High-Speed Camera Grabs First 3D Shots of Untouched Snowflakes

sciencehabit writes "Researchers have developed a camera system that shoots untouched flakes 'in the wild' as they fall from the sky. By grabbing a series of images of the tumbling crystals—its exposure time is one-40,000th of a second, compared with about one-200th in normal photography—the camera is revealing the true shape diversity of snowflakes. Besides providing beautiful real-time 3D snowflake photographs from a ski resort in Utah, the goal is to improve weather modeling. More accurate data on how fast snowflakes fall and how their shapes interacts with radar will improve predictions of when and where storms will dump snow and how much."

Comments

Did the Penny Drop!

[Anonymous Coward]

So who else found the penny?

Re:Did the Penny Drop!

[al0ha]

I figure they were flipping a coin to decide who has to stay outside and man the camera.

Re:Did the Penny Drop!

[KitFox]

Previous Story: "Show me the money"

This story: "Here's the snowflakes! And the money."

Just in case this page gets updated and the penny gets bumped off, hopefully the direct link to the beautiful high-speed photo of a falling penny will persist in the records and the annals of time forever.

Re:Did the Penny Drop!

[DanTheStone]

It's certainly an unusual shape for a snowflake untouched by human hands.

Re:Did the Penny Drop!

[ackthpt]

It's certainly an unusual shape for a snowflake untouched by human hands.

And just the other day I was looking a photograph of snowflakes caught on felt, which left them unaffected.

Sometimes you don't need uber technology, but the creativity of a person on a very limited budget.

Re:Did the Penny Drop!

[Fluffeh]

A penny for your thoughts?

Re:Did the Penny Drop!

[Cwix]

Go look at the pictures. Stop and go look at them. I guarantee they look nothing like he ones you saw on the felt. They flatten when they hit something, the ones in those pictures are not flat.

So yeah felt is a nice low tech way to take a look at them, but to really see what is happening the camera gives you so much more detail and dimension.

Re:Did the Penny Drop!

[Sulphur]

A penny for your thoughts?

A snowflake for your thoughts?

Re:Did the Penny Drop!

[Anonymous Coward]

It's a bitcoin, and did i win something?

Re:Did the Penny Drop!

[Synerg1y]

where did you think money came from? The sky of course!

Meanwhile, on the internet

[girlintraining]

...Moments later, the pictures were uploaded to Instagram with a vintage filter, and ceased being cool.

Re:Meanwhile, on the internet

[ackthpt]

...Moments later, the pictures were uploaded to Instagram with a vintage filter, and ceased being cool.

And then Mark Zuckerberg formed a new polical lobby for Snow Flake Preservation.

Re:Meanwhile, on the internet

[girlintraining]

And then Mark Zuckerberg formed a new polical lobby for Snow Flake Preservation.

Zynga then promptly released a game where you can buy virtual snowflakes to drop on your farms, but was sued by Apple for patent infringment over it's slide-to-drop snowflake technology. Microsoft released a snowflake player called 'Snune', but was widely panned by critics as being inferior to all other offers. A few weeks later, it quietly disappeared. Regular slashdot readers blamed stagnancy in cloud technologies for the lack of high performance snow, and girlintraining continued to snark the crap out of everything she comes across...

Re:Meanwhile, on the internet

[Jane Q. Public]

"...Moments later, the pictures were uploaded to Instagram with a vintage filter, and ceased being cool."

Actually, I don't think you need an instagram filter to make these pictures uncool.

I've seen lots of crystalline snowflakes under magnifiers and microscopes before, and they look nothing like these pictures. The ones I saw had nice sharply-defined edges, but these -- including the nice symmetrical ones -- are noticeably fuzzy around the edges.

No doubt, this is an improvement allowing us to get a look at lots more kinds of snowflakes. It's just that they're actually pretty shitty pictures.

Re: Meanwhile, on the internet

[Anonymous Coward]

Meanwhile, on the Internet, pictures of inappropriately touched snowflakes...

