Explaining neural networks for detection of tropical cyclones and atmospheric rivers in gridded atmospheric simulation data

Detection of atmospheric features in gridded datasets from numerical simulation models is typically done by means of rule-based algorithms. Recently, also the feasibility of learning feature detection tasks using supervised learning with convolutional neural networks (CNNs) has been demonstrated. This approach corresponds to semantic segmentation tasks widely investigated in computer vision. However, while in recent studies the performance of CNNs was shown to be comparable to human experts, CNNs are largely treated as a “black box”, and it remains unclear whether they learn the features for the correct reasons. Here we build on the recently published “ClimateNet” dataset that contains features of tropical cyclones and atmospheric rivers as detected by human experts. We adapt the explainable artificial intelligence technique “Layer-wise Relevance Propagation” (LRP) to the feature detection task and investigate which input information CNNs with the Context-Guided Network (CG-Net) and U-Net architectures use for feature detection. We find that both CNNs indeed consider plausible patterns in the input fields of atmospheric variables, which helps to build trust in the approach. We also demonstrate application of the approach for finding the most relevant input variables and evaluating detection robustness when changing the input domain. However, LRP in its current form cannot explain shape information used by the CNNs, and care needs to be taken regarding the normalization of input values, as LRP cannot explain the contribution of bias neurons, accounting for inputs close to zero. These shortcomings need to be addressed by future work to obtain a more complete explanation of CNNs for geoscientific feature detection.