OrbitalHub

In 2009, the Lunar Reconnaissance Orbiter finished its commissioning burns and started taking the highest-resolution pictures of the Moon anyone had ever made. Seventeen years later, on 10 September 2026, IBM and NASA published a free neural network that reads those pictures better than anything before it. The NASA-IBM Lunar Foundation Model, released under Apache 2.0 on Hugging Face, identifies lunar ice with 22 percent less error than the strongest dedicated algorithm, maps craters 19 percent more accurately at coarse resolution using half the labelled data, and traces volcanic deposits about 3 percent more cleanly with less fine-tuning (NASA Science and IBM Research, 10 September 2026). For the first time, planetary science has a single foundation model built specifically for Earth’s oldest companion.

The work matters because the next humans on the Moon are scheduled to land somewhere most maps refuse to draw. The Artemis program targets the lunar south pole, a region ringed by permanently shadowed regions — craters so deep that the Sun never rises more than a few degrees above their rims. Floor temperatures there bottom out near 40 kelvin, cold enough that water molecules arriving on the solar wind or from cometary impacts can survive for billions of years instead of sublimating away (Vasavada et al., Icarus, 1999). NASA estimates that the lunar north pole alone may hold roughly 600 million metric tons of water ice, enough to fill at least 240,000 Olympic swimming pools; the south pole is expected to hold comparable reserves (IBM Research, 10 September 2026). Ice is propellant, drinking water, and oxygen in waiting. Every kilogram of hydrogen and oxygen a crew can mine on the surface is a kilogram a Falcon Heavy does not have to lift off Earth. The economic case for Artemis depends on how cleanly geologists can rank those PSRs before a lander commits to one.

What the new model solves is a different problem that has quietly paralyzed the field: data fragmentation. LRO has produced more raw imaging data than every other NASA planetary mission combined (NASA Science, 10 September 2026). Some of it is camera imagery at one-meter-per-pixel ground sampling in seven spectral bands. GRAIL, in 2012, mapped the gravity field at roughly 20 kilometers per pixel — the shape of the interior rather than the surface. Japan’s SELENE (Kaguya) orbiter layered in mineralogical maps and laser altimetry. For sixty years, lunar science has lived in a stack of incompatible file formats. Combining them was a graduate student’s summer project, repeated for every paper.

The Foundation Model replaces that summer project with one download. IBM and NASA call the accompanying release SomBench, the first open-source, machine-learning-ready unified lunar dataset (Hugging Face, NASA-IBM-Lunar-Foundation-Model, 2026). SomBench aggregates more than 30 spatially aligned data layers from nine instruments across four missions — principally LRO and GRAIL plus complementary observations from SELENE — into roughly two million co-registered image tiles (CNET, 10 September 2026). The training corpus itself exceeds one million high-resolution images at one-meter resolution and close to 964,000 multispectral tiles at 100-meter resolution (NASA Science, 10 September 2026). Tiles have already been re-projected onto a common grid.

The people behind the release come from a partnership that reaches back to Apollo. IBM and NASA have worked together since the 1960s, when IBM guidance computers flew on Saturn V. Kevin Murphy, NASA’s chief science data officer, said in the joint release: “We also have to make data easier for scientists to explore and use” (NASA Science, 10 September 2026). Campbell Watson, the IBM Research senior manager who led the build, told CNET the team wanted a “shared foundation that researchers can adapt” (CNET, 10 September 2026). Juan Bernabé-Moreno of IBM Research put it as travel advice: “Experienced travelers know to get the lay of the land before setting out for a foreign destination” (IBM Research, 10 September 2026).

The machine-learning choices under the hood explain why the gains are real. The architecture is a Vision Transformer Base encoder-decoder — 768 hidden dimensions, 12 transformer layers, 12 attention heads — adapted from TerraMind, the Earth-observation foundation model IBM had previously built with ESA (Tech Times, 11 September 2026). The transformer self-attention operator at the heart of each block computes pairwise token affinities scaled by 1/sqrt(d_k), with d_k set to 64 in the 12-head configuration; the softmax of those affinities is what lets the network learn which patches of the input relate to which. TerraMind’s special sauce is its cross-modal pretraining, which forces the model to learn correlations between disparate sensor streams, so that a noisy radar channel can be reconstructed by attending to what a multispectral camera saw of the same surface. On the Moon, that trick becomes a way to fill in what the PSRs will not give an optical camera: a thermal map from the Diviner instrument over a shadowed crater can be cross-referenced with a gravity anomaly from GRAIL, and the model learns that combination as a single latent representation rather than two independent ones.

