AI-Powered Discovery: Mars Impact Crater Reveals New Insights into Planet's Interior

Mars spacecraft used AI to find a fresh impact crater on the surface - Earth.com

Mars continues to surprise scientists with its geological activity. While the Red Planet lacks the tectonic movements that shape Earth, it still experiences seismic activity caused by internal forces and meteoroid impacts. Recent discoveries, powered by artificial intelligence, have revealed a new crater on Mars that evidences how deeply impact-generated seismic waves can travel. This finding challenges existing models of the planet's interior and opens new avenues for understanding its geological history. A pair of recent studies published in Geophysical Research Letters detail how scientists linked a newly formed crater to seismic activity recorded by NASA's InSight lander. This marks a major breakthrough and shows that impact-related tremors can penetrate deeper into the Martian mantle than previously thought. With the help of data from the Mars Reconnaissance Orbiter (MRO), researchers can now refine their models of the planet's subsurface, which will enhance our knowledge of Mars and also our understanding of other rocky worlds, including Earth and its Moon. NASA's InSight mission was designed to probe the interior of Mars and offer an unprecedented look at the planet's crust, mantle, and core. The lander, which arrived in 2018, deployed the first seismometer ever placed on the Martian surface. Over its four-year mission, InSight detected more than 1,300 marsquakes, some caused by internal forces and others triggered by space rocks striking the planet. By studying how seismic waves move through different layers of Mars, scientists have been able to infer the planet's internal structure. Unlike Earth, which has a dynamic interior, Mars appears to have a colder, more rigid mantle. However, new findings suggest that some seismic waves travel deeper than previously assumed, indicating that the Martian mantle may be more complex than once thought. One of the most important recent discoveries involved a meteoroid impact near Cerberus Fossae, a geologically active region on Mars. The impact not only created a visible crater but also generated seismic waves that traveled in an unexpected way, revealing a new pathway for seismic energy through the planet's mantle. Scientists have long relied on seismic data to study the interiors of planets. On Earth, earthquakes help researchers understand how our planet's layers interact. On Mars, where seismic activity is much weaker, researchers use marsquakes and impact events to investigate the nature of the subsurface. The newly discovered Mars crater, which measures 71 feet (21.5 meters) in diameter, is located over 1,019 miles (1,640 km) from InSight's landing site. The seismic energy it produced was surprisingly strong for an impact at that distance. Scientists originally believed that Mars's crust would dampen seismic waves from impacts, thus weakening them before they reached the InSight lander. However, this event proved otherwise. "We used to think the energy detected from the vast majority of seismic events was stuck traveling within the Martian crust," said InSight team member Constantinos Charalambous of Imperial College London. "This finding shows a deeper, faster path - call it a seismic highway - through the mantle, allowing quakes to reach more distant regions of the planet." This new evidence forces scientists to rethink their models of how seismic waves travel on Mars. If impact-generated waves can reach deeper into the mantle, it means that Mars's internal structure may differ significantly from what was previously assumed. Tracking new impact craters on Mars has always been a labor-intensive task. Scientists rely on before-and-after images taken by MRO to detect changes on the planet's surface. In the past, this process required manually sifting through thousands of images, looking for telltale signs of fresh craters. To accelerate this process, NASA's Jet Propulsion Laboratory developed an AI-powered tool that can rapidly analyze images from MRO's Context Camera. This machine-learning algorithm scans tens of thousands of images in a matter of hours, identifying potential impact sites for further investigation. "Done manually, this would be years of work," said InSight team member Valentin Bickel of the University of Bern in Switzerland. "Using this tool, we went from tens of thousands of images to just a handful in a matter of days. It's not quite as good as a human, but it's super fast." The AI-assisted search focused on areas within 1,864 miles (3,000 km) of InSight's location, looking for craters that formed while the lander's seismometer was active. By comparing time-stamped images, scientists found 123 fresh craters. Of these, 49 showed possible matches with marsquake data. After additional filtering, they identified the impact crater in Cerberus Fossae as the most likely source of