Space Settlement An Easier Way

Space Settlement: An Easier Way

Al Globus, San Jose State University
Stephen Covey, DSI, Inc.
Daniel Faber, DSI, Inc.
July 2017
Copyright © 2015, 2016, 2017 Al Globus
“Come with me if you want to live,” Kyle Reese in The Terminator, 1984
Abstract
To survive in the long run, we must settle beyond Earth. We are taking the first steps now,
but there are major problems. Lunar or Martian settlements will be very far away and low
gravity is a serious issue for children. Free-space settlement designs have typically been
kilometer-scale spacecraft weighing millions of tons, requiring both large scale space
construction and extraterrestrial mining and materials processing infrastructure before the
first settler moves in. To utilize extraterrestrial materials these designs are typically located in
orbits very far from Earth, at least as far as the Moon. Being so massive and distant makes
construction impractical, at least for a long time to come.
Things are about to change. Two recent discoveries regarding space radiation and human
rotation tolerance suggest that the first free-space settlements may have a mass measured
in kilotons, rather than megatons, with dimensions around 100 m rather than half a kilometer
or more. First, space radiation computations suggest that Earth orbits below about 500 km
and close to the equator have radiation levels so low that little or no radiation shielding is
required [Globus 2017a]. Second, a careful examination of the literature suggests that
permanent settlers can tolerate much higher rotation rates than was commonly thought,
allowing much smaller settlements to provide 1g artificial gravity [Globus 2017b].
Between these two studies, the mass of early free-space settlement designs can be
hundreds of times less than previously believed and they can be located hundreds of times
closer to Earth, vastly simplifying construction and logistics. Furthermore, extraterrestrial
mining is no longer required to build the first settlements because of low mass and new,
large reusable rockets under development. Also, the first space settlements may be very
similar to large, advanced space hotels. Such facilities, in turn, may be developed
incrementally from smaller, less sophisticated hotels and stations starting with the ISS
(International Space Station), perhaps even at a profit. If the results of this paper are
confirmed by further research, a herculean task requiring monumental effort will instead
become a difficult but surmountable engineering challenge.
Acronyms
delta-v change in velocity
g gravity, 1g corresponds to gravity at the surface of the Earth
ELEO Equatorial Low Earth Orbit
GCR Galactic Cosmic Rays
HEO High Earth Orbit
ICRP International Commission on Radiological Protection
NSS Space Settlement Journal, July 2017 Page 1

ISS International Space Station

ITS Interplanetary Transportation System
kg kilogram
L2 Lagrange Point Two
L5 Lagrange Point Five
LEO Low Earth Orbit
m meter
mGy milli-Gray
MPA megaPascal
mSv milli-Sievert
NEO Near Earth Object
rpm rotations per minute
SSP Space Solar Power
T metric ton
Introduction
This paper is about the very first stages of life’s growth throughout the solar system and
beyond. It is about a radically easier, perhaps the easiest, path to the first space settlements.
For the purpose of this paper, a space settlement is a home in space where people go to
work, live, and, for those who wish, raise their kids. Raising children onboard distinguishes
space settlements from space stations, where astronauts go to work, and space hotels,
where people go to play.
In the 1970s a series of studies at Stanford University and NASA Ames Research Center led
by Princeton University physicist Dr. Gerard O’Neill suggested the feasibility of building and
living in free-space settlements located at the Earth-Moon L5 point or other orbits [Johnson
1975] [ O’Neill 1977]. Free-space settlements are essentially gigantic spacecraft in orbit, big
enough to live in. L5 is a point on the Moon’s orbit equidistant from Earth and the Moon.
These studies produced, among other things, the Stanford Torus design, a structure almost
two kilometers across, rotating at 1 rpm (rotations per minute) to provide 1g artificial gravity
at the rim.
There is a great deal of radiation in space from which settlers inside the Stanford Torus
would need to be protected by millions of tons of lunar regolith. This material would be
provided by a lunar mining operation involving an electromagnetic catapult on the Moon and
a catcher at L2. Progress towards this dream has been slow in large part because the
system is so big (kilometer scale with a mass of millions of tons) and it requires an extensive
extraterrestrial industrial base to take advantage of lunar materials before the first settler can
move in. These materials are hundreds of thousands of kilometers away from Earth.

Building and operating such space settlements is a massive challenge, worthy of a great
civilization. However, to date it has been too great a challenge to undertake with the
resources available. We propose a three pronged approach to make the first few settlements
much easier to develop:
1. Make them close.

2. Make them small.
3. Build up incrementally, ideally with profit at each step.
Building the first settlements close to home makes everything much easier. For the purpose
of this paper, closer means Equatorial Low Earth Orbit (ELEO). Equatorial Low Earth Orbit
refers to orbits close to Earth and at all points close to the equator 1. This is much closer than
the Moon, asteroids or Mars. A 500 km circular orbit is 760 times closer to Earth than the
Moon or L5 and 100,000 times closer than Mars at closest approach 2. Transportation cost,
time and risk are accordingly lower. Unhappy settlers can return to Earth in a few hours.
Telephone calls and video conferences can work since communication delays are short.
New settlers can get to their new home relatively easily. Evacuation is at least theoretically
possible if disaster strikes. The importance of distance is hard to overstate.
A key benefit of ELEO settlement is vastly reduced radiation levels [Globus 2017a]. The
mass of previous free-space settlement designs is dominated by radiation shielding.
Eliminating radiation shielding reduces the mass of cylindrical designs by a factor of at least
19 and toroidal designs by a factor of about 155 [Johnston 1975] (see Table 1). As we will
show computationally, settlements in ELEO may need very little or no dedicated radiation
shielding as they will be protected by the Earth’s magnetic field and the mass of the Earth
itself.
1
This is a region with far less radiation than other orbits, or the surface of Mars or the Moon.
2
A note on numbers: there are a lot of quantitative results in this paper. Almost all of them have been
rounded off as excessive precision in the data presented here is pointless and misleading. Thus, you
should expect to find many small, non-consequential inaccuracies.

Stanford/NASA Ames Mass Estimates
Mass model
based on 1975
study all mass values in metric tons
factor by which the
mass is reduced when
radiation shielding is
shape structure air shielding mass not necessary
multiple
dumbbells 75,000 37,000 9,900,000 89
multiple torus 100,000 10,400 9,700,000 89
banded torus 112,000 13,200 7,000,000 57
single torus 4,600 1,900 1,000,000 155
cylinder 775,000 299,000 19,400,000 19
sphere 64,600 35,200 3,300,000 34
dumbbell 400 200 1,400,000 2,334
Table 1: Mass estimates from [Johnson 1975]. The first column lists various shapes a
settlement could have. The next three are the estimated structural, air and radiation
shielding mass in tons. The last column was added by the authors to indicate the factor by
which the mass is reduced when radiation shielding is not necessary.
This huge reduction in total mass compensates for the greater energetic difficulty of
launching materials from Earth to ELEO as opposed to launching from the Moon to L5, the
design location of the Stanford Torus. In the early studies, the Earth-Moon L5 point was
chosen as the location of a settlement for the energetic advantage of launching materials
from the Moon. Going from the Moon to L5 requires a delta-v 3 of 2.3 km/sec, and going from
Earth to 500 km ELEO requires 10 km/sec [Cassell 2015]. Using the velocity squared as our
energy measure 4, Earth to ELEO requires 19 times more energy per unit mass. But Table 1
suggests that at least 19 times less mass is needed if no radiation shielding is required.
Thus, the energetic advantage to launching the mass of a settlement with deep space
radiation shielding from the Moon to L5 is balanced by launching far less mass from Earth if
no radiation shielding is necessary.
3
delta-v is change in velocity necessary to move from one orbit (or surface) to another.
4
One could also use the rocket equation to make the comparison, but that makes the result strongly
dependent on ISP and is much more complex. At high ISP (e.g., LOX/H2) the difference compared to
delta-v squared is not great.

Making the first settlements small directly reduces construction difficulty. A village is easier
to build than a city. The faster a settlement rotates, the smaller it can be and still provide 1g
of artificial gravity at the rim, as needed for children to grow up strong [Globus 2017b]. As we
will show from an examination of the literature, settlers can probably tolerate 4 or perhaps
even 6 rpm [Globus 2017b]. For the purpose of this paper we assume, based on the
literature, that a diameter of around 112 m (corresponding to rotation at 4 rpm for 1g of
artificial gravity) is adequate for settlement, as opposed to the 1790 m (1 rpm) or 450 m (2
rpm) diameter found in earlier designs. Since the mass of a torus or cylinder-shaped
settlement scales somewhere between the square and the cube of the diameter, increasing
the spin rate results in an enormous reduction in the total amount of material needed.
Building up capability incrementally is a time honored engineering tradition because it works.

For example, the 1969 Moon landing was built up in stages:
1. one man suborbital flight.

2. one man orbital flight.
3. two man orbital flight.
4. docking and space walks.
5. three man orbital flight.
6. circumnavigation of the Moon.
7. “That’s one small step for (a) man, one giant leap for mankind.”
Similarly, we propose starting with the International Space Station (ISS), then building more,
better, and larger revenue-generating space stations and hotels over time until the sizes
approximate that of a 4 rpm settlement (112 meters diameter). The economics of space
hotels should encourage operators to develop the life support to keep the crew and guests
breathing, drinking and eating with minimal resupply from Earth. Eventually, building and
operating the first space settlement may not be much more difficult than building another
hotel.
This is economically difficult using current expendable launch vehicles, which cost at least
$1,650 per kilogram [SpaceX 2017] from Earth to LEO and tens of millions of dollars to put a
passenger in orbit. However, SpaceX is developing the ITS—Interplanetary Transportation
System, a two-leg Earth to Mars transportation system based on large fully reusable vehicles
[Musk 2017]. Target costs are $140/kg and $200,000 per passenger to Mars. The first leg of
the ITS is to LEO and these costs should be significantly lower. If SpaceX is able to even
approximate the target costs, building small space settlement in ELEO may be affordable.
This paper will compare slightly modified 5 versions of two published space settlement
designs, the Stanford Torus [Johnson 1975], a 1 rpm design, and Kaplana One [Globus
2007], a 2 rpm cylinder, with versions having no radiation shielding and rotating at 3 or 4
rpm. Moving these designs to ELEO, making them as small as we dare, and launching all
materials from Earth results in a system mass two or three orders of magnitude smaller,
which will make such designs more practical.
5
We will assume modern materials and radiation shielding requirements, thus focusing on the
differences caused by location in ELEO and higher rotation rates.

A Tale of Two Studies

Our approach may have a chance to work because two recent studies strongly suggest that
free-space settlements in ELEO can be far less massive than previously thought. These
discoveries combined with improvements in launch vehicles necessary for any space
settlement to succeed may allow space settlements to be built with components launched
from Earth. This vastly simplifies the construction and operation of the first space
settlements. We will examine these studies in greater detail below, but first we take a high
level look at the results of each study and the implications for early settlement size.
The first of the two studies used sophisticated NASA radiation modeling software [OLTARIS
2011] to show that a settlement in ELEO below about 500 km may require little or no
dedicated radiation shielding, based on radiation limits for settlements derived from an
extensive literature search [Globus 2017a]. The Earth itself and the Earth’s magnetic field
protect this area from most space radiation. Staying close to the equator avoids passing
through the South Atlantic Anomaly so radiation levels are much lower than for the ISS.
Since radiation shielding is the vast majority (typically over 95%) of the mass of previous
space settlement designs, and since total mass is a fairly good proxy for difficulty, placing
settlements in ELEO makes construction of the first settlements much easier.
Figure 1 provides some ISS experimental data on radiation levels in LEO. Notice that a
spacecraft in orbit near the equator will only encounter blue areas on the map,
indicating low radiation levels. We will quantify these levels computationally below.

