Earth’s Natural Harbor

Starship will change everything. And I mean everything. Everything you know about spaceflight going back to Sputnik will be wrong.

The old paradigm is that you launch something to the place it’s going to need to be, on a rocket that you throw away. You accept the payload penalty of going to that place that is suboptimal, in terms of
inclination or altitude, relative to its nominal due-east performance
from its launch site. You also have to accept launch windows, and the
risk of missing them due to weather, or technical issues, which can be a real problem for a grand tour of the solar system that you only get one shot at in a decade or five, because that’s just the way it is, and the way it’s been forever.

That is about to change.

The easiest orbit to get to from the planet’s surface is equatorial. You get maximum advantage of earth’s rotation by launching due east at the equator. There are no launch windows to get there; you can launch any time of the day, every day, and you will be in the equatorial orbit plane. There is also little weather risk; hurricanes at the equator are almost unheard of (there’s too little coriolis there to spin things up).

So if you want a steady pipeline of stuff going into space (and coming
back) at minimal cost, it can be done from some number of spaceports on the equator.

There is another advantage to equatorial orbit, from a space
sustainability standpoint. If everything we had in space was in a single orbit plane, it would be much easier to do space traffic control because it is reduced to a two-dimensional problem, at low relative velocities, without all of the high-velocity conjunctions that occur in multiple orbit planes.

Finally, having a single orbit plane to deal with vastly simplifies
mission planning for beyond-LEO flights, whether to the moon or beyond cislunar space. This is where my harbor analogy comes in. Everything leaving the vicinity of the earth or returning to it, will do so via that orbit plane.

So why don’t we do that?

For two reasons: 1) You can’t get to an equatorial orbit from a
non-zero-latitude launch site without severe payload penalties (if you can get there at all), and all sovereign launch sites to date are non equatorial; and 2) Many satellites want to have inclinations greater than zero (e.g., remote sensing, and communications and navigation constellations), and the velocity cost of getting to those inclination in LEO from an equatorial orbit can be higher than the cost of getting into orbit in the first place. For example, if you wanted to do a plane change of sixty degrees in LEO of a payload that had been delivered there by a Falcon Heavy, you’d have to start by putting another Falcon Heavy in orbit to do it.

There are solutions to both of these problems, and they’re happening now.

The first problem is solved by building equatorial spaceports out of
which SH/Starship (or its I hope inevitable competitors) can operate to provide regularly scheduled service to and from ELEO. The idea would be to do everything that doesn’t require a different
inclination/altitude (e.g., weightless research and manufacturing, or
space tourism that accepts that only the equatorial region will be seen from orbit) in the equatorial plane.

The second is solved by moving things that do require different orbit planes and altitudes there by alternate means to chemical propulsion (e.g., electric thrusters). This can take a long time, but if there’s no rush, that’s not a problem; we’re already doing it with GEO comsats. But for those things that do require higher inclinations/altitudes and are in a hurry and can afford it (e.g., space tourists who want to see a lot more of the planet from above), this will continue to provide some business for existing launch sites.

But here is the key point. I am not proposing this for existing markets; it will require vast new ones. The goal is not to reduce the amount of stuff going into space via the old paradigm (though that will happen to a significant degree), but to dwarf that amount of activity with new activities enabled by the new one, with its vastly reduced costs of getting things into and out of orbit. I’ll quit here for now, and let people discuss in comments just what those things might be.

72 thoughts on “Earth’s Natural Harbor”

  1. There are a number of **POTENTIAL** launch sites within a degree or two of the equator – some have the additional benefit of being an island as well.

    Off the coast of Brazil. The Galapagos. The Banda Islands, the horn of Africa (although they’d fail the security evaluation.

    And some of those governments would be happy to get launch fees.

    1. Singapore is only very slightly north of the equator, which runs through Indonesia. Indonesia has coastal sites on several large islands. For only a slight penalty you could launch out of Papua New Guinea, though cannibalism might become a problem, probably leading to Reavers. Another ideal equatorial location would the the southern end of Somalia, but then you’d get a problem with space pirates. The mouth of the Amazon is right on the Equator. I assume Arianne didn’t go there because they wanted to be in an area under French control.

      Looking at a global windspeed database. Average winds at the equator are:

      13 mph at the mouth of the Amazon.
      16 mph at the southern end of Somalia
      9 mph in the Maldives
      8 mph throughout most of Indonesia and micronesia
      13 mph around Kiribati and Jarvis Island (US).
      7 to 16 mph in the Galapagos, depending on location.

      The more remote locations wouldn’t make financial sense because you’d have to be shipping everything across thousands of miles of ocean to support the launches.

