Evolution of machine learning in environmental science—A perspective
Open Access
- 13 April 2022
- journal article
- research article
- Published by Cambridge University Press (CUP) in Environmental Data Science
- Vol. 1
Abstract
The growth of machine learning (ML) in environmental science can be divided into a slow phase lasting till the mid-2010s and a fast phase thereafter. The rapid transition was brought about by the emergence of powerful new ML methods, allowing ML to successfully tackle many problems where numerical models and statistical models have been hampered. Deep convolutional neural network models greatly advanced the use of ML on 2D or 3D data. Transfer learning has allowed ML to progress in climate science, where data records are generally short for ML. ML and physics are also merging in new areas, for example: (a) using ML for general circulation model parametrization, (b) adding physics constraints in ML models, and (c) using ML in data assimilation.Keywords
This publication has 47 references indexed in Scilit:
- Deep learningNature, 2015
- Using Ensemble of Neural Networks to Learn Stochastic Convection Parameterizations for Climate and Numerical Weather Prediction Models from Data Simulated by a Cloud Resolving ModelAdvances in Artificial Neural Systems, 2013
- Decadal Climate Simulations Using Accurate and Fast Neural Network Emulation of Full, Longwave and Shortwave, RadiationMonthly Weather Review, 2008
- The impact of time-averaging on the detectability of nonlinear empirical relationsQuarterly Journal of the Royal Meteorological Society, 2002
- Statistical Modeling: The Two Cultures (with comments and a rejoinder by the author)Statistical Science, 2001
- Coupling Neural Networks to Incomplete Dynamical Systems via Variational Data AssimilationMonthly Weather Review, 2001
- Nonlinear principal component analysis by neural networksTellus, 2001
- Use of a neural‐network‐based long‐wave radiative‐transfer scheme in the ECMWF atmospheric modelQuarterly Journal of the Royal Meteorological Society, 2000
- Neural networks for the prediction and forecasting of water resources variables: a review of modelling issues and applicationsEnvironmental Modelling & Software, 2000
- Support-vector networksMachine Learning, 1995