Site Selection for Lunar Industrialization, Economic Development, and Settlement


The subject of a lunar landing site/outpost/base has been explored extensively. Due to the cost and complexity involved, until now this has been almost the exclusive domain of government. In the United States we have gone through at least three generations of work in this area since the Apollo era. The vast majority of these plans and projects have been science driven, and scientific priorities have governed site selection and architecture. The general purpose here is to develop something fundamentally different. The general question to be investigated is; what would a non- governmental lunar development look like, premised upon economic development, industrialization and settlement? The specific purpose here is to zero in on a location so that further development and cost estimation can begin soon.

The question of lunar development and its importance was discussed in an exceptional activity at a major Silicon Valley venture capital office in August of 2014. Over 40 participants, most of whom have spent their careers in engineering, the sciences, and finance, were gathered to discuss the subject of what would a privately financed installation, to the tune of $5 billion dollars, to be operational by the end of the year 2022 look like? The 2022 project milestone would be a permanently inhabited installation that would initially house at least ten people on extended tours. Discussions were held about cost, implementation, economic activity, and so forth and it was encouraging to see how quickly, with the parameters enumerated, the various participants came to a consensus on the path forward. This missive will discuss site selection within the larger context of the overall project goal.


For lunar/mars exploration a loose taxonomy has been defined for classes of installations. These definitions have shifted with the direction of the political winds for NASA’s exploration program and so a definition of terms is important.


A sortie mission is a short duration stay with little or no infrastructure at a single or multiple locations. The Apollo missions fall into this class.


An outpost is a semi-permanent facility. It has habitats for humans, equipment for facility development, scientific exploration capabilities, and the like. However, it is not normally continuously manned, and the stays by the crew are short term (days/weeks vs months/years). This is the general style of facility described by NASA for a presence on the Moon and Mars over the last decade, though the same term was used for a much more robust presence during the Space Exploration Initiative (SEI) in the 1980’s and early 90’s. Also the crew is small with all members brought in or leaving on one flight.


This term was used originally all the way back into the pre Apollo era (1959) U.S. Army Horizons report. The Von Braun authored report (vol 1), (vol 2)had a large initial crew (12 men), carried to the Moon via a large modular and reusable vehicle assembled in Earth orbit. A base is continually occupied, has many occupants, and is used for science and other activities (a military base in the Horizon report). This is also the class of facility that was illustrated in the science fiction television show “Space 1999”.


The term development, from a mental imagery standpoint goes well beyond anything that government studies have focused on. However, these have been defined extensively in science fiction. Usually the word base or outpost is used. However, these are limiting in their imagery to what is possible for a commercial site. However, it is also premature hubris to call a lunar development a city. The German rocket scientist Krafft Ehricke went further than anyone with his “Selenopolis” as a city on the Moon. Of the three words in the title, at this time, the favored word is development. Development has many connotations but the foremost mental image is one of growth. We do developments all the time on the Earth for many purposes, thus this term gives freedom and while a development can and should encompass the requirements of science; it is not defined by them.

There are many different types of developments on the Earth. There are housing, industrial, mining, retail, commerce, sports, basically any type of, well development. Since it is our goal not to overly constrain the vision of those who embark on this quest, this missive will use the term development, and here is the definition in the current context.

Mixed Use

There are many examples of mixed use developments on Earth. In our context the mixed use incorporates industry, commerce, settlement, science, communications, transportation, and settlement.


It is becoming increasingly clear that our space based civilization will have a population of robots far larger than the human population. The development will begin this way out of the necessity to constrain costs, but will continue with its human population after the initial site preparation and construction has been completed. This is really not much different than a terrestrial development as heavy equipment for construction is required here on the Earth well before a development has its initial occupancy. The initial occupancy will be as few as two people to start and grow to its initial stage (the 2022 deadline) of ten people. Beyond that the growth in humans will be organic, growing as demand for humans increases. My opinion is that the demand for local humans will be great and we could see as many as hundreds by the year 2030 and thousands of robots. Beyond that there will be new developments located in many places on the Moon.

Sustainable (Not Self Sufficient)

Sustainable means that the development does not require constant resupply of everything needed in order to thrive. This means the massive implementation of Situ Resource Utilization (ISRU) and in situ manufacturing/farming to reduce the logistics burden and offset costs. No Earthly city is entirely self sufficient but they are sustainable. A desirable definition would be to generate enough economic output to justify its existence. Using this definition the U.S. Antarctic base is not sustainable, thus falls within the taxonomy listed here for base. It is the goal of the development on the Moon to be sustainable in that the economic product output exceeds the operational cost. The lunar development does not have to do this initially, but it is the absolute goal to do so. A path to exponential growth is strongly desired. As in any investment the goal is to develop a thriving economic ecosystem. Thus in this work we are aiming for a development.

Economic Output

A development must have an economy.  Not just an internal economy that offsets costs, but one where there is economic output that justifies the investment.  This also integrates the definition of sustainable because at the end of the day, without economic output our development just becomes Detroit (an American city who’s economy and population collapsed as its economic output nose dived) on the Moon.  What economic output is in detail will be left up to follow ons to this missive, but can and should be extensive.  In 2004 when president George W. Bush gave his speech on the Vision for Space Exploration (VSE), he made an astounding statement that was little noticed, then ignored by NASA but bears repeating here.

Establishing an extended human presence on the moon could vastly reduce the costs of further space exploration, making possible ever more ambitious missions. Lifting heavy spacecraft and fuel out of the Earth’s gravity is expensive. Spacecraft assembled and provisioned on the moon could escape its far lower gravity using far less energy, and thus, far less cost. Also, the moon is home to abundant resources. Its soil contains raw materials that might be harvested and processed into rocket fuel or breathable air. We can use our time on the moon to develop and test new approaches and technologies and systems that will allow us to function in other, more challenging environments. The moon is a logical step toward further progress and achievement.

It is my strong opinion that the sustainable development of a colony on Mars (advocated by Elon Musk and others, including myself), if not impossible, is far more costly without the support of a industrial and manufacturing development on the Moon.  Thus the Moon becomes an integral part of the economic development of the solar system.

These are all farther term grand economic schemes, however, the lunar development must have near term economic development as well.  There are many ways to do this, that we will address in follow on missives.  The thing to be established is that the sense of purpose is not to just be a place like our Antarctic bases but to define it a place of commerce, industry, and growth.

Lunar Development Site Selection Requirements

Requirements Development

Each type of site as we defined in the previous section has its set of requirements but we are only focusing on our development. Our lunar development has a lot of implied requirements but these are all constrained and defined by the input parameters of the discussion at the team meeting at the VC firm. These are called in NASA parlance, level zero requirements, or the project goals. All lower level (usually level 1 requirements are the highest functional level and then the lower level requirements are 2,3,4 and so on) requirements must adhere to during the development of a detailed plan.   Thus the level zero requirements are as follows:

  • Approximately $5 billion in development costs.
  • Operational by the end of 2022.
  • Economically sustainable as a core value with exponential growth possible.