a recorded seismic event. One of the biggest challenges in planetary seismology is differentiating between quakes caused by internal processes and those triggered by meteoroid strikes. InSight's data has helped scientists improve their ability to classify these events, but this new discovery suggests that some previous assumptions may need revision. "We thought Cerberus Fossae produced lots of high-frequency seismic signals associated with internally generated quakes, but this suggests some of the activity does not originate there and could actually be from impacts instead," Charalambous said. This distinction is crucial for understanding Mars's long-term geological evolution. If some seismic signals attributed to tectonic activity were actually caused by impacts, it could reshape interpretations of Mars's internal dynamics. The success of AI in detecting impact craters is just one example of how machine learning is transforming Martian research and overall planetary science. AI tools are now being used to identify landslides, dust devils, and other surface changes on Mars. Similar techniques have been applied to analyze images of the Moon, and they have revealed previously unnoticed craters and geological features. "Now we have so many images from the Moon and Mars that the struggle is to process and analyze the data," Bickel said. "We've finally arrived in the big data era of planetary science." As missions like NASA's Perseverance rover and ESA's ExoMars continue to gather vast amounts of data, AI will play an increasingly vital role in processing and interpreting these findings. Scientists are now developing new machine-learning models that can automatically detect geological changes, allowing them to make discoveries faster than ever before. NASA's InSight mission was part of the agency's Discovery Program, managed by the Marshall Space Flight Center. Lockheed Martin Space designed and built the spacecraft, while numerous European institutions contributed to its scientific instruments. France's Centre National d'Études Spatiales (CNES) provided the Seismic Experiment for Interior Structure (SEIS), with other contributions from the Institut de Physique du Globe de Paris, the Max Planck Institute for Solar System Research, and ETH Zurich. Germany's DLR provided the Heat Flow and Physical Properties Package, while Spain's Centro de Astrobiología supplied temperature and wind sensors. The Mars Reconnaissance Orbiter, which played a crucial role in this discovery, is managed by NASA's Jet Propulsion Laboratory. The University of Arizona operated its high-resolution camera, HiRISE, while Malin Space Science Systems built and operates the Context Camera. The discovery of a fresh impact crater linked to seismic waves traveling through Mars's mantle marks a significant advance in planetary science. It not only challenges previous models but also highlights the importance of AI in accelerating discoveries. As researchers continue to analyze InSight's data, they will refine their understanding of how the Martian interior behaves. With AI-driven tools improving the efficiency of crater detection and seismic analysis, future missions will be better equipped to unlock the secrets hidden beneath the Red Planet's surface. Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates.

NASA's InSight Finds Marsquakes From Meteoroids Go Deeper Than Expected - NASA

With help from AI, scientists discovered a fresh crater made by an impact that shook material as deep as the Red Planet's mantle. Meteoroids striking Mars produce seismic signals that can reach deeper into the planet than previously known. That's the finding of a pair of new papers comparing marsquake data collected by NASA's InSight lander with impact craters spotted by the agency's Mars Reconnaissance Orbiter (MRO). The papers, published on Monday, Feb. 3, in Geophysical Research Letters (GRL), highlight how scientists continue to learn from InSight, which NASA retired in 2022 after a successful extended mission. InSight set the first seismometer on Mars, detecting more than 1,300 marsquakes, which are produced by shaking deep inside the planet (caused by rocks cracking under heat and pressure) and by space rocks striking the surface. By observing how seismic waves from those quakes change as they travel through the planet's crust, mantle, and core, scientists get a glimpse into Mars' interior, as well as a better understanding of how all rocky worlds form, including Earth and its Moon. Researchers have in the past taken images of new impact craters and found seismic data that matches the date and location of the craters' formation. But the two new studies represent the first time a fresh impact has been correlated with shaking detected in Cerberus Fossae, an especially quake-prone region of Mars that is 1,019 miles (1,640 kilometers) from InSight. The impact crater is 71 feet (21.5 meters) in diameter and much farther from InSight than scientists expected, based