusted R2 = 0.94–0.98). Allometric relationships for tops-and-branches comprised more variability, likely impacted by scan quality, having adjusted R2 values of ~0.52–0.59. Notably, the inclusion of height in allometric relationships for total above-stump volume precluded the need for decay class as a covariate, accounting for the variability inherent in each decay class. Importantly, this means that total above-stump allometric equations could be effective under different decay class systems or, more broadly, where no decay classes were measured, thus providing broad utility. Empirically-derived proportion-remaining volume of SDT components followed expected decreasing trends by decay class. Interestingly, proportion-remaining volume for tops-and-branches closely matched theoretically-derived values from a previous FIA-related study. Ultimately, terrestrial LiDAR was critical for efficiently measuring volume of southern pine SDT by decay class and for developing SDT-specific allometric relationships of volume and estimates of structural change by decay class. This study showcases the feasibility of LiDAR-derived, SDT-specific tools for improved accounting of SDT resources in FIA and other inventories.

中文翻译：

---

非破坏性地面激光雷达对立枯南方松树异速生长和结构体积变化的研究

摘要 重大干扰造成的树木死亡会大大增加森林地区的死木，影响火灾强度和行为、野生动物栖息地和碳动态。由美国林务局森林清查和分析 (FIA) 计划进行的准确量化区域立枯树 (SDT) 池仍然是一个突出的挑战。由于与 SDT 相关的测量和采样挑战，几乎没有做过关于衰减类别中 SDT 体积结构变化的实证工作。光探测和测距 (LiDAR) 代表一种遥感技术，具有提高采样效率和通过无损采样提供 SDT 体积估计的潜力。按照此，本研究的目的是探索使用陆地 LiDAR 通过衰减等级经验量化和评估南方松 SDT 结构体积的可行性。为了实现这一目标，我们解决了三个目标，1) 使用地面 LiDAR 和基于体素的体积计算算法，通过衰减类构建 SDT 的经验体积估计，该算法能够考虑遮挡和点云质量，2) 开发地上的异速生长关系按衰减类别划分的 SDT 组件体积并评估模型和预测中的误差，以及 3) 相对于预测的完整树木体积，量化来自地面 LiDAR 衍生体积的 SDT 组件的剩余比例。这项研究代表了第一个以经验为基础的研究，地面 LiDAR 衍生的异速生长体积关系和 SDT 的比例-剩余体积按衰减类别。结果表明，地面 LiDAR 衍生的 SDT 体积按衰变类别生成了可靠的异速生长方程，用于总树桩上方和茎加树皮成分（调整后的 R2 = 0.94-0.98）。顶部和分支的异速生长关系包含更多的可变性，可能受扫描质量的影响，调整后的 R2 值约为 0.52–0.59。值得注意的是，在树桩总体积的异速生长关系中包含高度排除了将衰变类作为协变量的需要，从而解释了每个衰变类中固有的可变性。重要的是，这意味着总的树桩上异速生长方程在不同的衰变等级系统下或者更广泛地说，在没有测量衰变等级的情况下可能是有效的，从而提供广泛的效用。经验得出的 SDT 成分的剩余比例遵循衰减类别的预期下降趋势。有趣的是，顶部和分支的剩余比例与之前 FIA 相关研究的理论推导值密切匹配。最终，地面 LiDAR 对于按衰减等级有效测量南方松 SDT 的体积以及开发 SDT 特定的体积异速生长关系以及按衰减等级估计结构变化至关重要。本研究展示了 LiDAR 衍生的 SDT 特定工具在 FIA 和其他库存中改进 SDT 资源核算的可行性。顶部和分支的剩余比例与之前 FIA 相关研究的理论推导值密切匹配。最终，地面 LiDAR 对于按衰减等级有效测量南方松 SDT 的体积以及开发 SDT 特定的体积异速生长关系以及按衰减等级估计结构变化至关重要。本研究展示了 LiDAR 衍生的 SDT 特定工具在 FIA 和其他库存中改进 SDT 资源核算的可行性。顶部和分支的剩余比例与之前 FIA 相关研究的理论推导值密切匹配。最终，地面 LiDAR 对于按衰减等级有效测量南方松 SDT 的体积以及开发 SDT 特定的体积异速生长关系以及按衰减等级估计结构变化至关重要。本研究展示了 LiDAR 衍生的 SDT 特定工具在 FIA 和其他库存中改进 SDT 资源核算的可行性。