Academic Journal
Προβολή σε EDS

Physics‐Informed Neural Networks for Modeling the Martian Induced Magnetosphere.

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: Physics‐Informed Neural Networks for Modeling the Martian Induced Magnetosphere.
Συγγραφείς: Gao, Jiawei1 (AUTHOR) jgao2@bu.edu, Dong, Chuanfei1,2 (AUTHOR) dcfy@bu.edu, Zhang, Chi1 (AUTHOR), Qin, Yilan1 (AUTHOR), Shekarpaz, Simin1 (AUTHOR), Li, Xinmin1 (AUTHOR), Wang, Liang1 (AUTHOR), Zhou, Hongyang1 (AUTHOR), Tadlock, Abigail1 (AUTHOR)
Πηγή: Geophysical Research Letters. 6/16/2026, Vol. 53 Issue 11, p1-14. 14p.
Θεματικοί όροι: *Solar wind, *Magnetosphere, *Space sciences, *Space environment, *Magnetospheric physics
Περίληψη: Understanding the magnetic field environment around Mars and its response to upstream solar wind conditions provide key insights into the processes driving atmospheric ion escape. To date, global models of Martian induced magnetosphere have been exclusively physics‐based, relying on computationally intensive simulations. For the first time, we develop a data‐driven model of the Martian induced magnetospheric magnetic field using Physics‐Informed Neural Network (PINN) combined with MAVEN observations and physical laws. Trained under varying solar wind conditions, the data‐driven model accurately reconstructs the three‐dimensional magnetic field configuration and its variability in response to upstream solar wind drivers. Based on the PINN results, we identify key dependencies of magnetic field configuration on solar wind parameters, including a more negative By ${B}_{y}$ component in the ‐E hemisphere near the Martian south pole. These findings offer valuable insights into how the configuration of the induced magnetosphere varies with upstream solar wind parameters. Plain Language Summary: Unlike Earth, Mars does not have a global magnetic field. Instead, the solar wind interacts directly with the Martian upper atmosphere, forming an induced magnetosphere. Understanding how this induced magnetosphere changes with solar wind conditions is essential for studying the space environment around Mars. Traditional methods, such as statistical analyses or numerical simulations, often struggle to capture the complexity of the magnetic field while satisfying physical laws, such as the zero divergence of magnetic field. In this study, we applied a machine learning approach called Physics‐Informed Neural Networks to model the three‐dimensional magnetic field around Mars. By combining MAVEN spacecraft observations with physical laws and boundary conditions, our data‐driven model accurately reproduces three‐dimensional magnetic field structures and their response to varying solar wind conditions. This work highlights the potential of machine learning methods to advance our understanding of planetary space environments. Key Points: The induced magnetic fields around Mars are modeled using Physics‐Informed Neural Networks (PINNs)A PINN model is developed by integrating MAVEN observations with physical laws and boundary conditionsThe model reveals how the magnetic field configuration in the induced magnetosphere varies with upstream parameters (BIMF ${B}_{\text{IMF}}$, PSW ${P}_{\text{SW}}$, and θcone ${\theta }_{\text{cone}}$) [ABSTRACT FROM AUTHOR]
Βάση Δεδομένων: Academic Search Index
FullText Text:
  Availability: 0
CustomLinks:
  – Url: https://resolver.ebsco.com/c/fiv2js/result?sid=EBSCO:asx&genre=article&issn=00948276&ISBN=&volume=53&issue=11&date=20260616&spage=1&pages=1-14&title=Geophysical Research Letters&atitle=Physics%E2%80%90Informed%20Neural%20Networks%20for%20Modeling%20the%20Martian%20Induced%20Magnetosphere.&aulast=Gao%2C%20Jiawei&id=DOI:10.1029/2025GL121532
    Name: Full Text Finder (for New FTF UI) (ns324271)
    Category: fullText
    Text: Full Text Finder
    MouseOverText: Full Text Finder
Header DbId: asx
DbLabel: Academic Search Index
An: 194450212
RelevancyScore: 1431
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 1430.76574707031
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Physics‐Informed Neural Networks for Modeling the Martian Induced Magnetosphere.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Gao%2C+Jiawei%22">Gao, Jiawei</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> jgao2@bu.edu</i><br /><searchLink fieldCode="AR" term="%22Dong%2C+Chuanfei%22">Dong, Chuanfei</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<i> dcfy@bu.edu</i><br /><searchLink fieldCode="AR" term="%22Zhang%2C+Chi%22">Zhang, Chi</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Qin%2C+Yilan%22">Qin, Yilan</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Shekarpaz%2C+Simin%22">Shekarpaz, Simin</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Li%2C+Xinmin%22">Li, Xinmin</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Wang%2C+Liang%22">Wang, Liang</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Zhou%2C+Hongyang%22">Zhou, Hongyang</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Tadlock%2C+Abigail%22">Tadlock, Abigail</searchLink><relatesTo>1</relatesTo> (AUTHOR)
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Geophysical+Research+Letters%22">Geophysical Research Letters</searchLink>. 6/16/2026, Vol. 53 Issue 11, p1-14. 14p.