on the quake's seismic energy. The Martian crust has unique properties thought to dampen seismic waves produced by impacts, and researchers' analysis of the Cerberus Fossae impact led them to conclude that the waves it produced took a more direct route through the planet's mantle. InSight's team will now have to reassess their models of the composition and structure of Mars' interior to explain how impact-generated seismic signals can go that deep. "We used to think the energy detected from the vast majority of seismic events was stuck traveling within the Martian crust," said InSight team member Constantinos Charalambous of Imperial College London. "This finding shows a deeper, faster path -- call it a seismic highway -- through the mantle, allowing quakes to reach more distant regions of the planet." A machine learning algorithm developed at NASA's Jet Propulsion Laboratory in Southern California to detect meteoroid impacts on Mars played a key role in discovering the Cerberus Fossae crater. In a matter of hours, the artificial intelligence tool can sift through tens of thousands of black-and-white images captured by MRO's Context Camera, detecting the blast zones around craters. The tool selects candidate images for examination by scientists practiced at telling which subtle colorations on Mars deserve more detailed imaging by MRO's High-Resolution Imaging Science Experiment (HiRISE) camera. "Done manually, this would be years of work," said InSight team member Valentin Bickel of the University of Bern in Switzerland. "Using this tool, we went from tens of thousands of images to just a handful in a matter of days. It's not quite as good as a human, but it's super fast." Bickel and his colleagues searched for craters within roughly 1,864 miles (3,000 kilometers) of InSight's location, hoping to find some that formed while the lander's seismometer was recording. By comparing before-and-after images from the Context Camera over a range of time, they found 123 fresh craters to cross-reference with InSight's data; 49 of those were potential matches with quakes detected by the lander's seismometer. Charalambous and other seismologists filtered that pool further to identify the 71-foot Cerberus Fossae impact crater. The more scientists study InSight's data, the better they become at distinguishing signals originating inside the planet from those caused by meteoroid strikes. The impact found in Cerberus Fossae will help them further refine how they tell these signals apart. "We thought Cerberus Fossae produced lots of high-frequency seismic signals associated with internally generated quakes, but this suggests some of the activity does not originate there and could actually be from impacts instead," Charalambous said. The findings also highlight how researchers are harnessing AI to improve planetary science by making better use of all the data gathered by NASA and ESA (European Space Agency) missions. In addition to studying Martian craters, Bickel has used AI to search for landslides, dust devils, and seasonal dark features that appear on steep slopes, called slope streaks or recurring slope linae. AI tools have been used to find craters and landslides on Earth's Moon as well. "Now we have so many images from the Moon and Mars that the struggle is to process and analyze the data," Bickel said. "We've finally arrived in the big data era of planetary science." JPL managed InSight for the agency's Science Mission Directorate. InSight was part of NASA's Discovery Program, managed by the agency's Marshall Space Flight Center in Huntsville, Alabama. Lockheed Martin Space in Denver built the InSight spacecraft, including its cruise stage and lander, and supported spacecraft operations for the mission. A number of European partners, including France's Centre National d'Études Spatiales (CNES) and the German Aerospace Center (DLR), supported the InSight mission. CNES provided the Seismic Experiment for Interior Structure (SEIS) instrument to NASA, with the principal investigator at IPGP (Institut de Physique du Globe de Paris). Significant contributions for SEIS came from IPGP; the Max Planck Institute for Solar System Research (MPS) in Germany; the Swiss Federal Institute of Technology (ETH Zurich) in Switzerland; Imperial College London and Oxford University in the United Kingdom; and JPL. DLR provided the Heat Flow and Physical Properties Package (HP) instrument, with significant contributions from the Space Research Center (CBK) of the Polish Academy of Sciences and Astronika in Poland. Spain's Centro de Astrobiología (CAB) supplied the temperature and wind sensors. A division of Caltech in Pasadena, California, JPL manages the Mars Reconnaissance Orbiter Project for NASA's Science Mission Directorate, Washington. The University of Arizona, in Tucson, operates HiRISE, which was built by BAE Systems in Boulder, Colorado. The Context Camera was built by, and is operated by, Malin Space Science Systems in San Diego.

