Using Probabilistic Machine Learning to Better Model Temporal Patterns in Parameterizations: a case study with the Lorenz 96 model
This work addresses the challenge of improving low-cost climate models for atmospheric simulations, though it is incremental as it builds on existing stochastic methods like red noise.
The paper tackled the problem of modeling small-scale processes in climate models by improving temporal pattern modeling in parameterizations, using a physically-informed recurrent neural network within a probabilistic framework. The result showed that the model is competitive and often superior to baselines, including a GAN, in the Lorenz 96 simulation, with better generalization to unseen scenarios.
The modelling of small-scale processes is a major source of error in climate models, hindering the accuracy of low-cost models which must approximate such processes through parameterization. Red noise is essential to many operational parameterization schemes, helping model temporal correlations. We show how to build on the successes of red noise by combining the known benefits of stochasticity with machine learning. This is done using a physically-informed recurrent neural network within a probabilistic framework. Our model is competitive and often superior to both a bespoke baseline and an existing probabilistic machine learning approach (GAN) when applied to the Lorenz 96 atmospheric simulation. This is due to its superior ability to model temporal patterns compared to standard first-order autoregressive schemes. It also generalises to unseen scenarios. We evaluate across a number of metrics from the literature, and also discuss the benefits of using the probabilistic metric of hold-out likelihood.