Map and measure a million Moon craters!

By Phil Plait | April 30, 2012 9:16 am

I give talks about asteroid impacts quite often, and sometimes people ask me why we should worry about them. I reply, "Go outside and look at the Moon. Then tell me we don’t need to worry about asteroid impacts!" The Moon is covered in craters, and it really brings home — literally — the fact that we need to understand impacts better.

I’m not being facetious, either. Looking at the Moon is a great way to learn about craters. By measuring their size, position, and shape, we can find out a lot about the history of impacts in the Earth-Moon system. The problem is there are so many craters — billions, if you look at high enough resolution. How on Earth — haha — can any scientist or team of scientists possibly look at them all?

Well, it depends on how big the team is. Enter citizen science: non-professional-science people who nevertheless love science. If you’re reading my blog — and you are — then that means you! is a group of astronomers, run by my friend Dr. Pamela Gay, who have created a series of projects where people like you can perform needed tasks that are real science… in this case, measuring craters on the Moon! Using MoonMappers, you can identify and measure craters using images from the Lunar Reconnaissance Orbiter, a spacecraft currently circling our Moon and taking thousands of high-resolution pictures.

I signed up and started right in, and find it somewhat addicting. You’ll need to register first through the CosmoQuest forum, which takes one minute and is free. Once you’ve done that, just go back to Moonmappers and dive in. I was able to identify dozens of craters in just a few minutes. Here’s a typical scene:

The blue circles are craters found using automated software. The green ones mark craters I found. The task is really simple: you can mark craters with your mouse, dragging the circle to match its size. If you miss a bit, you can easily adjust the circle’s position to re-center it. You only need to find craters bigger than 18 pixels in size, so it’s not an impossible chore! You can also flag odd features like linear cliffs, boulders, and so on, if you happen to see any. Several of the images I went through had them. One had lovely striations in an old lava flow, so you never know what you’ll see.

Sound like fun? It is! But hurry: right now, CosmoQuest has issued a Million Crater Challenge, to get 1,000,000 craters identified by full Moon, which is on May 5, just days away. As I write this they’re still a long way from their goal. How many can you find?

And remember: this isn’t just fooling around, this is real science. How are craters made? Why are they different shapes? How many are 10 meters across versus 20 versus 30 versus 100? All these questions are important in understanding impacts… especially that last one. Getting the scales of impacts, and how the numbers of them increase as the size gets smaller, is critical in being able to predict how often they happen. At some point, we’ll see a small asteroid headed toward Earth, and we’ll have to decide if it’s big enough to worry about and spend hundreds of millions of dollars deflecting it. The work you do here, quite seriously, can help inform that decision.

Related Posts:

Find cold, distant worlds with Ice Hunters
YOU can find extrasolar planets
Two exoplanets discovered by “citizen scientists”
GLOBE at Night wants you to look up!

CATEGORIZED UNDER: Astronomy, Cool stuff, Pretty pictures

Comments (16)

  1. JF

    You shouldn’t assume that only non-professional-science-types read your blog. I’m a particle physicist and still enjoy reading it!

  2. arabwhipmonk

    This is really cool. Seems like they’re doing the same for identifying Kuiper Belt objects. For the craters, I’m guessing they’ll match the same image from multiple users and find the most probable craters and their sizes (among other things). I wish I had that data to look at. Also, does the LRO collect height data? I recall reading something about it. All that combined could tell us about the frequency and common sizes of such collisions in this part of the galaxy, perhaps?

  3. arabwhipmonk

    This is really cool. I’m not sure about surface features, so I’m just focusing on craters for now. For the craters, I’m guessing they’ll use multiple user inputs to find the most probable craters and their most probable sizes. Is the data from LRO’s altimeter available for the global surface (maybe in the future)? All that together could perhaps tell us about frequency and sizes of collisions in this part of the galaxy? Is that data, once extracted, going to be open source? I’m guessing not. Seems like they’ve a similar project to identify possible Kuiper Belt objects for ICE.

  4. dcortesi

    Hey, how is this supposed to differ from — or integrate with — MoonZoo which has been doing exactly the same thing for like a year now? They have classified over three million images already.

    It’s not that the world can’t use two citizen-science moon crater surveys, but how come there’s no cooperation, or even recognition of the older project?

  5. josie

    At my work we’re running into a ‘counting problem’ as well. In my case it’s cells/nuclei/cytoplasm. I brought up the galaxy zoo citizen science projects at a meeting and how counting galaxies and Kuyper Belt objects is essentially the same thing as counting cells. Unfortunately, in-house counting with expensive software remains our preference apparently 😛

    I still love the idea of these projects though, I’ll sign up tonight!

