OrbitalHub

On the evening of July 5, 2026, a 600-kilogram box of solar panels and instruments zipped past a peanut-shaped rock 450 meters across at roughly five kilometers per second. The flyby lasted about as long as the time it took you to read this sentence. The spacecraft was JAXA’s Hayabusa2, on a deliberate dash past near-Earth asteroid 98943 Torifune, also known by its provisional designation 2001 CC21. By the time anyone on Earth saw a single pixel of the encounter, the spacecraft was already several thousand kilometers beyond the asteroid, never to swing back (JAXA, July 6, 2026).

That brevity is the point. Torifune is the first of three asteroid encounters the extended Hayabusa2 mission is supposed to make before the probe reaches its true final target, the tiny 11-meter asteroid 1998 KY26, in 2031. The spacecraft is a sample-return craft that has already done its main job. Every extra maneuver now is rehearsal for an encounter that will be the most technically punishing of the mission (The Planetary Society, July 7, 2026).

Two reasons. First, Torifune is a contact binary, a body made of two lobes stuck together like a snowman, and those shapes are common in the small-body population but hard to study up close. Ground-based radar and adaptive-optics images suggested it was elongated; until last week no spacecraft had ever confirmed the structure with its own camera.

Second, the flyby is a deliberate dress rehearsal for planetary defense. JAXA wanted to demonstrate the kind of precision terminal guidance a kinetic impactor mission like NASA’s DART would need, where the spacecraft picks a small, faint, irregularly shaped target and threads a narrow approach corridor without hitting it. Torifune is the right kind of object: small, dim, and only clearly resolved a few days before closest approach. The mission team could not plot the final approach trajectory until the very end, because Torifune is faint enough that its exact orbit was uncertain until then (The Planetary Society, July 7, 2026).

If you are trying to design a spacecraft that bumps a city-block-sized rock off course, this is the kind of practice you want.

Hayabusa2 launched in December 2014 on an H-IIA rocket and arrived at asteroid Ryugu in mid-2018. Over the next eighteen months it dropped a pair of tiny hopping rovers, lowered the German-French MASCOT lander onto the surface, fired a 2-kilogram copper projectile into Ryugu to expose fresh sub-surface material, and scooped up samples. The return capsule landed at Woomera, Australia, in December 2020. Inside, scientists found water-bearing minerals, organic molecules, and grains that had escaped billions of years of space weathering. The samples are still being analyzed and are among the cleanest primitive materials anyone has yet returned from a carbon-rich near-Earth asteroid (Sci.News, July 8, 2026).

Most sample-return missions end there. Hayabusa2 did not. After releasing its sample capsule, the main spacecraft fired its ion engines and began a new trajectory. The extended mission has two goals: a high-speed flyby of Torifune in 2026, two Earth gravity assists in 2027 and 2028, and a final rendezvous with 1998 KY26 in 2031 (ISAS, mission page).

The Torifune encounter happened on July 5 at 18:30 JST, give or take one second. The spacecraft passed about 800 meters from the asteroid’s center, well outside the body but close enough that its three imaging instruments could resolve surface detail. ONC-T, the visible-light Optical Navigation Camera, started tracking Torifune on June 20 and continued to use the asteroid as an optical navigation reference right up to closest approach. About an hour before the flyby, the spacecraft also switched on NIRS3, a near-infrared spectrometer that hunts for water and hydroxyl signatures; TIR, a thermal infrared imager; and LIDAR, which bounces a laser off the surface to measure distance (JAXA, July 6, 2026).

Closest approach came and went. Hayabusa2 could not look back. JAXA confirmed the spacecraft was healthy at 18:35 JST, five minutes after the flyby, and the first images and thermal maps arrived within a day (Sky & Telescope, July 6, 2026).

The visible-light image showed what ground-based observers had suspected. Torifune is two distinct lobes joined at a narrow neck, the textbook shape of a contact binary formed when two smaller asteroids collided gently and stuck rather than bouncing apart. At about 450 meters across, it is in the same size class as the near-Earth asteroid Apophis and roughly an order of magnitude larger than the object that exploded over Chelyabinsk in 2013 (The Planetary Society, July 7, 2026).

The naming fits. JAXA chose “Torifune” through a public contest; it is short for Ame-no-Torifune, a Japanese deity associated with safe, steady, high-speed travel. You can see why the mission team liked it.

A flyby at five kilometers per second does not leave much margin for error. A small miscalculation in the closest-approach distance translates into several hundred meters of horizontal miss, easily enough to fly by the wrong side of the asteroid. There is no second chance. Hayabusa2 had to arrive at a precise point in space at a precise time, then keep going.