Figure 1: Radiation measurements taken on the ISS. Note the high levels of
radiation over the South Atlantic and much of South America and very low levels
near the equator. The ISS orbit is around 400 km, somewhat below the altitude of
the computational data presented in this paper. Image credit NASA.
The second of these studies examined the literature on human tolerance of rotation [Globus
2017b]. Rotation is used in free-space settlement designs to provide something similar to
Earth-normal gravity to the settlers. In the earlier designs it was assumed that a rotation rate
of no more than 2 rpm is advisable as rotation can make people ill. Limiting rotation to 2 rpm
means that the diameter of the settlement must be at least 450 m to achieve Earth-normal
artificial gravity. Careful examination of the literature finds little basis for the 2 rpm limit.
While people do get sick when rotated, they adapt quickly and there is an effective training
regime. In fact, the literature suggests that 4 rpm (and perhaps more) is fine for settlers,
requiring only a few hours to a day or so of adaptation or perhap six hours training. Thus, a
112 m diameter is acceptable (4 times less than Kalpana One and 16 times less than the
Stanford Torus).
Note that the mass of a settlement design roughly scales by the inverse of the fourth to fifth
power of the rotation rate. Thus, cutting the rotation rate in half leads to a roughly 16-32 fold
reduction in mass, all else being equal. Why?
From the equation for centripetal acceleration

𝛼 = 𝜔2 𝑟
𝛼 is acceleration
𝜔 is rate of rotation
r is radius
the radius scales as the inverse square of the rotation rate given constant acceleration. The
mass of a settlement scales with the radius as follows:
● Hull mass scales as the cube
○ the surface area scales as the square and
○ the required strength of a pressure vessel 6 scales linearly
● the atmospheric mass scales as the cube
● the furnishings in the 1g living area scale as the square
Thus, the furnishing mass scales as the inverse fourth power of the rotation rate and
everything else scales as the inverse fifth power of the rotation rate.
Combining the mass reduction of eliminating the radiation shielding (19 times for Kalpana
One and 155 times for Stanford Torus, as shown in Table 1) and increasing rotation to 4 rpm
(a mass reduction of 16-32 times for Kalpana One and 64-128 times for the Stanford Torus),
we get a total mass reduction of 300-20,000 times. The large difference is because the
Stanford Torus is 1 rpm and Kalpana One 2 rpm. To check these analytic results, we have
built two simple models of settlement mass, one for cylinders and another for tori.
6
The settlement must hold an atmosphere!

Model comparisons of the mass of settlements similar to Kalpana One and the Stanford
Torus at various rotation rates with and without radiation shielding are found in Tables 2 and
3, and are very roughly consistent with the analytic results. Note that the models are not
quite the same as the original Kalpana One and Stanford Torus to make the comparisons
with smaller designs more direct. This is particularly noticeable for the population size, which
is quite a bit higher than in the original papers. A description of the models may be found in
Appendix B and the models themselves at
http://www.nss.org/settlement/journal/EasyModel.xls.
Kalpana One
Figure 2: Kalpana One. Image credit Bryan Versteeg, used by permission.

Kalpana One Mass Estimates

Measure Shielding No Shielding No Shielding
rotation (rpm) 2 3 4
radius/height (m) 224 99 56
Hull (kT) 50 3 0.5
Internal Struct (kT) 63 12.4 4
Radiation shield (kT) 3,870 0 0
Non-struct (kt) 54 10.9 3.4
Air (kt) 42 3.7 0.7
Total Mass (kT) 4,100 30 8.5
7
Falcon Heavy Launches 77,000 560 160
8
ITS launches 14,000 100 28
Mass ratio 1 136 480
Population 7,850 1,550 490
Mass/person (T) 519 19 17
2
drag mass/area (T/m ) 22 0.809 0.732
deorbit time 500 km (years) NA 166 150
Table 2: This table examines the effect of mass and size as the modified Kalpana One
design increases rotation rate, reducing radius, and sheds its radiation shielding. The
numbers in red are the reduction factor for mass. Flat end caps are assumed. The
population row assumes 1g residence area of 40 m2/person, roughly the floor area per
person of a smallish suburban house. Mass ratio compares the mass of each option to that
in high orbit at 2 rpm. Orbital lifetime calculations (last row) are based on [Panwar 1999] as
implemented by the Satellite Orbital Decay web site for a 500 km circular orbit
(http://www.lizard-tail.com/isana/lab/orbital_decay/). The cross section is based only on that
of the settlement, not solar arrays or thermal radiators. That will be the subject of a future
study and the time will go down accordingly.
The original Kalpana One design was a 2 rpm cylindrical settlement. Thus the diameter was
about 450 m. The original paper had a cylinder height of 325 m but for the purpose of the
calculations in this paper the cylinder height is equal to the radius. Table 2 suggests that if
we shrink Kalpana One down to 112 m diameter (4 rpm) and eliminate the radiation
shielding by placing it in ELEO, the total mass of the system is reduced by a factor of about
480, consistent with our analytic results. The population is also reduced by a factor of about
16. Reductions are even larger if we go to 6 rpm but it is not at all clear that a 50 m diameter
home can be a viable settlement. Indeed, now that it seems higher rotation rates are
possible, the minimum size of a settlement is probably determined by other factors, perhaps
psychological or social issues. This is an area that needs investigation.
7
The Falcon Heavy is a launch vehicle under development by SpaceX with a projected maximum
payload mass to LEO of 54 tons [SpaceX 2017].
8
ITS is a very large launcher under development by SpaceX with a projected maximum payload to
LEO of 300 tons in fully reusable mode [Musk 2017].

The last two rows of Table 2 (and Table 3) address a major safety issue with large LEO
satellites. If they are abandoned they will eventually enter the atmosphere and impact the
ground. We see that the time to deorbit, even if an ELEO settlement is completely
abandoned, is so long (over 166 years for a cylinder) that any reasonably capable
spacefaring civilization can easily address the threat. Thus, the only realistic situation in
which an ELEO settlement would enter the atmosphere is if our civilization collapses
completely. In this case, space settlements falling to Earth somewhere along the equator will
be the least of our worries.
Stanford Torus
Figure 3: Stanford Torus, image credit NASA.

Stanford Torus Mass Estimates

Measure Shielding No shielding No shielding
Rotation (rpm) 1 3 4
Outer radius (m) 895 99 56
Inner radius (m) 65 10 7
Hull (kT) 67 0.228 0.08
Internal Structure (kT) 145 2.5 0.98
Radiation Shield (kT) 12,740 0 0
Non-structure (kT) 127 2.1 0.86
Air (kT) 82 0.2 0.06
Total Mass (kT) 13,300 5.1 2.0
Falcon Heavy Launches 250,000 96 38
ITS launches 44,000 16 7
Mass ratio 1 2,600 6,650
Mass/person (T) 725 16.5 16.3
Population 18,000 310 123
drag mass/area (T/m2) 15

0.325 0.311
deorbit time 500 km (years) NA

67 63
Table 3: This table examines the effect of mass and size as the modified Stanford Torus
design increases rotation rate, reducing radius, and sheds its radiation shielding. The
numbers in red are the reduction in mass. The population row assumes 40 m2/person for a
residence. Orbital lifetime calculations (last row) are based on [Panwar 1999] as
implemented by the Satellite Orbital Decay web site for a 500 km circular orbit
(http://www.lizard-tail.com/isana/lab/orbital_decay/). The cross section is based only on that
of the settlement, not solar arrays or thermal radiators. That will be the subject of a future
study and the time will go down accordingly.
The Stanford Torus is a 1 rpm toroidal design with a 130 m inner diameter. The inner
diameters for the higher-rpm calculations are less. Because we are starting at 1 rpm
(Kalpana One was 2 rpm) and because a torus needs more radiation shielding for the same
living area, eliminating the shielding and increasing the rotation rate leads to an enormous
mass reduction. Note that the total mass of the 4 rpm version could be launched from Earth
with about seven ITS launches.
We will examine the results of the radiation and rotation studies that drive this enormous
improvement in the ease of building the first space settlements more carefully. But first we
must explain why ELEO is an easier target for early settlement than the Moon or Mars.

The Moon and Mars

We know empirically that the Moon and Mars will likely be much harder to settle than ELEO
because of our experience with space stations. The first space stations went into LEO in the
1970s. The ISS is up there now (Spring of 2017) and has been continuously inhabited since
October of 2000. No stations have been built on the Moon or Mars not because no one
wants them, but because they are much harder to access than LEO. This is primarily
because a LEO orbit (assuming 500 km) is about 760 x closer to Earth than the Moon and
100,000 x closer than Mars at closest approach. This is a massive logistical advantage for
LEO.
Note that there are at least three relevant ways to compare the difficulty of travel from Earth
to places in space: distance, delta-v, and time. Table 4 provides these for the places of
interest in this paper. The square of the delta-v is the energy necessary to change location
which correlates with the amount of fuel necessary. Delta-v appears in the exponent of the
rocket equation and so has an enormous effect on fuel requirements, although this can be
reduced with fuel depots. Time is important if only because it limits the number of trips a
single reusable vehicle can make.
Getting To and From ELEO is Relatively Easy

distance (km) delta-v (km/sec) time, one way
ELEO 500 10 hours
L5 382,000 14.1 3 days
Moon 382,000 16.4 3 days
Mars (at closest approach) 54,600,000 20.2 260 days

Table 4: Distance between the Earth and potential settlement locations by different
measures. The Mars distance actually varies a great deal over time. Here we list the
distance at closest approach. The times are for current or near-term technology. All figures
are approximate.
The Moon and Mars also have a big problem for early settlers. Children raised there will
almost certainly have very weak bones and muscles, and possibly other as-yet-unknown
maladies. This is because Mars and the Moon have a surface gravity much less than Earth
normal (1g). The lunar surface is at roughly 1/6g and Mars about 3/8g. Muscles and bones
develop in response to stress, and children raised in low-g cannot be expected to be strong
enough to visit Earth except in extremis, or perhaps after an incredibly vigorous exercise
program. Consider an individual who weighs 73 kg (160 pounds). If they went to a 2.7g
planet, the equivalent of moving from Mars to Earth, they would weigh 194 kg (about 420
pounds) and probably could not get out of bed without assistance.
For children raised on the Moon or Mars, attending college on Earth, competing in the
Olympics or performing at Carnegie Hall will be either extremely difficult or impossible. By
contrast, free-space settlements can rotate to provide artificial gravity at whatever level

desired. Although this is not true gravity it will load children’s bones and muscles and there is
every reason to believe they will grow up strong enough to return to Earth for a visit or even
to live [Globus 2017b]. Space settlement does not have to be a one-way permanent
commitment for the settlers or their children.
Mars and the Moon do have ample supplies of easily accessible bulk materials for radiation
shielding and other purposes. Many tons of material per square meter can be used to bury
living modules to provide a reasonable radiation environment, but such protection does not
appear to be needed in ELEO. So, for early settlement the total mass of an ELEO settlement
launched from Earth is roughly comparable to the mass that must be launched for a similarly
sized surface settlement. This will only change when space mining, processing, fabrication
and transport sufficient to build the more massive settlement components (e.g., pressurized
modules) can be developed.
Structures can be a fair bit less massive on the Moon and Mars as they are not exposed to a
full 1g. Also, oxygen for breathing can be extracted from the Martian atmosphere fairly
easily. However, power systems on the Moon and Mars will likely be more massive than at
ELEO. A lot of storage is needed for the two week lunar night and solar energy at Mars is
more than twice as diffuse than at Earth orbit. There are also long duration dust storms.
Nuclear power is possible but means settlements would be Earth-dependent for energy.
Trading away in situ energy to gain in situ materials may not be advantageous as materials
can be recycled but energy cannot.
Thus, radiation shielding being unnecessary in ELEO eliminates much, although not quite
all, of the materials advantage the Moon and Mars provides for the earliest space
settlements. The residual advantages may well be swamped by the fact that ELEO is 760
times closer than the Moon and 100,000 closer than Mars to the industrial might of Earth.
Finally, if humanity wants to grow and survive over the very long term, a civilization based on
free-space settlements has a large advantage over one on planetary bodies and moons. If
humanity were to completely settle the Moon and Mars we would be spread out over three
locations and would roughly double our living area. If we build free space settlements co-
orbiting with asteroids, there is enough material for living area of roughly 400 times the
surface area of Earth spread out among perhaps a hundred million settlements.
Furthermore, free-space settlements make excellent generation ships for traveling to nearby
stars, and on arrival do not need a planet, just some space junk. If your answer to Kyle
Reese’s imperative that introduced this paper is ‘I do,’ then free-space settlements are for
you.
We now examine in more detail the two studies that suggest that small ELEO space
settlements may be the easiest path to early space settlement. There is an extensive section
on radiation with the evidence suggesting that ELEO settlements below around 500 km may
need no radiation shielding. There is then an extensive section on rotation tolerance. These
are followed by a market-driven approach to solve the launch problem and learn how to build
small settlements in an incremental fashion