Re:Meanwhile, on the internet

[DerekLyons]

The ones I saw had nice sharply-defined edges, but these -- including the nice symmetrical ones -- are noticeably fuzzy around the edges.

The pictures are absolute crap, so there's no real way to tell if the (symmetrical) flakes are actually fuzzy around the edges, or if it's the horrid pixelation, or the focus/depth of field problems, or... well, you get the picture.

(So I wouldn't go on about Instagram, because you haven't a clue what you're looking at or talking about.)

Re:Meanwhile, on the internet

[Jane Q. Public]

"So I wouldn't go on about Instagram, because you haven't a clue what you're looking at or talking about."

What the hell are YOU talking about?

I wasn't "going on" about instagram at all. I was simply discussing the quality of these pictures versus OTHER pictures and micrographs I have seen of actual snowflakes. And I have seen many.

It was GP who was joking about instagram. If you have an issue, go talk to her.

Re:Meanwhile, on the internet

[Synerg1y]

You're a little too good at that, don't you think?

Re:Meanwhile, on the internet

[Synerg1y]

It's pretty crazy how complicated such a seemingly simple thing as photographing snowflakes can be.

Re:Meanwhile, on the internet

[DerekLyons]

It was GP who was joking about instagram. If you have an issue, go talk to her.

Try reading your own messages sometimes - I wouldn't have brought up instagram if you hadn't.
 

I was simply discussing the quality of these pictures versus OTHER pictures and micrographs I have seen of actual snowflakes.

Um, no. Try reading your own messages sometimes. You were discussing the features of the flakes as seen in these images.

Re:Meanwhile, on the internet

[Anonymous Coward]

Um, no. Try reading her post again...
"I've seen lots of crystalline snowflakes under magnifiers and microscopes before, and they look nothing like these pictures. The ones I saw had nice sharply-defined edges, but these -- including the nice symmetrical ones -- are noticeably fuzzy around the edges." She obviously means "the ones I saw..." as the prior images she had seen, and "fuzzy around the edges" as the images in the link. I wouldn't go on about reading comprehension because you haven't a clue what you are talking about

Re:Meanwhile, on the internet

[Jane Q. Public]

I agree. And I wasn't disparaging this technology, which may well turn out to be valuable. But let's face it... the pictures aren't of wonderful quality. No doubt that will improve.

Re:Meanwhile, on the internet

[Jane Q. Public]

"Try reading your own messages sometimes - I wouldn't have brought up instagram if you hadn't."

Jesus Christ. Do you have reading comprehension issues? Here is a quote of my first sentence. Emphasis added:

"Actually, I don't think you need an instagram filter to make these pictures uncool. "

This was in response to someone else's joke about Instagram. Are you reading the same words I am? Where in that sentence of mine is there anything negative about Instagram?

"Um, no. Try reading your own messages sometimes. You were discussing the features of the flakes as seen in these images."

Um, yes. Repeat, again:

"I've seen lots of crystalline snowflakes under magnifiers and microscopes before, and they look nothing like these pictures. The ones I saw had nice sharply-defined edges, but these ... -- including the nice symmetrical ones -- are noticeably fuzzy around the edges... It's just that they're actually pretty shitty pictures."

Repeat again: it is a discussion of these pictures VERSUS OTHER PICTURES of snowflakes I have seen. It is right there in plain English. Same as I wrote it the first time. In context, no changed words.

Re:Meanwhile, on the internet

[Synerg1y]

I was slightly amazed that the camera they're using... something way way above consumer grade wasn't able to get the job done to a level of high quality, I agree, the pictures are kind of grainy and its hard to tell the actual shape on some of them.

Nice pictures

[Anonymous Coward]

Now all they need is a camera to photograph leprechauns, fairies, unicorns, Nessie, the yeti, Bigfoot, aliens, and UFOs all in 3D.

