Collaborative Research: Predicting post-wildfire sedimentation of reservoirs: probabilistic modeling of debris flow generation and downstream sediment routing

在美国，林地为三分之二的人口提供水源。然而，在美国西部建造大部分水利基础设施的几十年里，野火的燃烧面积、频率和严重程度都大大增加。虽然野火会对供水的数量和质量产生短期影响，但严重烧伤后发生的侵蚀也会向河流和下游水库输送大量沉积物，减少供水的长期储存能力。此外，随着预计未来野火的增加，河流沉积物将增加。因此，在这个项目中，研究人员将开发新的基于计算机的建模工具，能够识别和量化野火后侵蚀对下游水利基础设施造成的风险。该建模框架的第一个应用将是整个犹他州的供水水库，犹他州是美国最干旱的州之一，其中将量化每个水库在野火造成的侵蚀和沉积风险方面的脆弱性。与全国各地的水坝水库类似，犹他州水库的沉积是一个日益关注的水基础设施老化，甚至在考虑到未来野火的预计增加。最后，研究人员将把他们的模型整合到在线开源程序中，使任何有兴趣将该模型应用于其他州或地区的个人或机构都可以使用这些资源。该项目的交付成果将为改进和更有针对性的森林管理提供关键信息和工具，帮助确定和保护脆弱的水资源，并填补预测野火后侵蚀对下游影响的关键知识空白。该项目将在两所大学合作，为一名博士后研究员（PI Murphy），两名博士生和至少六名本科生提供支持，以培训和发展他们在水文学，地貌学，数据分析和管理以及科学传播方面的技能。该项目将推进对预测野火后沉积物输送到下游水基础设施的位置和时间至关重要的基础知识。研究人员将把新的和现有的模型联系起来：1）预测野火后侵蚀的位置和程度，2）以基于物理和水文地貌敏感的方式通过河流网络将野火后沉积物输入下游，3）确定一系列潜在的体积沉积物输入到下游水库在一系列野火条件下。将这一新的建模框架应用于整个犹他州的133个主要水库，该项目将回答四个关键的研究问题：1）犹他州哪些供水水库最容易受到野火后的侵蚀？2)野火发生和下游水库蓄水损失之间的时间差是多少？3)哪种景观、火灾、水文和植被特征对上游储存和向水库输送火灾后沉积物具有最强的控制作用？4)什么样的景观、火灾和河流网络属性控制了野火后沉积物产量高于背景产量的相对增加？通过这项分析，研究人员将专门评估沉积物连通性对水库脆弱性的影响，以及粗沉积物输入对水库蓄水量减少的贡献。鉴于水坝的所有权和管理复杂，他们将聘请一个利益相关者咨询小组，跨越不同的所有权范围，包括公用事业部门，州和联邦森林管理机构，以及水坝运营商。此外，他们将与社区表面动力学建模系统（CSDMS）合作，将他们的模型集成到开源平台中，并创建一个公共平台来托管项目数据集，教育材料，技术报告和出版物。该项目代表了综合地球科学前沿的研究，这一新的建模框架填补了评估野火和水安全等紧迫社会问题所需工具的关键空白。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

In the United States, forested lands provide water supply for two-thirds of the population. However, in the decades since most water infrastructure was constructed in the western U.S., the burned area, frequency, and severity of wildfires has increased considerably. While wildfires can have short-term impacts on the quantity and quality of water supply, the erosion that occurs after severe burns can also deliver significant amounts of sediment to rivers and downstream reservoirs, reducing the long-term storage capacity of water supplies. Further, with projected future increases in wildfire, there will be increases in river sediment. Thus, in this project the researchers will develop new computer-based modeling tools capable of identifying and quantifying the risk that post-wildfire erosion poses to downstream water infrastructure. The first application of this modeling framework will be the water supply reservoirs throughout Utah, one of the driest states in the U.S., where the vulnerability of each reservoir will be quantified as to the erosion and sedimentation risk posed by wildfire. Similar to dammed reservoirs across the nation, sedimentation in Utah reservoirs is a growing concern for aging water infrastructure, even before accounting for the projected increases in future wildfire. Finally, the researchers will integrate their model into online, open-source programs, making these resources available to any person or agency interested in applying the model to other states or regions. The deliverables of this project will provide critical information and tools for improved and more targeted forest management, help identify and protect vulnerable water resources, and address crucial knowledge gaps for predicting downstream impacts from post-wildfire erosion. Collaborating across two universities, this project will provide support for one post-doctoral researcher (PI Murphy), two PhD students, and a minimum of six undergraduate students to train and develop their skills in hydrology, geomorphology, data analysis and management, and science communication. This project will advance fundamental knowledge critical for predicting the locations and timing of post-wildfire sediment delivery to downstream water infrastructure. The researchers will link new and existing models that: 1) predict the locations and magnitudes of post-wildfire erosion, 2) route post-wildfire sediment inputs downstream through river networks in a physics-based and hydro-geomorphically sensitive manner, and 3) determine a range of potential volumetric sediment inputs to downstream reservoirs under a range of wildfire conditions. Applying this new modeling framework to the 133 major reservoirs throughout Utah, this project will answer four key research questions: 1) Which water supply reservoirs in Utah are most vulnerable to post-wildfire erosion? 2) What is the time lag between occurrence of a wildfire and loss of reservoir storage downstream? 3) Which landscape, fire, hydrologic, and vegetation characteristics exert the strongest control on the upstream storage vs. delivery of post-fire sediment to reservoirs? 4) What landscape, fire and river network attributes control the relative increase in post-wildfire sediment yields above background yields? Through this analysis, the researchers will specifically assess the influence of sediment connectivity on reservoir vulnerability, as well as the contribution of coarse sediment inputs to the reductions in reservoir storage over longer transport timescales. Given the complex ownership and management of dams, they will engage a stakeholder advisory group that spans the diverse range of ownership and includes public utilities departments, state and federal forest management agencies, and dam operators. Further, they will work with the Community Surface Dynamics Modeling System (CSDMS) to integrate their models into open-source platforms, and create a public platform to host the project datasets, educational materials, technical reports, and publications. This project represents research at the frontier of integrated geosciences, and this new modeling framework fills a critical gap regarding the tools needed to assess urgent societal concerns regarding wildfire and water security.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.