A new data visualization illustrates how an experimental NASA technology can provide extra lead time to communities in the path of a tsunami. Called GUARDIAN (GNSS Upper Atmospheric Real-time Disaster Information and Alert Network), the software detects slight distortions in satellite navigation signals to spot hazards on the move.
The animation breaks down a real-life case study: 2025’s massive Kamchatka earthquake and the tsunami that it sent racing across the Pacific and towards Hawaii at more than 500 mph (805 kph).
The visualization shows the magnitude 8.8 earthquake (seen in purple) strike off the Russian coast on July 29, 2025, triggering the tsunami. The red, orange, yellow, and green ringlets represent real-time readings from ground stations tracking GPS and other navigational satellite signals. The disturbances were spotted by GUARDIAN’s artificial intelligence-powered detection algorithms as soon as eight minutes after the earthquake.
For the next several hours, signs of the tsunami were picked up by GUARDIAN across the Pacific Ocean in near real time. The system flagged an incoming wave off the coast of Kauai some 32 minutes before it made landfall and was detected by tide gauges (shown in blue).
The results highlight GUARDIAN’s potential to augment existing early warning systems, said Camille Martire, one of its developers at NASA’s Jet Propulsion Laboratory in Southern California.
Currently, determining whether an earthquake generated a tsunami remains a challenge. Forecasters rely on seismic data and computer simulations to make their best prediction, then wait for pressure sensors attached to the ocean floor to confirm a passing wave. Those sensors work well but are expensive and thinly dispersed. Gaps in coverage remain. And in those gaps, warning time disappears.
The GUARDIAN approach is complementary and cost effective because it monitors existing data from GPS and other constellations that make up the Global Navigation Satellite System. It’s also free to access, though for now best suited to analysts trained to interpret its findings.
How GUARDIAN works
All day, every day, geopositioning constellations transmit radio signals to ground stations around the globe. On the ground, the data is refined to sub-decimeter (less than 10 centimeters) positioning accuracy by JPL’s Global Differential GPS System. Before the signals get there, however, they must travel through an electrically charged skin of plasma called the ionosphere.
Solar storms and other space weather can wreak electrical mayhem in the ionosphere, and so can events on Earth. Tsunamis and earthquakes, by displacing large amount of air at Earth’s surface, unleash pressure waves that can slightly perturb the radio signals coming down from satellites. While systems are in place to correct for this “noise,” GUARDIAN considers it a useful signal.
Currently, GUARDIAN scours data from more than 350 GNSS ground stations around the Pacific Ring of Fire, a hotbed for the ocean’s deadliest waves. And the system is not confined to tsunamis. Earthquakes, volcanic eruptions, missile tests, spacecraft reentries, meteoroid splashdowns — anything that produces a large rumble on Earth is potentially fair game. While the Kamchatka event didn’t cause widespread damage to people or property, it showed how the next time disaster strikes, NASA science could give communities a few more minutes to act.
GUARDIAN is being developed at JPL by the GDGPS project, which is partially supported by NASA’s Space Geodesy Project.
A massive earthquake and subsequent tsunami off Russia in late July tested an experimental detection system that had deployed a critical component just the day before.
A recent tsunami triggered by a magnitude 8.8 earthquake off Russia’s Kamchatka Peninsula sent pressure waves to the upper layer of the atmosphere, NASA scientists have reported. While the tsunami did not wreak widespread damage, it was an early test for a detection system being developed at the agency’s Jet Propulsion Laboratory in Southern California.
Called GUARDIAN (GNSS Upper Atmospheric Real-time Disaster Information and Alert Network), the experimental technology “functioned to its full extent,” said Camille Martire, one of its developers at JPL. The system flagged distortions in the atmosphere and issued notifications to subscribed subject matter experts in as little as 20 minutes after the quake. It confirmed signs of the approaching tsunami about 30 to 40 minutes before waves made landfall in Hawaii and sites across the Pacific on July 29 (local time).
“Those extra minutes of knowing something is coming could make a real difference when it comes to warning communities in the path,” said JPL scientist Siddharth Krishnamoorthy.
Near-real-time outputs from GUARDIAN must be interpreted by experts trained to identify the signs of tsunamis. But already it’s one of the fastest monitoring tools of its kind: Within about 10 minutes of receiving data, it can produce a snapshot of a tsunami’s rumble reaching the upper atmosphere.
