cd /news/machine-learning/unveiling-atmospheric-circulation-on… · home topics machine-learning article
[ARTICLE · art-120181] src=eos.org ↗ pub= topic=machine-learning verified=true sentiment=· neutral

Unveiling Atmospheric Circulation on Mars with Magnetic Field Data

A new machine learning model called the Neural-Curlometer, developed by Delcourt and Mittelholz, uses magnetic field data from NASA's MAVEN satellite to reconstruct large-scale electrical currents in Mars' ionosphere, revealing a hemispheric, seasonally varying vortex system at altitudes of 125-220 km. The findings, published in AGU Advances, provide a data-driven proxy for atmospheric circulation on Mars and could serve as new inputs for Martian general circulation models.

read1 min views1 publishedSep 3, 2026
Unveiling Atmospheric Circulation on Mars with Magnetic Field Data
Image: Eos (auto-discovered)

Editors’ Highlights are summaries of recent papers by AGU’s journal editors.

Source: AGU Advances

AGU Advances

Electrons and ions respond differently to atmospheric winds and to magnetic fields. This generates a large-scale electrical current in the “dynamo” region of the Mars ionosphere, where the electrons preferentially gyrate around the magnetic field while the ions undergo collisions with the neutrals.

Delcourt and Mittelholz [2026] developed a “Neural-Curlometer” model – a new machine learning (neural network) computer model based on Ampere’s Law and Gauss’s Law, which they apply to analyze the magnetic field resulting from the large-scale dynamo currents and measured onboard NASA’s MAVEN satellite.

The authors find the reconstructed currents to form a hemispheric, seasonally varying vortex system in the altitude range of 125-220 km, with a transition in direction near 160 km and a correlation in the current density with the crustal magnetic field. Furthermore, the hemispheric vortex pattern is consistent with the predicted atmospheric wind circulation on Mars, as well as the transport driven by the Coriolis force and associated with seasonal CO2 condensation at the poles. The findings provide a data-driven proxy for atmospheric circulation at ionospheric altitudes and potentially serve as important new inputs for Martian general circulation models.

*Citation: Delcourt, T. & Mittelholz, A. (2026). Global Circulation of Martian Ionospheric Currents Revealed by Magnetometer Data. AGU Advances, 7, e2026AV002408. *https://doi.org/10.1029/2026AV002408

—Andrew Yau, Editor, AGU Advances

── more in #machine-learning 4 stories · sorted by recency
── more on @delcourt 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/unveiling-atmospheri…] indexed:0 read:1min 2026-09-03 ·