Training was masked-autoencoder pretraining on the full two-million-tile corpus, where the model is asked to reconstruct randomly masked patches and in doing so must learn the statistics of lunar terrain. To prevent the network from memorising a few well-imaged tiles, the team held out distinct geographic “wedges” of the Moon for testing (Time News, 11 September 2026). The technical report lists three downstream gains. For ice detection, the model cut root-mean-square error by up to 22 percent over a SwinV2-B transformer pretrained on ImageNet. For crater detection at 100-meter context, it outperformed the same baseline by nearly 19 percent with half the labelled training tiles; at meter scale it matches state-of-the-art head-to-head. For irregular mare patches — the subtle volcanic features whose ages are reshaping the chronology of lunar cooling — the model improved extent accuracy by about 3 percent at lower fine-tuning cost (NASA-IBM Lunar Foundation Model technical report, Hugging Face, 2026).

The physics gives those percentages meaning. Ice stability inside a PSR depends on local temperature integrated over a full precession cycle and on burial depth below the gardening reach of micrometeorite impacts. A model that has internalised those thermophysical correlations can rank candidate deposits the way a specialist would, across the whole Moon. Crater counts age a surface because the impact flux is approximately known, and a model trained on the full LRO catalogue has absorbed what an impact signature looks like at every illumination angle — critical for a sensor whose parent body has a 29.5-day day-night cycle. Even IMP detection improves because the model has learned that young volcanic features are defined by a bundle of morphological and compositional cues, not by any single texture.

What changes because of this release is the slope of the field. The model is not a substitute for a graduate-level lunar geologist; it lets researchers who previously spent six months pre-processing a single tile bundle ask questions across the whole catalogue in an afternoon. Apache 2.0 means any team — a Chinese lunar university, an Indian student, a Brazilian startup — can build on the same backbone. The Prithvi family IBM and NASA open-sourced for Earth observation will likely pick up a lunar cousin by year’s end, and the codebase sits on GitHub at NASA-IMPACT, where custom fine-tunes for new instruments (Lunar Trailblazer, India’s Chandrayaan follow-ons, China’s Chang’e 7 imaging) can be trained without revisiting the data-integration problem. If the 22 percent lead against SwinV2-B holds once independent teams retrain on their own data, the rest of the 2020s will inherit a Moon that, for the first time, is machine-readable at planetary scale.

On 15 June 2026, at 15:24 CEST, a spacecraft 100 million kilometers from Earth received a single command through ESA’s Estrack deep-space antennas: stop firing. After almost eight years of nearly continuous thrust, the four QinetiQ T6 ion engines on the Mercury Transfer Module shut down for the last time. With that, the European/Japanese BepiColombo mission crossed an invisible line. From that moment, the spacecraft was on a ballistic trajectory, coasting with no propulsion toward the deepest gravity well in the inner Solar System. Mercury orbit insertion is scheduled for 21 November 2026 (ESA BepiColombo mission timeline, July 2026 update). What used to be a braking problem became a falling problem, and falling problems don’t have throttle knobs.

Mercury is the least explored planet in the inner Solar System, and the hardest one to reach without surrendering. It sits deep inside the Sun’s gravity well, moving at 47 km/s on average, more than three times Earth’s orbital speed. To fall into orbit around it rather than whip past it, a spacecraft must shed almost all of the kinetic energy it gained falling toward the Sun from Earth. NASA managed the trick with Mariner 10 in 1974 and MESSENGER in 2011. The Soviets never made it. The Japanese never tried. If BepiColombo succeeds, it will be only the third spacecraft in history to enter orbit around Mercury, and the first from Europe or Japan (ESA BepiColombo overview).

The mission also carries the heaviest scientific payload ever sent to Mercury, sixteen instruments distributed across two orbiters stacked on a propulsion bus. The trajectory they are flying is, by itself, an engineering demonstration: a multi-year, multi-flyby spiral through nine planetary encounters that uses gravity assists from Earth, Venus, and Mercury itself to bleed off energy without spending propellant.

BepiColombo launched from Kourou on 19 October 2018 aboard Ariane 5, lifting a 4,100-kilogram composite spacecraft into a series of looping escape orbits. The plan was, in spirit, simple: spiral inward. In practice, the MTM had to do most of the braking. Its four T6 ion thrusters, each producing only about 145 millinewtons of thrust (about the weight of a sheet of paper resting on a hand), fired almost continuously for years, accumulating more than 15 kilometers per second of total delta-v.

Then, in April 2024, the system lost a fight with itself. A short circuit in the MTM power conditioning unit reduced available thrust to roughly 90 percent of nominal. The original December 2025 arrival was no longer reachable. ESA’s flight dynamics team at ESOC in Darmstadt had to rebuild the entire interplanetary approach from the ground up, slicing flyby altitudes closer to Mercury by about 35 kilometers to recover the energy deficit through sharper gravity assists. The sixth and final Mercury flyby, on 8 January 2025, came in at just 295 kilometers above the night-side surface. Close enough that the spacecraft’s M-CAM monitoring cameras caught permanently shadowed craters at the north pole that may hold frozen water (ESA, January 2025 flyby report).