The dots in this graph indicate wave disturbances in the ionosphere as measured between ground stations and navigation satellites. The initial spike shows the acoustic wave coming from the epicenter of the July 29 quake that caused the tsunami; the red squiggle shows the gravity wave the tsunami generated. (Image: NASA/JPL-Caltech)
The dots in this graph indicate wave disturbances in the ionosphere as measured between ground stations and navigation satellites. The initial spike shows the acoustic wave coming from the epicenter of the July 29 quake that caused the tsunami; the red squiggle shows the gravity wave the tsunami generated.
The goal of GUARDIAN is to augment existing early warning systems. A key question after a major undersea earthquake is whether a tsunami was generated. Today, forecasters use seismic data as a proxy to predict if and where a tsunami could occur, and they rely on sea-based instruments to confirm that a tsunami is passing by. Deep-ocean pressure sensors remain the gold standard when it comes to sizing up waves, but they are expensive and sparse in locations.
“NASA’s GUARDIAN can help fill the gaps,” said Christopher Moore, director of the National Oceanic and Atmospheric Administration Center for Tsunami Research. “It provides one more piece of information, one more valuable data point, that can help us determine, yes, we need to make the call to evacuate.”
Moore noted that GUARDIAN adds a unique perspective: It’s able to sense sea surface motion from high above Earth, globally and in near-real-time.
Bill Fry, chair of the United Nations technical working group responsible for tsunami early warning in the Pacific, said GUARDIAN is part of a technological “paradigm shift.” By directly observing ocean dynamics from space, “GUARDIAN is absolutely something that we in the early warning community are looking for to help underpin next generation forecasting.”
How GUARDIAN works
GUARDIAN takes advantage of tsunami physics. During a tsunami, many square miles of the ocean surface can rise and fall nearly in unison. This displaces a significant amount of air above it, sending low-frequency sound and gravity waves speeding upwards toward space. The waves interact with the charged particles of the upper atmosphere — the ionosphere — where they slightly distort the radio signals coming down to scientific ground stations of GPS and other positioning and timing satellites. These satellites are known collectively as the Global Navigation Satellite System (GNSS).
While GNSS processing methods on Earth correct for such distortions, GUARDIAN uses them as clues. The software scours a trove of data transmitted to more than 350 continuously operating GNSS ground stations around the world. It can potentially identify evidence of a tsunami up to about 745 miles (1,200 kilometers) from a given station. In ideal situations, vulnerable coastal communities near a GNSS station could know when a tsunami was heading their way and authorities would have as much as 1 hour and 20 minutes to evacuate the low-lying areas, thereby saving countless lives and property.
Key to this effort is the network of GNSS stations around the world supported by NASA’s Space Geodesy Project and Global GNSS Network, as well as JPL’s Global Differential GPS network that transmits the data in real time.
The Kamchatka event offered a timely case study for GUARDIAN. A day before the quake off Russia’s northeast coast, the team had deployed two new elements that were years in the making: an artificial intelligence to mine signals of interest and an accompanying prototype messaging system.
Both were put to the test when one of the strongest earthquakes ever recorded spawned a tsunami traveling hundreds of miles per hour across the Pacific Ocean. Having been trained to spot the kinds of atmospheric distortions caused by a tsunami, GUARDIAN flagged the signals for human review and notified subscribed subject matter experts.
Notably, tsunamis are most often caused by large undersea earthquakes, but not always. Volcanic eruptions, underwater landslides, and certain weather conditions in some geographic locations can all produce dangerous waves. An advantage of GUARDIAN is that it doesn’t require information on what caused a tsunami; rather, it can detect that one was generated and then can alert the authorities to help minimize the loss of life and property.
While there’s no silver bullet to stop a tsunami from making landfall, “GUARDIAN has real potential to help by providing open access to this data,” said Adrienne Moseley, co-director of the Joint Australian Tsunami Warning Centre. “Tsunamis don’t respect national boundaries. We need to be able to share data around the whole region to be able to make assessments about the threat for all exposed coastlines.
“Seen & Heard” is a monthly feature of GPS World magazine, traveling the world to capture interesting and unusual news stories involving the GNSS/PNT industry.
Bikes get tricked out
Image: Snik Bike
Snik Bike is a new app-paired tracking device designed to help users track their bikes if they are lost or stolen, reported North Shore News. Snik Bike Security Co-founder Fraser Vaage developed Snik Bike after having two of his bikes stolen. Snik equips any bike with a rechargeable location device that can be installed in five minutes or less. After pairing the device, the bike is automatically registered with Project 529, which is an online bike registration service. Vaage emphasized that while this device is not a ride-tracking software, such as Strava, it acts as an odometer, tracking overall mileage. The battery is only activated when a bike is stolen, Vaage said, therefore, it’s unlikely Snik will run out of battery. However, if it does, the device diverts to AirTag technology as a backup.