Machine learning reveals meteoroid impacts may play a bigger role in triggering marsquakes

Meteoroid impacts create seismic waves that cause Mars to shake more strongly and deeply than previously thought. This is shown by an investigation using artificial intelligence carried out by an international research team led by the University of Bern. Similarities were found between numerous meteoroid impacts on the surface of Mars and marsquakes recorded by NASA's Mars lander InSight. These findings open up a new perspective on the impact rate and seismic dynamics of the red planet. Meteoroid impacts have a significant influence on the landscape evolution of solid planetary bodies in our solar system, including Mars. By studying craters -- the visible remnants of these impacts -- important properties of the planet and its surface can be determined. Satellite images help to constrain the formation time of impact craters and thus provide valuable information on impact rates. A recent study led by Dr. Valentin Bickel from the Center for Space and Habitability at the University of Bern presents the first comprehensive catalog of impacts on the Martian surface that took place near NASA's Mars lander during the InSight mission between December 2018 and December 2022. Bickel is also an InSight science team member. The study has just been published in Geophysical Research Letters. Machine learning identifies new Martian impacts The impact events were cataloged using a machine learning approach. Tens of thousands of satellite images were searched for new craters that formed during the seismic monitoring by InSight. Using images from the High Resolution Imaging Science Experiment (HiRISE) and the Bernese Mars camera CaSSIS the craters were classified according to their size. "Next, we compared the distribution of the craters with the seismic recordings from InSight and looked for matches in space and time," explains first author Bickel. This innovative approach made it possible to identify a total of 123 previously unknown impacts. Based on their determined formation time, estimated magnitude and distance to InSight, the researchers found potential matches between 49 seismic events and one or more possible impact events. "Our data show that more impacts occur on Mars than were determined in previous studies using orbital images," says Bickel. The estimated impact rate is around 1.6 to 2.5 times higher than previously assumed. "Our observations show that some of the recorded marsquakes are actually caused by meteoroid impacts and not tectonic activity. This has far-reaching implications for estimates of the frequency of marsquakes and our understanding of the dynamics of the Martian surface in general." Wave propagation through the Martian mantle In a companion study, the research team focused on one of the newly discovered events, a 21.5-meter impact crater in the Cerberus Fossae region, which the team linked to a specific high-frequency marsquake. The Cerberus Fossae rift system is located in a young volcanic plain on Mars that is known for its tectonic activity. This discovery enables the first direct comparison between an impact-induced seismic signal and a signal caused by internal tectonic movements. The researchers compared the impact location and the time at which InSight registered the respective marsquake. They were able to show that some of the seismic waves propagated through the deeper Martian mantle and not, as previously assumed, only through the surface crust. "These findings challenge previous assumptions about the propagation of seismic waves and suggest that numerous recorded marsquakes were actually further away from the Mars lander InSight than previously thought," says Constantinos Charalambous, InSight science team member at Imperial College London and lead author of the companion study also published in Geophysical Research Letters. "In addition to re-locating the epicenters of a range of quakes, this also means that the internal structural model of Mars needs to be revised," says Charalambous. Searching for further similarities "Our results are not only important for the scientific community. For example, if you want to build a permanent infrastructure on Mars in the future, you need to be able to assess the risk of structural damage, such as caused by meteoroid impacts," emphasizes Bickel. The studies show that the combination of seismic data and orbital image information is crucial for understanding the geophysical properties of Mars. Further research on Mars will aim to refine estimates of marsquake frequency and impact rates. The studies are the result of an international, interdisciplinary collaboration between researchers from the University of Bern and other renowned institutions, including the NASA Jet Propulsion Laboratory (JPL), Imperial College London, Brown University, and ETH Zurich. "At the University of Bern, we are ideally positioned to conduct this type of research -- particularly because of our interdisciplinary expertise in planetary sciences and machine learning, as well as Bern's active participation in InSight, HiRISE and CaSSIS," concludes Bickel.