The spacecraft used a hybrid optical and radio navigation scheme. ONC-T took repeated images of Torifune against the background star field, and the team on the ground measured the asteroid’s apparent position to refine the predicted encounter point. Combined with deep-space network Doppler and ranging data, this let mission control update the trajectory within the last few days before closest approach. JAXA has not published the final navigation error budget, but the fact that ONC-T returned sharply focused images from 800 meters suggests the geometry worked (JAXA, July 6, 2026).

The thermal imager, TIR, observed Torifune in the 8 to 12 micrometer band, the thermal infrared where a sunlit asteroid radiates most of its heat. The brightness temperature, divided by the solar illumination, lets you back out the surface’s thermal inertia, a property set by regolith grain size, packing, and whether the surface is bare rock or blanketed in dust. For a small contact binary, the thermal data is the only way to learn what the surface is made of, because visible and near-infrared spectroscopy during a flyby captures too little reflected sunlight to break out mineral absorption signatures cleanly.

LIDAR is simpler but just as important. It rangefinds the surface with a 1.064-micrometer pulsed YAG laser, accurate from 30 meters out to 25 kilometers. During the Torifune encounter it produced a continuous altitude profile in the final minutes before closest approach, doubling as both a navigation input and a shape model. Combined with the ONC-T silhouette from a known viewing angle, the LIDAR track pins down the asteroid’s three-dimensional shape far better than the optical images alone (JAXA, July 6, 2026).

The instrument suite is what the spacecraft will carry into the 1998 KY26 encounter in 2031. KY26 is a different problem. At roughly 11 meters across, it is small enough that Hayabusa2 will not simply fly past; the extended mission plan calls for an actual orbit insertion and an attempted landing. The asteroid’s rapid rotation period, on the order of a few minutes, will make that landing extremely difficult, and the extended mission team is using Torifune to validate the optical and thermal mapping pipeline they will need for the harder job.

Hayabusa2 has now visited three asteroids across two missions. The first primary mission gave the world its best sample of a carbon-rich near-Earth asteroid. The extended mission is a chain of progressively harder technical demonstrations, and the Torifune flyby is the one with the lowest stakes but the cleanest test of the navigation approach JAXA wants to use in 2031. More data will downlink over the coming weeks.

Two years from now, the spacecraft will swing past Earth for a gravity assist. Three years after that, in 2031, it will arrive at a rock smaller than a house and try to land on it. Last week’s flyby is the first of the milestones the extended mission still owes the public. JAXA hit the geometry and returned a clean, recognizable image. The data will inform how we eventually deflect a real asteroid.

A Falcon 9 lifted off from Space Launch Complex 4E at Vandenberg Space Force Base at 11:32 a.m. PT on Thursday, October 1, 2026, carrying 130 satellites and hosted payloads into a sun-synchronous orbit on SpaceX’s eighteenth dedicated rideshare mission. It was one of three Falcon-family flights SpaceX flew that day, sandwiched between the Crew-13 astronaut launch from Florida in the morning and a National Reconnaissance Office Falcon Heavy from the Cape that night, and the size of the manifest made it the most diverse single launch of the year so far.

Transporter missions are SpaceX’s standing answer to the question “where do smallsats go?” The SmallSat Rideshare Program sells slots on quarterly Falcon 9 flights to SSO for as little as $350,000 per payload for up to 50 kg, with the cadence locked far enough in advance that customers can plan a satellite around a specific quarter instead of buying a dedicated rocket. The eighteenth such mission brought the program’s all-time tally past 1,800 payloads delivered to orbit, per SpaceX’s official mission page, a number that captures how the commercial smallsat industry is built as much on shared buses and shared launches as on the satellites themselves.

The first stage that flew Transporter-18 had already done this twice. It was the booster’s twenty-fifth flight, with a previous log that included NROL-126, Transporter-12, SPHEREx, NROL-57, USSF.366, and fourteen Starlink flights. Seven minutes and thirty-two seconds after liftoff it touched down on Landing Zone 4 at Vandenberg, returning to the same pad complex from which it had departed a few minutes earlier, in the kind of same-pad turnaround that has become routine for the West Coast booster fleet. Residents of Santa Barbara, San Luis Obispo, and Ventura counties may have heard one or more sonic booms as the booster came back through the denser lower atmosphere.

Inside the fairing, the manifest ran from cubesats to hosted payloads to two reentry vehicles and four orbital transfer vehicles that will deploy 41 of the 130 payloads at a later time. That architecture is the quiet structural innovation of recent rideshare flights: instead of forcing every customer into the same orbital plane on the same timeline, SpaceX has absorbed the role of mission manager. The OTV providers (D-Orbit, Exolaunch, Maverick Space Systems and others) now offer “deploy on a different day, at a different altitude” as a service to the payload builders. The Falco trade Via Satellite’s coverage of Transporter-18 describes a manifest where, for example, Tomorrow.io’s two weather satellites and Satellogic’s four optical Earth observation satellites ride along on the same Falcon 9 but can be dropped into different local times of day on station by their respective OTVs.