Re:Nice pictures

[Dahamma]

Analogy fail. You must live somewhere warm... hate to break it to you, but snowflakes are real ;)

why not post the text that comes with the pics!

[Anonymous Coward]

rather than the bs from sciencemag?

here it is:

"MASC Showcase: Snowflakes in Freefall

For more information about this University of Utah and National Science Foundation project please visit the Snowflake Stereography and Fallspeed home page or email Tim Garrett.

This is a gallery of snowflake images captured in freefall at Alta Ski Area using the University of Utah MASC (Multi Angle Snowflake Camera). When it is snowing, images of snowflakes captured live in free fall can be found at Alta's Snowflake Showcase.

Images are taken at f/5.6 with an exposure of up to 1/40,000th of a second using 1.2MP and 5MP industrial cameras with lenses ranging from 12 mm to 35 mm. The image resolution ranges from 9 micrometers to 40 micrometers.

Click on any image to see it in full resolution and to play a slide show.

Donations to continue the Snowflake Showcase at Alta Ski Area are welcomed.

Alta is terrible with rotten snow; don't go there

[TofuDog]

They're very unfriendly and the "Wasatch cement" is horrid and it's crowded. No one should go there.

Don Komarechka

[Flammon]

I have a cousin who's writing a book on this.

http://www.indiegogo.com/projects/sky-crystals-unraveling-the-mysteries-of-snowflakes

every time I see the "research" like this ....

[Anonymous Coward]

Only thing that comes to my mind: https://www.youtube.com/watch?v=N-NkX86uPI0

six-side snowflakes

[Anonymous Coward]

From the article:

The classic image of a snowflake is a fluke. That flat, six-sided crystal with delicate filigree patterns of sharp branches occurs in only about one in every 1000 flakes.

Wrong. The six-sided snowflake is a particular type of snow that occurs in specific conditions. When conditions are correct, you'll see endless amounts of six-sided snowflakes. Source: I live in the Canadian Prairies.

Re:six-side snowflakes

[sarysa]

A good portion of the snowflakes on the site were six sided or clearly used to be six sided. Expecting the level of precision (that they are implying) in nature is ridiculous, though....

Re:six-side snowflakes

[Anonymous Coward]

Most of what they are capturing are clusters of flakes which have smacked into each other during free-fall. It's kind of neat photography, but it doesn't really give us any way to legitimately claim that "the classic image is a fluke"... because we have no way to tell if any of those flakes bumped into any other flakes after/during the process of formation.

why are snowflakes symmetric?

[The_Rook]

what i've always wanted to know was why are snowflakes symmetric?

sure, a hexagonal crystal has bilateral symmetry etc. but snowflakes form long, complex arms with what can best be described as filigree. and yet, the filigree on opposite sides of the snowflake are also symmetric - that's shown even in these photos.

so what i want to know is how does crystal faces on opposite sides of the snowflake 'know' to grow symmetric filligree? what mechanism is there that allows one crystal face of a snowflake to 'know' what the other crystal faces are doing and so grow identical structures?

Re:why are snowflakes symmetric?

[exploder]

I think it's because (snowflakes being quite small) all sides experience nearly identical conditions of temperature, humidity, whatever-else-affects-crystal-growth at nearly identical times.

Re:why are snowflakes symmetric?

[Anonymous Coward]

I've always wondered the same. The best idea I can come up with is that as the arms form, the micro-environment they each experience is almost identical, yielding essentially the same formation. Looking at the photos of these natural snowflakes, none look nearly as symmetrical as the photos we typically see. I'm guessing that the typical perfectly-symmetrical snowflakes are made under very homogeneous lab conditions (or better yet, Photoshop).

Re:why are snowflakes symmetric?

[c0lo]

what i've always wanted to know was why are snowflakes symmetric?
how does crystal faces on opposite sides of the snowflake 'know' to grow symmetric filligree?