Study One: Radiation Protection

This section is a relatively brief summary of the results reported in [Globus 2017a], where
more detail and references may be found. That study is based on a series of radiation
calculations using OLTARIS, NASA’s web front end to their very sophisticated radiation
codes [OLTARIS 2011, OLTARIS 2014]. These calculations suggest that in ELEO below
about 500 km little or no dedicated radiation shielding is necessary. This reduces the mass
of settlement designs by at least a factor of 19.
Space residents will be exposed to significantly more radiation than most, although not all,
people on Earth. This radiation can be kept away from settlers by massive amounts of
radiation shielding, but placing settlements in ELEO takes advantage of the Earth’s magnetic
field which would prevent many charged particles from reaching a settlement.
The primary health concern is radiation induced cancer, which can be expected to be more
frequent in space settlers than on Earth. Interestingly, however, the closest approximation of
space radiation found on Earth, the low level exposure of radiation workers and people in
high background radiation areas, has not been shown to increase cancer rates in humans.
However, animal experiments and atomic bomb survivor studies have shown increased
cancer rates with much higher dose rates for short periods of time.
Much radiation in ELEO is low energy and can be stopped by minimal shielding, such as the
structural hull of a settlement. On very rare occasions there are severe solar storms,
consisting mostly of protons, that may require substantial additional shielding for a few hours
or days, particularly if coincident with geomagnetic storms. Further research is necessary to
evaluate the need for solar storm shelters in ELEO, which will not be addressed here.
However, if settlements feature low-g swimming pools wrapping around the axis of rotation,
the population could perhaps simply go swimming during storms since water is an excellent
radiation shielding material.
However, the most dangerous ELEO radiation comes from Galactic Cosmic Rays (GCR).
GCR consists of all kinds of particles at a wide variety of energies, but most of the light
weight, low energy particles are diverted by the Earth’s magnetic field leaving only heavy
(e.g., iron ions) high speed particles to threaten ELEO settlements. These are at a very low
level, but are very difficult to stop and come from every direction. Furthermore, small
amounts of shielding can increase the damage from GCR when high energy particles collide
with nuclei in the shielding, generating a shower of secondary particles.
To determine the amount of shielding necessary one must choose an acceptable level of
radiation. For the general population [Globus 2017a] chose 20 mSv9/year based on an
extensive examination of recommendations for radiation limits on Earth. For example, 20
mSv/year is the occupational limit for nuclear workers recommended by the International
Commission on Radiological Protection (ICRP) [Wrixon 2008]. Although the average
background radiation in the U.S. is only 3.1 mSv/year [Linnea 2010, NRC 2010], this is an
average and in many areas levels are higher. In Europe, large parts of Spain and Finland
9
The Sievert is a measure of the biological damage caused by radiation. mSv stands for milli-Sievert,
one one thousandths of a Sievert. The mSv is calculated from the incident radiation, measured in
Grays, weighted by the kind of radiation and the type of tissue affected.

experience more than 10 mSv/year [World Nuclear Association 2014] and in a few parts of
the world background radiation is far above 20 mSv/year [Ghiassi-nej 2002] with no apparent
negative effect. Thus, 20 mSv/year seems a reasonable and conservative requirement given
current knowledge.
However, there will be children and pregnant women on space settlements and a radiation
limit suitable for the general population may not be sufficient. The embryo and fetus are
known to be particularly susceptible to radiation damage. A careful study of the ICRP studies
on radiation and pregnancy suggests that 5 mGy 10/pregnancy, the equivalent of 6.6
mGy/year, may be a reasonable limit [Wrixon 2008] [ICRP 2000] [ICRP 2003]. There
appears to be no data to suggest that this level will cause damage, but the data are not very
complete. For example, for the fetus and embryo there is no accepted approach to
converting radiation of various types measured in mGy to biological effectiveness, measured
in mSv, due to lack of knowledge and the fact that radiation damage is much greater in
certain time windows than others.
Children are quite susceptible to radiation damage but [Globus 2017a] does not recommend
limits lower than 20 mSv/year. However, this is an area requiring additional research.
There is one major caveat to the 20 mSv/year-6.6 mGy/year limits adopted here: there is
very little data on the effects of high energy GCR at all, and that data is for much higher
levels per unit time. Most of the data underlying the above limits comes from atomic bomb
survivors, pregnant women receiving X-rays, and animals exposed to high energy particles
in accelerators, particularly the Brookhaven National Laboratory/NASA Space Radiation
Laboratory. The atomic bomb survivors and pregnant women receiving X-rays were exposed
to vastly higher radiation levels for a very short periods of time and the radiation was not that
characteristic of GCR. Although high energy iron and other heavy nuclei can be and are
generated in accelerators for animal experiments, for practical reasons it is difficult to run
experiments in an accelerator for long periods of time at low radiation levels, for example the
20 or so days needed for rodent gestation studies. For this reason, experimental protocols
involve much higher dose rates for very short periods of time. This is not necessarily an
accurate approach. For example, if you were to expose a person to a year’s worth of sunlight
in a few hours, their skin’s reaction would be quite different from that of a year round sun
bather.
We now examine the computational results from [Globus 2017a] that suggest that ELEO
settlements will require little or no dedicated radiation shielding. All of these data are based
on computations performed with OLTARIS. All calculations were done using the (simulated)
1977 Solar Minimum radiation levels. Since GCR levels near Earth tend to be higher during
solar minima as the Sun’s magnetic field is weaker, these results are conservative, i.e., they
report higher radiation levels than are usually seen. However, there have been some periods
of even lower solar activity. This will not matter much for the general population where the
limit is based on exposure over many years or even decades, but may be an issue for timing
pregnancies, i.e., it might be wise to avoid being pregnant near solar minima.
10
The Gray is a measure of radiation. A mGy is one one thousandths of a Gray.

Radiation Shielding in Deep Space

Large amounts of mass can be used to bring radiation levels inside a space settlement down
to acceptable levels. Table 5 shows computational data for polyethylene 11, water and lunar
regolith shielding above the Earth’s magnetic field. Polyethylene is an excellent radiation
shielding material, readily available, slightly more effective than water and much better than
lunar regolith. A bit over six tons of polyethylene for every square meter of hull is needed to
achieve our limits. This is a massive amount of material for a settlement of any significant
size.
Radiation Shielding Required Above the Radiation Belts
polyethylene water lunar regolith

2
tons/m mSv/yr mGy/yr mSv/yr mGy/yr mSv/yr mGy/yr
~0 462 128 462 128 462 128
1 194 85 200 86 281 110
2 137 52 147 54 275 82
3 91 31 101 34 240 62
4 57 18.5 67 21 194 48
5 35 10.9 43 12.5 149 37
6 21.0 6.3 26.5 7.5 109 28
7 12.3 3.6 16.1 4.4 77 20.9
8 52 15.1
9 34.9 10.5
10 22.8 7.1
11 14.5 4.7
Table 5: Comparison of shielding materials in deep space. The rows

indicate yearly radiation levels at a given shielding mass except the first
row which calculated the radiation for one millionth of a gram of lunar
regolith as a stand-in for no shielding at all (OLTARIS cannot calculate
zero shielding levels). The first column lists tons of shielding per square
meter; the other columns list different materials and measures. The mGy
columns are a (computational) measure of radiation taken inside
shielding; the mSv columns are biological effect. The red color indicates
that values are less than 20 mSv/year or 6.6 mGy/year. Note that
polyethylene is a bit more effective than water, and both are much more
effective than lunar regolith. All values are calculated by OLTARIS.
11
Polyethylene is made up of long carbon chains each with two hydrogens, except at the ends.

Notice that the radiation levels are monotonic, i.e., they always decrease with greater
shielding. This is because the shield is massive, the smallest amount being a ton per square
meter, so most secondary radiation is absorbed.
Radiation in ELEO
Table 6 presents computational data for a 500 km circular ELEO using polyethylene
shielding. Notice that even at 10 kg/m2 shielding, the equivalent of which is very likely to be
provided by any reasonable hull and furnishings, the 20 mSv/yr and 6.6 mGy/yr are met.
Indeed, with no shielding at all the general population limit is met and the pregnancy limit is
very nearly met. This has an interesting consequence: spacewalks in ELEO may be safe
enough from a radiation point of view to be a significant recreational activity. This is in stark
contrast to the surface of the Moon and Mars where radiation levels with no shielding are
quite a bit higher than ELEO, meaning walks around the surface must be limited to meet 20
mSv/year.
Notice, however, that unlike Table 5 the radiation levels are not monotonic, they sometimes
increase with increased shielding. This is primarily due to secondary radiation generated by
high-energy heavy GCR. However, the calculated radiation levels never exceed our limits at
higher shielding levels.

Radiation Shielding Requirements in ELEO
tons/m2 500 km mSv/yr mGy/yr 600 km mSv/yr mGy/yr
~0 17.7 10.2 40.1 1,559
0.01 17.1 3.6 29.8 101
0.025 16.4 3.7 24.4 50.6
0.05 15.3 3.9 19.8 21.8
0.075 14.4 4.0 17.4 12.5
0.1 13.7 4.0 15.9 8.9
0.15 12.7 4.1 14.1 6.1
0.2 12.0 4.2 13.1 5.3
0.25 11.7 4.3 12.6 4.9
0.5 11.9 4.6 12.6 4.9
0.75 12.7 4.8 13.4 5.1
1 13.3 4.9 14.0 5.2
1.25 13.6 5.0 14.3 5.2
1.5 13.8 4.9 14.4 5.2
1.75 13.7 4.8 14.4 5.1
2 13.5 4.7 14.1 4.9

Table 6: Yearly radiation levels calculated for circular equatorial orbits at
500 and 600 km altitude. The rows are for tons of polyethylene shielding
with the exception of the first row which calculated the radiation for one
millionth of a gram of lunar regolith as a stand-in for no shielding at all
(OLTARIS cannot calculate zero shielding levels). Red indicates that the
level meets the 20 mSv/year limit or the 6.6 mGy/year limit for pregnant
women. All calculations by OLTARIS.

Radiation in Non-equatorial LEO

Table 7 illustrates the importance of ELEO (equatorial orbits) as opposed to the most
commonly used LEO orbits at higher inclination. Notice the rapid rise in radiation as the
inclination increases. Note that the ISS is in a 51.6o inclination orbit.
Radiation Shielding Requirements at Different Inclinations

Mass 0° 15° 30° 45° 60° 75° 90°
(tons/m2) (mSv/yr) (mSv/yr) (mSv/yr) (mSv/yr) (mSv/yr) (mSv/yr) (mSv/yr)
0.25 14.6 262.8 636.0 424.1 345.0 335.5 334.0
0.5 12.1 112.1 253.5 178.2 164.0 168.7 170.3
1 13.2 43.4 88.7 79.1 90.3 100.2 102.7
2 14.0 21.6 36.7 45.5 58.9 66.4 68.3
3 12.6 16.4 25.1 33.2 42.4 47.1 48.3
4 10.3 12.3 17.6 23.5 29.4 32.2 32.9
5 7.9 8.9 12.1 16.1 19.6 21.3 21.7
6 5.7 6.3 8.2 10.7 12.7 13.7 13.9
Table 7. This shows the effect of inclination and shielding on radiation levels. The rows
indicate the amount of radiation inside a settlement with the given amount of water
shielding. Thus, the levels are not directly comparable to other tables in this paper but
the differences are small. The columns correspond to different orbital inclinations at a
600 km altitude. Red indicates that the level meets our 20 mSv/year limit for the general
population. All calculations by OLTARIS.
Radiation levels for equatorial orbits are low because these orbits do not pass through
the South Atlantic Anomaly, a region of high radiation over the South Atlantic and South
America. Figure 1 provides some experimental data on radiation levels in LEO which we
will repeat here as Figure 4 for your convenience.

Figure 4: Radiation measurements taken on the ISS. Note the high levels of
radiation over the South Atlantic and much of South America and very low levels
near the equator. These low levels are well below our 6.6 mGy/yr limit, but the ISS
orbit is around 400 km, somewhat below the altitude of the computational data
presented in the paper. Image credit NASA.
Thus, it appears that we can live in space settlements in ELEO below about 500 km, and
perhaps a bit higher, with no dedicated radiation shielding, as the structure, solar arrays,
micrometeor shields and so forth are likely to provide the small amount of shielding needed.
However, as mentioned above there are caveats:
● The limits chosen may not provide acceptable protection. Indeed, it is not clear what
acceptable protection is. The pregnancy limit rests on particularly weak ground.
Special limits for children may also be necessary.
● The radiation codes used are based on data that may not be sufficiently applicable.
Very little of the experimental data regarding biological effect reflects the
characteristics of GCR and what little is available is at much, much higher dose rates.
Fortunately, this last fact means our results are probably conservative, i.e., that the
space radiation environment may be less detrimental than we think.

