NASA’s Double Asteroid Redirection Test changed more than the motion of its immediate target. DART’s collision with Dimorphos also changed the movement of the moonlet and its larger companion, Didymos, around the Sun. Results published on March 6, 2026 put the change in the binary system’s orbital speed at about 11.7 micrometers per second, with a 0.15-second change in its roughly 770-day solar orbital period.
Rahil Makadia of the University of Illinois Urbana–Champaign and colleagues used twenty-two stellar occultations recorded between October 2022 and March 2025, alongside other observations. During an occultation, the asteroid passes in front of a distant star and briefly blocks its light. Volunteer observers supplied measurements from widely separated locations. Neither asteroid threatened Earth before or after DART’s impact. The new result concerned their heliocentric motion, following the earlier measured change in Dimorphos’s orbit around Didymos.
Launching the impact experiment
DART launched on November 24, 2021, at 1:21 a.m. EST aboard a SpaceX Falcon 9 from Space Launch Complex 4 East at Vandenberg Space Force Base in California. The Johns Hopkins Applied Physics Laboratory in Laurel, Maryland, designed, developed and managed the mission for NASA’s Planetary Defense Coordination Office. Its planned destination was a binary asteroid system whose moonlet’s orbital change could be measured with ground telescopes.
The spacecraft separated from the launch vehicle’s second stage at 2:17 a.m. EST. Operators received its first telemetry minutes later and began orienting it for solar-array deployment. Almost two hours later, both twenty-eight-foot roll-out arrays had unfurled. DART carried Roll-Out Solar Array technology into deep space and incorporated experimental solar concentrators. The power system was also intended to support the NEXT-C ion engine, a technology demonstration for future missions. The launch plan called for ten months of travel near Earth’s orbit before interception. The Italian Space Agency’s LICIACube rode with DART, with deployment planned before the collision so that it could photograph the ejecta and the side of Dimorphos that DART’s camera would not see.
Finding the smaller target
SMART Nav supplied the autonomous component of DART’s terminal navigation. The design called for the spacecraft to begin its final four hours with a four-megapixel view containing only faint points of light. At that stage, its target would still be between approximately 54,000 and 61,000 miles away. Dimorphos itself was expected to become visible only about an hour before collision. The navigation system had to distinguish the moonlet from the brighter primary asteroid and direct the spacecraft toward the smaller object.
The algorithms worked with the spacecraft’s guidance and navigation system to locate objects, select the appropriate asteroid, estimate corrections and command maneuvers. Mark Jensenius, an APL guidance and navigation control engineer, described the intended operation as carrying out the final four hours without human intervention. Rather than following only a prescribed orientation or a sequence of corrections sent from Earth, SMART Nav would use its changing camera view to make those decisions aboard the spacecraft. The July 2021 design account placed this autonomous targeting requirement within the planned high-speed encounter, when a small maneuvering error could cause DART to miss Dimorphos.
Rehearsing with Jupiter and Europa
The operations team tested SMART Nav in flight on July 1 and August 2, 2022, by pointing DART’s DRACO camera toward Jupiter. Europa emerged from behind the planet, offering a moving scene in which a small target separated visually from a much larger body. The team used the sequence to test detection and targeting before the asteroid encounter. Earlier SMART Nav testing had relied on ground simulations.
During the rehearsals, engineers examined object brightness, pixel counts and the consistency of identification as the targets moved across the detector. Software engineer Peter Ericksen said the tests also helped the team refine its monitoring displays. DRACO had already taken thousands of stellar images for spacecraft testing and rehearsals. Its high-resolution design drew on the New Horizons imager that observed Pluto and Arrokoth. DART was approximately twenty-six million kilometers from Earth during the Jupiter imaging, while Jupiter was about seven hundred million kilometers from the spacecraft. The planned terminal approach remained autonomous, with ground controllers monitoring the tracked objects and retaining prearranged contingencies for significant deviations that threatened the mission.