Possible explanation not involving "knowing": if they are asymmetric, they break due to difference in air drag forces over the asymmetrical branches (?!?? something like "snow flake evolutionary pressure"?)

Re:why are snowflakes symmetric?

[Anonymous Coward]

Are you saying that snowflakes have Intelligent Design?

Re:why are snowflakes symmetric?

[Anonymous Coward]

I think it's because (snowflakes being quite small) all sides experience nearly identical conditions of temperature, humidity, whatever-else-affects-crystal-growth at nearly identical times.

I'm a cloud physicist and you, sir, are correct, identifying not only the reason for the symmetry (uniform conditions across the crystal over relevant time scales) but also the two controlling factors of temperature and humidity (well, humidity above saturation). Well done. Let me buy you a virtual beer.

The diversity of shapes is what's really cool. As far as I know (and this isn't my area) we have a phenomonology for habit (we know what shapes are most likely at a given temperature and humidity, for example) but we don't have a good theory to explain why that's so.

Re:why are snowflakes symmetric?

[pspahn]

Just to add to this, there is a cave in Colorado that I found last summer (it is known of, but it sees very little visitors... maybe a dozen a year) that apparently develops very rare ice crystal shapes in the winter near the entrance. What I read is that it is one of only a handful of places in the world where this is known to occur.

Re:why are snowflakes symmetric?

[dbIII]

Less energy to do so.

Re:why are snowflakes symmetric?

[Jon Abbott]

Do you mind sharing where? I'm doing a photography book on Colorado and would like to include such a place. My email is i_love_junk_email at yahoo.com.

Re:why are snowflakes symmetric?

[Jon Abbott]

No need to worry, I get plenty of spam already. Probably not as much as most Hotmail accounts get though.

Re:why are snowflakes symmetric?

[pspahn]

Can't remember the name specifically, it is almost exactly at 39.518712, -106.629701

There are some yurts just to the west about a 1/4 mile. I suggest staying there (Hidden Treasure Yurts) and reading the great binder full of information on local caves. I read the article about this specific cave, and it's quite fascinating.

If caving in Colorado, please educate yourself on White Nose Syndrome. It is not prevalent here, and people would like to keep it that way.

Re:why are snowflakes symmetric?

[pspahn]

Ah, Devil's Den cave. Note, you'll probably not find much info online about it... there is a popular cave in Arkansas with the same name.

Re:why are snowflakes symmetric?

[funkify]

TFA mentions that a "classic" snowflake only occurs about once every 1000 flakes. Here in the Rockies, I have observed that many snowstorms produce no noticeable "classic" snowflakes, while in other snowstorms, nearly every flake that falls is a "classic" snowflake.

Can you, Mr. Cloud Physicist, enlighten us on this?

Re:why are snowflakes symmetric?

[Jon Abbott]

Great, thanks. It's definitely a drive from where I'm at but hopefully I'll make it out there sometime.

Re:why are snowflakes symmetric?

[pspahn]

If you go in winter, access by snowshoe, snowmobile, or skis... there's no other options.

did someone noticed a coin amongst the shots?

[Anonymous Coward]

did someone noticed a coin amongst the shots?

Bad summary

[TrumpetPower!]

First, 1/200s is a very common shutter speed, yes, but most cameras can shoot at at least 1/2000s and most high-end cameras can shoot at 1/8000s...assuming, of course, you have enough light.

Most high-speed stills photography is actually done with a slow shutter speed; perhaps even a shutter left open for a couple seconds. Motion is stopped by the short duration of the flash burst. And with, for example, a Canon 580 EX II flash, you can get a 1/35,000s flash duration. Granted, this will be at minimum power...but they're operating at macro distances, where you can put the flash head almost on top of your subject and still overpower the subject with light.

Don't get me worng; this team is doing some nifty stuff. But it's also something that most professional photographers could easily replicate with the equipment they already have -- and that anybody who specializes in macro photography will probably already plan on playing around with next winter after reading this article.