These are the only real requirements for the project, but they are profoundly different than any government directed lunar plan, most specifically bullet three. These requirements flow down to all the rest of the requirements and influence how the architecture unfolds. In this missive I only go as far as site selection underpinned by the philosophical approach. Site selection is crucial, thus outline of what is required there in order to provide context for site selection.

Factors Influencing Site Selection

By definition any site selection must be constrained by the level zero requirements. Thus right away nuclear reactors are out due to development costs. Economic sustainability implies considerable power, because without energy there is no economic activity. Energy is life, work, air, resource extraction, processing, and communications, everything that you must have for a development. Thus if nuclear reactors are out, and plentiful energy is a must, then this begins to dramatically constrain and inform the architecture, site, and design of the lunar development.

Energy as a Site Selection Constraint

A lunar Synodic month is 29.5 days. A synodic month is the time it takes the moon to return to the same position in the sky relative to the sun, and thus this is the time period of interest, not the 27 day lunar orbit around the Earth. This is ~708 hours, the most important number for energy generation for solar power, which is the only viable alternative to nuclear in a cost constrained development.

In all areas of the Moon except for the polar regions, a lunar day is equivalent to its month. With a 708 hour day we have a night time of 354 hours, thus in any non polar area we not only have to supply power during the day, we have to supply it from some form of storage at night and have enough power surplus during the next day to recharge the energy storage system for the next night. This requirement drove NASA to select nuclear power, and it is one of the things that helped drive the cost of the NASA Space Exploration Initiative (SEI) of the 1980’s over the edge of what congress would pay.  Table 1 shows the difference in power available.

Table 1: Available Power Calculation, Non Polar, North and South Pole Sites
Table 1: Available Power Calculation, Non Polar, North and South Pole Sites

The analysis above is with a 100 kilowatt power supply. We used known factors of day and night, and averaged over the 708 hour period with enough reserve to recharge the energy storage system (not presuming what the storage system is). The average power available is 23. 5 kilowatts for a non polar location. This is clearly inadequate when you consider that the average power on the ISS is about 25-30 kilowatts for six people and little energy intensive activity. Listed here is a selection of sites with known illumination at the lunar poles in the paper by [Mazarico et al., 2011].[i] Figure 1 shows the numbers associated with the sites:

Figure 1: Sites of Maximum Illumination for North (a) and South (b) Polar Regions
Figure 1: Sites of Maximum Illumination for North (a) and South (b) Polar Regions

Site 35 on the (a) map roughly indicates the north lunar pole. The southern lunar pole is to the upper right of site 4 along the crater rim of Shackleton crater. Based on previous work site 1 in the south (1S) and site 2 (2N) on the intersection of the rim of Whipple and Peary craters in the north were chosen.[i] These sites are representative of the best areas in their respective polar regions.

Site 2 North

It turns out that 2N in the north has a total illumination of 598.7 hours of illumination out of a total of 708 hours a total of 84.56% of the time. The time in darkness is only 109.32 hours, well less than a third of non polar sites. This makes a dramatic difference in the power profile. The total energy available rises from 35.4 megawatt hours to 59.9 megawatt hours at ground level and 61.2 megawatt hours a mere 10 meters above the surface. This raises the average hourly energy available to 66.5 kW/hr at ground level and 69.46 kW/hr at 10 meters altitude. This is almost three times the power of a non polar site for the same hardware.

Site 1 South

At site 1S in the south the numbers are better than in the north. The total sunlit time rises to 89.01%, which gives a total energy of 63.0 MW/hr on the ground and 65.9 MW/hr at 10 meters altitude. This gives an average power provided to the facility per hour of 73.68 kw/hr at ground level and 80.61 at ten meters altitude. This is considerably better than some of the previous studies on the subject and is 116% of the average at site 2 north and well over three times that of a non polar site.


If cost minimization and maximum power production are requirements, then it is not even a competition between the polar regions and other areas of the Moon. The difference between the north polar location 2 and the south polar location 1 is substantial, about 7.2 kW/hr at ground level and 11.2 kW/hr at ten meters. If the terrain in the southern region were better this would indicate a clear win for the south polar location. Figure 2 shows a notional “Power Lander” that would have solar arrays that would track the sun in a 360 degree angle at the poles:

Figure 2: Power Landers On the Moon
Figure 2: Power Landers On the Moon

The power lander shown in Figure 2 was designed for a commercial lunar architecture. The vehicle can land approximately 100 kilowatts of solar panels, batteries, and the power conditioning system (480 volts, 60 Hz AC) can simply be plugged into using flight qualified variations of today’s standard plugs. The launch vehicle for this would be a Delta IVH or a Falcon heavy launcher.

It is beyond the scope of this paper to go into detail but our analysis showed that for a viable development on the Moon that could generate significant economic output, at least seven of these would be needed. The average power at site 2N would be about 428 MW/hr total or and hourly average of 486 kW/hr. At site 1S it would be 461 MW/hr and 564 kW/hr respectively.  For a non polar site this same hardware would only generate 248 MW/hr total and 162.75 kW/hr average.

Some of these same locations at the poles that are in full illumination on a seasonal basis depending on where the Moon is at in its 18 year processional cycle. The Mazarico paper illustrates this feature. It has also been theorized that at certain heights above the terrain, above 100 meters, there is full time illumination at both sites. However, for the purposes of this study we will use the average numbers.

An argument can be made, and was at the event, that power is not everything, that resources are important as well. This is a true statement, but without power, resources are worthless. Indeed, your ability to obtain and process resources is directly proportional to your available power. You can always drive to the resources, which will be required no matter where you are sited, but fortunately the vast majority of the volatile resources of the Moon are nearby the sites chosen. We will deal with resources in a following section but next we deal with communications.


Communications is another subject for a lunar site that gets people excited, which usually ends up with bigger ideas which cost a lot more money. With two constraints, one being money, and the other time, what can we do to maximize communications and how does this effect the site selection process? Figure 3 is another graphic from the [Mazarico et al., 2011] paper:

Figure 3: Average Visibility of the Earth from Lunar Polar Regions (a) N, (b) S.
Figure 3: Average Visibility of the Earth from Lunar Polar Regions (a) N, (b) S.

As figure 3 indicates, neither 1S or 2N sites are particularly good from the standpoint of Earth communications. 22N in the north and 33 S in the south both have better visibility and are both in high illumination areas. One thing that is for sure, in this day and age, high bandwidth communications is a must. NASA recently tested the LADEE laser communications link in lunar orbit with a 622 megabits/s downlink data rate. This technology is maturing fast and can be considered the fat data pipe to and from the Earth. A conservative 10 gigabits/s would be for phase 1 (through ten people) of the lunar development.

There are basically two choices that the lunar development architect has when looking at the two sites in terms of communications. The first is to choose a ground based or a space based communications infrastructure. It is quite clear that there will be one dedicated lander at the development site that has the high bandwidth laser-com system. What is not clear at this level of effort, is whether that is the systems solution that lowers the overall cost of the development. The easiest answer is to just have a relay in Earth/Moon L1 and be done with it. That would work for either 1S or 2N. However, that is a single point systems failure that one would like to avoid. An interesting compromise would be to place the laser relay at 22N or 33S. Neither has 100% connectivity to the Earth but there is a side benefit. If you have a set of RF relays at both 1S/33S or 2N/22N you can cover hundreds of square km for local wireless communications and you get ranging as well.