      1. Indonesia and environs, including PNG, are security nightmares.

        Kourou has a battalion of French Foreign Legion guarding it, and that’s relatively benign.

        My personal choice is Genovesa Island, Galapagos. It even has a great natural harbor, and damned near nothing on the island.

        1. Well, you’d have about 700 miles to the coast of Ecuador, so plenty of space for a drone ship recovery. But that also means all supplies would have to ship 700 miles to the island.

          That island might be too small to hold everything. At slightly more than two miles across, it’s about the size of South Padre, and could almost fit between Pad 39A and Pad 39B.

          But it’s a beautiful looking harbor.

        2. “Indonesia and environs, including PNG, are security nightmares.”

          Who knows what the future will be like though? The rise and fall of various geographies is a story that isn’t finished.

      2. You’d have pirates around Singapore and Indonesia. The primary difference is Somalia adds hostage taking while the ones in East Indian and South China Seas just want you to hand over the goods.

        The more remote locations wouldn’t make financial sense because you’d have to be shipping everything across thousands of miles of ocean to support the launches.

        Don’t give NASA any ideas.

  2. I think the noise issue of SH is going to have Musk move to sea launch platforms anyway. If he’s doing that and they are big enough to land a Starship on, the you can get to them via P2P from anywhere on the planet. Noise abatement of Starship might be far easier than for SH, if sub-orbital, ballistic trajectories are good enough to get you to that sea-launch platform that will put you into ELEO. The other advantage of sea-launch is that your are in international waters and not subject to the whims (taxes, nationalization) of a host country. You might fly the flag of your nation of charter in order to get naval protection of the asset if needed.

    But what I’m curious about is how well does ELEO align as a starting point for departures to Moon or Mars? I should go back an study how the ecliptic aligns with the equatorial projection.

    1. Speaking of noise anyone know what kind of sonic booms to expect from a returning SH or orbital Starship? Will we get the same ‘double boom’ we get from returning F9 boosters? You don’t have to explain here for my benefit, a link will do. Don’t want to hijack the thread.

      1. The problem with the SS sonic boom is location, because you don’t want to lay them down over Orlando or Brownsville/Matamoros. There’s really no good reason not to build a spaceport at Jackass Flats, NV. It’s 90 miles from Las Vegas, and that’s what the Boring Company is for.

        1. Yeah, launching and returning say 3 or more times a day. I’m reminded of Admiral Boom from Mary Poppins.

        1. Is that true? I’m under the impression the entry boom in the approach corridor is bigger. I think Musk has said the launch noise of SuperHeavy will be less than Falcon Heavy due to the Raptors being quieter than Merlins, and when SS/SH goes supersonic post Max-Q it’s pretty much exo-atmospheric. Have you heard something to the contrary?

          1. As I understand it, the problem is that you are following a parabolic arc (or close enough), so the sonic booms are “focused” into a small area. Lethal booms are not unusual, as I recall.

          2. Starship makes its hypersonic entry over a long, flattish trajectory, then turns vertical when it reaches the top of the troposhere. That’s why the descent and landing tests are altitude limited: to stay subsonic on ascent and descent.

  3. I think the premise (an old one) is questionable, since the only “destination” above Earth’s equator is GEO. Otherwise, Earth’s obliquity of 23.44deg sees all points on the Earth’s surface tipping u and down in relation to all other destinations by that amount. The Moon doesn’t orbit above the equator, so it and all other destinations above GEO are bobbing up and down in the sky. Seen from Earth, the Moon declination varies from 28deg 36min to 18deg 20 min over a period of 18 or so years. Also, the equatorial “advantage” is only of high importance if you’re using marginal launch vehicles. Otherwise, the lost impulse is in the single digit percentiles. I’ve always advocated for a large facility at L1SE, which I hope is called Stardock (after the Fritz Leiber story).

    Also, let’s not forget you don’t change LEO orbital incliniation by brute force. You do it by lifting the apogee to supersynchronous, changing planes, then lowering at again. It’s not that expensive in terms of delta-vee.

    1. The fact that you can do a bi-elliptic for plane change is one of the things that will drive most traffic to equatorial, where you can get low-cost regularly scheduled rides. As Heinlein said, halfway to anywhere, and that applies to equatorial as well.

      1. You could be right, and it might result in equatorial launch sites as well. I didn’t think of the advantage for depots and tankers right off the bat, so their are at least two attractive destinations fixed over the equator (the other being GEO).

        On the other hand, one thing that’s virtually certain is,SuperHeavy will be superceded. The 9 meter version is driven by contruction, transportation, and launch infrastructure in the continental US. Musk keeps bringing up a 12 meter version. Back at the beginning of all this, I wondered if Musk might by Airlander (or come up with a muskified version of his own).