What the team is doing that's interesting isn't the photography. It's the 3D reconstruction and subsequent analysis and modeling. Making it seem that it's about the photography, which is the easy and inconsequential part, really detracts from the good stuff.

Cheers,

b&

Re:Bad summary

[Anonymous Coward]

The original researchers do point out that they are using off the shelf cameras for th individual views within the 3D setup, and that they're nothing particularly special (as opposed to say an off the shelf ICCD camera with exposure times in the nanoseconds). They discuss why they didn't go with high speed flashes and instead used cameras that could directly use such exposure times: low power so it can run off a small batter at,repetition rates of up to ~1 Hz for long periods of time and resilience in a cold, outdoor environment. The paper only spends about a paragraph on this, and the rest on triggering and setup of the whole system, and results.

Re:Bad summary

[Jon Abbott]

Agreed. Humorously enough, as I read your post, I saw the mention of Canon, then looked at your username and remembered seeing the same username over at CanonRumors.com. I also frequent that site... small world.

Re:Bad summary

[muridae]

That's something I can't figure out, though. You have a falling object, and a moving shutter that I hope is moving vertically downward. A sideways moving shutter should create some amount of distortion at 1/40,000 of a second. Even a vertical shutter should have some lag, unless I'm vastly mistaking the moving rate of a snowflake.

As a note for the non-photographers, high speed shutters do not open and close completely at 1/40000th. The normal method of off the shelf equipment is that the shutter opens, and while it is moving, the second closing shutter closes 1/40000th later. It still takes up to 1/400 or so (depending on the camera, look at the sync speed to get the right neighborhood of numbers) for the shutter to fully open and cover the shutter.

Re:Bad summary

[rgmoore]

The trick is that the shutter isn't doing the work; the flash is. It's possible to make very short flash pulses; I think you can make them even shorter than the 1/50,000 second mentioned in the article. As long as most of the light for the photograph comes from the brief but intense flash, the ability to freeze action depends on the flash speed rather than the shutter speed. You actually need to make sure the shutter speed is slow enough that the shutter is guaranteed to be all the way open when the flash triggers (X-sync speed or slower), or only the area behind the open part of the shutter will be exposed. Controlling things using the flash also guarantees that the multiple cameras used for 3D photography will all be taking their pictures at exactly the same instant.

Also note that the limitation you're talking about only applies to focal plane shutters (i.e. those right in front of the film or sensor). It's also possible to use a central shutter that's located right next to the iris of the lens. Central shutters open and close like the lens aperture, but block the lens completely when they're closed. Like the lens aperture, they block light to all parts of the focal plane more or less equally as they open and close, so they don't induce any of the motion effects that focal plane shutters do. Central shutters have their own problems- it's hard to make them work for very short shutter speeds, and they have limited efficiency when you use them that way because they're only completely open for part of the time- but they do eliminate focal plane shutter artifacts and allow you to flash sync at any available shutter speed.

Re:Bad summary

[Anonymous Coward]

maybe, but:
If you try this and get the flash too close to the snowflake subject, (especially with a conventional flash (i.e. the 580 EX II), the flash may melt (or partially melt) the subject...

One would probably be better off with something that has a cooler physical temperature - maybe an LED lighting rig.

Re:Bad summary

[Anonymous Coward]

Except in this case they explicitly say they are not using very short duration flashes, and are not using high speed photography flashes but instead just LED lamps. Cameras like the ones they use can easily have electronic shutters that operate at those speeds, so the discussion of the location is kind of moot, as it is right on the sensor (or more effectively, is the sensor just moving the accumulated charges from a light sensitive area to an area that blocks light before being read more slowly). They would take faster images, but would need brighter lights to do so. Even so, at 1/40,000th of a second, a typical fall speed of snow flakes at 1 m/s means it would only move 40 microns, and the resolutions of the cameras were either 27 microns per pixel or 9 microns per pixel. That was deemed acceptable, especially since there will be slower snow flakes too that could get better images. There was some independent measurement of fall speed that could be used to clean up the image a little too, although it looks like they just threw out any out of focus or bad images instead.