With one station you only get ranging distance to a rover, digger, or water harvester that may be in a permanently dark area, but with two with good separation (tens of km) you have a much more accurate system. Couple this with a technology such as ultra-wideband radio, which can operate at much higher powers on the Moon than the Earth and you get a very wide area communications and ranging system. This argues for two stations at 33S or 22N which can be used to extend the data rate to the Earth when needed, especially for tele-presence operations.

So, in terms of site selection communications is a draw. A much more detailed systems analysis is needed but to the first order a couple of communications stations, coupled with a single satellite in Earth/Moon L1, should deliver all the needed communications for the lunar development.


General Lunar Resources

The moon is rich with resources. Table 2 shows the bulk percentages:[iii]

Table 2: Composition of Lunar Landing Sites in Elemental Percentages
Table 2: Composition of Lunar Landing Sites in Elemental Percentages


The elemental compositions in table 2 are from the Apollo (A11-17), and Luna (16, 20, 24) landing sites. The AFHT in the table is from [Korotev et al., 2003] and represents an average of highlands material as derived from studying lunar meteorites. The problem is that these elements are bound to pesky oxygen molecules. Table 3 shows the Average in molecular percentages:[iv]

Table 3: Apollo/Lunar Composition of Lunar Soils, Rocks, and Minerals Fractions
Table 3: Apollo/Lunar Composition of Lunar Soils, Rocks, and Minerals Fractions

Finally, from the same chapter in Resources of Near Earth Space, in table 4 are the molecular constituents of highlands rocks:

Table 4: Lunar Composition of Lunar Soils, Rocks, and Minerals Fractions
Table 4: Lunar Composition of Lunar Soils, Rocks, and Minerals Fractions

Both polar regions of the moon are considered highlands regions so table 2 with AFHT and table 4 are the closest datasets that we have. These are just bulk compositions but they should inform the reader that there are far more resources on the Moon than just water. The issue with resources is having enough energy to do the work to separate the oxygen from the metals. Whether or not there is significant repositories of water ice and other hydrogen bound molecules of economic interest in the lunar polar regions, these metal oxides abound. There are also free metals on the moon from the impact of M class asteroids. Some Apollo samples had up to 1% metals in the regolith. This was a highlands sample from Apollo 16 and since both poles are highlands type sites, this will also be a resource, but it is also a draw from our perspective here in determining a site for the lunar development. There is no easy method of extracting metals from oxygen without a lot of electrical, thermal, or chemical energy. However, with the up to 1% free metals from meteorites in this highlands terrain, scooping up regolith and processing it for metals and volatiles is the safe bet for the first go at ISRU. Then there are the rich polar volatile resources that we are just learning about.

Volatiles in the Polar Regions

Figure 4 illustrates our state of knowledge about volatiles on the Moon:[v]

Figure 4: What We Know about Volatiles on the Moon
Figure 4: What We Know about Volatiles on the Moon

This is another thing that differentiates the polar region from any other area of the Moon. Water and or other volatiles are the game changer for building an economically sustainable lunar development. Water, or at least hydrogen which can then be bonded with oxygen to make water, is incredibly valuable on the Moon. Today it costs about $100,000 per kilo for a payload to the lunar surface. Sending water up from the ground has another issue, which is that it is the hydrogen that you want, not the oxygen due to the plenteous nature of oxygen on the Moon. Oxygen is 31 times heavier than hydrogen and thus if you ship water, you are wasting a lot of payload space. Hydrogen sent up has the problems with keeping liquid hydrogen cold at 20 degrees kelvin, which equals expensive, thus if you have hydrogen bearing molecules on the Moon costs can be dramatically reduced.

Rather than go into an extensive examination of the literature on the subject lets remain focused on the point here, which is to determine whether the north 2N, or the south 1S is superior from the perspective of access to what we think are the sources of H bearing molecules on the Moon. Figure 5 shows the extent of Permanently Shadowed Regions (PSRs), which are indicative of thermal conditions fostering the retention of hydrogen bearing molecules [Mazarico et al., 2011]:

Figure 5: Permanently Shadowed Regions (Average Over 4 Precession Cycles)
Figure 5: Permanently Shadowed Regions (Average Over 4 Precession Cycles)

According to the paper, the total PSR area in the northern polar region is 12,866 sq/km while in the south it is estimated at 16,055. The PSR area in the south is much higher than the previous estimates that our team has used [Bussey et al., 2003]. The PSR area in the south is 2.5 times larger in this paper than in the previous work so the discrepancy must be further investigated. Table 5 from the paper shows the size distribution of PSRs in each polar region:

Table 5: Size Distribution of PSRs For Both Poles
Table 5: Size Distribution of PSRs For Both Poles

The area of the PSRs is substantial at both poles. The much larger size of the south polar region PSRs could be checked against the Moon Mineralogy Mapper (M3) 3 micron absorption bands.[vi] There is another means whereby to test the theory about volatiles in the PSR regions of the Moon. Figure 6 shows the results of radar imaging from the Indian Chandryaan-1 Mini-SAR radar:[vii]

Figure 6: Circular Polarization Ratio (CPR) of the Northern Region of the Moon
Figure 6: Circular Polarization Ratio (CPR) of the Northern Region of the Moon

The above map showing the circular polarization return (CPR) of the Moon’s northern region [Spudis, et al., 2010] and this can be mapped against the PSR values. The Spudis paper differentiates between fresh craters that also would have high CPR values due to scattering of the radar beam by fresh material, and high CPR values from other craters that may be water ice or other hydrogen bearing molecules. The Mini-SAR data indicates its presence in the northern and southern polar regions. This also can be checked against data gathered by the M3 instrument [Pieters et al. 2009] that discovered mobile water and hydroxyls that drift from the lower to the higher latitudes of the Moon. A good study would be to harmonize the results of these works as part of the lunar development’s mission planning.

Besides these higher values for hydrogen bearing molecules in the polar regions in the PSR regions, the entire area has an elevated level of these resources (10-100x the equatorial regions [see table 3 hydrogen ppm]). These elevated numbers indicate that even modest processing of bulk regolith will provide between 1-10 kg of mostly water per square meter of regolith near the poles. While this may not be enough (without extensive regolith movement and cooking) for propulsion, it would easily be enough to provide the crew health and other water needs for the development while the infrastructure is still building up to acquire the higher order resources from the PSR sites in the craters.

