Multi‐Model Ensembles for Upper Atmosphere Models

• Published in: Space Weather Vol. 21; no. 3
• Main Authors: Elvidge, S., Granados, S. R., Angling, M. J., Brown, M. K., Themens, D. R., Wood, A. G.
• Format: Journal Article
• Language: English
• Published: Washington John Wiley & Sons, Inc 01.03.2023
• Subjects: Algorithms; Atmosphere; Atmospheric models; Least squares; Mean square errors; Modelling; Reduction; Solar cycle; Solar maximum; Solar minimum; Thermosphere; Upper atmosphere; Weighting
• ISSN: 1542-7390; 1539-4964; 1542-7390
• DOI: 10.1029/2022SW003356

Multi‐model ensembles (MMEs) are used to improve the forecasts of thermospheric neutral densities. A variety of algorithms for constructing the model weights for the MMEs are described and have been implemented including: performance weighting, independence weighting, and non‐negative least squares. Using both empirical and physics‐based models, compared against in situ Challenging Minisatellite Payload (CHAMP) observations, the skill of each MME weighting approach has been tested in both solar minimum and maximum conditions. In both cases the MME performs better than any individual model. A non‐negative least squares weighting for the MME on a set of bias corrected models provides a 68% and 50% reduction in the mean square error compared to the best model (Jacchia‐Bowman 2008) in the solar minimum and maximum cases, respectively. Plain Language Summary Combining multiple models of the neutral upper atmosphere (thermosphere) can lead to the cancellation of errors and improved short‐term forecasts of the environment. In this paper a number of different methods for creating these “multi‐model ensembles” (MMEs) are investigated, varying how the different models in the comparison are weighted and combined. Using both statistical and first‐principles models and compared to observations from the Challenging Minisatellite Payload (CHAMP) satellite, the skill of each MME approach has been tested in both solar minimum and maximum conditions. In both cases the MME performs better than any individual model. The best performing combination makes a 68% reduction in the mean square error compared to the best individual model at solar minimum and a 50% improvement at solar maximum. Key Points Multi‐model ensembles (MMEs) are used to reduce the error in specifying the thermosphere The MME performs better than any individual model in all test scenarios A non‐negative least squares weighting for the MME reduces the error by 68% at solar minimum and 50% at solar maximum