Several of the 130 payloads were firsts for their operators, which is what made Transporter-18 worth covering as a single story rather than as a pile of separate launches. Planet Labs flew the prototype satellite for Project Suncatcher, the joint effort with Google to test how Google machine-learning models and Google Tensor Processing Units behave when run in orbit rather than on the ground. Planet was the largest single customer on the flight, and in addition to the Suncatcher prototype it also flew the second Tanager methane-tracking satellite (Tanager being the methane mapper Planet developed with the CarbonMapper coalition), plus 18 SuperDove imagers manifested by SEOPS. StarCatcher flew Protostar, billed as a first-of-its-kind power-beaming node designed to demonstrate transferring electrical power between two free-flying spacecraft using optical links; a separate payload from Cowboy Space called Reason-1 is testing high-power laser energy transfer from low Earth orbit down to a ground receiver. Two demonstrations of moving watts around with light, on the same rocket, a few hours after the first ground-to-orbit ML inference test.

Starfish Space sent its first Otter servicing vehicle. It will run a NASA mission under the Small Spacecraft Propulsion and Inspection Capability acronym (SSPICY), and the flight is the company’s first operational satellite. CesiumAstro sent Element, its first full satellite built on the company’s own phased-array antennas, software-defined radios and onboard processors; Unseenlabs sent a pair of next-generation radio-frequency monitoring satellites, one to add capacity to its first-generation constellation and a BRO (BRO-32) to expand RF detection coverage across sea, land and space. Dhruva Space integrated LEAP-2, its second commercial mission, hosting multiple in-orbit demonstration payloads including units from Sodern ArianeGroup and the India-based Satleo Labs TAPAS-1.

The mission also carried two reentry vehicles from Varda Space Industries, continuing the company’s program of returning pharmaceuticals manufactured in microgravity, and the eight-satellite IRIDE Eaglet II cluster for Italy’s national Earth observation constellation. Constellation refreshes dominated the rest of the manifest: four ICEYE synthetic-aperture radar satellites, four Satellogic optical satellites, five Sateliot IoT nodes, two GHGSat methane monitors manifested by SFL Missions, and Tomorrow.io’s two weather satellites. According to the SpaceWatch.Global launch report and the NASASpaceFlight mission preview, the deployment sequence began at T+54 minutes and 22 seconds with the NASA-manifested Sprite cubesat and ran for the next hour through 89 individually sequenced deployments, with SpaceX confirming the deployment sequence had concluded at 3:41 p.m. ET.

What makes Transporter-18 worth thinking about beyond the manifest is the layering of three different kinds of space activity on the same rocket. There are constellation refreshes, which are now a steady, repetitive part of the smallsat industry the way cellular tower upgrades are a steady part of the telecoms industry. There are first-of-kind demonstrators: the Project Suncatcher ML-in-orbit prototype, the StarCatcher and Cowboy Space optical power-beaming tests, the Starfish Otter servicing mission. Their operational lifetimes may or may not turn into recurring revenue depending on what the data shows. And there are hosted payloads and secondary slots that let a university lab or a small country’s space agency get a satellite into orbit for the price of a rideshare ticket, which is what turns a quarterly launch cadence into infrastructure for everyone who builds the future of space hardware.

The next Transporter flight will not be far behind. Rideshare customers now plan their satellites around a known quarterly cadence from Vandenberg and the Cape, and a mission profile with 130 payloads, an OTV captain, and a twenty-fifth-flight booster is no longer an exception to the rule. It is what a normal Falcon 9 launch looks like in late 2026, and the question worth watching is not whether the next rideshare can carry this much, because the manifest will only grow. It is whether the long tail of secondary payloads can keep its quality bar as the cadence accelerates, and whether the experimental demos like Suncatcher and Protostar turn into recurring programs that share the same quarterly rocket on its own.

At 04:15 UTC on July 10, 2026, a 70-meter Chinese rocket climbed off the pad at the Wenchang Commercial Space Launch Site on Hainan island, lit seven kerosene engines, and headed southeast over the South China Sea. Roughly ten minutes later, the upper stage was still climbing toward orbit, and the booster had already turned around, fallen back through the sensible atmosphere, and snagged itself on a grid of tensioned steel cables strung across a ship called Linghangzhe. The booster came to rest hanging in mid-air over the deck, engines off, plumes of evaporating kerosene streaming off the warm tank. CASC, the state-owned China Aerospace Science and Technology Corporation, called the mission a “complete success.” It is the first time any Chinese rocket has been recovered after reaching orbit, and the first time any operator anywhere has used a net to do it.