In winding up this section on resources it seems that we still have no clear winner between a site at 1S or 2N. This was unexpected and the new results from figure 5 indicate that if you strictly look at PSR values the 1S site is better. However, we know from some of the other remote sensing that the north has the most hydrogen bearing molecules. One thing that might tip the scales to the north is the plentiful nature of small PSR regions. No one has ever built a rover or any other equipment that can operate for long periods of time in temperatures not much higher than liquid hydrogen. This is going to be probably the biggest technical challenge of the entire project. On the other hand the small PSR regions in the north could enable a “dine and dash” strategy where the water capturing equipment dashes into the PSR area grabs a large chunk of the resource and then dashes back out into the sunlight. This would reduce the energy required for heating the unit and could minimize the risk of the infrared heat of the equipment destabilizing the ices themselves.


To me operations is the most interesting aspect of a lunar development. After sorting through the issues of power, communications, and resources, operations makes or breaks the success of the effort. The term operations in this context, comprises all of the things necessary to make the development work. The bullet points covering the scope of operations are as follows.

  • Earth/Moon Transportation
  • Site Preparation and Buildup
  • Local Communications
  • Resource Acquisition
  • Energy Management
  • Long Term Growth

Site selection heavily influences all of the above operational issues and the purpose of early site selection is to lower development cost and maximize the potential upside. Optimizing for only one parameter is guaranteed to drive up the cost of the others. Examining each of the above we can arrive at a gestalt that informs our decision related to which site has the most potential for future lunar development.

Earth/Moon Transportation

Payloads and the transportation network between the Earth and the Moon dominate the capital cost of a lunar development. A significant level of buildup activity must occur before any significant revenue can be generated; this costs time and money. Another thing that costs time and money is scheduling. The polar regions are much better for transportation due to the orbital dynamics involved, as you can launch at any time from the Earth, landing three days later. For non polar sites the wait time averages two weeks for the Earth and the Moon to get properly aligned for a low energy (lower cost) mission. Ironically, we continue with the 50/50 split between the 1S and 2N sites as the energy to each location is identical.

Site Preparation and Buildup

Site preparation and buildup sets the pace for the future success of the lunar development. Mobility, the ability to move around while expending a minimum of energy, is a key determinate for early success. Figure 7 shows a 3X exaggerated terrain for both the north and south polar regions up to 2.5 degrees from the poles:

Figure 7a: LOLA 10 Meter Gridded Terrain from 87.5 Degrees to the North Pole
Figure 7a: LOLA 10 Meter Gridded Terrain from 87.5 Degrees to the North Pole
Figure 7b: LOLA 10 Meter Gridded Terrain from 87.5 Degrees to the South Pole
Figure 7b: LOLA 10 Meter Gridded Terrain from 87.5 Degrees to the South Pole

These LOLA gridded terrain maps were produced in a paper for a poster session at LPSC 2012 [Epps and Wingo, 2012] regarding tele-presence lunar rover traverses near the lunar north pole[viii]. In our investigation we rejected the south polar region for route traverses for a major reason, the roughness of the terrain. There are very few routes in the southern polar region for distance travel without excessive terrain excursions. The reason for this is that the area of the South Pole is within the rim of the south pole Aitken Basin, the oldest major basin on the Moon which dates from the Pre-Nectarian period [Stuart-Alexander, 1978; Wilhelms et al., 1979][ix],[x]. This, along with an abundance of other large craters, results in very difficult terrain for traverses. On the other hand, the northern polar region has much less difficult terrain, especially in the direction towards the better known near side Mare region (indicated by the arrow).   If you look at figure 7a, even with the 3X exaggeration, the terrain is flat for a distance of almost 80 km across the floor of the crater.

Referring to figure 7a again, the driving routes from the 2N development site to the nearest small PSRs in the floor of Peary crater are easily discerned. Figure 8a (left) shows the terrain and 8b (right) three driving routes from site 2N:

Figure 8a: Slopes in the Area of Peary/Whipple with 200 Meter Elevation Contours Figure 8b: Examples of Driving Routes from Site 2N to the First Small Peary PSRs
Figure 8a: Slopes in the Area of Peary/Whipple with 200 Meter Elevation Contours
Figure 8b: Examples of Driving Routes from Site 2N to the First Small Peary PSRs

With the availability of the highly accurate LOLA derived digital elevation models for the polar areas it is possible to accurately plan traverses that avoid steeply sloped terrain. Additionally, the LROC 1 meter resolution images will allow traverse planning that avoids small scale hazards such as small craters and boulders.

We have developed a set of tools for route planning in the polar regions. Candidate routes can be initialized by visual inspection of co-registered datasets based on their proximity to locations of interest, avoidance of hazards, and predicted capabilities of vehicles. In Figure 8b three routes are shown. Route one is a short distance, high slope angle route to small PSRs in the northern floor of Peary. Route two is a longer, moderate slope route. Route three provides an example of a significantly longer minimal slope route to the floor of Peary. Also, it passes small PSRs outside of Peary itself. Figure 9a (left) and 9b (right) shows how the LOLA data can be processed for route planning for route two to give distances and slope angles:

Figure 9a: Elevation Vs Distance Route 2:
Figure 9a: Elevation Vs Distance Route 2:
Figure 9b: Slope Histogram Route 2
Figure 9b: Slope Histogram Route 2

The tools that we have developed, based on LOLA, LROC, Lunar Orbiter, and other data sets, form a powerful basis for route planning, resource acquisition, energy management, and communications network planning for a surface development. In our initial deployment of these tools we have found the critical differentiation between the sites 1S and 2N, with 2N decisively better from an access to resources, local communications, and driving routes, including all the way to Mare Frigoris.

For a cost effective site buildup it is imperative that at the earliest possible moment the lunar development begins to use locally derived resources. This will begin with regolith handling, sintering landing pads, and road grading. The tools that are within our grasp now can allow us to accurately plan these activities and to simulate in virtual space and through analog sites on the ground the means to most cost effectively implement the lunar development.

Local Communications

Local communications will only be cursorily addressed in our focus on site selection. However, there are some interesting things to point out. Since most of the outside activity will be conducted with robotic agents of one type or another, high bandwidth local communications as well as ranging will be important. If ranging is done right, then it becomes possible to program swarm behavior for robotic systems, thus influencing the overall robotic systems design. Instead of a few large regolith movers, many small ones can do the task, and most probably, without human intervention after the system is set up.

This idea provides a set of implied requirements for communications and ranging. Here on the Earth we have cell towers and GPS for this, but on the Moon an orbital lunar GPS may not be cost effective, at least in the near term. Thus in the evolving the lunar development site, communications towers for advanced Wi-Fi and ranging will probably be needed. The same tools that allow for traverse planning can be used for communications network development.

It is also quite clear that the very first lander that goes to the development site carry payloads for communications, and some power provisioning as well for follow on systems. This also includes a beacon for automated landing of follow on payloads as well as a local Wi-fi/radar system to assist mobile robotic systems.

Resource Acquisition

Resource acquisition and its detailed implementation, is also beyond the scope of this document. However, site selection has been strongly influenced by the presence of polar water and highlands type resources, known only by the ground truth of the Apollo missions. We know what the minimums are and thus can begin to plan now for these minimum level resources to see if the level zero requirements can be met. Project success hinges on the scale of resources and our ability to obtain and process them. To help us to this end, there have been some amazing detailed studies done in the 1970’s and 80’s by a company called Eagle Engineering that must be examined and revisited with the technologies of today. At this time, with the probable water and metallic resources we can do a good job to scope the project and see where we would be at the end of the deadline. It is our strong opinion, based upon the evidence presented here, that site 2N in the north is our best candidate, though it is imperative to obtain ground truth as early as possible.