Radiation Risk Reduction

Greater understanding of radiation risks will require significant study of organisms exposed
to low levels of GCR for long periods of time. At the moment there appears to be no way to
do this on the ground as very long studies are impractical. It may be enough to study
radiation levels on the ISS, but the environment is not ideal. Radiation levels are much
higher as the ISS passes through the South Atlantic Anomaly, and for most of the time GCR
is lower since the ISS is generally at a lower altitude (around 400 km), deeper in the Earth’s
magnetic field and better protected by the Earth itself. Also, at present the ISS has no
centrifuge capable of maintaining 1g for rodents so the effects of weightlessness, which are
significant, would mask radiation induced changes. Smaller centrifuges suitable for smaller
animals, e.g., fruit flies, and tissue are available.
Particle accelerators on the ground such as the Brookhaven National Laboratory can
generate high energy iron particles very similar to the part of GCR of concern. Indeed, these
facilities are routinely used to study space radiation effects on tissue and animals, including
rodents. However, these accelerators are expensive, $7,000/hour in this case, and long term
studies are logistically difficult. Experimental protocols usually involve exposing subjects to
the same total quantity of radiation expected on, for example, a trip to Mars, in a very short
period of time, then assuming that this will have a somewhat similar effect. Unfortunately,
there is little if any evidence that they will have the same effect and some evidence to the
contrary. In some rodent experiments, animals exposed to a little radiation preceding a
larger dose had fewer negative consequences than animals exposed only to the larger dose
[ICRP 2003, paragraph 424].
To really explore the radiation effects at ELEO a space station in such an orbit is necessary
to expose subjects to exactly the target environment. Such a facility must either house
centrifuges to expose test subjects to 1g or the entire facility must rotate. This last option is
likely more expensive to build but staff can stay for longer tours as they will not be exposed
to weightlessness for long periods of time. The station can be relatively small, 25 m radius or
maybe less, as we shall see in the next section. Furthermore, data taken from the staff
themselves will be more meaningful if the facility rotates. The facility must study different
types of organisms, up to at least rodents, through multiple life cycles to throw light on
negative radiation effects on pregnant mammals.
It should be noted that the negative effects of space radiation may become treatable by
medical advances in the decades between now and the construction of the first settlements.
For example, radiation can induce cataracts but affected eyes can be repaired by modern
surgical techniques, and cancer treatment is rapidly improving.
Having examined the beneficial radiation effect of locating settlements in ELEO, we now turn
our attention to the rotation rate in an attempt to reduce the minimum size of a viable
settlement.

Study Two: Rotation Tolerance

This section is a brief summary of the results reported in [Globus 2017b] where more detail
and references may be found. That study examined the literature on human rotation
tolerance and found that settlement residents can easily tolerate up to 4 rpm, and 6 rpm is
probably acceptable. At these rates, visitors may experience discomfort and nausea for a
few hours or a day or two after arrival, but this is of minor consequence to settlers. However,
at 6 rpm very short term visitors (e.g., family visiting for a weekend) may spend much of their
visit feeling ill. There is an effective training regime that teach one to control their biological
responses to rotation using biofeedback and other techniques.
Increasing the rotation rate reduces the radius of a spinning settlement to achieve 1g
artificial gravity and thus reduces the size and mass of a settlement. For example, doubling
the rotation rate reduces the radius by 4 times. Thus, if other factors, such as psychological
and/or sociological issues, allow for smaller settlements, increasing the rotation rate can
reduce size and ease construction difficulty.
Table 8 shows diameter vs rotation rate to achieve 1g:
Rotation Rate and Diameter

Rate 1 2 3 4 5 6 7 8 9 10
(rpm)
Diameter 1790 448 200 112 72 50 38 28 22 18

(m)
Table 8: Rotation rate and diameter required to produce 1g of artificial gravity at the rim of a
space settlement. Notice that the diameter drops rapidly with increasing rpm and flattens out
because the diameter scales with the inverse square of the rotation rate.
Rotating space settlements can provide artificial gravity to the residents and avoid a wide
range of negative effects of exposure to microgravity. These negative effects include fluid
redistribution, fluid loss, electrolyte imbalances, cardiovascular changes, red blood cell loss,
muscle damage, bone damage, hypercalcemia, immune system changes and “aging,”
vertigo and spatial disorientation, space adaptation syndrome, loss of exercise capacity,
degraded vision, degraded smell and taste, weight loss, flatulence, changes in posture and
stature, and changes in coordination.
Various countermeasures have been tried to address microgravity symptoms in a piecemeal

fashion, but these do not resolve most of the issues and often have negative effects
themselves. Rotation addresses the underlying problem—that we have evolved to live
experiencing 1g acceleration nearly all the time. Artificial gravity by rotation is the only
practical countermeasure that gets at the underlying cause.
Experiments with artificial gravity on small animals and cell cultures have yielded
encouraging results. In the Soviet satellite Cosmos 936 in 1977, the lifespan of rats exposed
to centrifugation was significantly greater than that of non-centrifuged control animals
[Connors 1985]. In Spacelab D-1 in 1985, experiments showed that T-cell function—which is

severely hampered in microgravity—is preserved in artificial gravity via centrifugation

[Diamandis 1987].
However, rotation itself has negative effects. These include motion sickness, movement
errors, throwing errors and illusions. Motion sickness is by far the most serious for space
settlement. In experiments, and life, rotation can cause fatigue, stomach awareness, nausea
and even vomiting. The effects vary a great deal from person to person and are increased by
faster rotation, smaller radii of rotation and high g-level. It is possible to train people to
reduce the symptoms.
The space settlement rotation rate recommendations of [Globus 2017b] are:
● Up to 2 rpm should be no problem for residents and require little adaptation by

visitors.
● Up to 4 rpm should be no problem for residents but will require some training and/or
a few hours to perhaps a day or two of adaptation by visitors.
● Up to 6 rpm is unlikely to be a problem for residents but may require extensive visitor
training and/or adaptation over a few days. Some particularly susceptible individuals
may have a great deal of difficulty.
● For up to 30 rpm and beyond adaptation has been achieved with specific training.
However, the radius of a settlement at these rotation rates is so small (under ~20 m
for seven rpm) it’s hard to imagine anyone wanting to live there permanently, much
less raise children. Rotation at high rates, however, may be useful for a radiation
study station in ELEO.
Biofeedback augmented with autogenic training has been shown to give subjects control
over up to 20 physiological responses related to motion sickness. Subjects learn to control
the physiological systems affected by motion sickness and avoid discomfort [Cowings
2000][Cowings 2013]. Astronauts provided with several half hour sessions of similar training
up to 30 rpm suffered less space sickness.
Note that there are two classes of people that must be accommodated: residents and
visitors. For residents a few days of feeling ill at the beginning of a multi-year stay is of little
concern. However, if a settlement expects many short-term visitors it may be best to keep
the rotation rate under about 4 rpm.
Much of the data on the negative effects of rotation was gathered in rotating rooms,
experimental facilities where subjects could be exposed to rotation on the ground for days or
even weeks. Many of these studies were conducted in a facility including the Pensacola
Slow Rotation Room led by Dr. Graybiel.
In 1960 Graybiel published a paper [Graybiel 1960] that is a good example of the data
behind [Globus 2017b] recommendations for space settlement design. In this particular
study there were five regular subjects and one deaf subject who had lost otolith 12 function.
The deaf subject had no motion sickness symptoms at any time, suggesting that otolith
response is the driver in rotation tolerance. Subjects were tested in two-day runs at five
12
The otolith organs are structures in the inner ear that are sensitive to gravity and acceleration.

rotation rates. They were given a number of tests to complete, but there was no adaptation
procedure.
The results of this study may be summarized:

1. 1.71 rpm: very mild symptoms.
2. 2.2 rpm: one subject threw up (he had a history of seasickness) but otherwise
similar to 1.71 rpm.
3. 3.82 rpm: mild symptoms and subjects adapted within a day; adaptation was
longer for the less resistant subject.
4. 5.44 rpm: highly stressful (except for the deaf subject) but most adapted in a
day or so. Subjects with prior rotation experience did better than those
without.
5. 10 rpm: highly stressful (except for the deaf subject); subjects could not
complete all tasks. There was some adaptation over the two day run.
Note that these are very close to our recommendations. There are many other studies with
more-or-less similar results.
At least five authors have surveyed the literature and come to conclusions similar, but not
exactly, to ours. The results of these studies are summarized in Figure 5.
Figure 5. Note that we are only concerned with the 1g line. Green indicates
all authors agree this rotation environment is acceptable. Red means no

authors thought so. Intermediate colors indicate disagreement between the

authors [Globus 2017b].
None of these authors were looking at settlement related issues, but rather considering
rotating spacecraft for much shorter periods of time; for example, on a trip to Mars. Note that
most authors agree with our analysis that 4 rpm is acceptable, although there is less
agreement between them at 6 rpm.
There are some caveats to these results. First are factors which might make our results
more or less optimistic. These studies
1. have very few subjects, usually 10 or less.

2. show great variability in rotation tolerance from person to person.
3. sometimes chose subjects for higher than normal rotation tolerance.
4. have only adult subjects.
5. are only a few weeks or less in duration.
However, there is some reason to believe the ground based experimental results are
pessimistic for space, particularly settlement, applications:
1. the experiments were conducted with very short rotation radii, usually under 3 m, and
shorter rotation radii generally lead to more severe symptoms.
2. at least one very encouraging experiment strongly suggests that in orbit negative
rotation effects may be much less than on Earth.
In the 1970s, SKYLAB experiments with a rotating chair took data before, during, and after
flight. Eight crew members participated. Nausea levels were measured. All subjects
experienced much less nausea in microgravity, as opposed to pre and post flight in 1g. On
the ground many of the subjects could not complete the 150 prescribed head motions, but in
orbit all did with few or, in most cases, no symptoms [Graybiel 1977]. Similar results have
been obtained in parabolic aircraft flight.
The literature indicates that high-rotation-rate space settlements will be uncomfortable for
people when they first arrive, just as arriving in a high-altitude city can be uncomfortable.
However, within a few hours or days we can expect symptoms to pass, just as altitude
sickness does. This is certainly acceptable for permanent residents who will stay for years.
To really nail down the effects of rotation in orbit, additional experiments along the lines of
the SKYLAB experiment discussed above are in order. This could be supplemented by
parabolic aircraft flights. If the results reported above on a small number of subjects extend
to the general population, then high (4-6 rpm) rotation rates may be acceptable.
As noted above, the combination of the results of these radiation and rotation studies
strongly suggest that space settlements can be far less massive (480 times or more) and
much closer to Earth (760 times) than previously thought. Even with these enormous
reductions in size and distance, there is a still a major barrier that ELEO shares with all
space settlement schemes.