Lighthouses no more
Image: Wiltser/E+/Getty Images
With the wide adoption of GNSS, lighthouses on U.S. shorelines are no longer needed for navigation. To preserve these properties, the General Services Administration has been transferring ownership of the lighthouses to anyone willing to preserve them, reported The Guardian. This year, six lighthouses are being offered to federal, state or local government agencies, non-profits, educational organizations or anyone willing to make them publicly available for educational, cultural, or recreational purposes.
Location data ad weather resiliency
Image: DenisTangneyJr/E+/Getty Images
A Southern Methodist University research team, led by Nicos Markris, measured Dallas’ resilience by recording anonymous cell phone location data among residents in the Dallas metroplex before, during, and after the February 2021 North American winter storm. Measuring a city’s resilience is critical for planning responses to future events and uncovering potential vulnerabilities. By averaging location data, Makris and his team outlined the movement patterns of Dallas residents during a typical week. They compared the normal movement patterns to those during and after the week of the winter storm to determine when Dallas started getting back to normal.
GPS mitigates natural disasters
Image: Philip Thurston/E+/Getty Images
NASA’s Jet Propulsion Lab (JPL) is testing new ways to detect tsunami-like ocean waves before they cause catastrophic damage. The GNSS Upper Atmospheric Real-Time Disaster Information and Alert Network (GUARDIAN) is a new experimental monitoring system that can use data from clusters of GPS and other satellites to detect deadly waves triggered on Earth. Radio signals from GNSS are examined by scientific ground stations around the world. That data is then reviewed by the JPL’s Global Differential GPS network to help mitigate disasters. The GUARDIAN is still evolving and may be used in the future to develop early warning strategies, according to the United Nations’ International Committee on GNSS.
GNSS researchers presented hundreds of papers at the 2022 Institute of Navigation (ION) GNSS+ conference, which took place Sept. 19–23 in Denver, Colorado, and virtually. The following five papers focused on atmospheric effects on GNSS signals. The papers are available at www.ion.org/publications/browse.cfm.
Addressing Scintillation Error
Mitigating the scintillation effect at low latitude is a complex matter: several kinds of experimental data must be collected, realistic models must be developed, and, most importantly, useful real-time indices and alerts must be made available.
The authors introduce a prototype based on a patent owned by SpacEarth Technology to address scintillation error detection and mitigation, supporting precision GNSS-based services at low latitudes in any season and space weather conditions. The patent relates to a method of total electron content (TEC) and scintillation empirical forecasting, in particular short-term forecasting (seconds to minutes). The output of the method is necessary to feed mitigation algorithms aiming at improving accuracy on GNSS precise positioning techniques (RTK, NRTK, and PPP) under ionospheric harsh conditions.
The prototype is designed with a Central Elaborating Facility, which collects the data provided by a network of GNSS monitoring stations detecting scintillation events, and broadcasts foreseen scintillation parameters. Users with a rover mitigation device can apply the parameters from the central facility for scintillation error mitigation.
Vincenzo Romano, INGV and SpacEarth Technology; Claudio Cesaroni, INGV; Luca Spogli, Alessandro Fiorini, INGV and SpacEarth Technology; Marco Fermi, Gter; Lorenzo Benvenuto, Gter and University of Genoa; Tiziano Cosso, Gter; Marcin Grzesiak, SRC/PAS; Joao Francisco Galera Monico, Italo Tsuchiya, UNESP; Gabriel Oliveira, Marcos Guandalini; “Ionospheric Scintillation Mitigation at Low Latitude to Improve Navigation Quality.”
Ring of Fire GUARDIAN
Commonly, natural hazards release energy into the Earth’s atmosphere in the form of acoustic-gravity waves, which propagate up to the ionosphere. The resulting traveling ionospheric disturbances (TIDs) can be detected using GNSS signals, through the computation of the integrated total electron content (TEC) along the lines of sight between GNSS receivers and satellites. The global distribution of ground-based GNSS receivers constantly tracking multiple GNSS constellations (GPS, Galileo, GLONASS, BeiDou, and others) provides excellent spatio-temporal coverage around the world, including in areas of limited coverage by existing warning systems.
The authors present the operational GNSS-based Upper Atmospheric Real-time Disaster Information and Alert Network (GUARDIAN). Based on dual-frequency GNSS data from the Global Differential GPS (GDGPS) network of the Jet Propulsion Laboratory, the GUARDIAN architecture computes slant TEC time series in near real time.