  6. dcsohl

    I’ll just echo dcortesi here and say that, while there’s plenty of room for citizen science — indeed, we need *lots* more of it — it’s really disappointing to see a brand-new outfit trying to occupy the exact same space as a well-organized and long-running project like MoonZoo. Especially with the new outfit being run by the likes of Dr Pamela Gay and promoted by you, Phil.

    For example, the CosmoQuest “Ice Investigators” project is far more innovative and an important project, and one I’d be far more supportive of than Moon Mappers, if only because Ice Investigators is not exactly duplicating a Zooniverse project.

  7. Margrit McIntosh

    Wait – this sounds exactly like MoonZoo – ? Also using LRO images? Makes no sense.

  8. arabwhipmonk:

    LRO Altimetry data is already available. It’s just extremely sparse compared to these LRO-NAC images shown in Moon Mapper. A single LRO-NAC observation normally has about 5 LOLA tracks through it. However the above screenshot is just a crop from LRO-NAC, and probably no LOLA measurements exist for that crop. Thus, crater counting still seems to be the realm of 2D image processing.

  9. Regarding MoonMappers versus Moon Zoo, these are completely separate projects with different science teams and goals. Yes, at a very basic level, they do the same thing – identify craters on the Moon – but that’s where the similarities stop. Besides craters, Moon Zoo is looking at constraining the thickness of the lunar regolith (the “soil”), map the distribution of boulders across the surface, and identify and catalog unusual geologic features.

    Our starting projects for Moon Mappers are very different. Among other things, we’re looking initially at how we identify craters so that we can plan better in the future. For example, the Man vs. Machine interface (what Phil took a screenshot of) is not just identifying craters for our science questions that require the building of a crater catalog, but we’re studying how to fine-tune the crater detection algorithm we’re using and whether it’s faster to use this kind of interface or do everything on your own. Science-wise, one thing we’re trying to answer early on is what sun angles are best for identifying craters, and if there are systematic differences we can model if we don’t have that ideal angle. Beyond that, another early project is looking at refining the crater densities at the Apollo landing sites to help constrain lunar chronology, and we’re also exploring the enigmatic cryptomare region within the South Pole-Aitken Basin.

    We also presented our first results from MoonMappers last month at the Lunar and Planetary Science Conference, and you can check out the results and the poster we presented here: [link didn’t go …] (or, if the link doesn’t go through, navigate over to, go to the blog, and look for the post on March 20 entitled, “First MoonMappers Science Poster Being Presented Tonight”).

  10. mikel

    Anybody else get an Apollo landing site?

  11. Zenzan

    Puts me in mind of aBeetle’s song Phil:

    I read the news today, oh boy
    Four thousand holes in Blackburn Lancashire
    And though the holes were rather small
    They had to count them all
    Now I know how many holes it takes to fill the Albert Hall

    :) Nice post

  12. Phil didn’t even give us his username to count his referrals. It doesn’t appear to be “BadAstronomer”.

  13. Moffett Dude

    MoonMappers and MoonZoo will soon have some competition. Bob Richards told his Moon Express staff the other day that the company is shifting its focus to software and that their new product will be a service where people can name craters on the moon.

  14. Mike Moffett

    Looks like everyone’s opinions are shifting. Moon Express was making pro-mining noises last week. But then earlier this week they started to lay off some of their engineers and told the remaining employees that this hardware company is now a software company and that their new product is something that names moon craters.

  15. I’ve checked into a few of these citizen science sites like MoonZoo and GalaxyZoo. I think they would get a lot more participation if these each featured: 1) an easily printable black-on-white logbook so students could provide evidence of participation to teachers 2) suggestions about projects or lab-like activities students could do either as classroom assignments or as projects in a science fair. Observe 300 craters and plot this vs. that and look for… etc. From an educator’s perspective this holds limited possibilities because I have to use it to advance my specific curricular goals…which I can accomplish in a 5 minute demo. The repetition and attention these projects require for the scientists is not meeting the needs of teachers and students unless there’s a more immediate, tangible product to generate. But maybe it’s just my lack of creativity.

  16. Moon Express response to ’13. Moffett Dude’ & ’14. Mike Moffett’ comments:

    To set the record straight with some facts, there haven’t been any lay-offs at Moon Express, ever, and we have never told our staff we are becoming a software company. We continue to hire engineers and grow our staff and operations toward our short term goal of lunar robotic exploration and long term goals of lunar mining and resource development.

    The ME Team.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!


See More

Collapse bottom bar