        1. Anything that doesn’t need to be in a non-zero-inclination orbit will be done there. Orbital assembly, space manufacturing, propellant storage and transfer, space tourism that doesn’t require views of the upper latitudes, movie sound stages… All things that aren’t happening yet, but will quickly dwarf existing activity.

          1. I have a couple of reasons why this might not happen, which I’ll put at the bottom of comments to get them below some other comments I want to reference.

  4. The obvious one to me is shipyards in the harbor. Shipyards that primarily focus on craft or stations of various sizes that won’t ever land. Until we can start processing asteroids, it seems the equatorial route would be best for lifting all the material and machines needed for that kind of work.

    1. The theoretical advantage of a LEO equatorial orbit is that the spacecraft will pass over the launch site every orbit. But that doesn’t take into account the Earth’s hemispheric asymmetry (or the South Atlantic Anomaly, for that matter). I don’t actually know how significant the effect is.

      1. Not to mention that the Earth is an oblate spheroid, which would imply the atmosphere rises higher along the equator than at the poles. By how much I don’t know. Probably not significantly so.

        1. Or maybe that’s what you meant by ‘hemispheric asymmetry’?
          Other than mass/density?

        2. due to atmospheric heating and cooling, the ‘height’ of the atmosphere is constantly changing anyway, by a percent or two

        1. This is a point that Al Globus and Joe Strout have been making a lot, lately. They feel orbital settlement will begin in ELEO, even if it extends to higher orbits later when we enter the era of ET material use.

          1. I’m under the impression tides, including lunar tides, affect everything in the gravitational field. Certainly negligible on the scale of a human being (unless you’re doing a neutron star flyby), but it raises a number of issues. Aren’t their going to be some tidal or gavity gradients on a 120,000 metric ton tank of methane and liquid oxygen? Also, I was initially wondering about terresttrial gravitational effects on spacecraft orbits as masses of ocean water shift about below.

            (120,000 metric tons is only a tiny fraction of the size of Khufu’s Pyramid (approaching 6 million metric tons), but we’re starting to talk about objects in that ballpark.)

          2. I can’t see it as being anything unmanageable. I think that the gravity gradient from earth will be far more significant than that from the moon.

  5. Many satellites want to have inclinations greater than zero (e.g., remote sensing, and communications and navigation constellations), and the velocity cost of getting to those inclination in LEO from an equatorial orbit can be higher than the cost of getting into orbit in the first place.

    Why would you have to get to those inclinations from an equatorial orbit? Why not do what they do from Kourou now and launch directly into those inclinations from your equatorial launch sites?

    1. Because it may be cheaper. It will be a tradeoff between cost and time. For tourists who want to observe larger swathes of the planet, there will probably be 51.6 degree hotels (for historical reasons) that they’ll get to directly, for a higher cost.

  6. We need ocean settlements. And key to ocean settlements is cheap breakwaters and cheap anchoring.
    [[And I want freshwater lakes within the breakwaters.]]

    But more important [what I think has stopped this from already happening] is being able to buy real estate [land] which is ocean land- you own the sea floor {which you have anchor to] so you own ocean land and water above it.
    It seems it probably start at state level and state selling ocean owned by state and starting ocean settlement near existing coastal towns. I think a state could promote low incoming housing on the ocean- everyone can get cheap beach property

    1. I was about to chime in that Peter Thiel could step in with seasteading right about now. Really, a well-designed floating, moored island (or two) would go a long way toward facilitating us as a spacefaring species. There are a lot of people who would put everything they have into something like this, but only if it had the kind of freedoms that America was supposed to have been guaranteed by its Constitution.

    2. Great idea, but the concept of land ownership is only as strong as the laws you live under.

      European nations? Including their colonies? Not hardly at all. Asian countries? African? May be good, til the next coup.

  7. If you can put enough reliability into the system to diminish refurb after flight, then you can let the vehicle self transport. Then you just need a refinement facility to produce propellant. Probably sounds far fetched, but 20 years ago, reflight of a first stage booster 10 times in 30 months sounded far fetched. One cycle time was only 38 days and that was second best for the fleet.

  8. I guess you could launch off the top of Mt. Kenya and get the benefit of high altitude as well as proximity to the equator. It’s a little shorter than Kilimanjaro, but more than 2deg closer to the equator.

  9. Funny thing is, Von Braun made these same mistakes in “The Mars Project,” 70ish years ago, categorically stating that his Mars fleet would leave from the space station, but could not return to it, that there was no orbit such that that could be done. Both of those things are patently wrong, as Von Braun should have known, since Euler and Lagrange had this stuff worked out (and published) in the 1770s. Falling from infinity starting at the ecliptic plane, you can reach any Earth orbit you want, so could return to the space station. And in any case, the EML points are going to be right where you left them, when considered as earth-orbiting “objects.”