The collision and its witness
DART struck Dimorphos on September 26, 2022. Mission control at APL announced the successful impact at 7:14 p.m. EDT. The spacecraft had a mass of about 570 kilograms and hit at approximately 22,530 kilometers per hour. DRACO, its instrument, returned close-up images of the moonlet’s surface during the final seconds. The camera and guidance system had distinguished Dimorphos from Didymos and steered toward the smaller body.
LICIACube had separated fifteen days earlier. Its task was to observe the impact and the resulting cloud of material from a different vantage point. The small spacecraft’s limited antenna meant that images would be transmitted individually over the following weeks. At collision, the asteroid pair was approximately eleven million kilometers from Earth. Dozens of ground and space telescopes began observing the system to characterize the debris and determine the change in the moonlet’s orbit. NASA initially expected a reduction of about one percent, or roughly ten minutes. The impact announcement confirmed the collision; the orbital measurement followed after observers had collected additional data.
Timing the first orbital change
By October 11, 2022, observations showed that Dimorphos’s orbital period had fallen from eleven hours and fifty-five minutes to eleven hours and twenty-three minutes. The initial estimate was a reduction of thirty-two minutes, with an uncertainty of approximately two minutes. Before the encounter, NASA had set a successful period change at seventy-three seconds or more. The measured reduction exceeded that threshold by over twenty-five times.
The observing campaign combined several facilities:
- Goldstone planetary radar in California and Green Bank Observatory in West Virginia;
- the Swope Telescope at Las Campanas Observatory and the Danish Telescope at La Silla Observatory in Chile;
- Las Cumbres Observatory network telescopes in Chile and South Africa.
Researchers continued measuring the period to improve its precision. They also examined the ejecta’s contribution to the push against Dimorphos, using telescope observations and images from DART and LICIACube. The spacecraft images supported estimates of the asteroid’s mass and shape. These were early post-collision measurements, before subsequent refinements of the orbit and physical models.
Measuring the extra push from debris
By December 2022, investigators estimated that the collision had expelled more than a million kilograms of dusty rock. Material leaving the asteroid carried momentum, and its recoil enhanced the spacecraft’s direct push. At the American Geophysical Union’s Fall Meeting in Chicago on December 15, the team presented a preliminary interpretation of that momentum transfer. Assuming Didymos and Dimorphos had equal densities, they calculated an enhancement of roughly 3.6 compared with a collision in which the asteroid simply absorbed the spacecraft without ejecting material.
Observations before and after impact indicated similar compositions for the two asteroids, with material linked to ordinary chondrites, a common class of meteorite. Ejecta dominated the system’s reflected light in the days following collision. Telescope images also showed a tail extending tens of thousands of miles as sunlight pushed small particles away. APL investigation lead Andy Cheng emphasized momentum transfer as a quantity needed when estimating an impactor’s size and the warning time required for a future deflection mission. The numerical enhancement remained an estimate conditional on the assumed densities, while measurements of the debris and asteroid properties continued.
Following the expanding dust cloud
Hubble observed three overlapping stages of the ejecta’s development: a hollow cone, a rotating pattern shaped by the binary system and a long tail driven by sunlight. Jian-Yang Li of the Planetary Science Institute led the analysis published in March 2023. Hubble’s sequence began before impact, with its first post-impact image taken approximately two hours after collision. The debris cone contained long filaments, with material escaping the asteroid at a range of speeds.
Around seventeen hours after impact, interactions within the binary system distorted the cone into prominent pinwheel-like structures. Li associated those features with Didymos’s gravitational influence. Later, solar-radiation pressure swept small dust grains into a comet-like tail, with the lightest particles moving fastest and farthest. For several days, Hubble recorded a split tail. Both asteroids remained unresolved inside the image’s central bright spot. Thin spikes extending from that spot were optical artifacts of the telescope. Hubble followed the debris over a larger volume than LICIACube could cover during its brief flyby; the James Webb Space Telescope and the Lucy spacecraft also observed the impact and its aftermath.