The Big Problem: Launch

The proposed approach to developing the first space settlements depends heavily on
launching all the materials from Earth. The required improvements in launch for financially
viable ELEO settlements are significant. It should be noted, however, that all other
settlement locations also require much improved launch capabilities.
Space settlement is, at its core, a real estate business. We hope to build new communities in
order to sell housing in them. This must be done above cost or be subsidized. To keep the
subsidies to a minimum, we would like to be able to sell a house or apartment in space for
something close to the cost of a very high end location on Earth today. For example, the
average home price in Silicon Valley is well over $1 million today. Considering just the
launch costs, since these are the easiest to quantify, how much improvement is needed to
get a housing unit in a space settlement somewhere in the few million dollar per person
range?
There are two parts to the transportation cost: delivering the materials and components for
the settlement, and flying the proud owner to their new home.
Table 2 suggests that Kalpana One at 4 rpm requires 17 tons/person. The cheapest
advertised price today for delivering mass to orbit is the Falcon Heavy, in late-stage
development, at $90 million for 54 tons to LEO [SpaceX 2017], or $1.7 million per ton. For 17
tons that is about $29 million.
The cheapest advertised price to launch people to LEO is a bit over $26 million/seat on a
Falcon 9/Dragon which included a stay at a Bigelow space station, also in development. This
price was removed from the Bigelow website but it is still the best data we have. It is
consistent with the $20-50 million the Russians have reportedly charged to fly tourists to the
ISS.
Combining these two costs gives us (rounding up) $60 million per person. This does not
include materials, construction or resupply costs. We assume that government or space
tourism businesses (see below) will conduct most of the research and development cost
other than actually building a settlement.
To get the transportation costs to close to one million dollars, leaving some small number of
millions for everything else, we need to reduce the cost of launch by about a factor of 50 to
around $1.2M/person. There is a system at the very beginning of development with targets
in this range: the fully reusable SpaceX Interplanetary Transportation System (ITS), intended
to provide a transportation system to enable Mars settlement [Musk 2017]. It will use the
Raptor rocket engine, which is in development and ground test. In addition, a large carbon
composite cryogenic fuel tank has been fabricated and is undergoing test. This is no mere
paper study.
The system has two legs: Earth to LEO and LEO to Mars. The Earth to LEO leg involves two
vehicles. The “rocket booster” with 42 Raptor engines (up from 27 engines on the Falcon
Heavy) takes off from Earth, places a payload in orbit, and returns to the launch site ready
for refueling and another trip. This booster can place 550 tons of expendable payload in LEO

or 300 tons of “reusable payload.” The second vehicle is the “interplanetary spaceship” with
nine Raptor engines, cargo capacity of 300 tons to LEO (when launched with the rocket
booster) or, eventually, up to 100 passengers. The spaceship can take off from and land on
Earth. There are also hardware and systems for the LEO to Mars leg but they are not of
interest for ELEO settlements.
The ITS cost targets are $140/kg and $200,000 per passenger for transportation from the
Earth’s surface to the Martian surface. The first leg of this transportation system, from Earth
to LEO, should cost considerably less than half as a vehicle can make perhaps a trip per day
as opposed to many months for a single transit to and from Mars. If we conservatively
assume that the Earth to LEO leg is half the cost, that gives us $70/kg and $100,000 per
passenger. At these prices the transport of necessary materials and components to ELEO
will cost approximately $1.2 million and transport of the owner $100,000 for a total of $1.3
million, roughly the same as ELEO settlement requires. Note that the sunk cost is only the
$100,000 for the owner; the $1.2 million (plus materials and construction costs) could
conceivably be recouped or even increased by sale of the owner’s share of the settlement
and, presumably, a dwelling there.
Reducing launch cost by a factor of 50 will be extremely difficult, but would have fantastic
benefit for every use of space. The advantages of improved launch, well below a factor of
50, are recognized and are part of the reason we are in the midst of a golden era of launch
system development with SpaceX, NASA, United Launch Alliance, Virgin Galactic, Blue
Origin, Orbital ATK, the Europeans, Russia, India, Brazil, China and others all developing
new or improved launch vehicles. In particular, the SpaceX Falcon 9, a relatively new
vehicle, has reduced the launch price enough that their competitors are scrambling to follow
suit. However, to reduce launch price by a factor of 50 will require more than any of the
current efforts can possibly provide, with the notable exception of the ITS.
For any system to reach a 50 times price reduction, or anything close, will almost certainly
require fully reusable launch vehicles, much improved technology and a very high flight rate,
probably in the tens of thousands per year. The reusability and technology requirements are
generally recognized but for some reason flight rate is often ignored. However, with today’s
market of fewer than 100 launches per year, a single reusable vehicle capable of two flights
a week could, theoretically, satisfy the entire launch market! Even 1,000 flights per year
would only require 10 such vehicles. Large reductions in price will not come if vehicles are
built in such small numbers. Launch vehicles only make money when they fly, so we need a
very high flight rate, probably over 10,000 flights/year.
There are only two applications that, at the right price, could create a market requiring a
flight rate of tens of thousands or more per year: space solar power (SSP) 13 and tourism.
SSP requires a very large investment up front before any income is generated and is
vulnerable to terrestrial competition, particularly as batteries improve. So we turn our
attention to tourism.
13
For our purposes, SSP refers to gathering solar energy in space and beaming it to Earth.

Tourism: the Killer App

How do we get from where we are now to the construction of the first high-rpm space
settlement in ELEO? One word: tourism—a $2.3 trillion/year industry [Statista 2014]. Steady
growth in a for-profit space tourism market could provide most of the development needed to
enable construction of the first ELEO settlements.
Launch Rate
The first task is to drive the launch rate up to the point where prices are, very roughly, 50
times lower than today’s prices. There have been a number of studies of the space tourism
market [O’Neil 1998]. We present data from one of them in Table 9:
Relationship Between Space Tourism Price and Volume

price/ticket passengers/year Flights at 10 Vehicles Gross revenue
(2015 $) passengers per needed if flies ($ million)
vehicle twice per week
1,600 20 million 2 million 20,000 32,000
16,000 5 million 500 thousand 5,000 80,000
160,000 400 thousand 40 thousand 400 64,000
400,000 1,000 100 1 400
800,000 170 17 0.017 136
Table 9: From Crouch, G. I., “Researching the Space Tourism Market,” presented at the
annual Conference of the Travel and Tourism Research Association, June 2001. This study
attempted to determine global demand for tourist flights as a function of price. Prices were
converted to 2015 dollars and the last three columns added.
These are early estimates and require further research. However, there are three important
take away messages from Table 9 that are justified even if the data are not quite right.
1. Demand is a strong function of price. As the price goes down the number of
passengers per year increases very rapidly. This will reward those who can fly less
expensively.
2. The market may eventually be quite large, perhaps tens of billions of dollars per year.
3. There is some point—$160,000/trip here—where demand skyrockets.
4. Even with demand exploding at $160,000/trip the number of reusable vehicles
needed is not particularly large. Low vehicle price requires large volumes and to
require truly large numbers of vehicles we need a very low customer price, perhaps
in the tens of thousands of dollars per seat range.

The way to develop lower cost launch vehicles is to get the industry into a virtuous cycle
where
1. Lower launch prices stimulates increased demand.

2. Increased demand provides economies of scale to lower launch costs.
3. Back to #1.
What Table 9 indicates is that such a virtuous cycle is possible if the necessary vehicles can
be developed and we have a buyers market. The need to undercut the competition can drive
incremental technology development, and government research has been effective in
developing long term, breakthrough technology (the Internet is the classic case).
Note that the ITS, at $100,000 per seat may be in the vicinity of the price point that
corresponds to a rapid increase in the market. This is in the range of the reduction
necessary to build unsubsidized ELEO settlements. However, this may still not be enough to
require more than a few hundred launch vehicles.
Fortunately, the tourist market is not just a matter of surveys and speculation, there is
already a space tourism market with paying customers. The Russians have flown seven
paying customers to the ISS for a week or two each, one of them twice. Prices are believed
to have varied from about $20 to 50 million. Unfortunately, prices started some years ago at
$20 million and have since gone up, which can only mean that right now space tourism is a
sellers market. This is because in the last few years only one tourist seat into space has
been available.
While only a single seat has been for sale in a number of years, over 700 people have paid a
deposit for a short suborbital flight, at $250 thousand/seat, but the vehicles are not yet ready
and have not yet flown in space. Nor are any American crew-capable vehicles ready for
human launch to orbit, although two are in development. This leaves only the Russian Soyuz
and only when an extra seat is available beyond what is needed to staff the ISS. Thus the
$50 million/seat price. The cargo version of the Falcon 9 has reduced satellite launch prices
and stimulated an industry-wide scramble to lower prices. If the human-rated Falcon 9 under
development can do the same for tourist seats, our virtuous cycle may begin.
Space tourism can, at least in principle, provide the demand to develop the reusable, high-
launch rate vehicles necessary to successful ELEO space settlements, or anywhere else for
that matter. The tourist market can also drive the development of the technology and
infrastructure necessary to construction and operation of the first space settlements.
Facility Development: Space Hotels

One approach to developing ELEO settlements is to start with where we are and build more,
larger and better space stations and hotels to feed a growing market as space vacation
prices come down due to improvements in launch vehicles. Leasing space in stations and
providing hotel accommodations could provide near-term revenue to drive the market. The
owners of space hotels will have a strong incentive to recycle air, water, and even food to
reduce the cost of resupply. Higher volumes of tourist traffic will support larger and larger
hotels. Some space hotels may even rotate to provide fractional-g to allow staff longer stays
and make customer use of the toilet an easier task. Eventually, if all goes well, space hotels

may approach the size of a 4 rpm settlement (112 m diameter and 2-8.5 kT mass, see
Tables 2 and 3). At that point, building the first settlement may not be much more difficult
than building yet another hotel.
Figure 6: The International Space Station (ISS) in orbit. It is around 100 m in length, not
much less than the 112 m diameter required for a 4 rpm space settlement. The mass of the
ISS is about 0.42 kT [NASA 2015], or about 5-20 x less than the 4 rpm versions of the
Stanford Torus and Kalpana One (see Tables 2 and 3). The ISS has hosted 7 paying
tourists. Image credit: NASA.
First, let us consider where we are. The ISS typically hosts six people. It is around 100 m
long but does not rotate and is in a high-inclination, high-radiation orbit somewhat lower than
our preference (about 400 km rather than 500 km). The ISS has hosted seven space
tourists, one twice, for a price estimated at $20-50 million per person on trips of one to two
weeks. In addition to paying a very high price, these tourists needed to learn Russian and
train for many months in Star City in Russia. The reward has been fantastic views of Earth
and the incomparable experience of microgravity. Some of these ‘tourists’ have also
conducted important, and sometimes paid, experiments during their stay [Garriott 2014].
The U.S. government has no specific plans to develop additional space stations after the
planned decommissioning of the ISS in 2024. There are indications that NASA would like
private industry to develop commercial stations, but no firm commitment to help. The
Russian and Chinese governments have both indicated that they will build and operate
space stations of their own in the relatively near future. What might a U.S. led program to
replace the ISS look like?
For a variety of reasons, NASA cannot fly rich tourists. NASA can help develop space
stations though. One approach to getting started would be for the ISS to be augmented and
eventually replaced by private space stations, commercially owned and operated. The

Alliance for Space Development (ASD) and the National Space Society (NSS) proposed that
this be approached in a manner similar to to the COTS and CCDev programs [NSS 2015],
where NASA provides partial development funding and subsequently becomes an anchor
tenant, or at least an important customer. This approach successfully developed two cargo
launch vehicles at a small fraction of the usual rocket development cost (COTS program)
and is currently being used to develop passenger service to the ISS (CCDev).
Figure 7: The B330 space station. Image credit Bigelow Aerospace.
There are at least two private companies, Bigelow Aerospace and Axiom Space, that are
building space stations. Axiom Space is just getting started but Bigelow is more mature. The
Bigelow modules are based on inflatable technology acquired from NASA and subsequently
improved. Bigelow has flown two sub-sized test stations which are currently in orbit. Bigelow
also sent the BEAM (Bigelow Expandable Activity Module) to the ISS under a $17.3 million
contract with NASA for a two year trial. Finally, Bigelow is developing the B330 module, a six
person, 20 ton, 9.45 m length, 300 m3 volume, 20 year lifespan space station module that is
being marketed as a tourist destination and to national space programs and companies for
research and development [Bigelow 2017].
Two small companies making progress is hardly enough to declare victory and go home.
However, there are a number of aerospace companies with the capacity to build space
stations or hotels, and intelligent use of government development funds could both help
Bigelow Aerospace and Axiom Space succeed and bring others into competition to generate
a more vigorous market.
Let us assume, for the moment, that industry steps up to build and operate space stations in
response to current opportunities and perhaps government inducements. It will not escape
notice that space hotels are easier to build and operate than space stations. All that
expensive science equipment—furnaces, glove boxes, centrifuges and so forth—is not
needed. It is much simpler to have a large area for microgravity recreation and some first
class windows for viewing. If someone can build and operate a profitable space hotel, then
others will follow. Space hotels will need less expensive launchers to bring more and more

customers, and launch companies need high launch rate to reduce prices. Hotels are key to
to this virtuous cycle.
If a vigorous space hotel market is established, it is reasonable to expect that many of the
technical and infrastructure problems associated with ELEO settlement will experience
competitive pressure to be solved. For example,
1. Recycling air and water can lower costs significantly.