Energy Management

Energy management is absolutely crucial to the success of the lunar development. The tools that we have developed can be used to calculate the work required to drive a certain traverse so that energy vs distance vs load calculations can be developed. This will feed back into the power budget for the lunar development to see if what we have estimated back in table 1 is adequate. Indeed, we should be able to model a considerable amount of the effort a priori in a virtual environment to choreograph the development of the site. Developing these tools and the virtual environment would be a powerful first step towards validating the entire concept. This extends to the energy management of the habitats, food growing, resource extraction, and other activities.

Long Term Growth

It is obvious that we want this project and lunar development to extend past the end of 2022 date. Indeed even before that date we should be well on our way to revenue and evolving beyond the initial plan. It is not the purpose here to delve deeply into that but we can look at the site as per its value to longer term lunar development. Figure 10 shows something interesting:

Figure 11: Minimum Traverse Distance from Site 1S to the Nearside Lunar Mare
Figure 10: Long Distance Lunar Route Planning

In just a few days of driving the three Apollo lunar rovers with two crew persons traversed a total distance of over 90 kilometers. This was over all types of terrain, with single use 4×4 vehicles. In figure 10 above there is a mapped route that has been roughly validated using detailed Lunar Orbiter and Lunar Recon Orbiter images. With the increasing density of the LRO LOLA laser altimeter data we should be able to develop terrain maps with LROC image overlays to implement the same type of traverse planning as for the polar regions. An interesting fact is that the traverse above is only a little over 300 km from the development site at Whipple crater to the nearest area on the near side of the Moon (north of Mare Frigoris) that then gives access to the entire near side with very mild terrain from then on.

Surface level access to the entire near side of the Moon is the first exponential growth upside for the lunar development. Propellant is expensive, and to jump from one area of the Moon to another costs almost as much as going to orbit and back, thus is to be avoided. A lunar surface transportation network, allowing medium to high speed transport opens the entire resource base of the near side of the Moon to development. Doing this same thing from the south is more difficult as the distance from the 1S site to the Mare is over three times farther, as shown in figure 11:

Figure 11: Minimum Traverse Distance from Site 1S to the Nearside Lunar Mare
Figure 11: Minimum Traverse Distance from Site 1S to the Nearside Lunar Mare

Now with more work we may be able to lessen this distance but it is not feasible that it will be anywhere near as close as from the northern 2N site to the Mare.

With the near side open to development, the potential for exponential growth of the lunar development is high. Deposits of titanium, thorium, meteoric metals, all become available. The Moon was once called the slag heap of the solar system but it is clear from more recent data that there are very significant metals resources on the Moon. To access, process, and utilize them will require energy. Fortunately there are concentrated thorium resources on the Moon. Figure 12 shows a map of thorium concentrations on the lunar near side:

Figure 12: Lunar Nearside Concentrations of Thorium (From Spudis and NASA)
Figure 12: Lunar Nearside Concentrations of Thorium (From Spudis and NASA)

The concentrations of thorium noted in some of the craters is very interesting. The nearest concentration in the crater Aristillus is no more than 800 km from the polar development. Thorium reactors could be developed in situ on the Moon to provide tens of megawatts of power, plenty to begin real lunar industrialization. Plentiful lunar power enables the economic development of the entire solar system.

The Gestalt

Gestalt is a word from the German that basically means a whole that is greater than the sum of its parts. This missive has brought together many of the parts that make up the trades toward choosing a development site on the Moon. While we think that the north polar site is the best, there are still reasons to continue with plans to put landers in both areas and to explore both polar regions as well as non polar ones for their resource potential. However, it is our considered conclusion that by beginning at the northern lunar polar site 2N on the rim of Peary and Whipple, this provides the greatest leverage at the lowest cost for a commercial lunar development.

By covering the different high level trades and capabilities of both sites, we can start to get a feel for how a lunar development can be built out. It allows us to start bringing much higher fidelity to cost estimates, we can start to design the robotics, the rovers, and other heavy mobility systems based upon known terrain, something that was not possible before the LRO mission, still ongoing today. A concentrated study based on this site should be able to bring solid costing and a baseline of capabilities for a lunar development. This is where we would advocate some near term funding to pursue this, but without the requirements that a government style contract to do it NASA’s way would bring. For the first time in history we have at leas the minimum information necessary to do this task.

Next Steps

The meeting here in silicon valley was about doing something, not just getting together, talking, maybe writing a paper and then going on to the next shiny interesting space idea. The ideas presented in this missive are derived from our earlier work, and the output from the meeting. It was this idea, to pick a site, to explore what can be done there, that became a powerful element of consensus of the group. The evidence presented here does not preclude the 1S site, indeed it confirms it as an alternate or the phase II of the development. However, the level zero requirements, forces a triage of choices and a focusing of the effort. We don’t have unlimited amounts of money and time to send multiple missions to multiple locations, and then several years down the road make a decision. There is an old saying that the perfect is the enemy of the good. The 2N site is not perfect, but as we have gone through the factors that make for a good site, it would take a lot of work to find one better. That being said, if the first landing finds the site unsuitable, then so be it.

This brings us to our next steps. One thing that is dramatically different today than in previous efforts to develop lunar architectures, we know a lot more about the Moon, from multiple missions. The data products by the Chandrayaan-1, Kayuga, SMART-1, and now the LRO mission (and the scientists who interpret the data) have revolutionized our understanding of the Moon. Beyond that, the new missions have built upon and validated earlier efforts like Clementine, Lunar Prospector, and even Lunar Orbiter and Apollo, to allow a reinterpretation of data from that era in the light of our new knowledge. However, it is now time to do more, to go beyond. Following is a series of recommendations on near term and cost effective steps to pushing lunar development forward.

Data Set Integration

None of the currently existing software packages that have lunar data (Google Moon, Act-React, or the NASA LMMP project) lunar data interfaces is up to the task of what we want to do for mission planning for lunar site development. The tools that we have developed are suitable but need integration into a better global framework that incorporates surface and orbital activities. This would be a very good project, could be open source, and incorporate citizen science and student participation.

Virtual Environments

The global framework developed for data integration could itself be integrated into a virtual environment. Virtual environment technology is on its third iteration of its attempt to become mainstream and with modern computing power, cloud computing, and high bandwidth connections we may be at a point to where the promise of the VR world can finally be brought to bear to solve real world problems.

A lunar VR environment, incorporating the data sets knowledge bases could be a powerful tool for prospecting, operational scenario development, and refinement of designs from various teams. A further integration of real world engineering software for thermal environmental testing, structures, CAD/CAM, additive manufacturing and 3D printing could be a template for building a Tony Stark (the VR environment from the first Iron Man movie) type prototyping environment that could greatly advance the design of a lunar development as well as iron out some of the problems before we get there.

