Predicting post-fire hillslope erosion in forest lands of the western United States
Predicting post-fire hillslope erosion in forest lands of the western United States
复制标题
DOI：
10.1071/wf09142
复制
发表时间：
2011
影响因子：
3.1
通讯作者：
M. E. Miller;L. Macdonald;P. Robichaud;W. Elliot
中科院分区：
文献类型：
--
作者：
M. E. Miller;L. Macdonald;P. Robichaud;W. Elliot
由于燃料积聚和气候变化，许多森林及其相关水资源面临着越来越大的严重野火的风险。目前正在提议广泛的燃料治疗，但尚不清楚这种治疗应该集中在哪里。该项目的目标是：(1)预测美国西部森林和灌木丛火灾后潜在的侵蚀率，以帮助确定燃料处理的优先顺序；(2)评估模型的敏感性和准确性。利用历史火灾天气资料和一阶火灾效应模型对火灾后地面覆盖进行了预测。将扰动水蚀预测项目(WEPP)中的参数文件与GeoWEPP相结合，用于预测山坡尺度的火灾后侵蚀。预测的中位数年侵蚀率为0.1-2 mg ha-1年-1，太平洋沿岸湿润地区约10-40 mg ha-1，加利福尼亚州西北部高达100 mg ha-1年-1。敏感性分析表明，预测的侵蚀速率主要受降雨量的控制，而不是由地表覆盖率控制。有限的验证数据集显示了预测和测量的侵蚀速率之间的合理相关性(R2=0.61)，尽管预测的值比实测值要小得多。我们的结果证明了大范围预测火灾后侵蚀率的可行性。验证和敏感性分析表明，这些预测对于在当地而不是区域间范围内确定燃料减少处理的优先顺序最有用，它们还有助于确定模型改进和研究需求。
Many forests and their associated water resources are at increasing risk from large and severe wildfires due to high fuel accumulations and climate change. Extensive fuel treatments are being proposed, but it is not clear where such treatments should be focussed. The goals of this project were to: (1) predict potential post-fire erosion rates for forests and shrublands in the western United States to help prioritise fuel treatments; and (2) assess model sensitivity and accuracy. Post-fire ground cover was predicted using historical fire weather data and the First Order Fire Effects Model. Parameter files from the Disturbed Water Erosion Prediction Project (WEPP) were combined with GeoWEPP to predict post-fire erosion at the hillslope scale. Predicted median annual erosion rates were 0.1–2 Mg ha–1 year–1 for most of the intermountain west, ~10–40 Mg ha–1 year–1 for wetter areas along the Pacific Coast and up to 100 Mg ha–1 year–1 for north-western California. Sensitivity analyses showed the predicted erosion rates were predominantly controlled by the amount of precipitation rather than surface cover. The limited validation dataset showed a reasonable correlation between predicted and measured erosion rates (R2 = 0.61), although predictions were much less than measured values. Our results demonstrate the feasibility of predicting post-fire erosion rates on a large scale. The validation and sensitivity analysis indicated that the predictions are most useful for prioritising fuel reduction treatments on a local rather than interregional scale, and they also helped identify model improvements and research needs.