When a moderate geomagnetic storm struck Earth on 3 February 2022, its impact was immediate and costly. Shortly after launch, 38 of 49 newly deployed Starlink satellites were lost as increased atmospheric drag caused by the storm prevented them from reaching their operational orbit. The event highlighted a growing challenge: even relatively modest space weather disturbances can significantly affect critical infrastructure and technological systems.
As society becomes increasingly dependent on satellites, communications networks and power systems, improving our ability to monitor and forecast space weather has become a priority. At the same time, scientists are discovering that many signals generated by solar activity can resemble those produced by natural hazards such as earthquakes, volcanic eruptions and tsunamis. Distinguishing between these different sources of disturbance is essential for reliable monitoring and early warning systems.
This challenge lies at the heart of the Swarm-AWARE (Swarm Investigation of Space Weather and Natural Hazards Effects) project. First results from the project were presented at the European Geosciences Union (EGU) General Assembly 2026 in Vienna, and were further highlighted in a dedicated press release.
Monitoring space weather
The Swarm mission consists of three satellites that continuously measure Earth’s magnetic field, electric fields, plasma densities and temperatures, and a variety of other parameters in the near-Earth environment. These observations provide a unique opportunity to study how solar activity influences the magnetosphere, ionosphere and atmosphere.
Low Earth Orbit satellites such as Swarm are particularly well suited for monitoring space weather. They observe magnetic field variations and field-aligned currents that connect processes occurring in the solar wind and magnetosphere with effects experienced in the upper atmosphere and at Earth’s surface.
Recent research has demonstrated that Swarm data can be used effectively to monitor space weather impacts and assess associated risks. Building on these advances, Swarm-AWARE aims to develop methodologies that could eventually be applied across multiple satellite missions, contributing to a future global monitoring network for space weather, similar to those already used for meteorological forecasting.
When natural hazards leave their mark in space
Not all disturbances observed by satellites originate from the Sun. Powerful natural events can also affect the upper atmosphere and ionosphere.
One remarkable example is the January 2022 eruption of Hunga Tonga-Hunga Ha’apai. The eruption injected enormous quantities of water vapour into the stratosphere and generated atmospheric waves that propagated all the way into the ionosphere. These waves produced electric fields that travelled along Earth’s magnetic field lines, causing measurable disturbances thousands of kilometres away. Swarm satellites successfully detected these effects.
While Hunga Tonga represents an extreme case, scientists are increasingly interested in identifying much subtler signatures associated with ongoing volcanic and seismic activity. Active volcanoes continuously generate signals linked to magma movement, ground deformation, gas emissions and seismic processes. Some of these processes can also influence the local magnetic field.
These so-called volcanomagnetic signals typically appear as localised variations of only a few to a few tens of nanoteslas. Although small, they can provide valuable information about volcanic activity and may eventually contribute to eruption forecasting. Detecting such signals from space, however, remains a major scientific challenge because they are often masked by the much stronger effects of space weather.
Combining satellite observations and AI
The Swarm-AWARE project brings together observations from multiple sources to address this challenge. Researchers combine magnetic field measurements from Swarm with atmospheric observations from the Copernicus Sentinel-5P mission, particularly measurements of sulphur dioxide emissions associated with volcanic activity. Ground-based seismic and geophysical observations are also incorporated.
An example comes from the Central American Volcanic Arc (CAVA), where more than a decade of Swarm observations has revealed transient magnetic anomalies that appear independently of geomagnetic activity. One notable case occurred near Guatemala’s Santiaguito volcano on 19 July 2023, when unusual magnetic field variations coincided with enhanced sulphur dioxide emissions and regional seismic activity.
To investigate whether such anomalies are linked to volcanic processes, researchers are building a large database of satellite observations collected over active volcanic regions. The dataset already includes thousands of satellite passes from the Swarm constellation.
Artificial intelligence plays a central role in the analysis. Convolutional Neural Networks (CNNs) are being trained to recognise patterns in magnetic field measurements and classify different types of disturbances. By learning from large numbers of examples, these algorithms may eventually distinguish between signatures generated by space weather and those associated with natural hazards.
The classification process includes transient magnetic signals, elevated ultra-low-frequency background noise and accompanying plasma perturbations. The ultimate objective is to identify characteristic signatures that can be reliably associated with volcanic or seismic activity.
Towards new early warning capabilities
Although many studies have reported unusual magnetic and ionospheric signals near regions affected by natural hazards, important questions remain about the complex interactions linking Earth’s interior, atmosphere and ionosphere.
Swarm-AWARE seeks to fill some of these knowledge gaps. In the long term, researchers envision a future in which satellite magnetometers could act as a kind of “space-based seismometer”, detecting characteristic signatures of volcanic eruptions and other geophysical events in much the same way that seismologists identify seismic waves generated by earthquakes.
As our dependence on space-based technologies continues to grow, the ability to separate signals originating from the Sun from those generated on Earth may become increasingly important for protecting society and improving preparedness for future events.