    1. William, I’m pretty sure you’re mistaken on that. If the asymptote of the interplanetary orbit is not in Earth’s equatorial plane, but is instead at some angle phi from it, phi is the lowest inclination that can be reached with a non-lifting aerocapture.

      This image shows a zoomed-in view of Earth from Mars earlier today:

      If we assume a typical Mars-return trajectory has a view of Earth much like this, you can see that the sub-spacecraft point is well north of the equator. While you can choose which point on the horizon you aim for, you can see that no orbit inclination lower than the latitude of the s-s point can be reached without first capturing into an elliptical orbit, doing a plane change at high altitude, then aerobraking some more.

      Jon Goff’s analysis at gives more analysis of the inverse problem of departure.

      1. “Falling from infinity starting at the ecliptic plane, you can reach any Earth orbit you want, so could return to the space station. ”
        “William, I’m pretty sure you’re mistaken on that”

        Hmm, “if Falling from infinity starting at the ecliptic plane” {and you going to hit or get near hitting Earth- though with
        simple hohmann if coming from lower orbit, your orbital velocity is less, and if coming from higher orbit {returning from Mars to Earth] your velocity will faster than Earth’s orbital speed, and when get Earth’s gravity effect, if behind earth, the gravity increase your
        velocity even faster, making go further from the sun.
        Generally, without doing anything you going to miss hitting earth, and what you want is to closely “just” miss earth, it seem you can a choice of where on Earth you want to “just” miss earth. Or seems one can arrive bit earlier or later and miss close to either pole or in middle.
        But as far getting to orbit such where ISS is, there seems to be arguments about doing that. As in the “skipping off atmosphere kind of thing” but if want to spend weeks, months, one can spiral in {safely, or skipping atmosphere is considered risky and not tried}. Kind of like landing your first stage rocket to reuse it.

      2. Getting between ELEO and high inclination solar orbits is a problem that to some extent could be solved with Lunar flybys, as long as the velocity of the flyby isn’t too high.

      3. What is “non-lifting aerocapture?” Is that a ballistic entry from an incoming hyperbola? What would that look like? A terrifying plasma streak over Chelyabinsk?

        Humor aside, I was directly addressing Von Braun’s conclusion in “The Mars Project,” which involves nuclear thermal spacecraft making a propulsive return to LEO after a round trip to low-Mars orbit.

        Btw, anyone who hasn’t read “The Mars Project” should do so. It’s an important historical document. Make sure you get the blue paperback (a reprint from 1949) rather than the similarly titled science fiction book from the 50s.

        I like the guys at Selenian Boondocks and have read their interesting essays for many years, but in some ways they’ve become focused on aspects of space travel that are being rendered obsolute by Musk’s twin principles of cheap launch to LEO and propellant rich architectures. Aerocapture, momentum transfer tethers, space elevators, Aldrin cyclers, etc. etc. are simply not going to happen.

        I have to admit, I probably should avoid poetic language like “falling from infinity,” as the literal minded are going to assume it means no orbit shaping and no midcourse corrections.

        1. “but in some ways they’ve become focused on aspects of space travel that are being rendered obsolute by Musk’s twin principles of cheap launch to LEO and propellant rich architectures. Aerocapture, momentum transfer tethers, space elevators, Aldrin cyclers, etc. etc. are simply not going to happen.”
          Other than vertical takeoff and Landing, I think Musk has been following, Keep It Simple Stupid.
          And is true that Musk won’t use aerocapture?
          As it seems to me, that he has ship which could do it.
          Both for Mars use and returning from Moon to Earth.
          As for momentum transfer tethers and space elevators I always thought of them use pretty far in future type stuff rather short term use. And space elevator is to slow to get to orbit, but maybe one run power cable down one. But I never considered space elevators as pathway for CATS. The pathway of CATS is larger markets in Space {mining lunar water and Mars settlement OR simply time}.
          An Aldrin cycler, is interesting in terms of the fast ships which was suppose to dock to them- I never seen any details about this.
          I once said, NASA need depots in Mars orbit. Hmm, I was thinking mostly about lowering NASA Mars exploration cost and you could say, SpaceX will do that, and perhaps lower costs far lower and so not need the depots.
          I think if anyone want to compete with Musk in regards to being involved with Mars exploration program, using depots seems like a way to do it.

          1. An advantage of a Mars cycler is that 10 Starships each launching 3 times a day delivers people at the rate of 3,000/day and 90,000/month for the duration of your launch season. That is the equivalent of 30 Starships, which isn’t that hard for SpaceX to do but with 30 Starships launching 3 times a day, they could do 9,000/day and 270,000/month for the duration of the season.