2. Growing food on board can not only save cost, but high end customers may not
welcome freeze-dried food-like substances for dinner.
3. Rotation may be introduced, at low g-levels, to eliminate the customer training
necessary to use microgravity toilets, a tricky proposition with significant downsides
when things go wrong!
4. Modest artificial gravity levels might allow hotel staff to have longer tours of duty,
substantially reducing costs. Microgravity recreation would still be available near the
axis of rotation.
If the virtuous launch cost cycle can be established, over time it is reasonable to expect that
space hotels will evolve into facilities near the size of a 112 m diameter space settlement
with all or most of the necessary capacities: life support, artificial gravity, transportation,
communications and so forth. At that point building the first true space settlement, a place to
raise your kids, is not a lot different from building yet another hotel.
Thus, the first space settlement may not come from a heroic effort by an elite band of
adventurers, but rather from the natural outgrowth of a profitable commercial market
complete with the luxurious amenities expected by high-end tourists.
Immediate Tasks
The most important immediate tasks, as we have discussed above, are launcher
development and private, commercial follow-ons to the ISS. These can work out the
transportation, construction, and operational issues relevant to space settlement.
We have also indicated that an ELEO research station is necessary to investigate radiation
issues, and experiments rotating people in micro-g may may be important.
There are other issues that should be addressed soon: the relevant orbits are home to very
high velocity uncontrolled space junk, launch failure rates are too high for extensive tourism,
high launch rates may impact the atmosphere, and we do not know how small a viable
settlement can be.
Space Debris
The spacefaring nations of Earth have been creating debris in orbit for decades. As a result,
there are over 20,000 tracked Earth-orbiting objects larger than 10 cm and perhaps half a
million or more shrapnel fragments between 1 and 10 cm. These are traveling at speeds
around 27,000 km/hour [NSS 2016]. It has been reported that 10% of LEO satellites have
been hit and at least one communication satellite was destroyed by a debris collision,
creating thousands of new pieces of shrapnel. A collision with many of these bits of junk
could easily puncture the pressurized hull of a hotel or settlement. The problem is expected

to get much worse as the probability of a collision is superlinear with the number of pieces so
each collision significantly increases the probability of more collisions.
Figure 8 shows the number and mass of the debris as a function of altitude. Note that at the
altitude we expect for early ELEO settlements and hotels, around 400-500 km, there is not a
lot of debris. That is because debris at these altitudes decays and burns up in the
atmosphere fairly quickly, unlike large and heavy settlements with high mass per unit of drag
area. However, almost all of the debris generated in LEO will eventually pass through the
400-500 km altitude range and at that point constitute a serious threat to both hotels and
settlements, and, for that matter, all other satellites in LEO. Thus, it would be wise to prevent
additional space pollution and clean up the mess we have already made.
Figure 8: Number of tracked objects in LEO and the total mass as a function of altitude.
Image credit Darren McKnight and Patrick Dingman [McKnight 2012].
Fortunately, it is not necessary to clean up all the debris for ELEO settlements to be
practical. If there is a good system in place to detect a potential collision with some warning
it is possible to move a settlement. The ISS has made a number of collision avoidance
maneuvers to avoid potential collisions using docking thrusters, the Shuttle, and Progress
supply ships. These maneuvers also boost the ISS into a higher orbit so they can help keep
the station from deorbiting. The mass of the ISS is about 0.4 kTons or about 20 times
smaller than a 4 rpm Kalpana-class settlement. Scaling up to this boost capability is a
challenging engineering problem, but no more than that.

Launch Failures Rates

Launch vehicles have fairly high catastrophic failure rates. For example, the Space Shuttle
had a 1.5% failure rate. This is fairly low compared to most other launchers. [NASA 2011]
and [Lafleur 2010] report failure rates for a number of vehicles:
1. Shuttle: 135 launches with a 1.5% failure rate.

2. Soyuz: 1,698 launches with a 2% failure rate.
3. Delta: 347 launches with a 4% failure rate.
4. Ariane: 193 launches with a 5% failure rate.
5. Proton: 351 launches with a 11% failure rate.
6. Atlas: 347 launches with a 12% failure rate.
It should be noted that many of these failures were early in the vehicle’s lifetime and that
failure rates in the future are likely to be lower. For example, the most recent version of the
Atlas family, the Atlas V, has never had a catastrophic accident.
The maximum fatality rate a viable space tourism business can tolerate is unknown.
However, as early space tourism is definitely adventure travel, comparisons with other
adventure travel may be relevant. For example, the fatality rate for those climbing Mt.
Everest is about 2% [Explorersweb 2006]. Assuming that early space tourists can tolerate as
much risk as those climbing Mt. Everest the most reliable launchers appear to be good
enough, or at least almost good enough.
Of course, there are important differences between the space tourism experience and
climbing Mt. Everest. The climbers require great skill and must be in excellent physical
condition to even consider the attempt. Training and fitness requirements for space tourism
are much lower. Also, when climbing it is you against the mountain as opposed to being a
passenger whose fate is in others’ hands.
In the longer term, if we achieve the desired 10,000 launches per year or more a 2% failure
rate is completely unacceptable. That would be at least 200 fatal accidents a year or four per
week. Space launch failures will almost certainly be well publicized as explosions make for
high TV ratings. Thus, a vigorous program to reduce failure rates is imperative.
High Launch Rate Impact on the Atmosphere

Every space launch slightly alters the Earth’s atmosphere by inserting rocket exhaust at
various altitudes. Vehicles returning to Earth usually use atmospheric braking which involves
high temperatures and induces chemical reactions. These effects are particularly significant
at high altitudes where there is little matter to begin with.
At today’s very low launch rates, under 100 per year, there is little concern with atmospheric
pollution. The amounts are tiny compared to other sources. However, we have seen that
space settlement, at any location, requires tens of thousands of launches per year to bring
prices down to the point that very well off individuals might be able to finance their own
relocation into space. At these high launch levels it is not a given that atmospheric damage
is acceptable.

Fortunately, the cost of respecting the environment is usually low if environmental concerns
are addressed early before there is a large infrastructure based on polluting technology. In
the case of space launch, consider that H2/O2 is not only more powerful (higher ISP) than
carbon based fuels but the exhaust is only water, not more complex compounds and CO2.
A vigorous long range research program to understand potential threats to the atmosphere
and avoid them would be wise.
How Small Can a Settlement Be?

Our study of the rotation literature revealed that the criteria used for early space settlement
designs was unnecessarily conservative, although appropriate for those studies at that time.
In fact, settlers can tolerate much higher rotation rates and this translates into smaller initial
settlement size. However, rotation tolerance alone may not be sufficient to define the
minimum size of a viable settlement.
Most of the early settlers will likely come from large cities as maintaining a gigantic, complex
new machine such as a space settlement requires very high levels of technical expertise.
We will be asking these people to live in a much smaller social and physical environment.
How small can that be?
We don’t know, and it is important. It is important because the smaller the settlement the
easier it is to build, at least for the first few. In this paper we have assumed that a 4 rpm
settlement with a diameter of 112 m would be big enough, which seems reasonable but
there is very little data to back it up. Indeed, one might expect that the minimum size
depends a great deal on exactly who inhabits the settlement.
We conducted an internet survey to throw light on the minimum size issue. Just over 1,000
people took the survey and 95% were self-described space enthusiasts. Approximately 6%
of the respondents said they could be happy living in a space settlement the size of a large
cruise ship (100 m diameter, 50 m length cylinder) with 500 or less people and were willing
to devote up to 75% of their lifetime wealth to go. While hardly definitive, this suggests that
some people would at least consider living in space in a 4 rpm settlement and with
population of seven billion it should be easy to fill a settlement at the right price.
Settlement Growth Path

While the easiest approach to the first few settlements may be to use ELEO, this is not the
end goal but rather a very large first step. See Appendix A for an extremely long term
development path starting with the first ELEO settlement and continuing on to high Earth
orbit, the Moon, asteroids, Mars, and eventually generation ships to the stars.
Conclusion
To survive long term, humanity and life itself must diversify beyond Earth. Of all species only
homo sapiens is space faring and only humankind has the physical ability to settle space.
However, developing true space settlements—where residents live, work and raise their
children—is a daunting task. Surface settlements, on the Moon and Mars, are severely
problematic for raising children due to low gravity and are also extremely far away making
transport difficult. For forty years free-space settlements appeared to require mining millions
of tons of material from the Moon or asteroids and constructing kilometer scale space

systems at huge distances from Earth. In this paper we have shown how to make early
settlement construction and operation significantly less daunting by making those first
settlements close, small, and building up from the ISS incrementally.
We have seen that the task is much easier when the first settlements are built in ELEO
around 500 km above the Earth’s surface and directly over the equator. The radiation
environment in this region of space is relatively benign as we have seen from both
experiment and computation. In fact, a careful examination of the permissible radiation levels
combined with calculation strongly suggests that ELEO settlements may need no dedicated
radiation shielding, which is typically 95% or more of the mass of free-space settlement
designs. In space, mass is a good, although not perfect, proxy for difficulty.
We have seen that the literature suggests that human rotation tolerance is much greater
than the designers (including these authors) of earlier free-space settlements believed. This
permits the construction of much smaller settlements. In combination with eliminating the
radiation shielding the mass of free-space settlement designs can be reduced by at least two
orders of magnitude, which is a complete game changer. It means that lunar or asteroidal
mining, materials processing, and manufacture is no longer on the critical path to the first
settlement, although they will be able to exploit the markets created by ELEO settlements
launched from Earth.
We have seen that the primary barrier to space settlement is the cost of launch, as is true
with all large-scale space development. The near term cost to launch a single settler and
their share of the materials for an ELEO settlement is around $60 million, half just to launch
the settler. To reduce this cost into the range of an expensive house on Earth requires
somewhere around a 50 fold reduction in launch price. Fortunately, the ITS, a system in
early development by a launch company with a history of low-costs vehicles, SpaceX, has
cost targets that translate into transportation costs to ELEO settlements of around $1.3
million per settler, almost exactly what is needed, and only $100,000 of that cost is a seat for
the settler, the rest for materials and components [Musk 2017]. It should be noted that other
locations (Moon and Mars) have similar or greater mass requirements at least in the near
term, with access to materials being balanced by greater transportation cost.
While a 50 fold reduction in launch price will be difficult to achieve, there is a market
mechanism. The volume of space tourism is a strong function of price. Any reduction in
launch cost can substantially increase the number of paying customers, and this can provide
the increased volume needed to reduce the price, which can increase the number of
customers and so on in a virtuous cycle. The key is converting the current sellers market into
a buyers’ market. As an important side effect to this dynamic, tourists will need an ever
increasing number, size, and sophistication of space hotels, which may eventually evolve to
the point that building a small space settlement is not much different from building a large
hotel.
Thus, we see that the first major steps toward settlement of the solar system may not be
driven by elite astronauts, but rather by ordinary people taking the most amazing and
fantastic vacation of their lives.
Ad Astra!

Acknowledgements
Thanks to Al Anzaldua for the space debris information. Thanks to David Brandt-Erichsen,
Joe Carroll, Gary Oleson, Tom Marotta and Joe Strout for comments and editing.
Appendix A: Settlement Growth Path

While the easiest approach to the first few settlements may be to use ELEO, this is not the
end goal but rather a very large first step. In this appendix we sketch out an extremely long
term development path starting with the first ELEO settlement and continuing on to high
Earth orbit, the Moon, asteroids, Mars, the outer solar system and eventually generation
ships to the stars. The rest of this appendix will outline this expansion.
Fill ELEO with Settlements

Once a few ELEO settlements have been built and successfully operated the next order of
business is to fill ELEO with an industrial civilization. A 500 km altitude circular orbit at the
equator is about 43,000 km long. Assuming a spacing of 1,000 km, that means there is room
for 43 settlements. If we assume we can populate orbits every 50 km in altitude from 400 to
600 km we get about 200 settlements. While the first settlements should be as small as
possible to take advantage of space hotel development experience, most people, particularly
technical people, are accustomed to living in cities with large populations. Thus, it is
probably desirable to build larger settlements as quickly as possible and decommission the
early small ones once the slots fill up. If each settlement has at least a population of
10,000—the number chosen by the 1970’s studies [Johnson 1975] [O’Neill 1977]—that
makes for a total ELEO population of over two million.
ELEO Settlement Enables Lunar and/or Asteroidal Mining and Settlement