Reusability is the lever that turned SpaceX from an upstart into the dominant launch provider of the 2020s. A Falcon 9 booster that lands is a booster that flies again, and that one engineering change drives launch prices down by roughly an order of magnitude. China has watched that happen from the outside. Until last Friday, no Chinese rocket had ever returned to Earth intact after reaching orbit, which put a hard ceiling on how fast Beijing could build megaconstellations, crewed lunar hardware, or any of the other things reusable boosters make cheap.

The Long March 10B test changes that. It is also the third booster-recovery club to form on the planet, after SpaceX (Falcon 9 in 2015, Starship in 2024) and Blue Origin (New Glenn in late 2025). The catch is that China got there using a method neither of the other two has flown: catching the descending booster on a net, mid-air, instead of either landing it on legs or snatching it out of the air with a launch-tower arm like Starship does. If that approach scales, it shifts the trade-off between payload mass and reusability in a way the rest of the industry will have to think about.

Friday’s flight was the maiden launch of the Long March 10B, a two-stage medium-lift rocket built by the China Academy of Launch Vehicle Technology (CALT), a subsidiary of CASC. The booster is 5 meters in diameter and runs seven YF-100K engines burning kerosene and liquid oxygen. Unlike the older Long March workhorses, its upper stage burns methane and liquid oxygen, the same propellant combination SpaceX’s Raptor uses. The rocket was designed from the start to be recoverable.

The recovery hardware lives mostly on the ship, not the rocket. Linghangzhe (Navigator) carries a four-legged frame over which a tensioned cable grid is stretched. As the booster descends, four titanium grid fins steer it through the hypersonic and supersonic phases of reentry, the same aerodynamic trick Falcon 9 uses. Near the surface, the booster sheds its grid fins and extends small hooks that snag the cable grid. The cables carry the booster’s weight and arrest its downward velocity, and the engines shut down as it comes to rest suspended over the deck. CASC says the flight validated “multiple engine restarts with high-altitude ignition, high-precision navigation and control, and the first capture and recovery using a net system on a sea-based platform.”

While the booster was being caught, the upper stage kept climbing. About ten minutes after launch, it delivered a payload designated CX-26, built by China Satellite Network Group, into a roughly 800-kilometer low Earth orbit. Chinese officials have not publicly described the satellite’s mission, which is consistent with most classified or pre-operational Chinese payloads. SpaceNews reported that the payload may in fact be TJS-26A (Tongxin Jishu Shiyan-26A), a classified communications technology test satellite, but that identification has not been confirmed by CASC.

The flight is a stepping stone in a wider Long March 10 program. The Long March 10A, which shares the same first stage as the 10B, is intended as China’s next crewed launcher for low Earth orbit and is meant to evolve out of the older CZ-2F and CZ-7 families. A larger three-core Long March 10 will be the rocket that flies the Mengzhou crewed spacecraft toward the Moon. Data from the 10B’s recovery, and from its methane upper stage, feeds directly into both. China launched a scaled-down CZ-10A first stage in February 2026 to test the Mengzhou capsule’s launch abort system, and that booster performed a controlled splashdown about 200 meters from its recovery platform in the South China Sea. Friday’s net catch is the next step on that ladder.

Two design choices drive the net approach.

The first is mass. A Falcon 9 booster carries four deployable landing legs and four larger grid fins, plus the hardware to lock each leg into place. Mass saved is mass that can be payload. The 10B trades legs for hooks, which weigh very little, and offloads the heavy catch structure (frame, cables, winches, the frame’s support legs on the ship) to the boat. CASC’s own statements imply this is the main reason a 10B in reusable mode can still deliver 16 metric tons to a 200-kilometer low orbit, versus a single-use rocket of similar class.

The second is tolerance. A ship in the open ocean moves with the swells. A booster falling at terminal velocity has seconds to hit a moving target. The cable grid spans most of the frame, so the booster does not need to land on a small pad or balance on a single tower catchpoint the way Starship does at Boca Chica. The grid also absorbs a lot of energy, so the booster does not need a perfect engine-cutoff timing. That should, in principle, let the recovery team operate in worse weather and with less margin than leg-style landings.

The catch is not free. The booster has to be soft enough not to tear through the cables on contact, which constrains structural reinforcement and probably caps the size of recoverable boosters. The 10B is medium-lift. Falcon Heavy has three boosters, and the Super Heavy booster on Starship is far larger and heavier. Whether the net idea scales to heavy-lift, or only works for medium-lift, is an open question. CASC has not said whether the crew-rated 10A or the three-core Moon rocket will also use a net or fall back to legs.