            Of course you need Starships on both ends but maybe there will be a tipping point when a cycler makes economic sense or maybe by then, something new will have changed everything as much as the Falcon 9 did.

        2. Wodun, you’re describing a Mars cycler the size of an O’Neill colony.

          While I think such large habitats will be practical in this century (albeit far outside my lifetime, barring miracles). I think they’re much more likely to be built in cislunar space and “towed” to their operation site (Jovian Trojan asteroids, for example).

          I think in that latter day, interplanetary travel will be via “liners” model on the NautilusX, but much larger (some even number of rotating habitats strung on a spine, hosting maybe 5,000 passegers).

          The original Mars cycler idea was a way to do credible Mars exploration with OldSpace technology. Say you have Ares V expendables, 6-crew Orion capsule, and partially reusable Mars landers. You build an ISS class cycler in transit, able to host 18 crew, 9 of whom are Mars explorers. At the Earth end of the cycle you debark explorers and crew (and Mars samples) and embark more of same, along with supplies and equipment, so there are always 9 explorers on Mars and 9 cycler crew orbiting the sun, maintaining the cycler, refurbing landers, and doing a bit of space station science.

          As we keep noting, Starship will change all that. Ultimately, what we do where will be determined by the shape of cislunar space. I was in L5 along with probably alot of other oldsters.

          1. Oh, and what made it economical was, the spacecraft launching from Earth and Mars stayed with the cycler, rather than having to turn around and go back. The cycler was so you crew didn’t have to fly to Mars in a Winnebago (Spaceballs notwithstanding).

  10. We need to make a distinction between cargo and crew. Cargo would love ELEO for all of the reasons listed in this post. But most all passengers would hate ELEO. ELEO has very few views. By contrast, a decently inclined orbit would be able to view most all of the spectacular views including one’s home country. Everyone going to the Moon or Mars would love to die s a few weeks in a hotel viewing Earth so long as it was in an adequately inclined orbit. I imagine that practically all settlers would want to take this path. For that matter, I imagine that most Mars colonists would like to tour the Moon as well.

    1. “The Van Allen belts are most intense over the Equator and are effectively absent above the poles. No real gap exists between the two zones; they actually merge gradually, with the flux of charged particles showing two regions of maximum density. The inner region is centred approximately 3,000 km (1,860 miles) above the terrestrial surface. The outer region of maximum density is centred at an altitude of about 15,000 to 20,000 km (9,300 to 12,400 miles), though some estimates place it as far above the surface as six Earth radii (about 38,000 km [23,700 miles]).”

      “The Americans conducted a series of nuclear tests in the 1960s called Operation Dominic. It included a group of atmospheric tests called the Fishbowl events, designed to understand how nuclear weapon debris would interact with Earth’s magnetic field in the event of a nuclear war.
      The highest of the Fishbowl events was called Starfish Prime – a 1.4-megaton nuclear bomb detonated at 400 km on July 9, 1962.
      However, instead of clearing the inner Van Allen belt, it actually added more radiation to it. The Soviet tests in the same year increased the intensity of the inner belt by a million-times, and also destroyed several satellites that were already there.

      One of them was Telstar 1, which had been launched just the day after Starfish Prime. It relayed the first TV pictures, fax images and the first-ever transatlantic TV feeds. By October, the increased radiation from the Soviet test had burnt Telstar‘s transistors and it went out of service in 1963.

      The high-energy electrons injected into the lower Van Allen belt had decayed to one-twelfth of their post-test peak intensity only by 1969. …
      ….electrons below about 1 MeV were unlikely to be dangerous, as were protons below 10 MeV. For example, a proton with an energy of 3 MeV could penetrate
      about 6 mm of aluminum (a typical spacecraft material) whereas one of 100 MeV could penetrate up to 40 mm. So engineers fashioned shielding that consisted of a spacecraft hull and all the instrumentation lining the walls.

      Further, knowing the belts’ absence above the poles, the altitude of the lower edge of the inner belt being ~600 km (well above the LEO) and the location of the South Atlantic anomaly, where doses are at a high 40 mrads/day at an altitude of 210 km allowed NASA to design the Apollo translunar injection (TLI) orbit in a way that the spacecraft would avoid the belts’ most dangerous parts.”

      So what I looking was how high Van allen belts were at equator orbit, and didn’t really find answer, though above has diagram indicated strong field, and I wildly guess it seems higher than 2000 km- though I would imagine solar storms could push down further {I assume for shorter period of time} I would imagine if not in
      stronger field, one would get more radiation shielding from the more intense field.
      And so perhaps, if had a say 200 km by 2000 km equatorial orbit, when at 2000 km then one could get a better view of more of Planet Earth.