All of the current approaches to lunar and asteroidal mining suffer from the same weakness:
small markets in space. It is difficult for extraterrestrial materials to compete on Earth as the
same materials can generally be mined on Earth for a small fraction of the cost of developing
and operating mines in space and transporting the resulting materials to Earth. By far the
best markets for extraterrestrial materials are in space where transportation costs work
against Earth materials.
However, the current in-space market for materials consists of six people in the ISS and
about 1,300 robotic spacecraft, only one of which to date has been designed to be refueled
or repaired—certainly not enough to justify lunar or asteroidal materials to do so. The most
common market cited by space mining enthusiasts is oxygen and sometimes hydrogen for
use as rocket fuel. This market doesn’t exist at all right now, and even in rosy future
projections is quite small compared to tourism. One exception is the United Launch Alliance
proposal to build space solar power satellites using materials mined on the Moon and
transported by rocket fuel also mined from the Moon [ULA 2016].
Once extensive settlements in ELEO are established this will change. Instead of six people
in space there may be millions. These people will need water and atmosphere replacement,
reaction mass for orbit maintenance, and materials for manufacturing and for building
additional settlements. This may be enough to get lunar and asteroidal mining operations

started. Unless these mines can be entirely automated, including maintenance, a crew will
be necessary for each mine. These mining camps may, as they often have on Earth,
become the nucleus of new settlements.
Although lunar mines are the closest in distance, asteroidal materials may play a role at this
stage as some asteroids can be brought into lunar or even Earth orbit at low delta-v using
very efficient solar electric propulsion and lunar gravitation assist [Brophy 2012].
Thus, rather than space mines being essential to initial settlement construction, settlements
may create the conditions for successful extraterrestrial mining.
Lunar and/or Asteroidal Mines Enable High Earth Orbit (HEO) Settlement
Once there are lunar and/or asteroidal mines in operation it may become feasible to build
settlements beyond ELEO in the more intense HEO radiation environments that require
many tons of shielding per square meter of hull. Then the same problems faced by the
designers of the Stanford Torus will have to be overcome, but those tasks will be made
much easier by a large, experienced workforce in space, industrialization of ELEO, and
existing lunar and/or asteroidal mining operations that need to be scaled up, not created
from scratch.
Thus, ELEO settlements may enable the lunar and/or asteroidal mines which in turn enable
delivery of the materials required by HEO settlements. This space is vast and can support an
extremely large population—certainly in the billions, but there is more.
Asteroid Mining Enables Solar System Wide Settlement

Although the resources of Earth orbit (space and sunlight) and lunar materials are enormous
they should not hold humanity forever. Indeed, for survival it will be prudent to push beyond
Earth orbit, even HEO, to become independent of the Earth-Moon system. The obvious
approach is to build settlements co-orbiting with asteroids that supply the necessary
materials.
The first asteroid-based settlements will undoubtedly be close to Near Earth Objects (NEOs)
as they are the easiest to reach. At 2 rpm a cylindrical Kalpane-One-like settlement including
deep space radiation shielding has a mass of about 4 million tons according to Table 2. At a
representative asteroidal density (3 ton/m3 [Carry 2012]) a spherical 140 m diameter asteroid
has about that much mass. There are roughly 15,000 NEOs at least that large [NASA 2013]
and roughly 1.7 million main belt asteroids one km or greater across. More large large
asteroids can be found in Jupiter’s orbit, but that’s not all.
Outer Solar System

Beyond Neptune stretches a vast region of icy bodies including the Kuiper belt and Oort
Cloud. The furthest bodies in the Oort Cloud may reach halfway to the nearest star, Proxima
Centauri. Indeed, if Proxima Centauri has similar icy bodies they may overlap with those
orbiting Sol. In any case, by some estimates these bodies could support a population in the
trillions, or perhaps even more.

If a sizable fraction of the asteroids and icy bodies are developed into space settlements
mankind will have an enormous industrial base, perhaps sufficient to launch the next step.
Generation Ships Enable Galactic Settlement

If you and your family have lived for 500 generations in free-space settlements, how
important is proximity to Sol? For some, it may not matter much whether their settlement is
in orbit around the Sun or on the way to Proxima Centauri, four light years away. It does not
matter if Proxima Centauri has a habitable planet, as our distant descendants will only need
asteroids or comets to provide materials. If suitable propulsion, power source and extremely
efficient recycling can be developed in the next few tens of thousands of years 14, there is
every reason to believe that groups of co-traveling settlements (for redundancy should
something serious go wrong) may head out into the galaxy. As we are near the edge of the
galaxy where interstellar distances are relatively large, if we can get to Proxima Centauri we
can get anywhere in the galaxy given enough time [Globus 2012].
Appendix B: Free-Space Settlement Models

This section provides a description of the Microsoft Excel models that feed Table 2 (Kalpana
One cylinders) and Table 3 (Stanford Tori). The source spreadsheet may be found at
http://www.nss.org/settlement/journal/EasyModel.xls.
Both tables use a set of “knobs” that may be ‘turned’ to define the basic parameters driving
the calculations for a series of columns of the various size habitats. Some values are derived
from others, but of course may be overridden.
Note that for simplicity, the models hide certain rows and columns not needed for this
discussion, but interested parties may unhide them and try additional knobs and settlement
sizes. They are further documented in the spreadsheet.
The basic comparison values for the reduction factor going to ELEO merely change the
shield mass from 6 tonnes/m2 to 0. The other parameters are unchanged, yet are (in
general) different from the original settlement designs, so a direct comparison to, for
example, the original Stanford Torus, would require additional changes (such as the
population density, structural materials strength and density, etc.).
For the Kalpana One cylinders, the aspect ratio is fixed at 1.00 (versus the 1.3 of “Kalpana
One Revised” [Globus 2007]). For the Stanford Torus, the original 6.9 aspect ratio of the 1
RPM torus is preserved for comparison, but the other tori each have a knob for aspect ratio,
and they vary from 6 down to 2 for the various sizes, in an effort to provide minimally
acceptable populations. Note that a still smaller aspect of 1.5 (3:2, equivalent to a donut)
may be used, but less material is required for a cylinder.
The chosen hull material is a carbon fiber composite with a very high strength (2,400 MPA)
and a relatively low density (1.3 g/ml), an appropriate choice for Earth-launched habitats.
14
Or millions of years, if you are worried about insufficient time for R&D.

An internal structural mass value of 0.2 tonnes/m2 is sufficient to provide an extra level of
structure with walls and a ceiling that can support the weight above it (thus there is an outer
rim and a level above; one would be for residence and the other for equipment, storage,
agriculture, work areas, and the like).
A non-structural per-person mass allowance of 7 tonnes is specified, which includes:
1.5 of plants (per the NASA 1977 study) for food and recycling of air and water
+ 1.0 of H2O drinking & hygiene & recycling
+ 0.5 of H2O for recreation (pools) & aesthetics (fountains, streams, and ponds)
+ 1.0 for furniture & fixtures
+ 1.0 for lighting & equipment (includes plumbing & power & cooling)
+ 0.5 for paper & plastics & textiles
+ 1.5 for agriculture & recycling overhead and equipment
A rich allowance would be 14 (double everything). A minimum allowance would be 3.5 (half
everything). The population capacity is defined as 40 m2/person of projected 1g area.
There is also a design structural strength margin for the hull. This is initially set to 300%, and
a bare minimum of 50% may be acceptable. This margin is used to calculate the hull stress
requirement which includes the contributions from air pressure, the 1 g weights of the shield,
internal structures and the total non-structural masses in tonnes/m2 of hull. Note that air
pressure alone is 10 tonnes/m2. The margin applies to all of these.
Calculations are made for the drag cross section (m2) and density (tonnes/m2) used in an
external calculation for the lifetime of the habitat assuming no orbit-raising maneuvers and
an initial altitude of 500 kilometers.
Here are the knobs and resulting tables (a portion of the tables are used for Tables 2 and 3
above):
KNOBS
1.00 Cylinder aspect ratio (width to radius)
2
0 Shield mass (tonnes/m )
2400 Hull structural strength in MPA

3
1.3 Structural materials density (tons/m )
2
0.2 Internal rim structures mass (tons/m of cylinder rim)
2
0 Internal endcap structures mass (tons/m of cylinder end caps)
7 Non-structural mass in tons/resident

2
40.0 Cylinder area per resident (m ) (population density) - they don't all have to live along
the rim!
2
0.0 Endcap area per resident (m ) (population density) - additional population
300% Design Structural Strength Margin

Row Values & Formulas

RPM: an input value, is the only parameter defining the cylinders
Radius: 1/((pi*rpm/30)2/9.8) in meters for 1.0 G
Width: radius * cylinder_aspect_ratio
Design_Population: rim_area/cylinder_area_per_resident
+ endcap_area/endcap_area_per_resident
Shell_Mass: (paraphrased) density*2*pi*radius3*(1+aspect_ratio)
*design_stress/(hull_strength-density*(9.8*radius))
this is essentially the density * volume * stress * aspect_ratio_correction_factor
/ (strength - gravity_self_stress) (all in proper units)
from Space Settlements: A Design Study (NASA 1977) p66
For a torus, the correction factor of an infinitely long cylinder (1.5)
is used instead of (1+Ar), and the formula for the volume of a torus is
applied instead of the volume of a cylinder.
For both, the minimum mass is set as the minimum_thickness * hull_area
Hull_thickness is simply hull_mass / hull_area.
Internal_structures is the projected_1G_area * internal_rim_structures_mass
plus a similar endcap calculation for cylinders only
Shield is simply hull_area * shield_mass (in tonnes/m2)
Non_Structural_Mass is the population * non_structural_mass_allowance
Air_Mass is 1.2 kg * total_volume (as all, converted to kilotonnes or kT)
Total_Mass is simply hull+shield+internal+non_structural+air masses
Reduction_factor uses an iteration to compute the total mass including 6 t/m2 of shield for
the “original” settlement size (2 rpm for cylinders, 1 rpm for tori), then plug in that value for
the other calculations using other shield masses.
Drag_Cross_Section used an iteration to determine the angle presenting the maximum area
for drag, which turns out to vary slightly with aspect ratio but is close enough to 55o for our
purposes. Then the calculation is sin(55o)*pi*r2+cos(55o)*width*diameter
Drag_Mass_per_m2 is total_mass / drag_cross_section
The torus_aspect_ratio is another input value varying per torus.
Additionally, the mass, area, and volume for spokes and hubs is computed for each torus,
assuming six 2-meter-diameter spokes and a 4-meter-diameter central hub (for docking), all
with a thickness the same as the hull.
Kalpana One-style habitats
RPM 1.00 2.00 3.00 4.00 5.00 6.00
Radius m 895 224 99 56 36 25
Width m 895 224 99 56 36 25
Design Population 125,637 7,852 1,551 491 201 97
Shell (kT or kiloTons) 2,030.79 31.62 2.77 0.49 0.13 0.05
hull thickness (m) 0.155 0.039 0.017 0.010 0.006 0.005

Internal Structures (kT) 1,005.10 62.82 12.41 3.93 1.61 0.78
Shield (kT) 0.00 0.00 0.00 0.00 0.00 0.00
Non-structural mass kT 879.46 54.97 10.86 3.44 1.41 0.68
Air mass kT 2,698.72 42.17 3.70 0.66 0.17 0.06
Total Mass (kT) 6,614.07 191.57 29.74 8.51 3.32 1.56
Reduction Factor 1 21 137 479 1,230 2,613
Drag cross section m^2 2,977,356 186,085 36,757 11,630 4,764 2,297
Drag Mass/m^2 (tonnes) 2.221 1.029 0.809 0.732 0.696 0.680
Stanford Torus
RPM 1.00 2.00 3.00 4.00 5.00 6.00
Outer Radius m 895 224 99 56 36 25
Aspect Ratio 6.9 6 5 4 3 2
Tube Diameter m 130 37 20 14 12 12
Shield Mass kT 0 0 0 0 0 0
Population 18,208 1,309 310 123 67 48
Shell Mass kT 66.968 1.379 0.228 0.088 0.046 0.030
Thickness m 0.024 0.007 0.005 0.005 0.005 0.005
Internal Structures kT 145.666 10.470 2.482 0.982 0.536 0.388
Non-structural mass kT 127.458 9.161 2.172 0.859 0.469 0.339
Air mass kT 82.587 1.687 0.209 0.057 0.025 0.017
Spokes & Hub kT 1.069 0.079 0.026 0.015 0.010 0.007
TOTAL MASS kT 423.748 22.775 5.116 2.000 1.086 0.781
Reduction factor 31 580 2581 6602 12158 16901
Drag cross section (m^2) 871,357 64,259 15,741 6,431 3,596 2,610
Drag mass/m^2 (tonnes) 0.486 0.354 0.325 0.311 0.302 0.299