The wider context: at least two private Chinese companies have already tried similar landings. LandSpace flew its Zhuque-3 and attempted a downrange recovery in 2025, with mixed results. Space Pioneer and Galactic Energy are working on comparable vehicles. Whether the state-led 10B flight pushes the commercial sector toward a common architecture, or sends it looking for different solutions, is one of the things to watch over the next year.

The Friday flight did not change China’s space program in one stroke. It changed the launch industry in a small, concrete way: China now has a working, reusable orbital booster, and it has a working version of an idea nobody else had flown. The next 10B flights, expected later in 2026, will tell operators whether the catch system can fly again and again without rebuilding the cables. CASC’s stated plan is to gather flight data for the crewed 10A and the lunar Long March 10. If the recovery cadence holds, the third Chinese launch of 2027 may carry a Mengzhou capsule on a rocket whose first stage was already flown once.

On a Texas supercomputer this summer, a 370 Mpc/h cube of dark matter, gas, and stars finished a 13.5-billion-year journey. The run began at redshift z=99, when the cosmos was a thin hydrogen-helium plasma and the first stars had not yet ignited. It ended at z=0, with galaxy clusters, supermassive black holes, and the cosmic web we see today. The team calls the simulation ASTRID, and the final z=0 dataset was published in The Astrophysical Journal in February 2026, with a public release coordinated through the Texas Advanced Computing Center on August 27, 2026 (Zhou et al., ApJ, 2026).

The headline numbers are easy to recite. ASTRID evolves 2 × 5500³ ≈ 0.33 trillion particles in a 370 Mpc/h box. Each simulation snapshot totals 30 TB. The archived state across the run is measured in petabytes. Less obvious is what those numbers actually buy the cosmologist at the other end of the keyboard: a single dataset that connects the universe’s first black hole seeds to the billion-solar-mass monsters sitting in galactic centers today.

JWST keeps pulling ancient quasars and “Little Red Dots” out of the early universe that do not fit the old growth timelines. The first JWST surveys suggested that 10⁹ M☉ black holes already existed at z > 8, hard to reconcile with seed models from a decade ago (Naidu et al., 2024, and follow-ups). To make sense of those objects, theorists need a model universe that can track baryons, gas cooling, star formation, and black hole accretion from cosmic dawn to the local universe, all at high enough resolution to capture the small dark-matter halos where the first quasars lit up.

Cosmological simulations do that, but they are expensive. Each box has to be large enough to contain rare massive clusters, dense enough to track small galaxy halos, and long enough to evolve them across 13.5 Gyr. Few projects manage all three at once. ASTRID does.

Yihao Zhou, a PhD student at Carnegie Mellon University’s McWilliams Center for Cosmology and Astrophysics, led the run. His advisor, Tiziana Di Matteo, is the principal investigator and the center’s director. The team used the NSF-funded Frontera system at the Texas Advanced Computing Center (TACC) under a 2024 Leadership Resource Allocation (TACC press, August 27 2026).

The run sits in a crowded field. MillenniumTNG’s flagship covers 740 Mpc per side, roughly eight times the volume. IllustrisTNG-300 uses a similar box size with less dark-matter resolution. Eagle is smaller but uses higher resolution per particle. MassiveBlack-II covers 533 Mpc with far fewer particles and no full hydrodynamic run to z=0. ASTRID’s pitch is that no other published run combines this particle count, this box size, and a continuous evolution to z=0 with a complete black hole sub-population (Zhou et al., 2026).

That black hole population is the part astronomers care most about. ASTRID carries massive black holes (MBHs) spanning 4 × 10⁴ to 2 × 10¹¹ M☉, the range from putative seeds to the heaviest known monsters. Earlier ASTRID snapshots have already shown up in papers on Little Red Dot analogs (Hassan et al., Open Journal of Astrophysics, 2025) and on black hole merger rates that match LISA predictions. Mock JWST and Euclid light-cones built from the simulation give observers a template to compare against real survey catalogs, which is how the team plans to constrain the seed-black-hole mass in the next round of analyses.

ASTRID is a smoothed-particle hydrodynamics (SPH) simulation, which means the gas and dark matter are discretized into Lagrangian particles that move with the flow. The code solves the Euler fluid equations coupled to a particle-mesh gravity solver. Long-range forces use a coarse mesh; short-range forces use a hierarchical Barnes-Hut octree that drops subtrees whose multipole moments fall below an opening-angle tolerance θ.

