While answering a question about Starship versus skyhooks, I discovered that there is a big difference in payload between a Falcon 9 launch where the booster lands downrange on a drone ship and one where it returns to the launch site.

Performance Query NASA Launch Services Program Launch Vehicle Performance Website

For the 400 km orbit (lowest the tool supports for Falcon 9) the payload supported drops by a factor of...

$$RTLS/ASDS = 11700/15500 ~= 0.75$$

This got me thinking, is it possible that Starship could increase its payload if Superheavy landed "downrange" on a (perhaps artificially constructed) island or platform in the gulf of Mexico somewhere, assuming that SpaceX built a Mechazilla tower on said island or platform?

enter image description here

Is there a site that is both in a suitable location and shallow enough to build an island or platform on? If so, would it even make sense to land there, refill the booster, and fly it back to Boca Chica?

  • 1
    $\begingroup$ Not directly relevant to your question, but launching from East from Vandenberg and recovering into a land site with return by land or reflight has been proposed at various times, mostly limited by liability concerns than technical ones. The fun proposal would be to have launch sites around the globe, and have the first stages recovered one hop East each launch. $\endgroup$ Commented Jan 27 at 7:37
  • $\begingroup$ @phil1008 The link to "NASA Launch Services Program Launch Vehicle Performance Website" isn't working for me. $\endgroup$
    – Galerita
    Commented Apr 18 at 6:35
  • $\begingroup$ It still is for me. You then have to go to the Performance Query tab. $\endgroup$
    – phil1008
    Commented Apr 18 at 10:13
  • $\begingroup$ Most likely Not as super heavy doesn't have landing gear. If it had, it would require a gigantic barge and a much stronger octograbber robot to grab the bottom of the stage $\endgroup$ Commented May 22 at 13:55

2 Answers 2


TL;DR Yes, with increased payloads the SuperHeavy will be forced to land downrange, likely on a larger drone ship than is currently used for the Falcon 9 first stage. No, it does not make sense to refill the booster and fly it back to Boca Chica. Like the Falcon 9 first stage, the drone ship will return to a purpose built dock near Boca Chica where SuperHeavy will be unloaded and refurbished for relaunch.

Falcon 9 has 3 launch modes, dependent on the amount of fuel left at separation. Stage 1 will either:

  1. return to launch site (RTLS)
  2. land on an autonomous spaceport drone ship (ASDS)
  3. be expended somewhere in the Atlantic

Musk partly explains it here. Launches to GEO or Earth escape require too much fuel for the rocket to reverse its horizontal velocity and return to the launch site. Even for LEO the choice of landing site is payload and orbital height dependent, as in your graph.

An advantage of ASDS is the ship can be positioned near the likely landing site.

At the time of separation for the SpaceX Starship 3rd integrated test flight (T + 2:48), the rocket was at 71 km altitude ("space" is 100 km), travelling at 5700 km/h, and at an angle of approximately 30 degrees to the horizontal. That means it has to lose about 5700 × cos(30) ~ 5000 km/h of horizontal velocity and then fly in the opposite direction for RTLS.

For ASDS the rocket need only fly a ballistic trajectory and use thrust: initially to slow its vertical velocity then finally for a smooth ASDS landing.

After an ASDS landing, the Falcon 9 Stage 1 is shipped back to Port Canaveral near Starbase in Florida for refurbishment.

In the most extreme case the Falcon 9 first stage has burnt through all its fuel and has reached such a high velocity and altitude that it burns up on reentry.

The Falcon 9 payload capacities under the three categories for both LEO and GTO are in the side bar on Wikipedia under Capacity. For example the payload to GTO at full thrust (FT), where the first stage is expended is 8.3 t. For ASDS the payload is 5.5 t. And for RTLS the payload is only 3.5 t.

Similar options will be available for the integrated Starship/SuperHeavy rocket. Port of Brownsville is only 30 km from the Starbase production facilities. It is currently used as a port for the transfer of Starship components from Florida using the GO Discovery, which is a platform supply vessel, the same type of ship that supports ASDS operations.

I suspect a larger vessel will be required as an ASDS for the SuperHeavy booster, but then it will be shipped to Boca Chica for refurbishment, perhaps to a newly built dock or expanded facilities at Port of Brownsville. There is some discussion of this in chat groups, but I can find no "official" SpaceX communication. Maybe it's worth asking Musk directly on X.

It makes no sense to refuel the SuperHeavy and fly it back to Boca Chica. That would require the ASDS to be equipped with cryogenic oxygen and methane refuelling equipment. The cost in fuel of such a return flight would be much greater than shipping and it would result in additional wear and tear on the booster.

