Skillful statistical prediction of subseasonal temperature by training on dynamical model data. (23rd February 2023)

• Record Type: Journal Article
• Authors: Trenary, Laurie; DelSole, Timothy
• Is Part Of: Environmental data science. Volume 2(2022)
• Publication Date: 2023-02-23
• Subjects: CMIP6 -- lasso regression -- observations -- subseasonal prediction -- western United States; Environmental sciences -- Data processing -- Periodicals; 577.0285
• DOI: 10.1017/eds.2023.2
• Languages: English
• ISSNs: 2634-4602

Abstract

Abstract: This paper derives statistical models for predicting wintertime subseasonal temperature over the western US. The statistical models are trained on two separate datasets, namely observations and dynamical model simulations, and are based on least absolute shrinkage and selection operator (lasso). Surprisingly, statistical models trained on dynamical model simulations can predict observations better than observation-trained models. One reason for this is that simulations involve orders of magnitude more data than observational datasets.