Analog Sites

Current NASA analog sites like Desert RATS that has been done at NASA JSC and in the field in Arizona and the PISCES site in Hawaii are templates for this activity. Figure 12 shows some of our hardware that was at Desert RATS in 2010:

Figure 12: Solar Powered Satellite Communications and Power Infrastructure
Figure 12: Solar Powered Satellite Communications and Power Infrastructure

At Desert RATS in 2010 the NASA Ames center director, over a two hop (Ames-GEO-Earth-GEO-Desert RATS) internet connection, with a 2 second time delay, operated a NASA Ames rover from his office through our hardware shown in figure 12. This was done through a collaboration with the Challenger Center and NASA JSC at the event. Our analog of our Power Lander concept was used to provide power to the NASA habitation unit shown on the left while our communications system provided a link via satellite to a local Wi-Fi setup. We were able to remotely monitor and control our system at the site from a location in Maryland and from California.

These types of activities, when focused on the lunar development goals, can help to move development into rapid progress. Using a Maker community approach we can bring many stakeholders to the table and into the field to bring equipment to test and to validate operational procedures developed in the virtual environments. A good possible place would be at NASA Ames, at least for initial developments, and then into the mining country of California or Nevada.

Wheels on the Ground

It is absolutely imperative to get wheels on the ground, at site 2N as a first choice. Nothing substitutes for in situ data. Still to this day we are reliant on the Apollo and Luna ground truth samples to calibrate our orbital missions. We have excellent terrain and multispectral remote sensing (from orbit) data from the 2N and 1S sites and these must be validated and expanded upon, taking into account local conditions of the regolith.

There is nothing that substitutes for actual data derived from systems on site. This is why NASA has spent so much money on the Mars landers over the years and the scientific payoff has been large. We also know from the plethora of rovers on Mars that there are huge instances of metallic meteorites, a concentrated source of metals that will make building a Martian civilization much easier. It has been my thesis that these objects are on the Moon as well. We know that highlands terrain has more of this than the Mare regions from the Apollo ground truth. Knowing exactly what is there allows for a more intelligent planning for further operations. We know enough now to architect this lander, so this is a good first project and test of the intentions of the benefactors.

Mission and Project Development

Mission development guides, and is guided by, data integration, virtual environments, and analog activities. A well thought out plan that has been put through its paces in the virtual and analog environments has a far higher chance of success than one that does not take advantage of the latest developments in computer based engineering. It is from these activities that confidence can be built, problems spotted before they become expensive in flight failures, investment garnered, and public involvement fostered. When people see what is going on and when anyone can contribute in a meaningful way, the possible perception that this is just rich people’s folly can be mitigated. This must be a well funded effort, as it is impossible to get the fidelity needed to execute on a plan without some serious dedicated effort by a team who spends their available time on the project.

Final Thoughts

It is not our goal to build some kind of space utopia like in the movie Elysium. Our goal is to help develop an off planet economy and begin the development of the resources of the solar system for the benefit of all mankind, everywhere. Most of us who were at the event have the strong opinion that the Moon is the first significant step for mankind. Elon Musk believes in the colonization of Mars, a goal that we also share. However, for a robust colonization of Mars we also feel that a vibrant industrial economy on the Moon, and then the asteroids will bring together Mars and the Earth into an inner solar system wide economy. This is no longer the realm of science fiction. It is time to undertake these activities. It is also our concern that if our society only focuses on the resources of the Earth and keeping our gaze inward, we will lose sight that we are not the center of the universe and as we turn into snarling dogs, fighting for what we rightly feel is our piece of the Earthly pie, we all lose.

The pie is much larger than just the Earth and there is truly enough for everyone, we who have spent our lives in this realm have all the evidence we need to convince ourselves of this. There are also others of means who share our faith in the future and in the possibilities of technological development to bridge the gaps that too many well meaning people see no means of solving except through a regimented society. In a new book by Blake Masters and Peter Thiel called from “Zero to One” the difficulty of bringing imaginings into reality are explored. The book discusses the difference between horizontal progress, which is copying what works somewhere and the replicating that everywhere. Vertical progress which from the title of the book 0 to 1, is the jump in societal capability that technology brings. The premise is valid in that especially in space where building a self sustaining and then exponentially growing off planet civilization is the ultimate vertical progress.

The book’s premise is that our biggest successes as a global civilization in recent decades has been in horizontal progress. This is not a bad thing, and while it is necessary, it is not sufficient to bring what we need to continue in the progress of civilization. If we want to solve the problems of the twenty first century it is time to resume our vertical climb. Doing this will do more to make the world a better place and eliminate threats than any other activity we can engage in.

If done properly, and I believe it can, a lunar development conforming to the level zero requirements and goals outlined here can be achieved. This missive is not the plan, but it can be the beginning of the plan and I see absolutely nothing that precludes this from working. This could not have been said even five years ago, but recent advances in computers, robotics, 3D printing, and of course the launch vehicles of SpaceX has shifted the equation decisively into our favor.

The future is yet to be written…..

[i] Epps, A.D, Wingo, D.R.; Integrating LRO Data Products for Preliminary North Pole Rover Mission Planning, LPSC-2700, March 2012, Houston Texas

[ii] Mazarico, E. et. al; Illumination Conditions of the Lunar Polar Regions Using LOLA Topography, Icarus 211 (2011) 1066-1081

[iii] Prettyman, T.H. et. al; Elemental Composition of the Lunar Surface: Analysis of Gamma Ray Spectroscopy Data from Lunar Prospector, Journal of Geophysical Research, Vol 111, E12007, December 21, 2006

[iv] Waldron, R.D.; Production of Non-Volatile Materials on the Moon, Resources of Near Earth Space, University of Arizona Press, 1990, P262,

[v] Sanders, G.B. et. al; RESOLVE for Lunar Polar Ice/Volatile Characterization Mission, EPSC Abstracts, Vol. 6, EPSC-DPS2011-Preview, 2011 EPSC-DPS Joint Meeting 2001

[vi] McCord, T.B., et al., Sources and Physical Processes Responsible for OH/H2O in the Lunar Soil as Revealed by the Moon Mineralogy Mapper (M3), Journal of Geophysical Research, Vol. 116, E00G5, 2011.

[vii] Spudis, P.D., et al.; Initial Results for the North Pole of the Moon from Mini-SAR, Chandrayaan-2 Mission, Geophysical Research Letters, Vol. 37, L06204, 2010

[viii] Epps, A.D., Wingo, D.R.; Integrating LRO Data Products for Preliminary North Pole Rover Mission Planning, Poster 2700, LPSC 2012 Houston Texas, March 2012

[ix] Stuart-Alexander, D. E. (1978), Geologic map of the central far side of the Moon, Scale 1:5,000,000, U.S. Geol. Surv., I-1047.

[x] Wilhelms, D. E., K. A. Howard, and H. G. Wilshire (1979), Geologic Map of the South Side of the Moon, Scale 1:5,000,000, U.S. Geol. Surv. Misc. Invest. Series, I-1162.


