Using Deep Learning to Identify Initial Error Sensitivity of ENSO Forecasts

arXiv:2404.15419v1 Announce Type: cross
Abstract: We introduce a hybrid method that integrates deep learning with model-analog forecasting, a straightforward yet effective approach that generates forecasts from similar initial climate states in a repository of model simulations. This hybrid framework employs a convolutional neural network to estimate state-dependent weights to identify analog states. The advantage of our method lies in its physical interpretability, offering insights into initial-error-sensitive regions through estimated weights and the ability to trace the physically-based temporal evolution of …

abstract analog arxiv climate convolutional neural network cs.lg deep learning error forecasting framework hybrid identify network neural network physics.ao-ph sensitivity simulations state type