What happens between now and April 2027 is a four-stage separation sequence that ESA’s infographic captures cleanly. On 3 September 2026, still on approach, the MTM detaches and falls away, its job done and its xenon tanks empty. On 21 November 2026, a stack of two science orbiters fires a chemical-propellant braking burn and is captured into a high-altitude polar orbit. On 9–10 December 2026, ESA’s Mercury Planetary Orbiter (MPO) ejects the shielded Japanese Mio spacecraft onto its own elliptical polar trajectory. On 16 December 2026, MPO jettisons Mio’s protective sunshield (MOSIF). Over the next three months, MPO uses its own thrusters to lower itself to its working orbit of roughly 400 by 1,500 kilometers, arriving on 10 March 2027. Routine science begins on 6 April 2027 (ESA BepiColombo arrival timeline, July 2026).

The propulsion architecture is the story. The T6 ion thruster, built by QinetiQ Space in the UK, ionizes xenon and accelerates the ions through a grid biased at roughly 2,000 volts. Specific impulse runs around 4,300 seconds, about five times better than the best chemical bipropellant engine on the spacecraft. The trade is brutal: thrust is so low that BepiColombo had to spiral inward for eight years to make the maneuver work. The reward is propellant efficiency. The MTM carried only about 580 kilograms of xenon, and that modest tank delivered the equivalent delta-v of tens of metric tons of hydrazine.

Once the MTM separates, MPO has to handle the rest of the braking with its own chemical propulsion: a hydrazine-based system capable of producing about 1,000 newtons of total thrust across four 22-newton thrusters. To enter Mercury orbit, MPO must dissipate roughly 1.8 km/s of hyperbolic excess velocity in a series of maneuvers spread over weeks. Because Mercury’s gravity well is so deep, an error of even one meter per second in the insertion burn propagates into hundreds of kilometers of orbital drift by the end of the commissioning phase.

The two orbiters work at very different altitudes on purpose. Mio, the JAXA Mercury Magnetospheric Orbiter (formerly MMO), sits in a highly elliptical 400 by 12,000 kilometer orbit that swings it through the magnetosphere to sample solar wind interactions. MPO, in a tighter near-circular polar orbit, holds its instruments nadir-pointed at the surface. Eleven instruments on MPO and five on Mio together cover imaging (SIMBIO-SYS), X-ray fluorescence (MIXS), gamma-ray and neutron spectroscopy (MGNS), laser altimetry (BELA), magnetometry (MERMAG), and exospheric sampling (SERENA, PHEBUS). Temperatures range from about -180°C on the night side to +430°C on the day side, so most optical and infrared instruments sit behind thermal blankets and, where needed, gold-coated mirrors.

The water-in-shadow question is one of the cleaner open cases. Earth-based radar returns from Mercury’s polar craters are consistent with volatile ices buried under regolith. BepiColombo’s MGNS and MIXS instruments should, in principle, detect hydrogen directly through neutron moderation and detect elemental composition through characteristic X-ray emission. The permanently shadowed craters (Prokofiev, Kandinsky, Tolkien, and Gordimer at the north) were imaged by M-CAM 1 during the January 2025 flyby. Their rims just catch the sun, while their floors stay below 100 K year-round.

Three months from now, an interplanetary spacecraft that has spent almost a quarter of its life quietly thrusting through the inner Solar System is going to fall into the smallest planet’s gravity well. Whether the chemical insertion burn works depends on years of trajectory redesign, on the integrity of a 90-percent-power propulsion system that engineers coaxed across nine planetary encounters, and on a separation sequence that has to be executed cleanly four times in two months. If it works, BepiColombo will spend the next several Earth years mapping an airless, 430-degree-Celsius world whose permanently shadowed craters might, just maybe, hold enough water to matter for human exploration later this century.

If it doesn’t work, the spacecraft will whip around Mercury on a flyby trajectory, and the data already gathered by M-CAM during the six gravity assists will have to carry more of the mission’s scientific weight. As ESOC mission operators Ignacio Tanco and Nacho Clerigo have stressed in briefings, the arrival phase is not a single event but a six-month sequence of one-of-a-kind operations, each of which has to go right in the right order.

A fresh look at Mercury’s wrinkled face suggests the innermost planet is a tighter squeeze than anyone had measured. A team led by Gaku Nishiyama at the German Aerospace Center (DLR) reports on 10 September 2026 in Geophysical Research Letters that Mercury’s radius has shrunk by somewhere between 6.9 and 11.6 kilometres (about 4.3 to 7.2 miles) since the planet finished forming, up to 30% more shrinkage than the previous best estimate. The new number does not mean Mercury suddenly started cooling faster. It means the team has been able to count landforms that earlier surveys could not see, in places that earlier surveys did not look. (Phys.org, Space.com)

The cooling and contraction of a rocky planet are not abstract. They leave a written record on the surface. As the hot interior of a world like Mercury gives up heat to space, its iron core and silicate mantle both contract, and the cold brittle outer shell has to buckle and shorten to keep covering the smaller interior. On Mercury that shortening shows up as long curving cliffs called lobate scarps and as the smaller, bunched-up wrinkles that run across the volcanic plains. Each of those structures corresponds to a known amount of horizontal shortening, and the cumulative shortening of all of them tells you how much the planet’s radius has changed.