The empty weight of a SuperHeavy booster is 200 t. Prior to launch it is moved 4 km from the assembly Megabay1 to the launch pad using a multiwheeled transporter that fits on the 2-way Boca Chica Boulevard. Note the overhang - the SuperHeavy is 9 metres in diameter. The white-line to white-line width of Boca Chica Boulevard is about 8.5 m.

SuperHeavy Transporter

A video of the transport process can be viewed here, although in this case the SuperHeavy is being returned from the launch pad back to Megabay1.

There is a short link road from the Boca Chica Boulevard to the docks at Port of Brownsville, where objects such as this large methane tank have been seen unloading.

enter image description here

This Port Connector Road appears to have been built specifically for SpaceX.

There are also hints in job advertisements that SpaceX intends to implement "marine recovery" of the SuperHeavy.

  • $\begingroup$ a 30km road move from Port of Brownsville for Super Heavy seems very unlikely to me, especially at any kind of cadence. It's too large of a load. The Falcon cores are much easier to move around. $\endgroup$
    – Erin Anne
    Commented Apr 17 at 16:13
  • $\begingroup$ Look at the added text, the pictures and this video: youtube.com/watch?app=desktop&v=UHIZK--L_-4 You can judge the cadence there. They could easily build a parallel road-way if needed. I originally thought they might want to build a dock closer to the Starbase, but the nearby area is a wildlife reserve. They will not lose the opportunity for additional launch weight/altitude by keeping with the current return to launch site arrangement. $\endgroup$
    – Galerita
    Commented Apr 18 at 4:05
  • $\begingroup$ I'm aware that Super Heavy is already moved by road. It's already a huge pain to the locals to move it the short distance they already do by highway. Moving it 30km from the port of Brownsville is a much larger impact. The payload your last picture shows isn't a Super Heavy body (those are built on-site, not shipped in), it's a methane tank, as seen here youtube.com/watch?v=b01RIerMEk0 $\endgroup$
    – Erin Anne
    Commented Apr 18 at 4:40
  • $\begingroup$ @ErinAnne I've amended the text. The existing road is wide enough for the SuperHeavy transporter. I'm sure it's a relatively low cost exercise to duplicate the road. Given the Port of Bronwsville paid for the connector road, they'll probably help foot the bill for a duplicated Boca China Boulevard. Long distance road-legal road trains, weighing up to 200 tonnes, are common in Australia. I assume US roads can cope. To me the ASDS is a bigger problem. But given payload advantages of using one I don't doubt SpaceX will in time. $\endgroup$
    – Galerita
    Commented Apr 18 at 5:11
  • 1
    $\begingroup$ I think that the idea of drone ship that can both support a mechazilla tower and return to port won't be feasible. But possibly a larger platform with the mechazilla tower on it plus a smaller separate ship that is designed to carry the booster back might work. $\endgroup$
    – phil1008
    Commented Apr 18 at 10:32

It may be possible to land a Superheavy on an enormous ADSP but the problems should not be under estimated. There are two possibilities. Either modify Superheavy to have legs and at ~ 200 tonnes this would not be a simple "bolt on some legs effort" the legs would need to be very substantial and the legs would need extra propellant to launch, extra propellant to land and the landing propellant would also need launch propellant. All in all a considerable cost in launch mass and development time.

The other option is to build a catching tower on the ADSP. But in deep water with a tall tower the amount of sway, pitch and roll at 50-60m above sea level is not going to be insignificant even if the legs are partially flooded as changes at sea level will be magnified as the height increases above the platform. This will make catching the Superheavy safely even more challenging and reduce the window when launches are possible.

Then there is the delay in shipping the Superheavy back to the launch site which will not be that quick. Elon Musk seems set on very rapid reuse and virtually no refurbishment time. Extended sea shipment times would not fit well with rapid relaunch.

The gain in payload would have to be set against the costs in development, extra hardware requirements, delay and added risks. This would need to be compared against the option of launching two Starships from the same Superheavy cheaply and within a few hours. I may be wrong but if I was a betting man my money would be on RTLS every time.

PS re-tanking Superheavy at sea and flying back to the launch site would be even more problematic.

  • $\begingroup$ in sufficiently shallow water a jack-up barge might alleviate the sway/pitch/roll woes, though it introduces its own complications, and obviously doesn't address the rest of your points. +1 $\endgroup$
    – Erin Anne
    Commented May 19 at 21:09
  • $\begingroup$ Yes perhaps that might help a little, but the difficulties mount up as the water depth increases. In the case of Chevrons Petronius platform (reportedly the worlds tallest free standing structure):The compliant tower design was chosen for its ability to withstand hurricane conditions and operate in depths of 2,000ft (610m). The compliant tower design enables it to move within an envelope of 25ft sway (7.6m), and a 10ft (3m) rotation sway at the surface. But the water depth out in the deep Atlantic would be more like 3000m. $\endgroup$
    – Slarty
    Commented May 20 at 12:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.