Help Shape the Future of Space Exploration

Join The Planetary Society Now  arrow.png

Join our eNewsletter for updates & action alerts

    Please leave this field empty
Facebook Twitter Email RSS AddThis

Headshot of Emily Lakdawalla

The solar system at 1 kilometer per pixel: Can you identify these worlds? The answers

Posted By Emily Lakdawalla

30-09-2015 10:00 CDT

Topics: pretty pictures, scale comparisons, fun

On Friday I posted an image containing 18 samples of terrain, all shown at the same scale. The squares are about 500 kilometers on a side, and when fully enlarged, the images have a resolution of 1 kilometer per pixel. There are 18 solid-surfaced worlds in our solar system that are large enough to include here and for which we have imagery of sufficient resolution: Mercury, Venus (though only in radar), Earth, the Moon, Mars, Ceres*, Io, Europa, Ganymede, Callisto, Tethys, Dione, Rhea, Titan (though only in infrared), Iapetus, Triton, Pluto, and Charon*. I starred Ceres and Charon because we're a bit limited for our images of those at the moment; the images I included in the composite actually only cover about 430 kilometers square.

Were you able to figure out which square was which? Below the image I'll include a labeled version with the answers.

The solar system at 1 kilometer per pixel
The solar system at 1 kilometer per pixel

Ready? Here you go. As a bonus, I've included information on the instruments and missions that captured each image.

The solar system at 1 kilometer per pixel (labeled)
The solar system at 1 kilometer per pixel (labeled)

Some of the worlds' identities can be deduced from features that occur nowhere else: Europa with its bands, Ganymede with its grooved terrain, the smooth lunar maria, the funny dark splotches in Mars' flat-floored craters. You won't find rivers dammed to make reservoirs anywhere but Earth. (This particular piece of Earth is entirely contained within Spain, in case you were wondering.) People who are experienced observers of spacecraft imagery might recognize the fuzziness of Cassini's Titan images, and the peculiar patterns of light and dark in Magellan radar images of Venus. I would really like to know whether I would be able to tell Jupiter's moon Io and planet Venus apart if we were able to shoot them with the same camera.

After that, it gets a bit harder. I might possibly have recognized Ithaca Chasma on Tethys, but I don't think I would have been able to tell Dione and Rhea apart. And I think I would have been thrown by Ceres. Look how flat-floored its craters are! They look as much like Ganymede's or Mars' as they do like Tethys' or Rhea's.

The choice of 1 kilometer per pixel and 500-kilometer squares was not made at random. It was a deliberate choice based on my knowledge of what imagery is available and the sizes of solar system worlds, including the very new additions of photos from New Horizons at Pluto and Charon. The vast majority of cameras that have been sent to other worlds have detectors that are square and have 800 or 1024 pixels on a side. I needed to pick a square smaller than that, so that I could have some flexibility in resizing images to match their scales. There's also a limitation imposed by the solar system, a break in the size distribution below a diameter of about 1000 kilometers. The next-largest object in the solar system after Ceres that we have visited with a spacecraft is Vesta, which is barely more than half the size of Ceres. You can comfortably fit a 500-kilometer square with room to spare on a 1000-kilometer circle, so the choice of scale was basically made for me. In the future I may well do another of these comparisons looking at a much smaller scale, but I'll be limited by the availability of high-resolution images for many of the worlds shown above; it will take some work to identify the resolution that will allow me to include the highest possible number of places.

To locate these images, I went to a variety of sources. For Mercury, I began with the global mosaic from the MESSENGER website, which is available for download at a variety of resolutions. For Venus, I turned to some of the Magellan images that I had already processed for The Planetary Society. For Earth, I went to NASA's Visible Earth image gallery and filtered it by MODIS. For the Moon, I knew I could find the right resolution of imagery from the Lunar Reconnaissance Orbiter's Wide-angle Camera global lunar mosaics. For Mars, 1 kilometer per pixel is a kind of tricky in-between resolution that we haven't seen much of since early days, but I remembered that Mars Express had recently released an unusually wide-angle view of Mars that I could grab a crop from. For Ceres, there's no formally released data yet, so the best source was JPL's Planetary Photojournal for press-released Dawn Ceres images. For the moons of Jupiter, Saturn, and Neptune, I turned to the trusty Outer Planets Unified Search tool hosted by the Rings Node of the Planetary Data System. They have a relatively new feature that dramatically eases the search among Cassini images: if you click on "Cassini Surface Geometry" at the left, and click on "Cassini Target Name" and then select a target, you can specify a minimum and maximum pixel resolution to refine your search. And for Pluto and Charon, I went to my own index of publicly released New Horizons images.

