April 2022

Conference Paper

Multimodel Ensemble Predictions of Precipitation using Bayesian Neural Networks

By:
Fan, Ming ; Lu, Dan ; Rastogi, Deeksha
Journal Name:
The International Conference on Learning Representations
Page Number:
1-10
Issue Number:
1
Book Title:
AI for Earth Sciences: #ai4earth at virtual @ICLR2022 April 29
Publication Date:
April 2022
Publisher Location:
ICLR, California, United States of America
Conference Name:
The International Conference on Learning Representations (ICLR): AI for Earth Sciences
Conference Location:
Virtual, Tennessee, United States of America
Conference Sponsor:
multiple organizations

Abstract

Multimodel ensembling improves predictions and considers model uncertainties. In this study, we present a Bayesian Neural Network (BNN) ensemble approach for large-scale precipitation predictions based on a set of climate models. BNN infers spatiotemporally varying model weights and biases through the calibration against observations. This ensemble scheme of BNN sufficiently leverages individual model skill for accurate predictions as well as provides interpretability about which models contribute more to the ensemble prediction at which locations and times to inform model development. Additionally, BNN accurately quantifies epistemic uncertainty to avoid overconfident projections. We demonstrate BNN’s superior prediction performance to three state-of-the-art ensemble approaches and discuss its interpretability and uncertainty quantification.