References
[Bigelow 2017] “Bigelow Aerospace | B330.” http://bigelowaerospace.com/b330/. Accessed
July 10, 2017.
[Brophy 2012] Brophy, John, and Fred Culick. “Asteroid Retrieval Feasibility Study.” Keck
Institute for Space Studies, April 2, 2012.
[Carry 2012] Carry, Benoit. “Density of Asteroids.” Planetary and Space Science 73, no. 1
(2012): 98–118.
[Cassell 2015] Cassell, Chris. Personal communication based on calculations using AGI’s
STK.
[Connors 1985] Connors, Mary M.; Harrison, Albert A.; Akins, Faren R. Living Aloft: Human
Requirements for Extended Spaceflight (NASA SP-483). NASA Scientific and Technical
Information Branch.
[Cowings 2013] Cowings, Patricia S. “Psychophysiology of Spaceflight and Aviation.” Polish

Journal of Aviation Medical Psychology 19, no. 3 (2013): 9–16.
[Cowings 2000] Cowings, Patricia S., and William B. Toscano. “Autogenic-Feedback

Training Exercise Is Superior to Promethazine for Control of Motion Sickness Symptoms.”
The Journal of Clinical Pharmacology 40, no. 10 (2000): 1154–1165.
[Diamandis 1987] Diamandis, Peter H. “Reconsidering Artificial Gravity for Twenty-First

Century Space Habitats.” In Space Manufacturing 6: Nonterrestrial Resources, Biosciences,
and Space Engineering, B. Faughnan, G. Maryniak (Eds.), 6:55–68. AIAA, 1987.
[Explorersweb 2006] Explorersweb. “AdventureStats,” 2006.

http://www.adventurestats.com/tables/EverestAgeFat.shtml.
[Garriott 2014] Garriott, Richard. “Opening the Floodgates to Entrepreneurial Space

Exploration.” Luncheon talk presented at the 33rd International Space Development
Conference, Los Angeles, CA, May 14, 2014.
[Ghiassi-nej 2002] Ghiassi-nejad, M., S.M.J. Mortazavi, J.R. Cameron, A. Niroomand-rad,

and P.A. Karam. “Very High Background Radiation Areas of Ramsar, Iran: Preliminary
Biological Studies.” Health Physics 82, no. 1 (2002): 87–93.
[Globus 2007] Globus, Al, Nitin Arora, Ankur Bajoria, and Joe Strout. “The Kalpana One
Orbital Space Settlement Revised,” April 2007.
http://alglobus.net/NASAwork/papers/2007KalpanaOne.pdf accessed January 2017.

[Globus 2012] Globus, Al. “Our Solar System, Our Galaxy, then the Universe,” April 2012.
http://space.alglobus.net/papers/Expansion.pdf accessed December 2016.
[Globus 2017a] Globus, Al, and Joe Strout. “Orbital Space Settlement Radiation Shielding,”
NSS Space Settlement Journal, June 2017. http://www.nss.org/settlement/journal/NSS-
JOURNAL-Orbital-Space-Settlement-Radiation-Shielding.pdf accessed June 2017.
[Globus 2017b] Globus, Al, and Theodore Hall. “Space Settlement Population Rotation
Tolerance,” June 2017. http://space.alglobus.net/papers/RotationPaper.pdf accessed June
2017.
[Graybiel 1960] Graybiel, Ashton, Brant Clark, and J Zarriello. “Observations on Human
Subjects Living in a ‘Slow Rotation Room’ for Periods of Two Days.” Archives of Neurology
3, no. 1 (July 1960): 55–73. doi:10.1001/archneur.1960.00450010055006
[Graybiel 1977] Graybiel, Ashton, Earl Miller, and Jerry Homick. Biomedical Results From
SKYLAB: Experiment M131. Human Vestibular Function. NASA, 1977.
[ICRP 2000] Valentin, J. Pregnancy and Medical Radiation. Annals of the ICRP. International
Commission on Radiological Protection, 2000.
[ICRP 2003] Valentin, J. “Biological Effects after Prenatal Irradiation (Embryo and Fetus).”
Annals of the ICRP. International Commission on Radiological Protection, 2003.
[Johnson 1975] Johnson, Richard, and Charles Holbrow. Space Settlements: A Design
Study. NASA, 1975. http://space.alglobus.net/75SummerStudy/Design.html.
[Lafleur 2010] Lafleur, Claude. “Spacecraft Stats and Insights,” The Space Review, April 5,
2010. http://www.thespacereview.com/article/1598/1.
[Linnea 2010] Wahl, Linnea. “Environmental Radiation.” Health Physics Society, January
2010. http://hps.org/documents/environmental_radiation_fact_sheet.pdf accessed July 31,
2014.
[McKirdy 2014] McKirdy, Euan. “Fukushima Residents Cleared to Return Home amid
Ongoing Contamination Fears.” CNN, April 1, 2014.
http://www.cnn.com/2014/04/01/world/asia/fukushima-miyakoji-return/index.html.
[McKnight 2012] “We’ve Already Passed the Tipping Point for Orbital Debris” IEEE
Spectrum. http://spectrum.ieee.org//aerospace/satellites/weve-already-passed-the-tipping-
point-for-orbital-debris accessed September 24, 2015.

[Musk 2017] Musk, Elon. “Making Humans a Multi-Planetary Species” New Space. June
2017, 5(2): 46-61. https://doi.org/10.1089/space.2017.29009.emu presentation materials at
http://www.spacex.com/sites/spacex/files/mars_presentation.pdf accessed October 2016.
[NASA 2011] NASA. “Space Shuttle,” 2011.

http://www.nasa.gov/mission_pages/shuttle/shuttlemissions/index.html.
[NASA 2012] “NASA Orbital Debris FAQs.” Accessed September 26, 2015.
http://orbitaldebris.jsc.nasa.gov/faqs.html.
[NASA 2013] “Ten Thousandth Near-Earth Object Discovered,” June 2013, accessed 9
November 2015, http://science.nasa.gov/science-news/science-at-nasa/2013/24jun_neo/
[NASA 2015] “About the Space Station: Facts and Figures | NASA,” 2015.
http://www.nasa.gov/mission_pages/station/main/onthestation/facts_and_figures.html.
[NRC 2010] “NRC: Fact Sheet on Biological Effects of Radiation,” October 2011. Accessed
July 31, 2014. http://www.nrc.gov/reading-rm/doc-collections/fact-sheets/bio-effects-
radiation.html.
[NSS 2016] NSS. “Orbital Debris: Overcoming Challenges.” Position Paper. National Space
Society, May 2016.
http://www.nss.org/legislative/positions/NSS_Position_Paper_Orbital_Debris_2016.pdf
accessed January 2017.
[O’Neil 1998] O’Neil, Daniel, Ivan Bekey, John Mankins, Thomas F. Rogers, and Eric W.
Stallmet. “General Public Space Travel and Tourism - Volume 1 Executive Summary.” NASA
Marshall Space Flight Center, March 1998.
http://www.spacefuture.com/archive/general_public_space_travel_and_tourism.shtml.
[OLTARIS 2011] R Singleterry, S Blattnig, M Clowdsley, G Quallis, C Sandrige, L Simonsen,

T Slaba, et al. “OLTARIS: On-Line Tool for the Assessment of Radiation in Space.” Acta
Astronautica 68 (2011): 1086–97.
[OLTARIS 2014] “OLTARIS Home Page.” Accessed June 13, 2014.

https://oltaris.larc.nasa.gov/.
[O’Neill 1977] O’Neill, Gerard K. Space Resources and Space Settlements, NASA SP-428.
NASA, 1977. http://space.alglobus.net/spaceres/index.html.
[NSS 2015] “NSS Position Paper Next Generation Space_Stations 2015.” National Space
Society, April 2015.
http://www.nss.org/legislative/positions/NSS_Position_Paper_Next_Generation_Space_Stati
ons_2015.pdf.

[Panwar 1999] Panwar, Rakesh. “Satellite Orbital Decay Calculations.” The Australian Space
Weather Agency, 1999.
http://www.ips.gov.au/Category/Educational/Space%20Weather/Space%20Weather%20Effe
cts/SatelliteOrbitalDecayCalculations.pdf
[Space Adventures 2015] “Circumlunar Mission.” Space Adventures, Ltd. | Circumlunar

Mission. Accessed September 7, 2015.
http://www.spaceadventures.com/experiences/circumlunar-mission/.
[SpaceX 2017] “Falcon Heavy” 2017. http://www.spacex.com/falcon-heavy
[Statista 2014] Statista. “Travel & Tourism: Global Economic Impact 2014.” Accessed
September 28, 2015. http://www.statista.com/statistics/233223/travel-and-tourism--total-
economic-contribution-worldwide/.
[ULA 2016] United Launch Alliance, Transportation Enabling a Robust Cislunar Economy,
presentation,
http://www.ulalaunch.com/uploads/docs/Published_Papers/Commercial_Space/2016_Cislun
ar.pdf. Accessed July 10, 2017.
[World Nuclear Association 2014] “Naturally Occurring Radioactive Materials NORM,” June
2014. http://www.world-nuclear.org/info/Safety-and-Security/Radiation-and-Health/Naturally-
Occurring-Radioactive-Materials-NORM/.
[Wrixon 2008] Wrixon, A D. “New ICRP Recommendations.” Journal of Radiological

Protection 28, no. 2 (June 1, 2008): 161–68. doi:10.1088/0952-4746/28/2/R02.
http://iopscience.iop.org/0952-4746/28/2/R02/pdf/0952-4746_28_2_R02.pdf

Space Settlement An Easier Way

Uploaded by

Copyright:

Available Formats

Space Settlement An Easier Way

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Space Settlement An Easier Way

Uploaded by

Copyright:

Available Formats

Space Settlement: An Easier Way

Space Settlement: An Easier Way

NSS Space Settlement Journal, July 2017 Page 1

ISS International Space Station

NSS Space Settlement Journal, July 2017 Page 2

1. Make them close.

NSS Space Settlement Journal, July 2017 Page 3

Stanford/NASA Ames Mass Estimates

multiple torus 100,000 10,400 9,700,000 89

banded torus 112,000 13,200 7,000,000 57

single torus 4,600 1,900 1,000,000 155

cylinder 775,000 299,000 19,400,000 19

sphere 64,600 35,200 3,300,000 34

dumbbell 400 200 1,400,000 2,334

NSS Space Settlement Journal, July 2017 Page 4

Building up capability incrementally is a time honored engineering tradition because it works.

1. one man suborbital flight.

NSS Space Settlement Journal, July 2017 Page 5

A Tale of Two Studies

NSS Space Settlement Journal, July 2017 Page 6

From the equation for centripetal acceleration

NSS Space Settlement Journal, July 2017 Page 7

Figure 2: Kalpana One. Image credit Bryan Versteeg, used by permission.

NSS Space Settlement Journal, July 2017 Page 8

Kalpana One Mass Estimates

NSS Space Settlement Journal, July 2017 Page 9

Figure 3: Stanford Torus, image credit NASA.

NSS Space Settlement Journal, July 2017 Page 10

Stanford Torus Mass Estimates

Mass/person (T) 725 16.5 16.3

Population 18,000 310 123

drag mass/area (T/m2) 15

deorbit time 500 km (years) NA

NSS Space Settlement Journal, July 2017 Page 11

The Moon and Mars

Getting To and From ELEO is Relatively Easy

ELEO 500 10 hours

L5 382,000 14.1 3 days

Moon 382,000 16.4 3 days

Mars (at closest approach) 54,600,000 20.2 260 days

NSS Space Settlement Journal, July 2017 Page 12

NSS Space Settlement Journal, July 2017 Page 13

Study One: Radiation Protection

NSS Space Settlement Journal, July 2017 Page 14

NSS Space Settlement Journal, July 2017 Page 15

Radiation Shielding in Deep Space

Radiation Shielding Required Above the Radiation Belts

polyethylene water lunar regolith

Table 5: Comparison of shielding materials in deep space. The rows

NSS Space Settlement Journal, July 2017 Page 16

NSS Space Settlement Journal, July 2017 Page 17

Radiation Shielding Requirements in ELEO

tons/m2 500 km mSv/yr mGy/yr 600 km mSv/yr mGy/yr

~0 17.7 10.2 40.1 1,559

0.01 17.1 3.6 29.8 101

0.025 16.4 3.7 24.4 50.6

0.05 15.3 3.9 19.8 21.8

0.075 14.4 4.0 17.4 12.5

0.1 13.7 4.0 15.9 8.9

0.15 12.7 4.1 14.1 6.1