The sub-grid physics is where most of the engineering sits. Gas above a density threshold around 0.1 cm⁻³ is converted into stars using a Kennicutt-Schmidt-style surface-density law, Σ_SFR ∝ Σ_gas^n with n ≈ 1.4. Black holes are seeded in dark-matter halos above ~5 × 10¹⁰ h⁻¹ M☉ and accrete via a Bondi-Hoyle-Lyttleton prescription capped at the Eddington rate:

Mdot_BH = min( Mdot_Bondi, Mdot_Edd ) = min( 4π G² M_BH² ρ / (c_s² + v²), L_Edd / (ε c²) )

Feedback comes in two flavors. Thermal quasar-mode kicks in at high accretion (≳1% of Eddington), kinetic jet-mode at low accretion. That thermal-versus-kinetic split is what produces the bimodal color sequence between star-forming blue galaxies and quiescent red ones.

The dynamical-friction submodel is the piece most relevant to LISA. When two galaxies merge, their central black holes are flung into a wide orbit. Chandrasekhar’s formula determines how fast each black hole sinks into the merger remnant:

df/dt = -4π G² M_BH² ρ(r) ln(Λ) f(v) / v²

The Coulomb logarithm Λ runs ~3-5 for galaxy mergers. Without this term, ASTRID would smear the merger-rate signal that LISA wants to hear.

LISA’s sensitivity band runs from ~10⁻⁴ Hz to ~10⁻¹ Hz, the regime populated by MBH mergers of 10⁴ to 10⁷ M☉ across cosmic time. Di Matteo’s earlier work (Ni et al. 2022) projected ~30 MBH mergers per year detectable by LISA out to z ≈ 10. The full z=0 ASTRID data set should sharpen that number, and provide a sky-position catalog that future gravitational-wave observers can cross-match against electromagnetic transients (Zhou et al., 2026).

On the engineering side, Frontera is the largest NSF open-science cluster in the United States. Built around Intel Cascade Lake Xeon nodes with ~60 PB of Lustre storage, it absorbed ASTRID’s 30 TB snapshots without complaint (TACC user guide, August 4 2026). The team’s next target is TACC’s Horizon system, the NSF Leadership-Class Computing Facility machine deploying in two phases (GPU first, then CPU) later in 2026. The ASTRID code is currently CPU-only; Zhou and colleagues are rewriting the gravitational and black-hole feedback kernels to run on NVIDIA GPUs.

For years, astrophysicists have asked how supermassive black holes got so big, so fast. ASTRID does not answer that on its own, but it gives the community a 0.33-trillion-particle laboratory to test the answers against.

The next year or two will bring a GPU port to Horizon, plus companion runs at higher resolution or with varied cosmological parameters. By the time LISA Pathfinder’s successor launches around 2035, theorists will want catalogs like ASTRID’s to interpret the gravitational-wave sky.

What comes out of a simulation like this is rarely a single discovery. It is a reference. A way for theorists to ask whether their model matches what we should see, with enough resolution to be honest about the mismatch. ASTRID is now that reference for the next decade of SMBH and gravitational-wave cosmology.

Image source: NASA Image and Video Library. Photograph by NASA, public domain.

Crew Dragon Grace docked to the forward port of the International Space Station’s Harmony module at 7:05 PM Eastern time on Thursday, October 1, completing the fastest launch-to-ISS-docking transit in the history of U.S. human spaceflight. The four Crew-13 astronauts — NASA’s Jessica Watkins and Luke Delaney, Canadian Space Agency astronaut Joshua Kutryk, and Roscosmos cosmonaut Sergey Teteryatnikov — arrived at the orbital outpost seven hours and fifty-five minutes after their Falcon 9 lifted off from Cape Canaveral, with the spacecraft traveling 263 miles above the northeastern Pacific Ocean at contact. A welcome ceremony at the station’s hatch followed at 8:45 PM Eastern.

The fast-rendezvous profile is the headline. Most Crew Dragon flights to the ISS take 24 to 28 hours and complete roughly eighteen orbits before contact; the typical transit requires Dragon to make a series of calculated engine burns to gradually raise its orbit to match the station’s altitude and phase. The Crew-13 profile completed the same work in fewer orbits by setting Dragon’s initial orbit to a period slightly shorter than the station’s, allowing Dragon to catch up faster — a more propellant-intensive but faster approach. Spaceflight Now noted in pre-launch coverage that the choice of profile depends on orbital geometry at launch time, and that the specific 11:10 AM Eastern launch window was selected in part because it put the ISS in the right relative position to support the short rendezvous. NASA’s launch-day release called the transit the fastest launch-to-docking by a U.S. spacecraft in ISS history; the previous mark was set on Crew-4 in 2022 at roughly seventeen hours.