The problem is that some of those wrinkle ridges are buried under rubble. Mercury is the most heavily cratered surface in the Solar System besides the Moon, and each impact spreads a blanket of broken rock over a wide area. Where the surface is rough, the short scarps get masked by impact debris and are no longer visible from orbit. Earlier global surveys counted only the ridges they could see in smooth volcanic plains and used that visible population to estimate the global total. Nishiyama’s team did something different: they overlaid Mercury’s contraction-related fault map with a high-resolution surface roughness map, found that the roughest patches are systematically missing the scarps the smoother patches have, and used that correlation to estimate how much contraction the rough terrain is hiding.

The corrected total is 4.3 to 7.2 km of radial contraction, depending on how the unmapped regions are extrapolated. The upper end is about 30% larger than the previous widely cited figure, and the lower end is broadly consistent with the older estimate. Either way, the planet’s interior has gone through more total cooling than the existing record implied. That matters for what scientists think is going on inside Mercury today. A planet that has lost more heat over its lifetime is consistent with an iron core that has fewer light elements mixed into it, or with a higher starting temperature, or both. Lead author Nishiyama, in statements accompanying the paper release published via the American Geophysical Union’s media release, lays the implications out plainly: “More shrinking means Mercury could have a larger metal core, less light elements like silicon mixed into the metal core, or a higher starting temperature.” (Time)

The story behind the paper is at least as interesting as the headline number, because the data set the team used is mostly the same one that supported the earlier, smaller contraction estimate. NASA’s MESSENGER spacecraft orbited Mercury from 2011 until it ran out of fuel in 2015, and its Mercury Dual Imaging System returned the global monochrome and colour mosaics that almost every modern Mercury paper still leans on. The MESSENGER Laser Altimeter (MLA) produced a topographic map of the northern hemisphere but could not see all the way to the south. The team behind the new paper combined MESSENGER imagery with a recent roughness map derived from those same images, cross-checked with the older Byrne et al. 2014 estimate from Carnegie Institution and Smithsonian, and identified which kinds of terrain were systematically underrepresented in the historical contraction count.

The surface structures this work is about are dramatic at human scale even if they are subtle in orbit. The Wikipedia entry for Carnegie Rupes, a textbook lobate scarp in Mercury’s northern volcanic plains, describes an escarpment about 267 kilometres long with several kilometres of vertical relief. The image NASA’s MESSENGER spacecraft captured of that scarp cutting through the small crater Duccio (PIA19279, public domain, NASA / Johns Hopkins Applied Physics Laboratory / Carnegie Institution) is reproduced below. The curving bright ridge in the middle of the frame is the scarp itself. Each kilometre of vertical throw on a scarp like that one corresponds to a known amount of horizontal shortening, and the cumulative throw across thousands of such features is what gets converted, at the end, into a single radius change for the whole planet.

The instrument that will close the question is now on its way. ESA’s BepiColombo mission, launched in October 2018 and scheduled to enter orbit around Mercury in late 2026, dropped its cruising platform on 3 September 2026 after a flyby and is closing in for orbit insertion in November. It carries the BepiColombo Laser Altimeter (BELA), a German-Swiss instrument led by the same DLR group that authored the contraction paper. BELA is designed to map Mercury’s topography from the equator to the south pole at roughly ten times the vertical resolution of MESSENGER’s MLA, and Nishiyama is a member of the BepiColombo science team. He has been explicit about what the new data will buy. “Future data from laser altimetry on BepiColombo will collect more information on planetary contraction by measuring topography more precisely,” he said in the paper’s release materials, and the broader BepiColombo topographic record will let the team test whether the corrected 6.9 to 11.6 km figure is the final answer or itself an underestimate. (Phys.org)

The deeper question, beyond Mercury’s own radius, is what kind of planet Mercury is. It is the smallest planet in the Solar System by radius, the second-densest after Earth, and the only one with a fully liquid iron core that generates an active magnetic field. Its high density combined with its low mass implies an oversized metal core, and the surface contraction is one of the cleanest windows into how that core has been cooling for the past 4.5 billion years. A planet that has lost more heat than its visible faults suggest is a planet whose interior has been more efficient at giving up its primordial warmth. For models of planetary dynamos, of core composition, and of the thermal history of rocky worlds in general, that 30% revision is a meaningful step.

The lesson is also methodological. Planetary geologists working on Mars, the Moon, and the icy satellites of the outer planets face the same problem of counting only the faults that happen to be visible on the surface. Mercury’s case, where the obscuring layer is impact debris rather than vegetation or ice, is unusually clean. The team’s correlation between contraction feature density and surface roughness gives them, and anyone who follows them on other worlds, a concrete way to estimate how much of a planet’s tectonic record is hidden in plain sight. That kind of correction does not invalidate the earlier surveys; it shows that those surveys were working from the right kind of data and the right kind of reasoning, but with incomplete coverage.