– Name: Subject
  Label: Subject Terms
  Group: Su
  Data: *<searchLink fieldCode="DE" term="%22Solar+wind%22">Solar wind</searchLink><br />*<searchLink fieldCode="DE" term="%22Magnetosphere%22">Magnetosphere</searchLink><br />*<searchLink fieldCode="DE" term="%22Space+sciences%22">Space sciences</searchLink><br />*<searchLink fieldCode="DE" term="%22Space+environment%22">Space environment</searchLink><br />*<searchLink fieldCode="DE" term="%22Magnetospheric+physics%22">Magnetospheric physics</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Understanding the magnetic field environment around Mars and its response to upstream solar wind conditions provide key insights into the processes driving atmospheric ion escape. To date, global models of Martian induced magnetosphere have been exclusively physics‐based, relying on computationally intensive simulations. For the first time, we develop a data‐driven model of the Martian induced magnetospheric magnetic field using Physics‐Informed Neural Network (PINN) combined with MAVEN observations and physical laws. Trained under varying solar wind conditions, the data‐driven model accurately reconstructs the three‐dimensional magnetic field configuration and its variability in response to upstream solar wind drivers. Based on the PINN results, we identify key dependencies of magnetic field configuration on solar wind parameters, including a more negative By ${B}_{y}$ component in the ‐E hemisphere near the Martian south pole. These findings offer valuable insights into how the configuration of the induced magnetosphere varies with upstream solar wind parameters. Plain Language Summary: Unlike Earth, Mars does not have a global magnetic field. Instead, the solar wind interacts directly with the Martian upper atmosphere, forming an induced magnetosphere. Understanding how this induced magnetosphere changes with solar wind conditions is essential for studying the space environment around Mars. Traditional methods, such as statistical analyses or numerical simulations, often struggle to capture the complexity of the magnetic field while satisfying physical laws, such as the zero divergence of magnetic field. In this study, we applied a machine learning approach called Physics‐Informed Neural Networks to model the three‐dimensional magnetic field around Mars. By combining MAVEN spacecraft observations with physical laws and boundary conditions, our data‐driven model accurately reproduces three‐dimensional magnetic field structures and their response to varying solar wind conditions. This work highlights the potential of machine learning methods to advance our understanding of planetary space environments. Key Points: The induced magnetic fields around Mars are modeled using Physics‐Informed Neural Networks (PINNs)A PINN model is developed by integrating MAVEN observations with physical laws and boundary conditionsThe model reveals how the magnetic field configuration in the induced magnetosphere varies with upstream parameters (BIMF ${B}_{\text{IMF}}$, PSW ${P}_{\text{SW}}$, and θcone ${\theta }_{\text{cone}}$) [ABSTRACT FROM AUTHOR]
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=asx&AN=194450212
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1029/2025GL121532
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 14
        StartPage: 1
    Subjects:
      – SubjectFull: Solar wind
        Type: general
      – SubjectFull: Magnetosphere
        Type: general
      – SubjectFull: Space sciences
        Type: general
      – SubjectFull: Space environment
        Type: general
      – SubjectFull: Magnetospheric physics
        Type: general
    Titles:
      – TitleFull: Physics‐Informed Neural Networks for Modeling the Martian Induced Magnetosphere.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Gao, Jiawei
      – PersonEntity:
          Name:
            NameFull: Dong, Chuanfei
      – PersonEntity:
          Name:
            NameFull: Zhang, Chi
      – PersonEntity:
          Name:
            NameFull: Qin, Yilan
      – PersonEntity:
          Name:
            NameFull: Shekarpaz, Simin
      – PersonEntity:
          Name:
            NameFull: Li, Xinmin
      – PersonEntity:
          Name:
            NameFull: Wang, Liang
      – PersonEntity:
          Name:
            NameFull: Zhou, Hongyang
      – PersonEntity:
          Name:
            NameFull: Tadlock, Abigail
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 16
              M: 06
              Text: 6/16/2026
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 00948276
          Numbering:
            – Type: volume
              Value: 53
            – Type: issue
              Value: 11
          Titles:
            – TitleFull: Geophysical Research Letters
              Type: main
ResultId 1