Detection and quantification of local methane sources using novel high-resolution satellite data

人为的温室气体正在增加地球的温度。在这些温室气体中，甲烷(CH4)的重要性仅次于二氧化碳(CO2)。虽然甲烷最终被氧化成二氧化碳的时间尺度为9年，但在其生命周期内，其每个分子的温室效应要大得多。大气CH4水平在经历了1990年代和2000年代初的短暂增长停滞后正在上升，这令人惊讶(例如，McNorton等人。2016)。恢复增长似乎与人为排放的持续增加相一致，但对这些变化还没有出现明确的解释。有多种人为的CH4来源过程，包括农业、湿地、稻田和化石燃料生产的排放--例如煤炭开采或水力压裂。了解甲烷来源的大小和位置对于采取政策行动限制这种短期温室气体的排放至关重要，这种温室气体也与通过臭氧产生而导致的当地空气质量恶化有关。自20世纪80年代以来，大气甲烷浓度一直在全球站点网络中进行现场测量，主要由美国国家海洋和大气管理局(NOAA)协调。最近，卫星对大气CH4的估计已经成为可能，但这种数据的空间覆盖范围仍然相对粗略。Sentinel 5P卫星上的欧空局Tropomi仪器彻底改变了这种情况，该仪器自2017年10月以来一直在记录大气CH4的观测。该项目旨在利用Tropomi卫星的高空间和时间分辨率(每日覆盖，图1)，这是完全前所未有的。结合使用代表大气CH4经历的化学和物理传输过程的模型，这一新数据开启了比以前更详细地解析CH4来源的可能性，将它们与生物或人类活动等特定过程联系起来。到目前为止，Tropomi数据很少被用来确定人为排放的“热点”，或更好地了解各种源过程的变化，从而更全面地了解全球CH4循环。图1.来自Tropomi的北非上空平均大气柱甲烷混合比的一个例子。来自Hu等人。(2018).为了充分利用这些新的甲烷观测，有必要也用高分辨率表示模拟的大气传输。该项目将开发相对粗略的大气输送模式TOMCAT(Chipperfield等人。2006)构建了一个版本，该版本允许在特定感兴趣区域上插入--或“嵌套”--有限的高分辨率网格。由于TOMCAT模型在正向和反向模式下均可工作，因此新的嵌套高分辨率模型将用于识别和量化特定感兴趣区域的CH4源。最初的工作将包括使用当前版本的TOMCAT模型和Tropomi数据之间的比较，以便在学生继续开发TOMCAT模型中的嵌套网格之前，得出关于全球CH4循环最近变化的大规模结论。一旦完成这项工作，对感兴趣的特定地点的关注将使人们能够通过使用TOMCAT的反向版本对一个特别鲜为人知的甲烷来源区域进行高分辨率的案例研究(Wilson等人，2014年)。作为项目的一部分，学生将在英国气象局工作，以熟悉更多种类的大气建模技术，并利用那里的专业知识，以及在该组织工作的经验。

Anthropogenic greenhouse gases are increasing the Earth's temperature. Among these greenhouse gases, methane (CH4) is second in importance after carbon dioxide (CO2). While CH4 is eventually oxidised to CO2 on a timescale of 9 years, during its lifetime its greenhouse warming effect per molecule is much larger. Atmospheric CH4 levels are rising, following a brief period of stalled growth during the 1990s and early 2000s, which came as a surprise (e.g. McNorton et al. 2016). The resumed growth seems consistent with continuing increases in anthropogenic emissions but clear explanations for these changes have not yet emerged. There is a wide range of anthropogenic CH4 source processes, including emissions from agriculture, wetlands, rice paddies and fossil fuel production - e.g. from coal mining or fracking. Understanding the magnitudes and locations of CH4 sources is essential for informing policy action to limit emissions of this short-lived greenhouse gas, which is also linked to degradation in local air quality through ozone production.Atmospheric CH4 concentrations have been measured in situ at a global network of sites since the 1980s, coordinated primarily by the U.S. National Oceanic and Atmospheric Administration (NOAA). More recently, satellite estimates of atmospheric CH4 have become available, but spatial coverage of such data were still relatively coarse. This situation has been revolutionised with the ESA TROPOMI instrument on the Sentinel 5P satellite which has been recording observations of atmospheric CH4 since October 2017.This project aims to exploit the high spatial and time resolution (daily coverage, Fig. 1) of the TROPOMI satellite, which is entirely unprecedented. Together with the use of models which represent the chemical and physical transport processes undergone by atmospheric CH4, this new data opens up the possibility to resolve CH4 sources in much greater detail than before, relating them to specific processes like biogenic or anthropogenic activity. So far, TROPOMI data have been little exploited either for identifying anthropogenic emissions 'hotspots' or to obtain a better understanding of changes in various source processes and thus to gain a more complete understanding of the global CH4 cycle.Figure 1. An example of mean atmospheric column methane mixing ratio over northern Africa from TROPOMI. From Hu et al. (2018).To take full advantage of these new methane observations it is necessary that the modelled atmospheric transport is also represented in high resolution. This project will develop the relatively coarse atmospheric transport model TOMCAT (Chipperfield et al. 2006) to build a version which allows the insertion - or 'nesting' - of a limited high resolution grid over specific regions of interest. Since the TOMCAT model works in both a 'forwards' and 'inverse' mode, the new nested high resolution model will then be applied to identify and quantify CH4 sources for specific regions of interest.The initial work will involve using comparisons between the current version of the TOMCAT model and the TROPOMI data in order to draw large-scale conclusions about recent changes in the global CH4 cycle, before the student moves on to develop the nested grid within the TOMCAT model. Once this has been completed, a focus on a particular location of interest will allow for a high-resolution case study of a particularly poorly understood methane source region through use of the inverse version of TOMCAT (Wilson et al., 2014). As part of the project the student will spend time at the UK Met Office to gain familiarity with a greater variety of atmospheric modelling techniques and to take advantage of the expertise there, along with experience of working within the organisation.