Trapped lee waves as a source of low-level drag on the atmosphere

用于数值天气预报和气候预估的大气模式在表示造成模拟大气环流误差的小尺度过程方面具有很大的不确定性。事实上，政府间气候变化专门委员会（IPCC）的最新报告认为，根据大气模型推断出的全球变暖情景下对流层环流变化的预测可信度很低。一个对大气环流有重大贡献但在大气模式中没有充分体现的特殊过程是小山脉产生的阻力（即摩擦力）。在这个项目中，你将使用理论、高分辨率数值模型模拟和观测来研究这一缺失过程的重要性。当大气流过山脉时，会产生内部大气波（称为地形重力波），对大气施加阻力，使局地和远距离的大尺度环流减速。这种阻力的很大一部分是由水平尺度上的山脉造成的，而这些山脉在通常用于天气预报或气候和季节预测的模式中部分或完全没有表现出来，因此它们对环流的影响是通过称为参数化的近似来解释的。现有的参数化集中在垂直传播的地形重力波产生的阻力上，这种阻力通常在高海拔的破波区域内发生作用，有时达到平流层（海拔约10公里以上的大气层）甚至更高。然而，其他的地形重力波，被称为被困的背风波，也会对大气产生阻力。这些被困的背风波具有更小的水平尺度，并在较低的高度水平传播（通过云层排列可见-见下图），但在大多数天气和气候模型中没有被参数化。在强稳定分层的大气边界层（常见于陆地上空的高纬度地区），湍流与重力波共存。为了尽量减少预测地表温度和地表风角（例如，影响温带气旋衰减率）的误差，湍流混合常常不切实际地增加。这种增加有一个负面的副作用，即对边界层深度和低层急流的高度（夜间边界层的一个特征）估计过高，其因子多达两个，并可能导致对最低和最高温度的预测有偏差。造成这些问题的一个可能原因是忽略了捕获的背风波阻力，并将其错误地表示为湍流形式的阻力，这必然会降低预测的质量，因为每个过程对流动参数的依赖是不同的。本项目旨在通过理论、数值模拟和观测，阐明捕获的背风波对施加在大气上的低空阻力的贡献。假设具有代表性层状结构的大气的理论将用于建立阻力对基本流动参数的粗略依赖关系。使用英国气象局的业务天气预报模型（统一模型，MetUM）的高分辨率（1公里网格间距）数值模拟将用于在理想化和现实情况下测试该理论，并改进由理论提出的用于天气和气候预测的捕获背风波的表示。低分辨率模拟将用于在操作环境中测试这些想法。现有的被困背风波观测资料，包括FAAM研究飞机在英国进行的飞机观测资料，将为理论和数值模拟结果的验证提供一个基础事实。

Atmospheric models used for numerical weather prediction and climate projections have significant uncertainties in their representation of small-scale processes that contribute to errors in the modelled atmospheric circulation. Indeed, the latest report from the Intergovernmental Panel on Climate Change (IPCC) places low confidence on projected tropospheric circulation changes under global warming scenarios, deduced from atmospheric models. One particular process which both contributes significantly to the atmospheric circulation and is not fully represented in atmospheric models is the drag (i.e. frictional) force produced by small-scale mountains. In this project you will investigate the importance of this missing process using theory, high-resolution numerical model simulations and observations.When the atmosphere flows over mountains, internal atmospheric waves (known as orographic gravity waves) are generated that exert a drag force on the atmosphere, acting to decelerate the large-scale circulation both locally and remotely. A large proportion of this drag is caused by mountains at horizontal scales that are either partially or totally unrepresented in models typically used for weather forecasting or climate and seasonal projection, so their influence on the circulation is accounted for through approximations called parameterizations. Existing parametrizations focus on the drag produced by vertically propagating orographic gravity waves, which typically acts within wave breaking regions at high altitudes, sometimes reaching as high as the stratosphere (the atmospheric layer above about 10 km altitude) or even above. However, other orographic gravity waves, known as trapped lee waves, are also known to exert a drag on the atmosphere. These trapped lee waves have even smaller horizontal scales and propagate horizontally at lower levels (being made visible by cloud alignments - see figure below), but are not parametrized in most weather and climate models.In strongly stably stratified atmospheric boundary layers (common at high latitudes over land), turbulence co-exists with gravity waves. In an effort to minimize errors in the predicted surface temperature and surface wind angle (which affects, for example, the decay rate of extratropical cyclones), turbulent mixing is often unrealistically increased. This increase has the negative side-effect of overestimating the boundary layer depth and the height of the low-level jet (a feature of nocturnal boundary layers) by factors as large as two, and can result in biased predictions of minimum and maximum temperatures. A likely reason for these problems is the omission of trapped lee wave drag and its misrepresentation as turbulent form drag, which necessarily degrades forecasts, as the dependence of each of these processes on the flow parameters is different.This project aims to clarify the contribution of trapped lee waves to low-level drag exerted on the atmosphere using theory, numerical simulations and observations. Theory assuming atmospheres with a representative layered structure will be used to establish the rough dependence of the drag on basic flow parameters. Very high-resolution (1km grid spacing) numerical simulations using the Met Office's operational weather forecast model (the Unified Model, MetUM) will be used to test the theory in idealized and realistic cases, and improve the representations of trapped lee waves suggested by theory for weather and climate prediction. Lower-resolution simulations will be used to test these ideas in an operational context. A ground-truth for verification of results from theory and numerical simulations will be provided by available observations of trapped lee waves, including aircraft observations made in the UK by the FAAM research aircraft.