The booster that lofted Grace into orbit made its own contribution to SpaceX’s reusability ledger. B1101 returned to Cape Canaveral and touched down at Landing Zone 40 less than eight minutes after launch — the seventh landing at LZ-40 and SpaceX’s 664th Falcon booster landing overall. The booster was on its third flight: it had previously launched the Starlink 6-88 mission and Crew-12 in February 2026. B1101’s reuse profile, plus the use of a previously-flown Crew Dragon capsule, continues to be one of the structural arguments SpaceX makes for its pricing. Recent Dragon crew rotation flights cost NASA approximately $288 million per mission, or roughly $72 million per seat. Boeing’s Starliner was priced at roughly $90 to $99 million per seat under its original contract.

The flight carried more than just crew to orbit. Grace (tail number C-213) had flown before — she was the same Dragon that flew NASA’s SpaceX Crew-4 mission in April 2022, the mission on which Watkins made her first trip to space. SpaceX’s design certification for Crew Dragon, last reviewed against the manifest in 2024, allows each capsule to be reflown up to fifteen times across its service life. The Crew-13 mission flew with no public announcement of any hardware change to the capsule. The SpaceX contract for the flight did, however, expand: in a May 2026 announcement, NASA formally added six more Dragon commercial crew missions to its Commercial Crew Program manifest, replacing flights that had originally been awarded to Boeing’s Starliner. That addition, combined with a separate $946 million contract for three flights, means SpaceX now holds contracts for all remaining NASA ISS crew rotation missions through the station’s planned retirement around 2030.

The single-source-of-supply arrangement is not what NASA’s planners wanted. The original Commercial Crew Program was structured to fund two providers — SpaceX and Boeing — so the U.S. would not be wholly dependent on a single company for crew transport. That redundancy is now absent because of Boeing’s 2024 Crew Flight Test failure. In February 2026, NASA formally classified that failure as a “Type A mishap,” the agency’s most serious category, after multiple propulsion anomalies during transit — most notably a string of helium leaks in the service module and the failure of five of eight reaction-control thrusters in quick succession. NASA returned the vehicle uncrewed and brought the test crew home on a Dragon in March 2025. Boeing’s next Starliner mission is currently targeted as an uncrewed cargo flight no earlier than the fourth quarter of 2026, but Boeing’s own SEC filings describe NASA discussions as having an “uncertain” outcome.

Once aboard the ISS, Crew-13 will join Expedition 75 alongside the crew already on station. The handover will play out over the next week, after which Crew-12 — NASA’s Jessica Meir and Jack Hathaway, ESA’s Sophie Adenot, and Roscosmos’s Andrey Fedyaev — will board Dragon Freedom for the trip back to Earth. Watkins, Delaney, Kutryk, and Teteryatnikov are then expected to stay aboard through approximately next March and April, rotating through Expedition 75 and 76 as the station hands off twice during their stay. Their science agenda includes biology, materials science, Earth observation, and human physiology studies — including work that will feed into the cardiovascular, bone-density, vision, and immune-function research programs that inform Artemis-era lunar surface planning.

One piece of context that isn’t in NASA’s release but does change how the mission should be read: Grace is, according to a separate secondary-outlet report published today, the fifth and final Dragon capsule SpaceX has built. The company is not planning to construct another. The implication for ISS operations is significant — every operational NASA crew rotation to the station through 2030 will use one of those five Dragons, including a capsule that has been aloft for some of its life cycle. NASA’s Commercial Crew Program and Boeing’s recovery of Starliner will both be factors in how that plays out, but the operational constraint is set: there are five Crew Dragons, and there will be no more.

The next major event in the Crew-13 timeline is the handover, which is expected to take about a week and concludes when Crew-12 departs. NASA’s next crewed launch of any kind is currently scheduled to be the second lunar-flyby Orion mission, in late-stage processing at Kennedy. Crew-14 has not yet been publicly assigned.

For the moment, though, the four Crew-13 astronauts are aboard. Hatch opened at approximately 8:45 PM Eastern. The welcome ceremony was broadcast on NASA TV.

Image source: NASA Image and Video Library. Photograph by NASA/Joel Kowsky, public domain.

A SpaceX Falcon 9 lifted off from Space Launch Complex 40 at Cape Canaveral Space Force Station at 11:10 AM Eastern time on Thursday, October 1, carrying four astronauts on NASA’s Crew-13 mission to the International Space Station. About seven hours and fifty minutes after launch, Crew Dragon Grace is expected to dock autonomously to the forward port of the station’s Harmony module at approximately 7:00 PM Eastern. NASA’s launch-day release described the planned transit as a record-setting flight duration for any U.S. spacecraft heading to the station.

The four crew members are Jessica Watkins and Luke Delaney of NASA, Joshua Kutryk of the Canadian Space Agency, and Sergey Teteryatnikov of Roscosmos. Watkins commands the mission and is flying her second spaceflight, becoming the first NASA astronaut to fly twice on a SpaceX Dragon. Delaney pilots in his first spaceflight; Kutryk and Teteryatnikov are mission specialists, each also on a first flight. The launch was originally targeted for September 12 and was delayed by a propellant oxidizer leak discovered on the Dragon before that attempt. The crew entered final quarantine on September 17 and waited out the slip until the October 1 window, which weather officials rated at 70 percent go.

