I got in a discussion about possible solutions for climate change and this Wiki-article on sun-shading was proposed as a alternate solution with some ideas I'd not heard of.

The cloud of small spacecraft with a cost estimate of 5 trillion seems the most realistic and most accurately described, even if it's a less than ideal solution.

The 1000 km across, few molecules thick Fresnel lens idea strikes me as ludicrous and impossible, with the estimated cost (20 billion) a joke.

Am I wrong on that, or is this something that real scientists are actually looking at as a possible thing to build? What would be a rough estimate for the cost to build and launch something like that and how much time it might take to build. Would it even be possible?


3 Answers 3


It comes down to launch costs. All the proposed technologies involve putting something at L1 to interfere with sunlight over an area of very loosely 1 million square kilometers. The next consideration is the mass per unit area -- the swarm is basically plastic a few microns thick, so say $10^{-2} kg/m^2$ the Fresnel lens is described as millimeters thick, but I think this is a mistake. This source suggests 10 microns as the thickness, which seems more reasonable. The diffraction grating could be much lighter, but probably much more expensive to make, because it's mostly empty space.

So the swarm and the fresnel lens both mass about $10^{10} kg$. This would require 100 thousand launches of a superheavy launcher, costing trillions of dollars at today prices (not to mention the heat and $CO_2$ released by the launches). This suggests that these solutions would require a totally new technology such as an electromagnetic catapult or fabrication from lunar or asteroidal materials.

The mesh is claimed to be closer to $3 \times 10^6 kg$ or about 30 superheavy class launches, which is certainly doable and costs only a couple of billion dollars for launch costs. On the other hand it would be by far the most complex of the systems to design deploy and operate and I don't think technology is advanced enough to let us seriously assess the costs and challenges of doing that.


A sunshade is a feasible direction. And it has two distinct variables - what to make and how to make.

Those variables define the cost and feasibility, and possibly the location and potential complexity of implementation so as the usefulness of the end solution(as an example use that energy in space for something useful)

In 2016 China announced some plan worth 50 trillion to "displacing fossil fuels" to the 2050 - so 5 trillion is 10% of that, which is relevant to any country on the planet, this or another way.

Simple solutions like just foil like shreds may affect the situation not necessarily in a good way, or effects may be different for different countries and that is one of the problems of simple solutions, the lack of flexibility of real-time adjusting of the system. So it more theoretical solutions which show some direction, but not necessarily is an implementation worth pursuing.

But 2% of energy which we need to block and potentially can use in the L1 location. It is a lot of energy(it's huge - with something like 40% efficiency 2% it is about 1380TW). To build an equivalent source of energy here on earth, with price 1B per GW it is 1380 trillion. So if there is some way to build something more useful there in L1 for a fraction of that cost, it means such a solution not only can negate certain aspects of warming problem but also deliver some services to earth economies as a 1380 trillion worth installation.

In regard of combining solving one issue with some useful work, ServerSky looks like an interesting proposal in the context. Considering Generative design such a computing power could be used to address designs so as processes of production and potencially distribution(more variables more complex is the problem and it is essential enough problem to consider such exotic ways to address it(to understand the ways of solving it) like this one Slime design mimics Tokyo's rail system: Efficient methods of a slime mold could inform human engineers)

So as more wide and profound use of deep learning algorithms for different tasks, which are on the rise today, and when teaching them is a somewhat expensive problem and takes week's and more to accomplish on cloud systems, and costs significant sum.

So connecting solving one problem with what may be a solution for some of our demands, it may allow to look differently on the question what is the reasonable price, and scale of efforts worth taking, and complexity of the task in terms of R&D.

So changing a focus from some issue which is a spot on the picture and to think about the situation more broadly, thinking not about patch solution but about creating a useful system, may provide us with an understanding of opportunities in the case.

