文章基本信息

标题：Deep Emulators for Differentiation, Forecasting, and Parametrization in Earth Science Simulators
本地全文：下载
作者：Marcel Nonnenmacher ; David S. Greenberg
期刊名称：Journal of Advances in Modeling Earth Systems
电子版ISSN：1942-2466
出版年度：2021
卷号：13
期号：7
页码：e2021MS002554
DOI：10.1029/2021MS002554
出版社：John Wiley & Sons, Ltd.
摘要：To understand and predict large, complex, and chaotic systems, Earth scientists build simulators from physical laws. Simulators generalize better to new scenarios, require fewer tunable parameters, and are more interpretable than nonphysical deep learning, but procedures for obtaining their derivatives with respect to their inputs are often unavailable. These missing derivatives limit the application of many important tools for forecasting, model tuning, sensitivity analysis, or subgrid-scale parametrization. Here, we propose to overcome this limitation with deep emulator networks that learn to calculate the missing derivatives. By training directly on simulation data without analyzing source code or equations, this approach supports simulators in any programming language on any hardware without specialized routines for each case. To demonstrate the effectiveness of our approach, we train emulators on complete or partial system states of the chaotic Lorenz-96 simulator and evaluate the accuracy of their dynamics and derivatives as a function of integration time and training data set size. We further demonstrate that emulator-derived derivatives enable accurate 4D-Var data assimilation and closed-loop training of parametrizations. These results provide a basis for further combining the parsimony and generality of physical models with the power and flexibility of machine learning.
关键词：machine learning;deep learning;data assimilation;parametrization tuning;model Jacobians;Lorenz-96