      1. More stuff:
        “Figure 37: At the highest electron energies measured — above 1 MeV — researchers saw electrons in the outer belt only (image credit: NASA/GSFC, Duberstein)”
        “Figure 38: The radiation belts look much different at the lowest electron energy levels measured, about 0.1 MeV. Here, the inner belt is much larger than in the traditional picture, expanding into the region that has long been considered part of the empty slot region. The outer belt is diminished and doesn’t expand as far in these lower electron energies (image credit: NASA/GSFC, Duberstein)”
        Looks like place without the 0.1 MeV is equatorial low orbit.
        Figure 39: During geomagnetic storms, the empty region between the two belts can fill in completely with lower-energy electrons. Traditionally, scientists thought this slot region filled in only during the most extreme geomagnetic storms happening about once every 10 years. However, new data shows it’s not uncommon for lower-energy electrons — up to 0.8 MeV — to fill this space during almost all geomagnetic storms (image credit: NASA/GSFC, Duberstein)”
        Again during storm only place which does have up 0.08 MeV
        is Equatorial LEO
        So, it seems early diagram mentioned last post was wrong/or misleading. Or with a modest amount shielding one should able to go higher than 2000 km, though idea the intense region near equator providing more shielding {from say, GCR} appears to be wrong. Or I imagine generally, further one gets from the mass of Earth, the more GRC one gets. But going to Moon or somewhere, you would have some shielding for it. And issue is in terms weeks/months rather hours.

        1. Mistyping.

          “Again during storm only place which does have…” should have been:
          “Again during storm only place which doesn’t have up 0.08 MeV, is the Equatorial LEO.”

          Anyhow, I wonder if can go high enough without radiation issue to do the Jon Goff’s analysis, type thing.

    2. ELEO will be a waypoint for people in transit to other locations, including higher inclinations. For those going to the latter, they’ll get to see the earth from a great distance before arriving at the LEO destination.

      1. Think of ELEO as a hub–the DFW of space. It’s faster to fly direct, but you have a lot more options and lower cost if you go through a hub.

  11. Until there is substantial cargo traffic from the Moon to the Earth that is worth shipping propellant to the Moon to bring back, almost all trips to the Moon should be one-way trips. Especially if you are delivering pressurized habitation space; Instead of 100t of inflatables, you get 100t of inflatables as cargo + an 85t Starship that has life-support capability for 100 people, solar panels, etc. Even returning just the engines from early flights while they still cost $1m each probably is not best, at 1.5t that means Moon return cargo has to drop to less than $700/kg which will not happen until propellant on the Moon is cheap and plentiful. If a Starship costs $10 million to build at first, that’s only $118 per kg of Starship.

    The same is true for sending cargo to Mars.

  12. Another aspect of this: Orbital Traffic Control, which Rand touched on briefly when he mentioned the ease of keeping the orbital hub in one plane and thus collapsing a 3D problem into a 2D problem.

    Air Traffic Control for commercial air traffic is a complex web of air transport scheduling and real time control. As I understand it, the reason we ended up with hub-and-spoke airports was driven by of all things a *software* algorithm, known as the Simplex Algorithm which was one of the first programs written to solve the “Traveling Salesman Problem” which aligns quite well with the airline scheduling problem and is a linear problem solver.

    Here I suppose we have similar issue EXCEPT I’m thinking the “spokes” on the ELEO hub orbit vary in size over time, unlike their ground-based counterpart airports. Thus there is also a time variance that has to factor into the scheduling algorithm. Not all orbits are achievable at any given time if you want to minimize things like fuel consumption (ie cost). I don’t know if this helps or makes Simplex worthless for this application.

    Also probably will want to have really good imaging RADAR satellites in higher orbits (GEO?) to track the traffic. Stacking in the hub orbit might be a 2D problem but routing to a spoke orbit will still be a 3D problem just as it is today, but the flight paths will be arcs not the pseudo line tracks that ATCs deal with today because the Earth’s curvature can be ignored for this purpose.

    Unless you presume most non-ELE orbits are direct and not transfer. But then if you are scheduling cargo and passengers in space traffic that dwarfs today’s imagination, the simplicity of routing through ELEO may be unavoidable as Rand points out.

    Let’s also not forget those sub-orbital P2P arcs passing in and out of the top of the atmosphere as well.

    Could be another book on how to do OTC here…. *hint*

  13. A few thoughts:

    Re: orbital traffic control, usable ELEO is only about 2000km deep (air at the bottom, inner Van Allen at the top), and that’s right where the LEO comsat contellations will be. We might be talking hundreds of thousands of satellites (at least).