All that to say, that in some cases it worth to think about the moon as a palace which may help to implement such change(which btw produces other options to solve/fight some consequences of GW - acidity, sea fertilizing etc)

But in the case assessing the price of a solution is a quite complex topic, and obviously classical ways of establishing some infrastructure on the moon on the scale required, may be too expensive. Classical means bringing all the production equipment form the earth. But there are options in a direction like that Affordable, Rapid Bootstrapping of the Space Industry and Solar System Civilization - but one of the problems defending feasibility here also is a problem, but it rather feasible.

And if we may consider investing 5trillions in something(meaning it is a practical sum), it can be more than enough for developing and implementing of quite complex solutions in space. Our today's achievements in automation, in electronics and software they allow us to implement different ways to achieve the sunshade solution - with an ability to provide service and indirectly and directly affect how we do production and designs on earth - that indirect(and direct) positive effects on economy from space programs NASA talks so often.

Using moon may reshape the problem, in different ways, but in terms of delivery, the most obvious changing the problem from Earth to L1, to Moon to L1 which allows thinking about different massdriver solutions without a stretch and the problems earth atmosphere creates for those solutions.

The topic definitely needs more public discourse and more consideration than just theoretical direction, and it can be that game-changing space program we all would like to see. There are problems to solve, but they also bring different solutions which are worth to pursue.

Unfortunately, estimations at this point are quite useless, I could probably try to defend 1trillion investment, but also 20 billion do not look for me like a joke, it just much harder to defend as it needs more details as a plan/road to be able to do so. 100 billion may be a realistic and defendable position, which may give much more than just a sun shade.

I think it is an important problem or a change in a perception of the problem, so you may ping me on discord, if you have some questions or wish to discuss it.

  • 1
    $\begingroup$ This is a really interesting answer, put part of the sun shade to work as useful real-estate. $\endgroup$
    – uhoh
    Mar 8, 2019 at 12:59

Based on how we have developed other technologies, the most feasible approach is to deploy sun-shades incrementally and use the ongoing experience to improve the durability, effectiveness and cost-effectiveness of future deployments, and this needs to be done as a coordinated international collaboration. There is no reason this has to be done in one operation as a permanent solution, but there is value in getting started as soon as possible in implementing what we can now.

The inspiration I have wanted to share is an array of rotating mirrors that reflect infrared light back towards the sun while being transparent to visible light, reflecting at an angle to use the pressure of the reflected infrared to maintain orbital position. If the elimination of infrared from the light reaching the earth is enough to mitigate global warming, it would allow plant growth and photovoltaic energy production to continue unimpaired. I don't know how close we can come to this ideal, but it is the best target I can imagine for a space sunshade.

I agree that having the full, final deployment be manufactured on earth and lifted into space may not be the most economical solution, and that when experiment has narrowed down the best approaches, we might do better to manufacture some of the materials in space or on the moon, or even to fetch substances such as methane from outer planets for the making of plastic, to avoid the exorbitant cost of lifting bulk material into orbit. All this requires a lot of deliberation, and stackexchange is a great model for deliberation.

I noticed another post that suggested using the space sunshade for communication. An even more useful opportunity, if the filtering activity of the sunshade could be modulated, would be to control the amount of power attenuated from reaching the earth in real time, possibly even controlling where on earth it is attenuated. This might allow control of weather or adjustment of seasons, to further mitigate the impact of humankind's environmental disregard.

As a disclaimer, I don't consider the idea of a sunshade to be a complete solution for the problems of population control, resource depletion and CO2 pollution. People still have to learn to care for themselves and each other, but this idea gives us more time to do so.

  • $\begingroup$ While an interesting musings this question is about realistic costs/feasibility of a sunshade, so it would be a stronger answer if the mass of such a material was included (I think something similar is offered for house windows), along with any available information on how it is made - ie could it be made from the moon or asteroids or does it require materials only available on earth (plastics etc). $\endgroup$ Dec 30, 2022 at 22:21

Your Answer

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

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