Causally-Informed Deep Learning to Improve Climate Models and Projections

Climate models are essential to understand and project climate change, yet long-standing biases and uncertainties in their projections remain. This is largely associated with the representation of subgrid-scale processes, particularly clouds and convection. Deep learning can learn these subgrid-scale processes from computationally expensive storm-resolving models while retaining many features at a fraction of computational cost. Yet, climate simulations with embedded neural network parameterizations are still challenging and highly depend on the deep learning solution. This is likely associated with spurious non-physical correlations learned by the neural networks due to the complexity of the physical dynamical system. Here, we show that the combination of causality with deep learning helps removing spurious correlations and optimizing the neural network algorithm. To resolve this, we apply a causal discovery method to unveil causal drivers in the set of input predictors of atmospheric subgrid-scale processes of a superparameterized climate model in which deep convection is explicitly resolved. The resulting causally-informed neural networks are coupled to the climate model, hence, replacing the superparameterization and radiation scheme. We show that the climate simulations with causally-informed neural network parameterizations retain many convection-related properties and accurately generate the climate of the original high-resolution climate model, while retaining similar generalization capabilities to unseen climates compared to the non-causal approach. The combination of causal discovery and deep learning is a new and promising approach that leads to stable and more trustworthy climate simulations and paves the way toward more physically-based causal deep learning approaches also in other scientific disciplines. Climate models have biases compared to observations that have been present for a long time because certain processes, like convection, are only approximated using simplified methods. Neural networks can better represent these processes, but often learn incorrect connections leading to unreliable results and climate model crashes. To solve this, we used a new method that informs neural networks with causal drivers, therefore, respecting the underlying physical processes. By doing so, we developed more reliable and trustworthy neural networks, allowing us to accurately represent the climate of the original high-resolution simulation on which these neural networks were trained. Causal discovery unveils causal drivers of subgrid-scale processes across climates The causally-informed hybrid model runs stably and generates a climate close to the original high-resolution simulation Spurious correlations are evident in the non-causal parameterization, leading to underestimate feature importance of physical drivers