How did you do at guessing which world was which?

See other posts from September 2015


Read more blog entries about: pretty pictures, scale comparisons, fun


Mewo: 09/30/2015 11:03 CDT

I was sure about the Moon, Earth, Ceres, Pluto, Io, Charon, Iapetus, and Europa. All the others I had to guess, but guessed correctly, except for not being able to tell Dione, Tethys, and Rhea apart.

Jonathan Ursin: 09/30/2015 11:59 CDT

I got Europa and Ganyede. I'm giving myself half credit for swapping a few similar looking ones though :)

billeman: 09/30/2015 12:54 CDT

Pluto was easy. IO, got that one. Europa and Venus, still possible. For the rest, I didn't get any right, not even home Earth. Proves that pluto was really worth waiting for, big awe and background image on my PC for a long time.

Roger: 09/30/2015 02:35 CDT

The airless cratered worlds did me in.... and that includes Mars, which isn't exactly airless. Anyway pure guesses for Ceres, Ganymede, Callisto, Mercury, Rhea, Dione, and Mars. Correctly guessed Ceres, but that's it, for 12/18. Several hours of research, although fun, netted me nothing. I would like to develop some basis for distinguishing these and I'd like to hear comments that could help with that. I'm still thinking "no way" on that Swiss cheesy Mars photo. Great skull practice. Live forever, people....

Chris: 09/30/2015 03:23 CDT

I relied on first impressions and got most of them wrong. I identified most of Saturn's moon as such but mixed them up. I didn't even consider our Moon and Earth which I thought was Mars. Having Pluto on the brain helped me on those two frames. Io looked like Titan and Venus like Mars. The little ice (?) patches peeking through helped me with Ceres.

Øyvind G: 09/30/2015 04:47 CDT

Most of them were pretty easy, but I switched Rhea and Dione and got 16/18 correct without checking any maps or pictures. Nice to have the spacecrafts included, too, although it was not part of the "game". Got only about 14/18 correct. Assumed the Earth picture was taken by a Landsat, Mars by a Viking orbiter and Ganymede by Galileo. Guessed Moon (LRO) and Callisto (Voyager) right by pure luck, so it shouldn't count.

Ellen: 10/01/2015 02:39 CDT

The noticeable (but possibly illusory) curvature of some of the small worlds like Ceres and Tethys might have been a giveaway. 500km is a good fraction of those surfaces.

Roger: 10/01/2015 11:57 CDT

Øyvind, or someone, please share how you were able to identify some of these heavily cratered worlds. I recognized Ithaca Chasma on Tethys, but I had to guess at seven others (see my post above), including Mars and Mercury, and got just one of those correct. For any math geeks here, the expected number of right guesses on such a problem is precisely 1, whether you’re distributing 7 guesses, as I did, or 5, or all 18. While semi-obvious, I still think it’s interesting. It’s the same expected return you would get if you simply made the same guess for each one, i.e. I could have put Ganymede down for each of my 7 guesses and have an expected return of 1, although in this case, I would be _guaranteed_ 1 correct. Now, computing the _probability_ of getting 0, 1, 2, etc. correct out of 7 is way more complex, although I did discover that the answer lies in the area of math called derangements. I became somewhat deranged after an hour or so of researching the issue and just had to let it go. If anyone is interested though, I can supply a good link to the (nontrivial) solution of the problem.