19 thoughts on “Site Selection for Lunar Industrialization, Economic Development, and Settlement

  1. NOTE TO READER: On Sunday October 19th I did an update to this article. I added links to the U.S. Army Horizon report from 1959, provided some minor edits, and added a new sub definition to the “development” to explicitly include economic output as a definition to the term.

  2. Five billion dollars would be extraordinary to get a manned base on the Moon. Helpful though would be some notional idea of how the base would look, landers needed, mass that needed to be transported to the lunar surface, etc.

    Note that Jeff Greason has argued that Moon and Mars missions can be mounted for orders of magnitude lower costs than the NASA estimates.

    This is important because supporters of a “Mars first” approach such as Robert Zubrin take that stance because of the idea a lunar return would cost hundreds of billions of dollars. But this is by following the usual governmental financing approach. By following commercial space and not
    requiring all new “giga”-rockets, it could be done for a fraction of those estimates.

    Then all Mars supporters would come on board to support a return to the Moon.

    NewSpace 2014 – Pathways to Human Exploration: Are there alternatives to
    NASA? – YouTube.

    Bob Clark

    1. Bob, thanks for the comment. First, there is no way that “all” Mars supporters would ever come on board a return to the Moon. They are fixated on their goal, and that is a fine thing, would not seek to change their minds. However, as Mars One is showing, the current thoughts on how to do Mars cheaply show more the limitations than the enabling ability to execute.

      I would agree that putting the rest of it on the table is a good thing. I can do it but I these missives like this take time and research to execute. I have it in my head, just don’t have time to put it on paper/electrons right now. I have some ways to get early revenue as well but I am tired of companies picking up on my ideas and raising money on them… Thus I am in a quandary right now on what to write next….

      Some general principles though to set the stage.


      100% of the mass that goes to the Moon, gets used, reused, abused, and then recycled. This means paying attention at the design stage to the materials used. A small example of this would be to use boron metal struts for secondary structure. Boron is used in fishing rods today and works just fine. The boron when on the Moon becomes a feedstock for the dopants for making local silicon solar cells. A little bit of boron goes a long way and you essentially get it to the Moon for free.

      This principle has a wider application as well. Use carbon/carbon composites for the main load bearing struts in order to get carbon for food production and possibly the carbothermal iron reduction process. The valves on the propulsion system can be reused. The computer, sensors, and even wiring can be reused on the Moon. It does not matter if that reuse is years in the future, design for it now and you can essentially double your throw weight to the lunar surface.


      ISRU is the absolute key to making the $5 billion number work. It has to be there almost from the beginning. At Whipple this is pretty easy to do as the volatile content is high everywhere. Solar thermal is going to be just as important as solar electric in this regard.


      Everyone focuses on the means to get there but that is backwards. Focus on how how to minimize the transportation network by maximizing local resource use, that is how you make it work. We are making amazing advances in robotics these days, put that to use.

  3. A key reason for believing as Greason does that we can do a return to the Moon orders of magnitude more cheaply than the NASA estimates is the case of the lunar lander. A manned lander derived from NASA’s Project Morpheus lander or one derived from Masten’s Xeus lander can be developed for costs in the ten’s of millions of dollars range. This is in contrast to the $10 billion development cost for the Altair lunar lander.

    If you use say the Falcon Heavy, fully paid for by private financing, then that cuts another huge expense from the development costs since you don’t need the multi-billion dollar SLS.

    Taking these facts into account a return to the Moon privately financed could well be done in the single digit billions, or even less.

    Bob Clark

  4. …This is in contrast to the $10 billion development cost for the Altair lunar lander…….

    The architecture that led to the design of the Altair lander was a mistake. The Altair had to do the Lunar Orbit Insertion burn in order to get the weight down on the Orion vehicle that was going to be launched on the Ares 1. As soon as the CST-100 or manned Dragon are flying we can go back to the Moon with an SLS launch with a modest lander. This is a variation of Von Braun’s Earth Orbit Rendezvous and if you launched the Dragon on a Falcon heavy and the modest lander on the SLS with its upper stage, you could rendezvous above 4,000 km and get to the Moon fairly easily and land with a modest lander.

    That would be a NASA version. For a non NASA version a manned mission could be done with two Falcon heavies. One throwing an ultralight lunar lander (Like Gilruith’s lunar Gemini open cockpit lander) into lunar orbit and deliver the humans (two of them) in another Falcon heavy/dragon.

    Instead of staying for days, the crew stays for six months to a year and supervises/fixes the robots and does all of the other stuff that the robots cannot do. However, before you do that you have to land a lot of hardware, starting with the energy system. You really do need a minimum of 500 kW for this to work, but that can be supplied with solar and the power landers that I showed in the article. Energy = work and Energy + robots = lots of work.

    The first real construction project beside landing pads and roads are buildings for human habitation. Which means that you need to bring up a load of Ammonia. You want the nitrogen for the atmosphere and then you mix the hydrogen with oxygen for water, even before you do water harvesting on the Moon.

  5. I have a couple of questions concerning polar sites. First, how does the delta v of landing at or lifting off from near a pole differ from the landings on the near side of the Moon during the Apollo era?

    Second, any idea how much solar power is available at these polar sites for scaling up in the future?

    1. Karl

      That is a systems engineering question. The difference is a couple hundred meters per second if you are going from KSC. If you are going from Baikanour it is 100 meters/second LESS to go to polar orbit. My ideal system begins at ISS and thus has the advantage of the high inclination. A second and far more important factor is that from a polar site you have an anytime return capability, which you don’t really have from an equatorial site.

      Additionally, the cost to setup an equatorial site, in terms of the most important resource, energy, is far higher.

      There is square km of area at both polar sites, thus there is plenty of room for scaling to the megawatt level.

      1. Thank you, Dennis. I was curious about the relative difficulty of supplying and expanding these polar sites. I had a naive impression that polar sites would be relatively difficult to supply due to higher delta v.

        I see also that your paper gives a great deal of thought in addressing communication issues with Earth as well as local coverage on the Moon. That answers another question I was thinking of asking about.

        I’ve long been in favor of lunar development, but this definitely is a strong argument for prioritizing these polar sites over any other location.

  6. I like the idea of low cost missions, but I’ve got a couple of concerns about some of the long-term aspects of this plan. One is an issue I see quite often in discussions of ISRU; the industrial process that would be needed to process bulk ores into useful metals, particularly lightweight spacecraft grade ones generate very large amounts of waste heat, meaning any such processes done in a locale with a hard vacuum will need massive radiators.

    Next as far as I’ve ever learned in the topic of geology, the Lunar surface will not have very many concentrated ores since it had a far shorter time for differentiation, and a lack concentrating factors like water and an atmosphere. Metallic asteroid impact sites may be a rare exception to this but I’m not sure that their common compositions are particularly useful for the construction of spacecraft.

    1. David, I think that the waste heat issue is a big one myself. Fortunately there are the permanently dark areas where the large radiators can be placed and thus are always radiating to the 4k background. That waste heat can be recycled for other uses.