By the end of November 2026, BELA will start returning the topographic map that turns Nishiyama’s 6.9 to 11.6 km range into a single number with much smaller error bars. If the BepiColombo topographic record confirms the upper end of the contraction range, the standard thermal-history models for Mercury will need to be revised in much the same way the contraction estimate itself just was. If it confirms the lower end, the new paper becomes a careful quantification of how much a single observational bias was understating a real measurement. Either way, the figure that planetary scientists will quote for Mercury’s shrinkage from now on is not the number they were quoting last week.

On the morning of July 10, 2026, a 63-meter rocket climbed off a brand-new launch pad on Hainan Island, deposited an unnamed payload into a low orbit, and then — six minutes later — fell back out of the sky and let a steel net on a ship catch it. The vehicle was the Long March 10B, designated CZ-10B in China’s rocket catalog, and the flight was its maiden voyage. Both halves of the mission succeeded, which made China only the second country, after the United States, to propulsively recover an orbital-class booster. The second thing that mattered was that the recovery hardware was nothing like SpaceX’s drone-ship landings: no deployable legs, no tower catch, no hover-slam. The booster carried four hooks and used them to latch onto a cross-shaped steel cable net strung across the deck of a ship called Linghangzhe. Chinese state media called it the world’s first successful wire-arrestment recovery of a carrier rocket (Xinhua, July 10, 2026). For an industry that has spent the last decade copying the Falcon 9 playbook, the choice to do something different is the story.

Reusability is no longer optional in the commercial launch market. Reusing a first stage is the single largest lever on launch cost, and the cost gap between an expendable booster and a flight-proven one is roughly a factor of two on Falcon 9 (SpaceX stated figures cited in SpaceNews, 2025). The industry consensus for the last decade has been that vertical propulsive landing is the only path: relight a few engines, fall ballistically through the upper atmosphere, use grid fins to keep the stack on course, throttle down, set down on legs. China has just demonstrated a working alternative, which suggests the engineers at CALT made a deliberate trade.

The arguments are mostly about mass. Landing legs, deployment hinges, hydraulic dampers, the extra beef in the engine bay to absorb landing loads, and the reserve propellant for a hoverslam all eat into payload. Falcon 9 sets aside on the order of 6% of its first-stage propellant load for a drone-ship landing and roughly 30% for a return-to-launch-site profile (SpaceX public estimates; Flight Club reconstructions). A booster that does not have to carry legs can carry more propellant or more payload. CALT’s published commentary frames the net as a way to “shift the structural burden from the vehicle to the sea-based platform.” The ship, not the rocket, holds the arrestment load.

Geography helps. Wenchang sits on the east coast of Hainan, and the standard launch azimuths send boosters downrange over the South China Sea. A recovery vessel can sit offshore and the booster has a flat ocean to fall into, with no populated land downrange and no steep boostback burn. Linghangzhe is the first ship purpose-built for this. It is 144 meters long, 50 meters wide, displaces 25,000 tonnes at full load, has DP2 dynamic positioning, and carries the arrestment net on a steel frame taller than a six-story building (CASC, July 2026). LiDAR units at the corners of the frame feed an automated capture loop. No one on deck is steering.

Liftoff came at 12:15 p.m. Beijing time on July 10. Seven YF-100K engines, each rated at about 1,250 kilonewtons of sea-level thrust, lit together to push the 760-tonne stack off the pad (Xinhua, July 10, 2026; CALT engine specifications, 2024). The seven-engine cluster is notable: this is the LOX-kerosene YF-100 in its latest pump-back-swing variant, with the thrust chamber gimbaled while the turbopumps stay fixed. That rearrangement is what lets CALT fit seven of them into a 5-meter core. About 150 seconds into the flight, at roughly 100 kilometers of altitude, the first stage separated and the second stage’s single YF-219 methane-LOX engine took over to deliver the satellite. The first stage then began a six-minute return: unpowered coast to apogee, attitude turn using cold-gas thrusters, a reentry burn to scrub velocity, aerodynamic deceleration with titanium grid fins steering, and then a final descent burn that cut off just before the hooks met the cables.

The flight was an operational test of the recovery system and engines. The booster caught in the net has to be inspected, refurbished, and flown again. CASC said it intends to refly this exact first stage before the end of 2026. If that happens, China will have skipped the long Falcon 9 march from 2013 to 2017 that SpaceX needed to land and re-fly the same booster.

The CZ-10B is the middle child of a three-rocket family. The crew-rated CZ-10A will launch astronauts in the Mengzhou spacecraft to the Tiangong station. The heavy-lift CZ-10 — three cores bundled together, 92.5 meters tall, 21 engines — will eventually carry the Mengzhou and Lanyue lander to the Moon with roughly 27 tonnes on a trans-lunar trajectory. The 10B is the test bed: same diameter, same first-stage engines, with a methane upper stage optimized for commercial missions rather than crew.

