Development and application of an adjoint version for a multi-phase air quality model

数学模型在塑造我们对空气污染的理解以及确定将其降低到可接受水平的可能解决方案方面发挥着越来越重要的作用。科学家、监管者和政策制定者使用空气质量模型来洞察空气质量，并做出明智的决定，以了解人类活动对空气质量的影响。用于这类目的的一种重要方法是敏感性分析。敏感性分析工具可以为空气污染的主要问题提供答案：排放的变化将如何影响空气质量？已经开发了复杂的数学工具来计算空气质量模型中的灵敏度信息。其中一些方法是以污染源为基础的，提供了有关排放源的详细敏感性信息。其他的则是基于受体的，因此被认为是解决政策问题的理想选择。在这个项目中，目标是开发一组基于受体的工具，称为伴随工具，用于广泛使用的空气质量模型。它旨在将开发的伴随模型用于各种应用，包括：1)改进空气质量预报，2)更好地估计模型中使用的排放清单，3)设计实现空气质量目标的最佳污染控制策略并进行成本效益分析，以及4)对跨省、跨境和洲际污染运输的定量评估。

Mathematical models play an increasingly significant role in shaping our understanding of air pollution, and in identifying possible solutions for reducing it to acceptable levels. Scientists, regulators, and policymakers use air quality models to gain insight and make informed decisions about the quality of air and ways that it can be impacted by human activities. An important method that is often used for such purposes is sensitivity analysis. Sensitivity analysis tools can provide answers to the main question with respect to air pollution: how air quality will be affected by changes in emissions? Sophisticated mathematical tools have been developed to calculate sensitivity information in air quality models. Some of these methods are source-based and provide detailed sensitivity information about emission sources. Others are receptor-based and are, therefore, deemed to be ideal for addressing policy questions. In this project, the goal is to develop one group of such receptor-based tools, referred to as an adjoint, for a widely used air quality model. It is intended to use the developed adjoint model for various applications that include: 1) improved air quality forecasting, 2) developing better estimation of emission inventories that are used in the models, 3) design of optimal pollution control strategies for reaching air quality objectives and cost-benefit analysis, and 4) quantitative evaluation of inter-provincial, trans-border, and intercontinental transport of pollution.