As part of the GDGPS network, 78 stations around the Pacific ring of fire monitor the four GNSS constellations: GPS, Galileo, GLONASS and BeiDou. Cycle slips are corrected and the time series are filtered, both in real time. The resulting data stream is output live to a user-friendly public website, benefitting the general public and the scientific community.
The current GUARDIAN focuses on the Pacific region. However, the architecture can readily be extended to a worldwide coverage.
Léo Martire, S. Krishnamoorthy, L. J. Romans, B. Szilágyi, P. Vergados, A. W. Moore, A. Komjáthy, Y. E. Bar-Sever, A. B. Craddock, NASA Jet Propulsion Laboratory, California Institute of Technology; “GUARDIAN: A Near Real-Time Ionospheric Monitoring System for Natural Hazards Early Warnings.”
Civil Aviation Interference
The authors provide a survey on GNSS receiver architectures with emphasis on new carrier-tracking techniques for mitigating the adverse effect of ionospheric scintillation within the context of civil aviation. The survey is complemented by results gathered from simulations on the impact of ionospheric scintillation in conventional receiver architectures. A review on scintillation mitigation techniques is carried out, covering several “technique families,” highlighting their potential for performance improvement, as well as their shortcomings and challenges in implementation.
A semi-analytical simulation campaign is carried out for different modulations: L1, L5 for GPS, and E1, E5a for Galileo. Here, the performance of a standard receiver tracking a set of GPS and Galileo satellites affected by ionospheric scintillation is analyzed to pinpoint existing vulnerabilities to this effect.
The simulation results show that ionospheric scintillations are responsible for large variations in carrier-to-noise ratio, which in turn can be responsible for losses of lock and large phase variations, increasing phase RMSE and in some cases leading to cycle slips of the phase estimation. Thus, the adopted solution must be robust to signal power fluctuations and the occurrence of cycle slips and able to maintain phase lock.
António Negrinho, GMV-PT Pedro Boto, GMV-PT Marta Cueto, GMV-ES Mikael Mabilleau, EUSPA Claudia Paparini, EUSPA Ettore Canestri, EUSPA; “Survey on Signal Processing Techniques for GNSS Ionospheric Scintillation Mitigation.”
Tonga Eruption Data Analyzed
Extreme natural disasters, such as volcanic eruptions, can create visible pressure waves in the atmosphere and trigger observable ionospheric wave responses that can travel hundreds of kilometers in the ionosphere. The acoustic and gravity waves generated can cause ionospheric TEC perturbations and variations. The TEC determines the GNSS ionospheric delay and can cause significant positioning errors, which may affect the performance of GNSS-based applications.
The researchers processed GNSS data collected from the Hong Kong Satellite Positioning Reference Station Network to analyze the ionospheric activity and positioning performance responding to the Tonga volcanic eruption on Jan. 15, 2022. To detect and repair cycle-slip jumps, the researchers applied theTEC rate and Melbourne Wubbena Wide Lane (MWWL) linear combinations. A Savitzky-Golay low-pass filter with a 30s window was used to improve the TEC accuracy.
The team investigated the changes in TEC, Rate of TEC index (ROTI) and positioning errors in the eastward, northward and upward directions after the anomalous ionospheric propagation to Hong Kong between 11:30 and 14:30. The team found the ionospheric anomaly could generate large changes in the three parameters, with peaks up to three times the calm period. Their prompt research contributes to a better understanding of the coupling of extreme ionospheric activities and dynamics caused by volcanic eruptions.
Xiaojia Chang, Kai Guo, Zhipeng Wang, Kun Fang, Hongxia Wang, Beihang University; Hailong Chen, China Academy of Aerospace Electronics Technology; “Ionospheric Anomaly and GNSS Positioning Responses to the January 2022 Tonga Volcanic Eruption.”
Toolbox for Monitor Network
The MONITORtoolbox is a set of Python-coded software tools to perform automatized large-scale processing of data from the Monitor network of the European Space Agency (ESA). The Monitor network aims to continuously monitor ionospheric scintillation events from multiple ground stations strategically located around the globe. It accommodates a repository with a large number of GNSS measurements containing scintillation events for users to analyze scintillation data or for research purposes.