The trick that makes the architecture work is the propulsion. Specific impulse is the engine-design figure of merit: how many seconds of thrust one kilogram of propellant produces. The YF-100K runs an oxidizer-rich staged combustion cycle at a chamber pressure near 18 megapascals, with a sea-level specific impulse around 302 seconds and a vacuum number near 338 seconds (CALT YF-100K specification sheet, 2024). For comparison, the Merlin 1D on Falcon 9 produces about 282 seconds at sea level. Higher Isp means less propellant mass buys a given delta-v, which the rocket equation makes unforgiving:

Δv = v_e * ln(m_0 / m_f)

Every kilogram of dead mass stripped off the first stage translates, through that logarithm, into more kilograms of payload at orbit.

Methane on the upper stage is the second bet. LOX-methane engines run cooler, leave less coking residue, and are easier to turn around between flights than kerolox. That matters more for an upper stage than for a booster, because the upper stage burns for several minutes and accumulates more thermal soak-back. The YF-219 has been a long-running CALT development, and the 10B flight is its first orbital demonstration.

The recovery system is the engineering story most worth lingering on. The cross-shaped net is built from high-strength steel cables with hydraulic buffers at the four corners that absorb kinetic energy. Four hooks on the booster engage the cables at low speed (the descent burn throttles the booster down to a few meters per second before capture). Once the cables take the load, hydraulic winches stabilize the booster against wind and waves, and an automated locking platform clamps it for the trip back to port. The sequence is unmanned: LiDAR tracks the booster, the position-keeping system holds the ship steady, and the capture happens in software.

The CZ-10B was a single Friday-morning launch, but the chain it pulls on is longer. The 10A will use the same first stage (minus the net hardware) to put Mengzhou crews into low orbit; the heavy CZ-10 will use it again, three at a time, to support China’s crewed lunar program before 2030. If CASC can fly, catch, inspect, and re-fly a first stage in 2026, the company buys a launch cadence for the Guowang megaconstellation and Tiangong that costs a fraction of an expendable fleet. The United States spent the better part of a decade proving booster reuse was physically possible. China has, in a single afternoon, suggested there is more than one way.

The lesson for the rest of the industry is straightforward. Reusability is the destination, not the route. The path that SpaceX blazed — vertical landing on legs — is one road. A net on a ship is another. Both work. Both have trade-offs. The only thing that no longer works is to keep throwing away 70% of a rocket.

A 51.7-metre rocket lifted off in the dark hours of Friday morning at Sriharikota, on India’s southeast coast, and put a roughly 2,367-kilogram Earth-observation satellite into a Sub-Geosynchronous Transfer Orbit. ISRO’s GSLV-F17 mission was scheduled for 2:55 am IST on 4 September 2026 from the Second Launch Pad at Satish Dhawan Space Centre (ISRO mission page), and the launch itself went according to plan. What made it more than routine was that EOS-05 is the country’s first imaging satellite to operate from a geosynchronous perch, and it flew on a launcher that India’s space agency has spent more than two decades trying to make reliable.

The Geostationary Imaging Satellite programme goes back nearly a decade, to an earlier concept called GISAT-1 that was meant to give India a continuous, fixed-eye view of the subcontinent from 36,000 km up. The original GISAT-1 launch in 2020 was called off because of a technical issue with the cryogenic stage, and a second attempt the following year failed to ignite the cryogenic engine in flight. That delay had already pushed India’s geosynchronous imaging ambition by roughly five years before EOS-05 picked up the work as a direct replacement, with design updates informed by the failure analysis (TechTimes coverage). This time the cryogenic stage behaved. The third stage, designated CUS15, burned for about 814 seconds with the indigenous CE-7.5 engine, which ISRO developed to replace a Russian KVD-1 design after the Missile Technology Control Regime complications of the early 2000s (CE-7.5 Wikipedia entry). The launch matters because India’s earth-observation fleet has until now operated almost entirely from low Earth orbit, where a satellite passes over any given patch of ground for only a few minutes per orbit. From 36,000 km up, EOS-05 will sit over a fixed longitude and stare down at the same region continuously, day and night, gathering visible, near-infrared and shortwave-infrared imagery across roughly the whole visible disc (Wikipedia: EOS-05). That is a qualitatively different tool. A polar-orbiting imager gives you high spatial resolution at the cost of infrequent revisits; a geostationary imager gives you continuous revisits at coarser spatial resolution. EOS-05 is aimed at the second regime, with a 700 mm Ritchey-Chrétien optical telescope tuned for agricultural monitoring, cyclone tracking, and rapid disaster mapping over India, its 7,500-kilometre coastline and the surrounding maritime zones (Times of India mission summary).

