Artemis 4 is currently scheduled to mark America’s return to the surface of the Moon in 2028. However, as I’ve noted in covering the programme in this column, there are several technical matters which need to be addressed – particularly if SpaceX remains the supplier of choice for the mission’s lunar lander – which could see the mission slip back to 2029. But now questions on exactly where Artemis 4 should be going.
Taken as The Artemis Project aims to send humans to the Lunar South Polar Region with the aim of establishing a base there, on the basis that base parts of the Moon’s polar regions remain in complete shadow all the time – notably the bottoms of craters – then it is possible there is accessible water ice to be found which could make lunar operations somewhat easier (providing water for drinking / cooling systems, air, ingredients for rocket motor propellants).
The overall goal of the mission is to primarily check-out what will be largely untested hardware, equipment and systems – both in terms of the lander vehicles and the space suits developed for Artemis. But do these necessarily need to be tested using a mission to the Lunar South Polar region? Could they not equally be carried out using a more equatorial landing zone, one which involves less complexity in its execution – and potentially less risk?
Landing at the lunar equator offers predictable solar illumination, simplified orbital dynamics, and reduced communication blackout risks compared to the high-latitude Lunar South Polar Region. While much of the latter does experience near-continuous daylight, its chaotic topography with dramatic elevation changes coupled with a low-angle solar elevation (thus resulting in areas remaining in deep shadow, even outside of craters) present severe hazards for landing craft and extravehicular operations.
Thus, by targeting the more “known” regions of the Moon, such as Mare Imbrium (in which Apollo 17 landed) and Mare Tranquillitatis (Apollo 11), Artemis 4 could operate in a somewhat more predictable operational envelope than would be the case with a South Polar Region landing. This would potentially allow for better initial testing of guidance, control, automated landing systems and communications systems.
The latest voice to be heard advocating for this approach is that of Artemis 2 pilot, Victor Glover, who likened the switch to a more stepwise approach to lunar operations. For example, in the 1960s, NASA initially tested the Apollo Command and Service Module in Earth orbit (Apollo 7), then in a flight around the Moon (Apollo 8), then the lunar lander was tested in Earth orbit (Apollo 9) and then on a non-landing voyage to the Moon where its decent / ascent capabilities could be tested in lunar conditions (Apollo 10). Further, the Apollo missions grew in complexity, making targeting more difficult landing sites more acceptable as experience was gained from earlier missions and the capabilities of the vehicles and their systems became more clearly understood.
By contrast, Artemis effective jumps: from an uncrewed cislunar test of the Orion Earth-Moon-Earth vehicle (Artemis 1) to a crewed cislunar test of Orion (Artemis 2) to a semi-orbital test of the landing vehicles in Earth orbit (Artemis 3 – which will not even be a test of either vehicle’s flight and propulsion systems) to attempting a landing in a very challenging environment; so there is much to be said for revising efforts.
As well as lowering the technical complexity of the initial mission, Glover raises a health reason. What if a crew member suffers and injury or medical situation requiring evacuation to Earth? With Apollo, the Command and Service Module orbited between 100 and 300 km above the lunar surface. This made getting back to it aboard the lunar lander in an emergency, and it would be in a position to rendezvous with the ascend lander every few orbits, and likewise be able to more readily to break orbit and return to Earth.
Missions to the Lunar South Polar Regions require a more complex, propellant-intensive flight, culminating in an Elliptical Polar Orbit with Coplanar Line of Apsides (EPO/coLA). Because more propellant is required, this orbit must vary between 100 and 6,500 km above the Earth, as the high altitude is required for Orion to be able to break orbit and return to Earth; lower than this, and it could have insufficient fuel remaining to escape the Moon.
Because of its range, this EPO/coLA orbit means that a rendezvous between Orion and an ascending lander can only happen once every couple of days or so. By switching to a more equatorial landing zone, propellant usage is reduced, making Orion more capable of entering a more circular, lower orbit of the Moon, thus presenting more opportunities for an emergency rendezvous with the lander should this be required. Of course, this doesn’t solve the issue of emergency evacuations on missions are Artemis 4; what it does mean is that the risks associated with the “first time” nature of Artemis 4 might be further reduced.
In terms of missions beyond Artemis 4, SpaceX have already suggested that their massive Starship-derived land could actually take Orion to the Moon if the two were to dock in Earth orbit. This would conserve Orion’s propellant, potentially allowing the elliptical nature of the EPO/coLA orbit to be reduced, again offering more opportunities for emergency rendezvous in lunar orbit. Similarly, the switch to using a modified Vulcan-Centaur booster and the upper stage of the SLS rocket, replacing the current Interim Cryogenic Propulsion Stage (ICPS) could also allow the EPO/coLA orbit range to be reduced, again allowing windows for Orion / lander rendezvous to be increased.
Finally, Glover (and others) note that the political onus on Artemis 4 is increasingly about “beating the Chinese” rather than achieving science / engineering goals. If this is the case, then a mission to land within the Moon’s more familiar equatorial regions, building on all of the knowledge gained on such mission during Apollo, would help make that goal easier by removing several levels of complexity to Artemis 4.
Thus far, NASA management has given no indication any such shift will be made, but in many respects it does fit with a more step-by-step approach to lunar operations they have adopted since the start of 2026 .
Starship IFT-13: More of the Same, but with Twists
On July 24th, SpaceX completed the 13th integrated flight test (IFT) of its Starship / Super Heavy booster combination, flying the “version 3” of each vehicle for the second time, the fist having been in May 2026.