This is the thirteenth NASA Commercial Crew rotation flight, and the twenty-first time a Crew Dragon has carried a crew to orbit. The mission will run approximately six months, ending with a Pacific Ocean splashdown no earlier than March 2027. Once aboard, the four crew members will join Expedition 75 alongside Anil Menon of NASA and Pyotr Dubrov and Anna Kikina of Roscosmos, who are already on the station. After a brief handover, NASA’s SpaceX Crew-12 mission — astronauts Jessica Meir and Jack Hathaway of NASA, Sophie Adenot of ESA, and Andrey Fedyaev of Roscosmos — will return to Earth, restoring the station to a seven-person complement.

The crew backgrounds are deliberately diverse. Watkins is a geologist who previously flew on SpaceX Crew-4 in 2022; her second flight makes her the first NASA astronaut to ride a Dragon twice. Delaney is a former U.S. Navy aviator; Kutryk is a Royal Canadian Air Force fighter pilot; Teteryatnikov is a Russian submariner. Kutryk’s flight is a small but real first: he is the first Canadian to fly as part of NASA’s Commercial Crew Program. The earlier Canadian on a Dragon was Mark Pathy, who flew privately on Axiom Mission 1 in 2022.

NASA announced the crew assignments on April 23, 2026. The pairing of an experienced commander with three first-time flyers, including two from agencies that have not previously flown on Commercial Crew, is consistent with how NASA has used early-rotation flights to broaden the program. A Roscosmos backup crew member, Harutyun Kviryan, was assigned to the mission and remains in crew-support rotation on the ground.

The science planned for the six-month expedition leans heavily on microgravity biology. The crew will work with human stem-cell-derived tissues, an area of research NASA has invested in heavily in the last several expedition cycles, with the goal of building disease models and pharmaceutical testbeds that are hard to assemble on Earth. The crew will also work with a mostly autonomous plant-growth system to test crop production outside dedicated growth facilities — an effort aimed at the longer question of feeding crews on deep-space missions. Two additional threads continue work from earlier expeditions: studies of how blood-flow regulation responds to microgravity, and a hand-held diagnostic device intended to give crews more autonomy on long flights.

The mission comes at a transitional moment for the U.S. human-spaceflight program. NASA’s commercial crew cadence has settled into a roughly twice-yearly rhythm, and Crew-13 is the third operational mission of 2026. At the same time, the agency is preparing for the next test of its deep-space crewed vehicle: Orion, which flew its second crewed mission in April. Crew-13’s fast rendezvous profile reflects improvements in Dragon’s on-orbit phasing software and ground-control procedures; the previous record for a U.S. crewed launch-to-ISS-docking transit was set on Crew-4 in 2022 at roughly seventeen hours.

The launch comes three days after SpaceX’s first orbital Starship flight on September 28. NASA’s launch administrator, Jared Isaacman, framed the back-to-back events in his remarks at Kennedy Space Center. He called Crew-13 “another demonstration of America’s unmatched capability in human spaceflight,” and credited the agency and SpaceX teams for the launch. Dr. Lori Glaze, NASA’s associate administrator for the Human Spaceflight Mission Directorate, tied the mission into the broader exploration narrative: the ISS is “sustaining our presence in Earth’s orbit and sharpening the tools we’ll take forward to the Moon and Mars.”

Docking is scheduled for 7:00 PM Eastern. NASA’s live arrival coverage begins at 5:20 PM Eastern for the rendezvous, docking, and hatch-opening sequence, after which the four astronauts will change out of their spacesuits, transfer cargo, and open the hatch into Harmony. The handover period between Crew-13 and Crew-12 is expected to run roughly a week, with Crew-12’s return targeted for later in October.

The Crew Dragon capsule flying this mission is named Grace. That name is not on NASA’s standard naming rotation for Crew Dragon; it was applied for this specific vehicle. The choice has not been widely discussed publicly, but the spacecraft’s flight record makes the name appropriate: Grace flew this mission after a propellant leak in late August forced a multi-week delay, and the recovery from that delay is what the crew and ground teams are credited with in NASA’s launch-day release.

Once Crew-13 is aboard, the station will continue its mixed program of microgravity research, maintenance spacewalks, and visiting-vehicle traffic through the spring of 2027. NASA’s next crewed flight of any kind is expected to be the second lunar-flyby mission of the Orion program, currently in late-stage processing at Kennedy. Crew-14, the next operational Crew Dragon rotation, has not yet been publicly assigned.