This was the 19th flight of the GSLV and only the fifth to use the Mk II configuration with the indigenous cryogenic upper stage, which is the part of the rocket that has historically given ISRO the most trouble. The Mk II stands 51.7 metres tall, weighs 420.5 tonnes at lift-off, and uses a stack of three stages wrapped by a four-metre composite ogive fairing. The first stage is the S139 solid core with four L40H liquid strap-ons clustered around it; the second is the GL40HT liquid-propellant stage; and the third is the cryogenic CUS15 (ISRO mission brochure summary). That third stage is what makes GSLV comparable to other medium-lift geosynchronous launchers in the world: liquid hydrogen and liquid oxygen are notoriously hard to handle because they sit at around -253 °C, which is why cryogenic upper-stage development is often the single largest non-recurring engineering cost for any space agency building a geosynchronous launcher.

The mission was also a deliberate break from a difficult 18 months for the Indian launch programme. PSLV, the workhorse polar-orbit rocket, suffered consecutive failures in May 2025 and again in January 2026, and the investigations into both kept most of ISRO’s launch manifest on the ground (Indian Express on the PSLV investigations). NDTV called GSLV the “Naughty Boy” of the Indian fleet in its post-launch coverage, an affectionate reference to the launcher’s long history of partial failures and stand-downs (NDTV: How ISRO’s ‘Naughty Boy’ Rocket Delivered A Big Win For India). Putting EOS-05 into a precise Sub-GTO is exactly the kind of success that rebuilds confidence launch by launch; ISRO Chairman V. Narayanan confirmed after the mission that the cryogenic upper stage had performed as designed and that the satellite’s Liquid Apogee Motor would now do the work of raising the orbit to its final geosynchronous station.

The CE-7.5 engine that powered that critical third stage is itself the product of a long-running technology development programme. It is a regeneratively cooled, fuel-rich, staged-combustion-cycle engine, which is the same thermodynamic cycle used by the Space Shuttle’s RS-25 main engines (CE-7.5 details). The engine is about 2.14 m long and 1.56 m in diameter, runs at variable thrust, and has a specific impulse of around 454 seconds in vacuum. Staged combustion is the most efficient chemical cycle in practical use, because the propellants are partially burned in a preburner before the main chamber, which extracts more energy from the same mass of fuel. The trade-off is that the engine is harder to build and to test; the preburner runs at extremely high pressures and temperatures, and any leak of the hydrogen-rich preburner gas into the wrong part of the engine can damage it. That is why ISRO’s early GSLV flights used the Russian-supplied KVD-1 instead, which uses a simpler gas-generator cycle at the cost of lower efficiency. By replacing the Russian engine with a domestic one, ISRO removed a long-standing supply-chain risk and aligned the upper stage with India’s existing manufacturing base. It also brought the GSLV’s payload capacity to geosynchronous transfer orbit up to roughly 2,500 kg, which is enough for EOS-05 and for the heavier navigation and communications payloads ISRO has planned for the rest of the decade. The fact that EOS-05 was the first GEO imaging payload to fly on it is not coincidental; ISRO held the satellite back until it had a launcher it trusted enough to put an irreplaceable asset on.

For the broader commercial and strategic picture, EOS-05 also closes a gap that India’s civilian and military users have lived with for years. The Indian National Satellite System’s INSAT fleet has carried transponders for communications and weather imaging for decades, but it has not had a dedicated hyperspectral imager staring down from geostationary orbit. Hyperspectral means the imager splits incoming light into many narrow bands, which lets it distinguish, say, healthy crops from drought-stressed ones in ways that a conventional three-band imager cannot. Continuous coverage from geosynchronous orbit, in turn, lets disaster managers watch a cyclone’s eye wall reorganise in near-real-time or track the spread of a flood across a state hour by hour. The payload does not replace India’s existing low-Earth-orbit earth-observation satellites, which still do most of the heavy lifting on high-resolution mapping. It complements them.

The next test will be whether the satellite reaches its final station and powers up its instruments cleanly. ISRO’s GSLV-F15 in January 2025 placed a navigation satellite into GTO correctly but the satellite’s own Liquid Apogee Motor failed due to an oxidiser valve malfunction, leaving that spacecraft stuck in a transfer orbit (NDTV: What Back-to-Back PSLV Failure Means For ISRO). EOS-05 has the benefit of a fresh cryogenic upper-stage performance record and of a launch-to-orbit precision that should let its own propulsion system do less work. If all goes well, the satellite should be operational at its assigned geostationary slot above the equator within a few months, and India’s weather, agriculture, and disaster-management agencies will have their first continuous watch over the subcontinent.

On Thursday afternoon at 14:00 CEST, a stack of four spacecraft modules that left Earth on an Ariane 5 in October 2018 will, very briefly, become three and then two. ESA’s Mercury Transfer Module is scheduled to detach from BepiColombo’s two science orbiters as the mission’s controllers at the European Space Operations Centre in Darmstadt look on, with the first signal from the freed composite expected back on Earth by 15:53 CEST if everything goes to plan (ESA, 27 August 2026). After eight years and nine gravity assists, the part of the spacecraft that made the trip possible is no longer needed, and the part that has to do the science is, ungainly and for the first time, on its own.