      For the other issue, there are other methods of concentrating metals, and I have written extensively about this. Metal meteorites for example, or the concentrations of Titanium in the Mare regions, which are much higher than similar concentrations in terrestrial ores. If you examine geology today, you will find that most of our commercial high grade ores are exhausted and that we are using ores of not much better grade than what is on the Moon.

  7. Several questions / comments here. Excellent article, BTW.

    Could we land a lander at a highly sunlit location, telerobotically set up broad swaths of solar panels along the landscape, and then either shoot or drape a wire down the edge of a crater down to where the volatiles would be at higher concentration and then the lander boosts off and hops down to the end of the wire is and connects. In this way, we wouldn’t have to make dashes into the shadow and try to come back into the sunlight before the batteries ran out. Rather, power from the rim would be continuously available to the shadowed work-site. The icy regolith from around the lander could be worked and the teleoperated ice harvesters could have ready access to recharging their batteries to keep them appropriately warmed. Also, we would never have to drive up and down the wall of the crater. If we could do this then the proximity of the sunlit ridge to the high-volatile, shadowed work-site might be a factor in the calculations. In this case, smaller craters might be better than larger ones.

    Also, I would make the distinction between different types of sustainability. There’s economic sustainability which means producing more income than expenses. There’s technical sustainability which means that you are utilizing in situ resources to make the program increasingly independent from outside supply. There’s what I’ll call programmatic sustainability which means having sufficient long-term political and hence budgetary support which ensures that the program continues (this would be in the context of a government program with the ISS being an example). Finally there is environmental sustainability which I’m not sure applies here very much. The beauty of lunar resource development, especially harvesting ice and processing it for propellant, is that this approach has the potential to achieve multiple types of sustainability.

    Finally, let me preface the following with the admission that I am neither trained nor experience to make the following statement. None-the-less, I think that either a strictly commercial or a public-private return to the Moon could be done for well less than $5 billion. A single launch of a Falcon Heavy would cost about $125 million (1). Dave Masten has publicly stated (2) that, for about $50 million, he should be able to modify the ULA-donated Centaur upper stages to be able to be a reusable cryogenic lunar lander (i.e. the Xeus lander) — or a non-ULA-related equivalent. Centaurs supposedly cost about $30 million (3). If a non-funded SAA could be obtained to be able to use the Space Power Facility or equivalent, lunar surface ice harvesting equipment could be designed and developed iteratively until it functioned reliably at cryogenic temperatures. I would think that this equipment could be developed as less cost than the COTS cost of developing launchers (4) that had to go through Max-Q and all. So let’s say no more than $200 million. At this point it comes to $405 million. Certainly there would be operational and other costs.

    LCROSS showed that lunar polar ice at at least one location is at one part per 18 (5). It would take 21 tonnes of propellant to refuel the modified Centaur so one would need to process at least 18 times that (i.e. 378 tonnes). You don’t need to transport this amount even back to the lander but could steam it out either in situ (sort of like an oil well) if the volatiles are deep or excavate the “fluffy” (6) icy regolith into the body of the harvester and then steam it out clothes dryer style and then “poop” out the dry regolith and then move forward a couple of meters if the volatiles are superficial.

    If the lander could be refueled and the lander begins transporting water to LLO (e.g. for future interplanetary shielding) or beyond or if the lander could begin providing propulsion service then that initial launch could start to generate revenue for as long as the components of the system continue to work. Later launches could send the necessary spare parts and spare equipment and later and inflatable habitat, greenhouse, supplies in prep for human return. Each later FH launch, lander construction, and surface equipment would cost in the neighborhood of $200 million. So about 13 missions could be done within the first $3 billion which I think is more than necessary before humans could be sent because each lander would be reusable and so would get a lot of flight experience.

    After that, humans are sent to the rim where the solar panels are, where they repair the worn-out equipment hopped up from the floor using the lander, and then they begin extracting metals from the regolith, producing the bulky parts of more harvesters, and hence greatly expanding the teleoperated workforce that’s controlled 24/7 from the Earth. Since they would be producing their own water, oxygen, CO2 & nitrogen from the ice — hence food, metals, ceramics, glass, and solar cells (7), and since stockpiles of high-tech items such computer chips, cameras, and radio equipment can be delivered en mass from Earth, it seems to me that the base could become 90+% mass independent from Earth fairly early on.

    (6) (Schultz et al., 2010) from

  8. Great article – thanks. I see now a lot of opportunities and struggles to implement these. And I wanna add some insight at integration of this concept into whole solar system explotation.
    First of all article mentioned building the loonar base from scratch by 5billion $ investment, why so high? We have already hudreds of tonnes of space waste and debries witch can be used for robotic exploitaition of very small space asteroids, rocks and comets.
    There are a lot of NEO, objects like Icar and others, constantly approaching Earth and then going much closer to Sun, where we can place a lot of solar chemical industry and produce a lot of stored chemicial energy. Thats robotic expoitation – so no humans needed, as well as no high rate communication without delays. Moon is too far away from Sun, Mercury is a better candidate for the task of solar energy production, while asteroids and NEO has lower requirements 4 propulsion by a thousand factor. Much easier to move chemically or even nuclear(radioactive/antimatter) energy itself from one place (free space, near Mercury orbits) than trying ineffectively
    And its easier to move whole industries to space than transport all energy to Earth/Moon/etc.
    Thats why Lunar space elevator is a more effective way to plant robotic seeds on its surface, collect bulk mass of same element and move it to space factory for production, than plant exploration bases only where is sufficient need of human presence, and remove it when its no longer needed.
    So I made an insight on specially build seed spacecraft production plant that turn space debris into small robotic seeds and sensors that fly into any and all nearest objects be it NEO, comets, asteroids in belts or small planets, even Moon itself.(Solar sails and cyclers as a propulsion means). Those seeds selfreplicate to a certain degree, constantly evolving by Earth commands/designs to certain huge systems of robotic colonies and constantly exploring, creating comm networks and prepare human habitation on the way. While humans constantly teleoperate this sorties and move from nearest locations in those constructed simple habitat to farvest, freely changing enviroments and learning new ways of living, constructing space based industries, colonies and high energy installations for solar-antimatter storages.
    So most gravitational wells like Mars, Moon, Mercury and others is useless for humanity, its a waste of effort to transform conditions on most planets for people, while effective radiation protection and pure spacebased colonies and exploration ships are enough for constantly moving and transforming humanity. Extraction of planets and moons mass for these is huge endevour, while some planets will be benefical in some way for transformed humans for being there.
    Right now – what we can do without 5 billion dollars? Scalar space robotics use swarms of very small robots that share energy source, propulsion platform and sources of pure/waste/raw material in the vicinity. Scalar means by joining different kinds of them they can extract materials/energy and reproduce different kinds of needed construction/work/exploration/new robots. Some bots are improved on earth, and reproduced in space, some are just constantly joining to scale production of needed chemical/material or huge infrastructure. We can add to them some new materials by rocket from Earth, or other colonies.
    Then we will puts seeds bot colonies everywhere possible, and create space based sustainable economy, than produce habitats for people to remove connection lags and communication times.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s