Correcting Coarse‐Grid Weather and Climate Models by Machine Learning From Global Storm‐Resolving Simulations
Correcting Coarse‐Grid Weather and Climate Models by Machine Learning From Global Storm‐Resolving Simulations
复制标题
DOI：
10.1029/2021ms002794
复制
发表时间：
2021-09
影响因子：
6.8
通讯作者：
C. Bretherton;B. Henn;Anna Kwa;Noah D. Brenowitz;Oliver Watt‐Meyer;J. McGibbon;W. A. Perkins;Spencer K. Clark;Lucas Harris
中科院分区：
文献类型：
--
作者：
C. Bretherton;B. Henn;Anna Kwa;Noah D. Brenowitz;Oliver Watt‐Meyer;J. McGibbon;W. A. Perkins;Spencer K. Clark;Lucas Harris
水平网格间距小于5 km的全球大气“风暴解析”模式可以解析复杂地形下的深积云对流和流动。它们有望成为参考模型，可用于改进计算负担得起的粗网格全球气候模型，涵盖一系列气候，减少区域降水和温度趋势的不确定性。在这里，将轻推趋势作为列状态函数的机器学习用于纠正真实地理粗网格模型（带有200公里网格的FV3GFS）中的温度、湿度和可选风的物理参数化趋势，使其更接近使用X‐SHiELD（带有3公里网格的FV3GFS的修改版本）进行的40天参考模拟。两种模拟都指定了相同的历史海面温度场。该方法建立在先前的研究基础上，使用全球观测分析作为参考。没有机器学习校正的粗网格模型的云层太少，导致地面白天加热过多，从而产生过多的地表潜热通量和降雨。通过从细网格模型中学习下流辐射通量，可以避免这种偏差。最好的配置使用学习到的温度和湿度的轻推趋势，而不是风。神经网络的表现略优于随机森林。通过应用ML校正，850 hPa温度预报提前3-7天提高了18小时的技能，时间平均降水模式提高了30%。加入机器学习的风向可以提高前5天的500 hPa高度技能，但会降低对流层上层时间平均温度和此后的纬向风型。
Global atmospheric “storm‐resolving” models with horizontal grid spacing of less than 5 km resolve deep cumulus convection and flow in complex terrain. They promise to be reference models that could be used to improve computationally affordable coarse‐grid global climate models across a range of climates, reducing uncertainties in regional precipitation and temperature trends. Here, machine learning of nudging tendencies as functions of column state is used to correct the physical parameterization tendencies of temperature, humidity, and optionally winds, in a real‐geography coarse‐grid model (FV3GFS with a 200 km grid) to be closer to those of a 40‐day reference simulation using X‐SHiELD, a modified version of FV3GFS with a 3 km grid. Both simulations specify the same historical sea‐surface temperature fields. This methodology builds on a prior study using a global observational analysis as the reference. The coarse‐grid model without machine learning corrections has too few clouds, causing too much daytime heating of land surfaces that creates excessive surface latent heat flux and rainfall. This bias is avoided by learning downwelling radiative flux from the fine‐grid model. The best configuration uses learned nudging tendencies for temperature and humidity but not winds. Neural nets slightly outperform random forests. Forecasts of 850 hPa temperature gain 18 hr of skill at 3–7 days leads and time‐mean precipitation patterns are improved 30% by applying the ML correction. Adding machine‐learned wind tendencies improves 500 hPa height skill for the first five days of forecasts but degrades time‐mean upper tropospheric temperature and zonal wind patterns thereafter.