    In that context, how large are, at a minimum, the depots likely to be? A hundred Starships of current design need 120,000 metric tons of fuel. If Musk’s profesied Synodic Mars Fleet is a thousand ships, will it need ten such depots?

    Secondly, it may be cheaper to simply do away with depots and/or tankers entirely, by simply building a TSTO resable LV whose payeload is a fully fueled 1500 ton starship. And Musk has at least once mention an 18 meter diameter for a SuperHeavy follow-on. About the right size?

    Back at the beginning of all this, when Musk announced Mars Colonial Transport, I visualized a 3-stage vehicle that could lift off from Earth, fly to Mars, drop off a hundred tons of whatever, and fly back, no ISRU or tankers needed. It was a bridge too far, so to speak, but maybe not for the next generation.

    1. usable ELEO is only about 2000km deep (air at the bottom, inner Van Allen at the top), and that’s right where the LEO comsat contellations will be. We might be talking hundreds of thousands of satellites (at least).

      I think you might be a little overly pessimistic here. Starlink constellation is operating in a narrow range of 300-580km orbits. From my own personal off-and-on observations of the Starlink system using the webesite I referenced in the “Starlink” thread (about two weeks back if you search this blog) new satellites orbit at about 380km until they get to their proper spacing and orbital alignment and then rise to about 580km to give a larger ground footprint. They stay more or less stable at that altitude until their batteries indicate end-of-life approaching. At that point I presume they begin a decent back to the 300km orbit perhaps to extend battery life at the closer range but eventually are finally de-orbited. The system is too new for me to see satellites being decommissioned by tracking any particular one, although the website IIRC has tracked early sats that weren’t in service I suppose most if not all those have been de-orbited. So I don’t know what is considered reasonable spacing in LEO but the plane crossings at least for Starlink are only in a range of approximately 280km or ~1/10 of the available orbits. Of course that’s not all the LEO satellites either. And within that 200km band it’s getting more dense, almost by the week. Certainly by the month. So you have the issue of what’s usable, above, below and within the band.

      Your 100k’s of satellites presumes maybe 3 to 4 competitors to Starlink. I think that may be optimistic. You’d need either a system comparable to F9 launchers, with only BO on the distant horizon there or companies contracting from SpaceX, which might be forced to launch their payloads to avoid anti-trust. It’s gonna be an interesting century.

  14. Musk has said he envisions the final Starlink constellation involving 47,000 satellites. Maybe China builds another one that size. Maybe Europe puts up a smaller one, maybe other players (who knows?)… Anyways, we’re talking intermediate far future (post 2050ish?) for Rand’s ELEO vision? For years, discussion groups talked about the “killer app for space.” That would be a description of Starlink. It wasn’t unknown before Musk came along (including at least one big failure), but Musk supplied the enabling technology (cheap launch, at a minimum). Everyone stuck on DSL, GEO broadband, cellular, or dialup (!) is going to move to Starlink. In time, only urban and dense suburban and some exurban will stay on cable (etc.). Time will tell. As for me, I’m looking forward to having broadband 60 times faster than my current rural DSL (which only came in 5 years ago, before that I had HughesNet, ten times slower still).

    1. “… Anyways, we’re talking intermediate far future (post 2050ish?) for Rand’s ELEO vision?”
      Interesting question.
      I was sort of thinking +2025

      What could happen before 2025.
      An operating Starship must happen before 2025 or I think some else will done instead. Starship has proven not to work, for some reason.
      But tend to think it is shown to work before the end of 2021. Though this happens in somewhere in 2022. Then it seems Starship will be improving, though probably not a dramatic as Falcon-9 did.
      Or we can probably mostly ignore things done to make it work better, it can expect but not going fundamentally different, unless does not work {as above}. And this means rather than late, lunar exploration could be earlier, and maybe better than what hoped or expected NASA would do.
      Let’s see when next Mars window, Sept 2024. So some number of Starships will leave before this. I guess it’s one, and NASA may not buy it. And I rule out NASA buying 3 or more which go to Mars.
      So leaves starships for Moon. And how many would NASA buy before the end of 2024. Or not counting refueling. But lets just ask how many operating starships by end of 2024. 3 lunar starships and 5 starships? And wants mass production of them by the next window- 2026 Nov.
      As understand it, Musk going to do ocean launch near Texas- seems unlikely that going to happen before 2026.
      So equator anything, is probably +2027
      And no one else could do it before 2027.
      But I expect other parties are already trying to catch up with Falcon-9 and in couple years from now work something relating to Starship.
      It seems equator launch will start before 2030.

