Science Through Machine Learning: Quantification of Post‐Storm Thermospheric Cooling
Abstract Machine learning (ML) models are universal function approximators and—if used correctly—can summarize the information content of observational data sets in a functional form for scientific and engineering applications. A benefit to ML over parametric models is that there are no a priori ass...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2022-09-01
|
| Series: | Space Weather |
| Subjects: | |
| Online Access: | https://doi.org/10.1029/2022SW003189 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841536303260237824 |
|---|---|
| author | Richard J. Licata Piyush M. Mehta Daniel R. Weimer Douglas P. Drob W. Kent Tobiska Jean Yoshii |
| author_facet | Richard J. Licata Piyush M. Mehta Daniel R. Weimer Douglas P. Drob W. Kent Tobiska Jean Yoshii |
| author_sort | Richard J. Licata |
| collection | DOAJ |
| description | Abstract Machine learning (ML) models are universal function approximators and—if used correctly—can summarize the information content of observational data sets in a functional form for scientific and engineering applications. A benefit to ML over parametric models is that there are no a priori assumptions about particular basis functions which can potentially limit the phenomena that can be modeled. In this work, we develop ML models on three data sets: the Space Environment Technologies High Accuracy Satellite Drag Model (HASDM) density database, a spatiotemporally matched data set of outputs from the Jacchia‐Bowman 2008 Empirical Thermospheric Density Model (JB2008), and an accelerometer‐derived density data set from CHAllenging Minisatellite Payload (CHAMP). These ML models are compared to the Naval Research Laboratory Mass Spectrometer and Incoherent Scatter radar (NRLMSIS 2.0) model to study the presence of post‐storm cooling in the middle‐thermosphere. We find that both NRLMSIS 2.0 and JB2008‐ML do not account for post‐storm cooling and consequently perform poorly in periods following strong geomagnetic storms (e.g., the 2003 Halloween storms). Conversely, HASDM‐ML and CHAMP‐ML do show evidence of post‐storm cooling indicating that this phenomenon is present in the original data sets. Results show that density reductions up to 40% can occur 1–3 days post‐storm depending on the location and strength of the storm. |
| format | Article |
| id | doaj-art-0932f80646f6434b97589c264c3fc41d |
| institution | Kabale University |
| issn | 1542-7390 |
| language | English |
| publishDate | 2022-09-01 |
| publisher | Wiley |
| record_format | Article |
| series | Space Weather |
| spelling | doaj-art-0932f80646f6434b97589c264c3fc41d2025-01-14T16:31:12ZengWileySpace Weather1542-73902022-09-01209n/an/a10.1029/2022SW003189Science Through Machine Learning: Quantification of Post‐Storm Thermospheric CoolingRichard J. Licata0Piyush M. Mehta1Daniel R. Weimer2Douglas P. Drob3W. Kent Tobiska4Jean Yoshii5Department of Mechanical and Aerospace Engineering West Virginia University Morgantown WV USADepartment of Mechanical and Aerospace Engineering West Virginia University Morgantown WV USACenter for Space Science and Engineering Research Virginia Tech Blacksburg VA USASpace Science Division US Naval Research Laboratory Washington DC USASpace Environment Technologies Pacific Palisades CA USASpace Environment Technologies Pacific Palisades CA USAAbstract Machine learning (ML) models are universal function approximators and—if used correctly—can summarize the information content of observational data sets in a functional form for scientific and engineering applications. A benefit to ML over parametric models is that there are no a priori assumptions about particular basis functions which can potentially limit the phenomena that can be modeled. In this work, we develop ML models on three data sets: the Space Environment Technologies High Accuracy Satellite Drag Model (HASDM) density database, a spatiotemporally matched data set of outputs from the Jacchia‐Bowman 2008 Empirical Thermospheric Density Model (JB2008), and an accelerometer‐derived density data set from CHAllenging Minisatellite Payload (CHAMP). These ML models are compared to the Naval Research Laboratory Mass Spectrometer and Incoherent Scatter radar (NRLMSIS 2.0) model to study the presence of post‐storm cooling in the middle‐thermosphere. We find that both NRLMSIS 2.0 and JB2008‐ML do not account for post‐storm cooling and consequently perform poorly in periods following strong geomagnetic storms (e.g., the 2003 Halloween storms). Conversely, HASDM‐ML and CHAMP‐ML do show evidence of post‐storm cooling indicating that this phenomenon is present in the original data sets. Results show that density reductions up to 40% can occur 1–3 days post‐storm depending on the location and strength of the storm.https://doi.org/10.1029/2022SW003189machine learningthermospherecoolinggeomagnetic storm |
| spellingShingle | Richard J. Licata Piyush M. Mehta Daniel R. Weimer Douglas P. Drob W. Kent Tobiska Jean Yoshii Science Through Machine Learning: Quantification of Post‐Storm Thermospheric Cooling Space Weather machine learning thermosphere cooling geomagnetic storm |
| title | Science Through Machine Learning: Quantification of Post‐Storm Thermospheric Cooling |
| title_full | Science Through Machine Learning: Quantification of Post‐Storm Thermospheric Cooling |
| title_fullStr | Science Through Machine Learning: Quantification of Post‐Storm Thermospheric Cooling |
| title_full_unstemmed | Science Through Machine Learning: Quantification of Post‐Storm Thermospheric Cooling |
| title_short | Science Through Machine Learning: Quantification of Post‐Storm Thermospheric Cooling |
| title_sort | science through machine learning quantification of post storm thermospheric cooling |
| topic | machine learning thermosphere cooling geomagnetic storm |
| url | https://doi.org/10.1029/2022SW003189 |
| work_keys_str_mv | AT richardjlicata sciencethroughmachinelearningquantificationofpoststormthermosphericcooling AT piyushmmehta sciencethroughmachinelearningquantificationofpoststormthermosphericcooling AT danielrweimer sciencethroughmachinelearningquantificationofpoststormthermosphericcooling AT douglaspdrob sciencethroughmachinelearningquantificationofpoststormthermosphericcooling AT wkenttobiska sciencethroughmachinelearningquantificationofpoststormthermosphericcooling AT jeanyoshii sciencethroughmachinelearningquantificationofpoststormthermosphericcooling |