Using Machine Learning to Parameterize Moist Convection: Potential for Modeling of Climate, Climate Change, and Extreme Events
Using Machine Learning to Parameterize Moist Convection: Potential for Modeling of Climate, Climate Change, and Extreme Events
复制标题
DOI：
10.1029/2018ms001351
复制
发表时间：
2018-10-01
影响因子：
6.8
通讯作者：
Dwyer, John G.
中科院分区：
文献类型：
--
作者：
O'Gorman, Paul A.;Dwyer, John G.
湿对流的参数化导致了气候模拟和数值天气预报的不确定性。机器学习（ML）可用于直接从高分辨率模型输出中学习新的参数化方法，但对于这种参数化方法在完全耦合到一个大气环流模式（GCM）中时的表现，以及它们对于气候变化或极端事件模拟是否有用，人们仍知之甚少。在此，我们利用理想化测试来关注这些问题，在测试中，一个基于ML的参数化方法是根据传统参数化方法的输出进行训练的，并在GCM模拟中评估其性能。我们使用决策树集合（随机森林）作为ML算法，它的优势在于能自动确保能量守恒和地表降水非负性。具有ML对流参数化的GCM运行稳定，并能准确捕捉包括极端降水在内的重要气候统计数据，且无需针对极端情况进行特殊训练。如果ML参数化仅在对照气候下进行训练，则无法捕捉对照气候和暖气候之间的气候变化，但如果训练包括来自两种气候的样本，则可以捕捉到气候变化。值得注意的是，仅在暖气候下进行训练时也能捕捉到气候变化，这是因为暖气候的温带地区为对照气候的热带地区提供了训练样本。除了可能对气候模拟有用之外，我们还表明可以对ML参数化进行探究，以提供对流与大尺度环境之间相互作用的诊断信息。
通俗语言概要：云等小尺度特征在气候模型中通常由简化的物理模型来表示，这些简化模型会引入误差和不确定性。一种有前景的替代方法是使用机器学习来训练一个统计模型，该模型基于能更好地表示小尺度过程的昂贵的基于物理的模型的输出来表示小尺度过程。在此我们利用理想化测试来探究在气候模型中纳入一个大气对流的机器学习模型的影响。我们发现这种方法能够对平均气候和强降雨事件进行准确模拟。如果机器学习模型仅在当前气候下进行训练，那么它在全球变暖问题上效果不佳。然而，如果在当前气候和更暖的气候下都进行训练，它在全球变暖问题上效果良好，并且如果仅在更暖的气候下进行训练，它的效果也出奇地好。我们还表明机器学习模型可用于更好地理解潜在的物理过程。
The parameterization of moist convection contributes to uncertainty in climate modeling and numerical weather prediction. Machine learning (ML) can be used to learn new parameterizations directly from high-resolution model output, but it remains poorly understood how such parameterizations behave when fully coupled in a general circulation model (GCM) and whether they are useful for simulations of climate change or extreme events. Here we focus on these issues using idealized tests in which an ML-based parameterization is trained on output from a conventional parameterization and its performance is assessed in simulations with a GCM. We use an ensemble of decision trees (random forest) as the ML algorithm, and this has the advantage that it automatically ensures conservation of energy and nonnegativity of surface precipitation. The GCM with the ML convective parameterization runs stably and accurately captures important climate statistics including precipitation extremes without the need for special training on extremes. Climate change between a control climate and a warm climate is not captured if the ML parameterization is only trained on the control climate, but it is captured if the training includes samples from both climates. Remarkably, climate change is also captured when training only on the warm climate, and this is because the extratropics of the warm climate provides training samples for the tropics of the control climate. In addition to being potentially useful for the simulation of climate, we show that ML parameterizations can be interrogated to provide diagnostics of the interaction between convection and the large-scale environment.Plain Language Summary Small-scale features such as clouds are typically represented in climate models by simplified physical models, and these simplified models introduce errors and uncertainties. A promising alternative approach is to use machine learning to train a statistical model to represent small-scale processes based on output from expensive physics-based models that better represent the small-scale processes. Here we use idealized tests to explore the implications of incorporating a machine-learning model of atmospheric convection in a climate model. We find that such an approach can give accurate simulations of mean climate and heavy rainfall events. The machine-learning model does not work well for global warming if it is only trained on the current climate. However, it does work well for global warming if trained on both the current and warmer climates, and it works surprisingly well if only trained on the warmer climate. We also show that the machine-learning model can be used to better understand the underlying physical processes.