In broad terms, the flight was something of re-run of the past few missions: launch, get Starship to sub-orbital velocity / altitude, return the Super Heavy booster for a controlled splashdown in the Gulf of Mexico, with the Starship deploying a payload before initiating a test-restart of one of its engines and then making a re-entry and splashdown in the Indian Ocean not all of which had been achieved through IFT-10 through 12.
An initial attempt to launch the flight had been made on July 16th, but was halted by on-board systems when four of the booster’s 33 Raptor engines failed to ignite. Subsequent analysis revealed that due to the way the engines are chilled, ice had formed within those four engines, choking them at start-up. This resulted in two of the engines having to be swapped-out on the pad. A second Attempt on July 23rd was scrubbed as cloud would have obscured observations of the vehicle’s ascent.
On July 24th, things went remarkably smoothly, with the raw power of the Raptor 3 engines clearly seen – it took just sixty seconds for the booster and ship to go from the launch platform to supersonic speed and then pass through max-Q (the point of highest dynamic stress on the vehicles during ascent).
Hot staging took place with the booster shutting down all but five of its engines as the ship fired-up its six engines. On separation, the booster immediately vectored away from the ascending Starship, avoiding the issue which led to a failure with the boost-back engine burn seen with IFT-12 in May. However, the booster did suffer a engine re-start failure just before a planned splashdown in the Gulf of Mexico, when only 9 of the anticipated 13 engines restarted – and four of those immediately shut down again and the booster hit the water at an estimated 160 km/h. This and the ice issues of July 16th suggest the Raptor 3 still needs refinement.
Starship S40, however, went on to achieve sub-orbital flight, deployed its payload, finally carried out a restart of an engine (something the last several flights have tried and failed to achieve, and which forms a vital part of vehicle operations) and then made a perfect re-entry into, and passage through, the denser atmosphere. Finally, it successfully re-lit three engines, flipped itself vertically to make a gentle landing in the Indian Ocean. So gentle, in fact, it didn’t explode; it gently tipped over on its back to float on the ocean, small fires licking around the hull caused by vented gases. This obliging allowed the aerial drone patrolling the landing zone to capture video of the vehicle’s re-entry heat shield for analysis.
The key difference between this mission and the earlier payload carrying sub-orbital flights lay in the fact that this time, and actual working payload was carried aloft in the form of 20 Starlink v3 satellites.
At up to 2 tonnes each, and with a span of 15 metres with their solar arrays open, the Starlink v3 is a veritable monster compared to the v1 (300 kg) and v2 (600 kg) variants and offers gigabit levels of data transfer rates.
These satellites are seen as vital to Starlink’s future, and are wholly dependent on the Starship / Super Heavy combination to launch them, as they are too big and bulky to be launched en-masse by Falcon 9 (hence why SpaceX is almost solely focused on getting Starship working with a Starlink deployment system rather than building it capable of delivering a range of payloads to orbit – in fact, there is an argument to make that Starlink is Starship’s only viable payload, but that’s for another article).
None of the deployed Starlink units survived long – just 20 minutes given they were sub-orbital as well. However, this was long enough for SpaceX to test their ion engines and communications capabilities, hooking them into the main Starlink network and testing their laser communications systems among themselves.
Six of the satellites were fitted with cameras and spotlights, and were tasked with following the starship vehicle down through re-entry, filming its manoeuvres and interactions with the upper atmosphere. At the time of writing, none of that footage, if successfully relayed to Earth, has been released.
Overall, IFT-13 was a step forward from the rinse-wash-repeat status of the last few Starship / Super Heavy flights. The system still has a long way to go before it can be said to have met even its baseline requirements (launch, deploy, return and capture for re-use), and even longer to go in terms of actually proving it is viable for anything other than Starlink launches. However, subject to the on-going issues with the Raptor 3 motors, SpaceX have indicated the 14th flight might actually be an attempt to reach orbit – possibly with an attempt to catch the orbital vehicle on its return.
Psyche’s Time-Lapse of Mars
On May 15th, 2026 NASA’s Psyche spacecraft swung by Mars, using the planet in a gravity-assist manoeuvre to help propel it on its way to its intended target, the M-type (indicating it has a high metallic content) asteroid 16 Psyche, which it will reach in 2029.
The fly-by marked the mid-point in the spacecraft’s journey, and I covered it and the overall Psyche mission just after it happened. However, earlier in July NASA issued a time-lapse video of the encounter, stitched together from individual images captured by the spacecraft’s imagining system between May 2nd and May 31st.
Approaching from a “high phase” angle relative to the Sun, Psyche initially reveals Mars as a thin crescent, only around 4% of its surface visible. This rapidly grows as the spacecraft sweeps down towards the planet’s northern hemisphere, features somewhat blurred, but with craters and clouds visible before Psyche arcs over the very heavily cratered southern hemisphere, closing to within 4,609 km of the planet’s surface before heading back into space, speeding towards the asteroid belt and 16 Psyche with 1,600 km/h more velocity than when it approached at Mars.
While the fly-by didn’t add much to the masses of data already gathered on Mars, it did serve a second important function for the Psyche mission: by aiming Psyche’s suite of instruments at a well-studied world, mission engineers were able to properly calibrate and validate the spacecraft’s major science instruments ahead of its encounter with 16 Psyche.
It’s been a good while since I offered any updates on the work of NASA’s Curiosity rover on Mars, which is a bit of a shame given it was my reporting on Curiosity’s arrival and mission on Mars which eventually morphed into Space Sunday.
Inara Pey: Living in a Modemworld 