Getting to Mercury is the kind of orbital mechanics problem where the obvious solution is forbidden. Falling toward the Sun does not slow a spacecraft down; it speeds it up, pulled by solar gravity, so the usual chemical-brake trick used at Mars is not enough to be captured into orbit around the smallest planet. BepiColombo has to match Mercury’s velocity almost exactly, and that takes far more energy than reaching even the distant gas giants. The mission’s answer was to spend years bleeding speed against an ion engine instead of staging a single dramatic burn, and the upcoming separation is the moment the engine is, finally, switched off and thrown away. As ESA’s mission manager Santa Martinez put it to Euronews on 1 September, “Mercury is a challenging destination. It’s very difficult to reach, it’s very difficult to operate there.”

Why a separation matters this much, and why the next two months are unusually tense, comes down to what changes about the spacecraft on Thursday. For the cruise the four modules sat one on top of the other inside a tight stack, with their instruments and cameras physically blocked from space by the boxy MTM at the base. Once the MTM floats off, the upper composite begins an entirely new mission with an entirely new thermal and radiation budget. Solar flux near Mercury is roughly ten times what it is near Earth, and the MTM had been there partly to shield the orbiters during the inner solar system transit. The flight director, Ignacio Tanco, has described the moment to Euronews as working “with a very hot pizza oven running right on your back” and “on a knife edge,” because the MPO’s solar panels have to be tilted within fractions of a degree of the Sun. Push them slightly away and the spacecraft starts losing power; angle them slightly closer and they start to bake. There is no second chance to find the right pointing before the thermal balance tips.

The spacecraft itself carries a long memory. BepiColombo launched on Ariane flight VA245 from Kourou on 20 October 2018 and reached escape velocity the same day, then spent almost exactly eight years climbing and falling through the inner solar system using four QinetiQ T6 ion thrusters on the MTM, with a maximum combined thrust of about 290 millinewtons, the most powerful ion engine array ever operated in space. The cruise used one Earth flyby in April 2020, two Venus flybys in October 2020 and August 2021, and six Mercury flybys between October 2021 and January 2025, the last of which imaged the permanently shadowed north polar craters Prokofiev, Kandinsky, Tolkien and Gordimer. The mission’s overall cost is estimated at 1.65 billion euros, and the arrival has already slipped about eleven months from its original December 2025 date because of a power-related anomaly that limited the available thrust before the fourth Mercury flyby in September 2024. On 15 June 2026 at 13:24 UTC the MTM’s solar electric propulsion was switched off for good, and from then on the mission is ballistic: BepiColombo is, in engineering language, “weakly captured” by Mercury’s gravity and following a free-falling trajectory through space.

What happens after 3 September is a sequence of three distinct steps rather than one event. From the separation through late November, the MPO and Mio remain mechanically and electrically joined, with the Magnetospheric Orbiter Sunshield and Interface (MOSIF) still attached, and the two orbiters fly free of the propulsion module for the first time in their lives. Then, between late November and early December 2026, a series of chemical thruster burns raises the orbit’s apocentre and lets Mercury’s gravity do the rest, putting the composite into polar orbit. Once captured, MPO and Mio separate from each other and head to their distinct orbits, MPO in a lower polar orbit for surface and interior science and Mio in a higher polar orbit at 590 by 11,640 kilometres for magnetospheric work. Engineers then spend roughly four months commissioning instruments on both orbiters, and only in April 2027 does the dedicated science campaign begin. The waiting list, when it eventually opens, is unusually long. Mercury’s iron core makes up roughly sixty percent of the planet’s volume, and the mission’s principal question is why a planet that small should pack such a dense metallic interior, and what that interior is doing to the unexpectedly long-lived magnetic field that makes Mercury (along with Earth) the odd rocky planet out. The two orbiters will “taste” the planet’s tenuous exosphere of hydrogen, helium, oxygen, sodium, potassium and calcium at the same time that MPO maps the surface, measuring surface temperatures down to a few hundred metres per pixel and sounding the crust with the RIME ice-penetrating radar that was deployed once in 2023 and has waited seven years to be used on Mercury. Separation on Thursday is the first of a small handful of moments that decide whether any of that science happens, and it is the one the ESA team can already see clearly. After MTM separation, mission controllers listen for the post-separation acquisition of signal and then go through the first health checks. If the composite is in the expected attitude, the high-gain antenna locking on Earth, and the temperatures inside the expected range, MPO and Mio spend the next three months coasting together in what ESA calls a Mercury approach configuration, conservatively waiting for the November capture window. The MTM, the workhorse that brought them there, has no attitude control after separation, so ESA’s flight dynamics manager Frank Budnik has been clear that “you cannot control it, you cannot manoeuvre it,” and the module will begin to tumble away in its own orbit past Mercury rather than around it. After eight years, the most powerful ion engine array ever flown is, for a few million kilometres at least, just a derelict tumbling through interplanetary space with the Sun in its face and no way to turn around.