Patteroast: 10/01/2015 03:18 CDT

I got the exact same as Øyvind, everything except I swapped Rhea and Dione. For reference, this was my thought process on each: 1) Dark smooth plains with a variety of craters. Moon. 2) Hmm. Medium grey, a variety of craters, bright spots, maybe a bit of curvature? Ceres? Really a guess, but not enough impact saturation to be one of Saturn's moons. 3) Snowy mountains and rivers. Earth for sure. 4) That contrast of blinding white and deep black could only be Iapetus. 5) Grooved bands with complex craters. Ganymede. 6) Bright surface, heavily cratered, and hey that's Ithaca Chasma. Tethys. 7) Hmm. Dark with bright craters mixed with fracture lines. Probably Callisto, but I checked that nothing else could be before I was sure. 8) Sputnik Planum is too fresh in my mind not to notice. Pluto. 9) Flat grey, complex craters with lots of pits, and some hints of scarps. Mercury. 10) Yeah, that fuzziness could only be Titan. 11) That cantaloupe terrain could only be Triton. 12) This is tough. Saturated craters, color almost has a graphite quality to it, and those fractures! The fractures remind me more of Dione than Rhea, so that's why I got that one wrong. 13) Volcanoes everywhere with a smooth background. Hard to mistake Io. 14) The same kind of saturation cratering you only really get on icy moons. Less fracturing led me astray to call this Rhea instead of Dione. 15) Smooth flat surface with crazy linea all over. Europa's gotta be the easiest one of all. 16) Faulted mountains with mixed craters and plains. Still fresh enough that I recognized it right away as Charon, but I think it'll remain pretty distinctive in my mind even later. 17) Huh, those are VERY flat craters with flat plains in-between. What is up with the dark patches? Almost like... wind-blown dust. Oh! Mars! Took me a while. 18) There's a quality about the sharpness of radar imagery that makes it stand out. Huge volcanic flows with faulting EVERYWHERE. Gotta be Venus.

Roger: 10/01/2015 08:29 CDT

Patteroast, thank you for your most thoughtful post. I will review it multiple times; if there's a next time, I will do better. Within 20 years I hope we'll have Titania, Oberon, Ariel, and Umbriel to distinguish as well. It's only 50/50 that I'll be here in 20 years, but that's how that other game is played. Alas, I suspicion -- as in "we suspicion the job was done by a cop" from West Side Story -- that Eris, Sedna, Quaoar, Makemake, and such will remain beyond our reach for many decades, maybe even a century or two, eek! Emily, this was an awesome quiz, thanks.

Øyvind G: 10/01/2015 08:44 CDT

I agree, very good explanations posted by Patteroast. I would add that the bright crater rims and ridges on a dark surface are characteristic for Callisto. You may notice that the inner, streaky walls on the biggest craters on both Dione and Rhea are quite similar to each other, and to what you may find on several other icy bodies. It looks like material has slided down into messy but relatively flat crater floors. As far as I know, you’ll not find (m)any craters with this combination of walls and floors on e.g. the moon or Mercury. The clear curvature on the Ceres picture is not that obvious as one might expect on the only slightly larger Dione, which made that moon hard to distinguish from Rhea. I too thought the fracture through the big Rhea crater represented part of the wispy terrain on Dione. Without Ithaca Chasma (or Odyssey Crater, not visible here), it would be close to impossible to distinguish Tethys from these two, too, despite the brighter surface, which could be a random image processing result. I’ll not repeat everything that’s already been said, except that Earth (lakes and snow), Iapetus, Triton, Io and Europa and were absolutely undoubted. Pluto and Charon, too. In part because these pictures were only a few days old and the terrain easily recognizable. The same reason helped me guess the somewhat older Ceres picture right. Ganymede, Titan, Mars and Venus were pretty obvious, too, from reasons already explained. Titan and Venus much because of the technical characteristic of the pictures themselves (narrowband IR- filter through thick fog and better-than-Cassini SAR image, respectively.) Just for the record: The picture of Earth is not entirely contained within Spain; it contains about one third of Portugal’s land mass, too.

Øyvind G: 10/01/2015 08:53 CDT

And of course, one million thanks to Emily for this and countless other outstanding blogposts :-)

Roger: 10/02/2015 06:15 CDT

Øyvind, thanks for your additions to the comments by Patteroast. In hindsight I should have _read_ about the moons I was uncertain about rather than looking at pictures of them. In that event I would likely have nailed Ganymede and Callisto. And I noted the black spots on Mars, but didn't make the right connection. Obviously, I take this stuff seriously -- time well spent though. Survive....

Leave a Comment:

You must be logged in to submit a comment. Log in now.

Space in Images

Pretty pictures and
awe-inspiring science.

See More

Join The Planetary Society

Let’s explore the cosmos together!

Become a Member

Connect With Us

Facebook, Twitter, YouTube and more…
Continue the conversation with our online community!