CAREER: Forecasting Impacts to Reduce Exposure to Smoke (FIRES) - Modeling wildfire smoke transport in the western U.S.

美国最大的野火发生在该国西部，该地区人口超过7700万。野火产生的烟雾对人类有害，减少与烟雾相关的疾病需要限制受火灾影响地区的户外工作和其他活动。这些影响区域可以延伸到距离大火数百英里的地方，因此准确的预测对保护人类健康至关重要。目前，天气和空气质量模型被用来向公众提供烟雾警告。然而，迫切需要对这些模型进行改进，以正确预测山区上空的烟雾传输。这项研究项目将对这些模型进行改造，并推进山区动态大气模拟的最新科学。新模型将用于开发在线烟雾预测工具，在美国西部发生野火事件时提供警告，以保护人类健康。该项目将通过开发在线计算建模课程来加强本科STEM教育，填补现有课程的空白。总之，这些教育产品将增强我们通过准确的野火预报保护人类健康的能力，同时培训下一代预报员并提高国家的科学素养。由于大气流动的复杂性和估计烟羽喷射高度的困难，模拟野火烟雾传输的区域尺度空气模式在山区地形上有很大的不确定性。为了改进烟羽预报，迫切需要开发新的模型，以减少与气象条件和排放模型相关的不确定性。这项研究的目的是使用跨学科的方法来提高我们对野火烟雾羽流动力学和控制山区地形上烟雾传输的复杂大气流动的基本理解。这项工作的成功完成有可能转变我们对大气-火-人类系统耦合的知识，并极大地提高我们保护人类健康的能力。其核心假设是，为平坦地形开发的大气湍流参数不能正确模拟山地地形上的行星边界层结构，导致野火烟羽的化学输送模拟存在偏差。本项目的四个研究目标是：1)改进山区垂直混合模式，以减少模拟烟羽输送的不确定性；2)开发一个新的烟羽喷射高度模式，以改善野火烟雾排放浓度的垂直分布和后续的区域输送；3)将大气混合和野火排放的模式改进纳入化学输送模式，以改进美国西部的烟羽预报；4)利用改进的区域尺度烟羽输送模式，创建一个在线烟雾预报工具。该项目的教育和推广目标是通过开发一门专注于数值天气预测建模和高性能计算的在线课程，为本科生提供更多的数值教育机会，并加强与当地利益攸关方的合作，为内华达州与野火科学相关的本科生提供独特的服务学习风格的研究体验。这项研究的主要产品将是与数据可视化和通信专家合作设计的在线烟雾预报工具。这一在线工具将与当地利益相关者共享，通过发布有助于减少野火烟雾暴露的警告，帮助公众了解糟糕的空气质量事件。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The largest wildfires in the United States occur in the western part of the country; an area with a population of over 77 million people. Smoke from wildfires is harmful to humans and reducing smoke-related illnesses requires limiting outdoor work and other activities in areas impacted by fire. These areas of impact can extend several hundreds of miles away from the fire, thus accurate prediction is critical to protect human health. Currently, weather and air quality models are used to provide smoke warnings to the public. However, there is a critical need to improve these models to correctly forecast the smoke transport over mountains. This research project will transform these models and advance the state-of-the-science of modeling the dynamic atmosphere over mountainous terrain. The new model will be used to develop an online smoke forecasting tool that provides warnings to protect human health during wildfire events in the western U.S. This project will enhance undergraduate STEM education by developing an online computational modeling course that fills a gap in existing curricula. Together, these educational products will enhance our ability to protect human health through accurate wildfire forecasting, while training the next generation of forecasters and increasing the scientific literacy of the Nation. Regional scale air models that simulate wildfire smoke transport have significant uncertainties over mountainous terrain due to the complexities of the atmospheric flows and difficulties in estimating smoke plume injection heights. To improve smoke plume forecasts there is a critical need to develop new models that reduce uncertainties associated with both meteorological conditions and emissions modeling. The aim of this research is to use a cross-disciplinary approach to improve our fundamental understanding of wildfire smoke plume dynamics and complex atmospheric flows governing smoke transport over mountainous terrain. Successful completion of this work has the potential to transform our knowledge of the coupled atmosphere-fire-human system and greatly improve our ability to protect human health. The central hypothesis is that atmospheric turbulence parameterizations developed for flat terrain do not correctly simulate the planetary boundary layer structure over mountainous terrain leading to biases in the chemical transport modeling of wildfire smoke plumes. The four research objectives of this project are to: 1) improve models for vertical mixing over mountains to reduce the uncertainties in modeling smoke plume transport, 2) develop a novel model for smoke plume injection height to improve the vertical distribution of wildfire smoke emissions concentrations and subsequent regional transport, 3) incorporate model improvements for atmospheric mixing and wildfire emissions into a chemical transport model to improve smoke plume forecasts in the western U.S., and 4) create an online smoke forecasting tool using the improved regional scale smoke plume transport model. The education and outreach objectives of this project are to enhance numerical educational opportunities for undergraduate students by developing an online course focusing on numerical weather prediction modeling and high-performance computing and increase collaboration with local stakeholders to provide unique service-learning style research experiences for undergraduate students related to wildfire science in Nevada. A primary product of this research will be an online smoke forecasting tool designed in collaboration with a data visualization and communication expert. This online tool will be shared with local stakeholders to help inform the public about poor air quality events by issuing warnings that aid in reducing wildfire smoke exposure.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.