This paper shows the potential of the MONITORtoolbox for providing access to a large amount of data that otherwise, without a systematic processing, becomes practically useless. The software developed implements the means to collect data and store it in a local database for quick offline access. It detects the presence of scintillation events based on certain conditions and criteria defined by the user and identifies its properties in terms of duration, time of occurrence, intensity and satellite location. It implements the tools to compute relevant statistics, providing insights on ionospheric scintillation phenomena.
Sergi Locubiche-Serra, Alejandro Pérez-Conesa, Diego Fraile-Parra, Gonzalo Seco-Granados, José A. López-Salcedo, Universitat Autònoma de Barcelona, IEEC-CERES; Juan M. Parro-Jiménez, Raúl Orús-Pérez, ESTEC, European Space Agency; “MONITORtoolbox — Software Tool for the Analysis of Ionospheric Scintillation Data from the ESA Monitor Network.”
On Jan. 15, Hunga-Tonga-Hunga-Ha’apai, an uninhabited volcanic island on the Tongan archipelago in the South Pacific Ocean, erupted with spectacular force, churning ocean waters halfway across the globe.
GNSS engineers also detected its effects hundreds of miles above, in the ionosphere. The GNSS community is now moving from such after-the-fact detection to real-time monitoring using NASA’s Global Differential GPS (GDGPS) system, according to a team with the Tracking Systems and Application Section at NASA’s Jet Propulsion Laboratory (JPL) in Southern California.
“We monitored, in real time, four GNSS satellite constellations from numerous stations around the world using the GDGPS network. In particular, the three stations closest to the volcano, in Samoa, Fiji and Tahiti,” said postdoctoral associate Leo Martire. “We could see extremely high and strong signals in the ionosphere, which is very unusual. As a function of radial distance from the eruption, the first detected ionospheric perturbation likely originated directly from the explosion. Then we see patterns propagating at increasing distances at different radial propagation speeds.”
Monitoring such events adds information to the catalog of signals from natural hazards, pointed out Siddharth Krishnamoorthy, a research technologist who manages JPL’s GUARDIAN near-real-time tsunami warning system, currently under development. “That is useful because, in the future, if you want to be able to spot natural hazards and issue alerts, you need to know what the signal looks like. There have been reports of a tsunami in Tonga due to this event, so we will look at potential tsunami-induced signatures in the ionosphere. We are trying to get to a place where we pick up a signal like this and we are able to say, ‘This is a tsunami propagating at this speed and in this direction.’”
Chart: Jet Propulsion Laboratory
Before being detected in the ionosphere, signals from natural hazards must travel all the way from the surface. For tsunamis, this usually takes more than 10 to 20 minutes, but the volcanic eruption only took a couple of minutes to reach the ionosphere because it shot straight up. “We do not know yet, based on observations, how exactly different events on the surface caused by natural hazards couple with the atmosphere,” said research technologist Panagiotis Vergados. “Every event is unique in its spectral properties.”
The event did not affect the quality of GDGPS’s GNSS positions or orbits, because dual-frequency measurements remove significant ionospheric effects. “Instead of looking at the direct effects on the position of our available reference stations, which is what our traditional real-time monitoring does and which was basically negligible, imagine the links from each of those stations to a dozen or more satellites,” said Larry Romans, GDGPS chief technologist. “Every time one of those many links pierces the ionosphere, we can monitor that signal for ripples as waves go by. So, this is an incredibly powerful method for seeing disturbances, just in terms of the density of data. It is very complementary to position-based natural-hazards monitoring because the data is much richer.”
In addition to volcanoes and tsunamis, several other natural events, such as earthquakes and very large thunderstorms, also produce these effects. “These natural forcings cause large-scale, low-frequency pressure perturbations that tend to travel up and be visible in the ionosphere,” Krishnamoorthy said. “There are also perturbations of the ionosphere due to events from outside the Earth, such as solar flares or bolide impacts.”
Many of these perturbations start from the troposphere, which ranges between 10 km and 15 km in altitude — including hurricanes, which overshoot gravity waves all the way to the ionosphere, and thermal tides that have been observed to go all the way up to 600 km, said Vergados. “There are also geomagnetic storms and sub-storms that, during electron precipitation, can change the ionization of the ionosphere. So, the coupling can happen from either below or above or simultaneously, and then the effect can be dramatically enhanced.”
Most of the perturbations that come from below are of a pressure nature — that is, they start out as mechanical waves — while most of those that come from above are electromagnetic. “Aside from nuclear explosions, very large chemical ones, such as the 2020 Beirut explosion, also cause a signature on the ionosphere because they create very large pressure waves,” Krishnamoorthy said.
Photo: Tonga Meteorological Services, Government of Tonga