Separating boulders from the dust
Hubble images taken on December 19, 2022, revealed thirty-seven boulders moving away from the asteroid. Photometry indicated sizes ranging from about three to twenty-two feet across. Their combined estimated mass was approximately 0.1 percent of Dimorphos’s mass, and they drifted away at a little over half a mile per hour. David Jewitt of the University of California, Los Angeles, interpreted their numbers and sizes as consistent with material knocked from the moonlet’s surface.
The large rocks were probably pre-existing surface boulders rather than fragments created by shattering the entire asteroid. DART’s last complete image had already shown boulders scattered over the surface. Jewitt estimated that the collision displaced about two percent of that surface population. Their release mechanism remained uncertain: they could have traveled with the ejecta plume or been shaken loose by a seismic wave passing through the asteroid. Further observations were intended to refine their trajectories and identify the directions in which they left the surface. Jewitt expected the slowly dispersing cloud to remain present when Hera reached the system for its planned survey in late 2026.
Reconstructing a changed shape
A March 2024 analysis led by JPL navigation engineer Shantanu Naidu combined DRACO images, Goldstone radar and light curves. Dips in the reflected light occurred when one asteroid crossed in front of, passed behind or cast a shadow on the other. The timings of these mutual events constrained the orbit’s geometry. The models indicated that Dimorphos had changed from a symmetrical, flattened body into a more elongated triaxial ellipsoid. Its previously circular orbit had also become slightly eccentric.
The modelled orbital period shortened further during the weeks after collision as additional rocky material escaped. The final estimate was eleven hours, twenty-two minutes and three seconds, a reduction of thirty-three minutes and fifteen seconds, accurate to about one and a half seconds. The mean orbital separation fell from approximately 1,189 meters to 1,152 meters. The models also indicated that Dimorphos rocked back and forth during its orbit. Coauthor Steve Chesley linked the inferred shape and orbit to the timing of the light-curve dips. These results described a later reconstruction with more information than the initial October 2022 period measurement.
Reading the asteroid surfaces
Detailed geological analyses in 2024 examined boulder distributions, craters and terrain on both asteroids. Olivier Barnouin and Ronald-Louis Ballouz of APL found smoother terrain at lower elevations on Didymos and rockier terrain higher up. Didymos also had more craters than Dimorphos. The team inferred that the moonlet probably formed from material shed by a spinning Didymos. Their estimated surface ages were about 12.5 million years for Didymos and less than 300,000 years for Dimorphos.
Other investigators examined different properties of the surfaces. Maurizio Pajola and colleagues at the Italian National Institute for Astrophysics studied boulder shapes, sizes and distributions, interpreting them as evidence of formation in stages from the parent asteroid’s material. Alice Lucchetti’s team investigated thermal fatigue, in which repeated heating gradually weakens and cracks rocks. Jeanne Bigot and Pauline Lombardo, working with Naomi Murdoch at ISAE-SUPAERO, estimated extremely low bearing capacity on Didymos. Colas Robin’s group compared Dimorphos’s boulders with those on Itokawa, Ryugu and Bennu. They associated elongated boulders near DART’s impact site with earlier impact processing. These were geological interpretations of observed surface features and physical models.
Estimating material hidden in the plume
LICIACube supplied a close view of the first minutes after impact. During its subsequent flyby, the small spacecraft had about sixty seconds for its critical observations, taking images approximately every three seconds. Its camera turned as it passed the asteroids, viewing the plume from changing angles.
Ramin Lolachi’s team analyzed eighteen images for an August 2025 study. The plume appeared bright during the approach, then dimmed as LICIACube looked back through it. The researchers inferred that much of the material consisted of comparatively large particles, about a millimeter or more across. The cloud’s inner region was opaque, so models and laboratory experiments helped estimate material that the camera could not see directly.
The resulting estimate was approximately sixteen million kilograms of expelled dust and rock, less than half a percent of Dimorphos’s total mass. Hidden material accounted for almost forty-five percent of the estimated plume mass. Planetary scientist Timothy Stubbs cautioned that stronger, more compact asteroids could respond differently from the loosely assembled Dimorphos. The mass estimate depended on modelling the particles and the obscured region.
Leave a comment