Model Development for Two-Phase Thin Layer Flows and Validation Using Debris Flows at Tungurahua, Ecuador and Ruapehu, New Zealand

泥流是最常见的自然灾害之一;它们经常与暴雨、地震或其他异常现象有关。它们往往造成巨大的生命和财产损失。这项研究将利用2007年3月在新西兰发生的一次实际的大型火山泥流的实地研究。建议开发和测试一种计算机模型，模拟固体颗粒和水的混合物的流动。 这项研究的一个重要部分是确定未来泥石流预测的确定性或不确定性程度。实地测量的分析将为测试和改进计算机模型提供基础。将进行一系列仔细的计算机模拟，以连续再现对实际流动进行的现场测量。可以对这一过程的结果进行分析，以提高预测世界其他地方类似和较小泥石流的置信度。该项目的第二个目标将是预测厄瓜多尔一座火山的较小泥石流，由于强降雨，该火山目前几乎每周都会产生类似的破坏性泥石流。本研究产生的模型和分析将提供更好的方法来对未来类似流动的灾害风险进行合理的预测。与间隙液混合的地质物质的流动是一个复杂的过程，在许多长度和时间尺度上激活物理学。该项目将扩展TITAN 2D计算环境，以包括流体-固体流动，并使用从新西兰Ruapehu的实地研究收集的数据来测试这些模型。统计分析将探讨对参数的敏感性问题，以及模型和测量误差，并将为预测提供定量基础。这项工作的一个特别重点是模型和校准2007年3月在新西兰鲁阿佩胡火山的火山口湖爆发流。该团队将与梅西大学的同事密切合作，利用他们在2007年爆发研究中收集的数据进一步完善数学和统计模型。梅西的同事们对Ruapehu进行了检测，这使他们能够从这一重要的流动事件中收集到特殊的数据集。这些数据包括事件前和事件后的航空激光雷达和数字摄影调查，显示了通道形态，沉积物侵蚀和再分布的后期变化。此外，用于机械、电磁、振动和压力检测系统的创新应用的仪器提供了对该事件中快速移动的沉积物-水泥浆内的速度、沉积物分布、流动行为和侵蚀/沉积过程的测量。这一独特的数据集将实现无与伦比的流量模型测试和校准。再加上统计方法，这些结果将作为一个标准，新一代的数值和物理质量流模型可以校准和完善。研究与其他地方类似火山爆发有关的更大规模危害问题的团体将对该项目的数据特别感兴趣。改进的TITAN模型预测较小的流量事件计划通过测试它在经常性的事件（在每周的时间范围内）在通古拉瓦火山在厄瓜多尔。

Mudflows are one of the most common natural hazards; they occur frequently associated with heavy rainfall, earthquakes or other unusual phenomena. They often cause a great loss of life and property. This research will use field studies of an actual large volcanic mudflow that occurred in New Zealand in March 2007. It is proposed to develop and test a computer model that simulates such flows that are a mixture of solid particles and water. An important part of this study is to determine the degree of certainty or uncertainty that is possible in forecasts of future mudflows. Analysis of field measurements will provide the basis for testing and improving the computer model. A careful series of computer simulations will be done to sequentially reproduce the field measurements taken of the actual flow. The results of this process can be analyzed to increase the confidence level in forecasting of similar and smaller mudflows elsewhere in the world. A secondary goal of the project will be to forecast smaller mudflows at a volcano in Ecuador that is currently producing similar destructive flows nearly every week due to heavy rainfall. The model and analysis resulting from this study will allow a better means to make sound predictions of hazard risk for similar flows in the future.The flow of geologic material mixed with interstitial fluid is a complex process activating physics across many length and time scales. This project will extend the TITAN2D computational environment to include fluid-solid flows, and to test these models using data collected from field studies at Ruapehu, New Zealand. Statistical analysis will probe questions of sensitivity to parameters, and of model and measurement error, and will provide a quantitative basis for prediction. A special focus of this effort is to model and calibrate the March 2007 breakout flow of the crater lake at Ruapehu Volcano in New Zealand. Working closely with colleagues at Massey University, the team will use data they collected in their study of the 2007 breakout to further refine mathematical and statistical models. The colleagues at Massey had instrumented Ruapehu, which allowed them to collect an exceptional data set from this important flow event. These data include pre- and post-event aerial LiDAR and digital photographic surveys that show postevent changes in channel morphology, sediment erosion, and redistribution. In addition, instruments for innovative applications of mechanical, electro-magnetic, vibration, and pressure detection systems provide measurements of velocities, sediment distribution, flow behavior, and erosion/deposition processes within the rapidly moving sediment-water slurries of this event. This unique dataset will enable unparalleled testing and calibration of flow models. Coupled with statistical methodology, these results will serve as a standard against which a new generation of numerical and physical mass-flow models can be calibrated and refined. Data from this project will be of particular interest to groups studying the broader-scale problem of hazards related to eruptions of similar volcanoes elsewhere. Refinement of the TITAN model to forecast smaller flow events is planned by testing it at the recurrent events (on a weekly time-frame) at Tungurahua Volcano in Ecuador.