Probabilistic Solar Proxy Forecasting With Neural Network Ensembles
Abstract Space weather indices are used commonly to drive forecasts of thermosphere density, which affects objects in low‐Earth orbit (LEO) through atmospheric drag. One commonly used space weather proxy, F10.7cm, correlates well with solar extreme ultra‐violet (EUV) energy deposition into the therm...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2023-09-01
|
| Series: | Space Weather |
| Subjects: | |
| Online Access: | https://doi.org/10.1029/2023SW003675 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841536342145630208 |
|---|---|
| author | Joshua D. Daniell Piyush M. Mehta |
| author_facet | Joshua D. Daniell Piyush M. Mehta |
| author_sort | Joshua D. Daniell |
| collection | DOAJ |
| description | Abstract Space weather indices are used commonly to drive forecasts of thermosphere density, which affects objects in low‐Earth orbit (LEO) through atmospheric drag. One commonly used space weather proxy, F10.7cm, correlates well with solar extreme ultra‐violet (EUV) energy deposition into the thermosphere. Currently, the USAF contracts Space Environment Technologies (SET), which uses a linear algorithm to forecast F10.7cm. In this work, we introduce methods using neural network ensembles with multi‐layer perceptrons (MLPs) and long‐short term memory (LSTMs) to improve on the SET predictions. We make predictions only from historical F10.7cm values. We investigate data manipulation methods (backwards averaging and lookback) as well as multi step and dynamic forecasting. This work shows an improvement over the popular persistence and the operational SET model when using ensemble methods. The best models found in this work are ensemble approaches using multi step or a combination of multi step and dynamic predictions. Nearly all approaches offer an improvement, with the best models improving between 48% and 59% on relative MSE with respect to persistence. Other relative error metrics were shown to improve greatly when ensembles methods were used. We were also able to leverage the ensemble approach to provide a distribution of predicted values; allowing an investigation into forecast uncertainty. Our work found models that produced less biased predictions at elevated and high solar activity levels. Uncertainty was also investigated through the use of a calibration error score metric (CES), our best ensemble reached similar CES as other work. |
| format | Article |
| id | doaj-art-943a1ebd604e4ef5a9df0f9fdc4ffbfa |
| institution | Kabale University |
| issn | 1542-7390 |
| language | English |
| publishDate | 2023-09-01 |
| publisher | Wiley |
| record_format | Article |
| series | Space Weather |
| spelling | doaj-art-943a1ebd604e4ef5a9df0f9fdc4ffbfa2025-01-14T16:31:22ZengWileySpace Weather1542-73902023-09-01219n/an/a10.1029/2023SW003675Probabilistic Solar Proxy Forecasting With Neural Network EnsemblesJoshua D. Daniell0Piyush M. Mehta1Department of Mechanical and Aerospace Engineering West Virginia University Morgantown Morgantown WV USADepartment of Mechanical and Aerospace Engineering West Virginia University Morgantown Morgantown WV USAAbstract Space weather indices are used commonly to drive forecasts of thermosphere density, which affects objects in low‐Earth orbit (LEO) through atmospheric drag. One commonly used space weather proxy, F10.7cm, correlates well with solar extreme ultra‐violet (EUV) energy deposition into the thermosphere. Currently, the USAF contracts Space Environment Technologies (SET), which uses a linear algorithm to forecast F10.7cm. In this work, we introduce methods using neural network ensembles with multi‐layer perceptrons (MLPs) and long‐short term memory (LSTMs) to improve on the SET predictions. We make predictions only from historical F10.7cm values. We investigate data manipulation methods (backwards averaging and lookback) as well as multi step and dynamic forecasting. This work shows an improvement over the popular persistence and the operational SET model when using ensemble methods. The best models found in this work are ensemble approaches using multi step or a combination of multi step and dynamic predictions. Nearly all approaches offer an improvement, with the best models improving between 48% and 59% on relative MSE with respect to persistence. Other relative error metrics were shown to improve greatly when ensembles methods were used. We were also able to leverage the ensemble approach to provide a distribution of predicted values; allowing an investigation into forecast uncertainty. Our work found models that produced less biased predictions at elevated and high solar activity levels. Uncertainty was also investigated through the use of a calibration error score metric (CES), our best ensemble reached similar CES as other work.https://doi.org/10.1029/2023SW003675machine learningneural networksforecastingspace weathersolar proxyuncertainty estimation |
| spellingShingle | Joshua D. Daniell Piyush M. Mehta Probabilistic Solar Proxy Forecasting With Neural Network Ensembles Space Weather machine learning neural networks forecasting space weather solar proxy uncertainty estimation |
| title | Probabilistic Solar Proxy Forecasting With Neural Network Ensembles |
| title_full | Probabilistic Solar Proxy Forecasting With Neural Network Ensembles |
| title_fullStr | Probabilistic Solar Proxy Forecasting With Neural Network Ensembles |
| title_full_unstemmed | Probabilistic Solar Proxy Forecasting With Neural Network Ensembles |
| title_short | Probabilistic Solar Proxy Forecasting With Neural Network Ensembles |
| title_sort | probabilistic solar proxy forecasting with neural network ensembles |
| topic | machine learning neural networks forecasting space weather solar proxy uncertainty estimation |
| url | https://doi.org/10.1029/2023SW003675 |
| work_keys_str_mv | AT joshuaddaniell probabilisticsolarproxyforecastingwithneuralnetworkensembles AT piyushmmehta probabilisticsolarproxyforecastingwithneuralnetworkensembles |