Air pollution is of major concern throughout the world and the use of modeling tools to analyze and forecast the pollutant concentrations in complex orographic areas remains challenging. This work proposes an exhaustive framework to analyze the ability of models to simulate the air quality over the French Alps up to 1.2 km resolution over Grenoble and the Arve Valley. The on-line coupled suite of models CHIMERE-WRF is used in its recent version to analyze a 1 month episode in November–December 2013. As expected, an improved resolution increases the concentrations close to the emission areas and reduced the negative bias for Particulate Matter that is the usual weakness of air quality models. However, the nitrate concentrations seem overestimated with at the same time an overestimation of surface temperature in the morning by WRF. Different WRF settings found in the literature are tested to improve the results, particularly the ability of the meteorological model to simulate the strong thermal inversions in the morning. Wood burning is one of the main contributor of air pollution during the period ranging from 80 to 90% of the Organic Matter. The activation of the on-line coupling has a moderate impact on the background concentrations but surprisingly a change of Particulate Matter (PM) concentrations in the valley will affect more the meteorology nearby high altitude areas than in the valley. This phenomenon is the result of a chain of processes involving the radiative effects and the water vapor column gradients in complex orographic areas. At last, the model confirms that the surrounding glaciers are largely impacted by long range transport of desert dust. However, in wintertime some outbreaks of anthropogenic pollution from the valley when the synoptic situation changes can be advected up to the nearby high altitude areas, then deposited.

中文翻译：

---

在法国阿尔卑斯山上通过在线CHIMERE-WRF模型进行的高分辨率化学物质运输建模—对山区气象学中表面颗粒物浓度的反馈进行的分析

空气污染是世界范围内最为关注的问题，使用建模工具来分析和预测复杂地形区域中的污染物浓度仍然具有挑战性。这项工作提出了一个详尽的框架，以分析模型对法国阿尔卑斯山的空气质量进行模拟的能力，该模型在格勒诺布尔和阿尔维谷地的分辨率高达1.2 km。CHIMERE-WRF模型的在线耦合套件在其最新版本中用于分析2013年11月至12月的1个月情节。如预期的那样，提高的分辨率增加了靠近发射区的浓度并减少了对微粒的负偏差。这是空气质量模型通常的缺点。但是，硝酸盐浓度似乎高估了，同时WRF早晨高估了地表温度。测试了文献中发现的不同WRF设置以改善结果，尤其是气象模型模拟早晨强烈的热反演的能力。在这段时间内，木材燃烧是造成空气污染的主要因素之一，占有机物的80％至90％。在线耦合的激活对背景浓度有中等程度的影响，但令人惊讶的是，山谷中颗粒物（PM）浓度的变化将比山谷中对高海拔地区的气象影响更大。这种现象是一系列过程的结果，这些过程涉及复杂地形区域中的辐射效应和水蒸气柱梯度。最后，该模型证实了周围冰川受到沙漠尘埃远距离迁移的很大影响。然而，