      1. Musk’s statement has been to double the Mars fleet every synod, so 2 in 2024, 4 in 2026, 8 in 2028, on the current schedule. I still think the equastorial launch will be delayed by not only the need for space traffic control (look-down telescopes), but also the lack of launch sites on the equator. There’s Kourou and Alcantara, both limited, but with resorces such that SpaceX could come in (ITAR notwithstanding). Sticking to east coast sites for now (otherwise no rotational benefit), Kenya, Malaya, Borneo and some tiny islands. Until we have inland launch sites, that’s a bottleneck, even for non-ELEO and BLEO flights. Considering an 18 meter HyperHeavy, Jackass Flats, Semipalatinsk, Novaya Zemlya (Arctic, no Van Allens!)… Von Braun wanted to set up his launch complex at Johnston Island. Equatorial, and if it’s good enough for atomic bomb testing… His three-stage reusable rocket, fueled with turpentine and white nitric acid, was 4x the weight of a Saturn V with a 20,000 pound payload to LEO. He anticipated 900 launches to build the space station and Mars fleet.

  15. More on orbital traffic control: a 120,000 metric ton depot isn’t going to be doing much fancy manuevering if it needs to dodge an out of control comsat or even a 5,000 metric ton orbital hotel full of honeymooning millionnaires. Imagine the headlines? Better yet, imagine standing in your backyard, watching it happen overhead!

  16. Launching from a high elevation above sea level can bring even greater advantages to rocket performance than launching from an equatorial location.

    There are a few places on the Earth which combine both advantages. Near to Quito, the capitol city of Ecuador, is one such location.

    1. There are LNG ships which transport 120,000–140,000 cubic meter liquid Methane. Which also could transport LOX. And if launching 3 Starship a day you going to need such volume {which can be cheaply shipped to launch sites in the Ocean]. With a mountain launch site, you going to need pipelines. Which could be done. But everything can be shipped fairly cheaply by ocean.
      And by using floating breakwater one could build airport on Ocean.
      So can have conventional and sub-orbital air travel to ocean spaceport.

      1. Musk has talked about making methane and lox at the launch site from atmospheric gasses. More expensive than pipelines and tankers, but maximum convenience. He already showed a version of this when he set up a lox plant at Kwaj.

  17. I also wonder if crewed LEO as a whole will eventually turn out to be a spent novelty. Views from LEO are not a lot more spectacular than views from airliners. When I was young, I insisted on a window seat, but later on, on tedious business trips I wanted the aisle for comfort and covenience. As a middle aged man I was tall and fat, so I wanted to be able to lean into empty space. Of course, one day, flying from east coast to west, I fell asleep in my aisle seat and my hand fell to the floor. I woke up when the stewardess ran it over with the drinks cart.

    It still makes sense to me to imagine a time not many decades down the road when fully fueled starships are delivered to orbit and lug their passengers to large, rotating habitats at the EML and SEL points, where they board liners for the other planets, moons, and asteroids, and their associated large, rotating habitats.

    1. The view.
      I am not very interested in going to space.
      The value of view to me, is it lives with you forever- I like my views.
      But at moment, I don’t like idea, that I might get sick. If that is not fixed, maybe people will go because they want to get sick.
      Microgravity seems like it could be quite fun {if don’t get sick}.
      And swimming and flying in space might better than Earth. Mars has to have flying. Hotels must have swimming pool. Big caves on Moon and Mars might worth the trip.

      1. I have a background history with nuclear submarines (as a shipyard mechanic, not crew) so I have an attraction to the idea of being a techie crewman on a long-haul spacecraft. The Moon and Mars are kind of drab. the one dark gray, the other like a dead version of Arizona. But I was always attracted to the potential beauty of the Saturn system. I published a story in Asimov’s about 20 years ago called “Down in the Dark,” about a repairman attached to an expedition to Titan stranded there when Earth civilization is destroyed by a large asteroid strike. My guesses about Titan weren’t that far off, though I did erroneously imagine the seas would be covered by a layer of hydrocarbon sludge.

        1. A boat that goes forever.
          When will someone take one and what type of boat will it be?

          The oldest creatures in the world are the bottom of a big dark sea.
          If googled:
          “Sponges are some of the simplest and most ancient of animals, though they don’t look like animals as we usually know them.
          Large ones provide ecosystem services such as filtering seawater, recycling nutrients on reefs and providing habitat for other species, and are estimated to be able to live for more than 2300 years.”

          No one wants to be a sponge, no gravity or low gravity is considered to problem that we might not overcome.
          We live on planet which is giant fission nuclear reactor.
          Is the earth a giant nuclear reactor?
          Some say yes and some say no. And so think we need a certain amount of radioactivity {a some low level}.

          But no gravity or low gravity, and no radiation or some radiation might related to living forever.
          Does the human beings need war- in terms of microbial war that we are in?

Comments are closed.