Re:Bad summary

[Anonymous Coward]

Or they could just use an electronic shutter as is typical on industrial machine vision cameras like the ones they describe using. Those simply have double the number of pixels they need essentially (or a couple extra transistors per pixel for CMOS), with half of them are not sensitive to light. After a short exposure, then every light sensative pixel at the same time is transfered to a non-sensitive pixel, where it can sit for some time to wait to be read serially without distortion or further exposure. Such electronic shutters can go much faster than mechanical ones, with exposures down to at least a microsecond (maybe even faster ones are possible... but I'm used to needing ICCD cameras that have an intensifier in front of the CCD at that point, which can be gated, basically turning on and off the amplification on the nanosecond time scales).

Still facsinating

[Ogive17]

I'm 33 and on occasion I'll see one of those perfectly shaped snowflakes land on my coat, after all these years I still think they are cool.

I just wish its trillions of friends would get off my lawn (and driveway).

Found in the gallery

[wbr1]

I wish we had these snowflakes here: http://www.inscc.utah.edu/~tgarrett/Snowflakes/Gallery/2012.11.08_17.53.16_flake_1_cam_0.png
I'd be rich!

Quantum..something..something...

[Anonymous Coward]

The answer to all our questions is in The Cloud!

FRACTALS !!!

[Taco Cowboy]

Wow !

The 3D structures of the snowflakes are fractals !!

Hopefully someone can come up with a fractal software that can let users experimenting with various 3D snowflake shapes

And with the availability of 3D printing, who knows what kind of new art forms this will bring forth ??

Re:FRACTALS !!!

[Internal Modem]

The photos are only 2D -- in some cases as low resolution as 1.2 megapixels. It is only the actual snowflakes themselves that are 3D as opposed to the standard 2D-like images of snowflakes we normally see. To print actual snowflakes in 3D will require different equipment and another method of photography.

As far as #D fractal software, there doesn't seem to be a shortage.

Care to share the link ?

[Anonymous Coward]

Thank you for the informative texts

Would you kindly share with us the link now ?

Thanks again !!

Re:Care to share the link ?

[Anonymous Coward]

2nd link in the summary.

Who would have thought?

[AndyKron]

Who would have thought you could photograph a snowflake?

If the snowflakes are "intelligently designed" ...

[Taco Cowboy]

Are you saying that snowflakes have Intelligent Design?

If the snowflakes are "intelligently designed" they may look like flying saucers

Re:Alta is terrible with rotten snow; don't go the

[TofuDog]

Exactly. Thank you. I won't be seeing you on the slopes.

Snowflake smithing

[dbIII]

I've tried, but it keeps coming up looking like a half grown girl with a pointy hat for some reason. I'm Aching for an answer.
Maybe some sort of Landover Wave Function will work instead.

Good diagrams and photo-tables in that article!

[girlinatrainingbra]

Thanks for the doi link! I'm amazed I can download the PDF without being at a university connection. The "habit diagram" in figure 2 of that paper is really cool. And figure 5 with Temperature vs. "Ice supersaturation" (? not fully understanding that) with the illustrations of the crystals that form at different variations of X vs. Y in the atmosphere or in their "diffusion" chamber. Very nice article.

Re:Good diagrams and photo-tables in that article!

[Anonymous Coward]

GP here. That article was in the Journal of the Atmospheric Sciences, one of about six top-tier journals published by the American Meteorological Association. A few years ago the AMS decided to require subscriptions to only the most recent two years (I think) of journals and to open up all older articles to everyone. It's a great idea. Most top-tier journals in the atmospheric sciences are put out by academic societies, and a couple are open access, but the AMS is the only publisher I know with this very enlightened approach.

And yes, those shapes are very cool.

Image quality

[Kenoli]

All of these images are low res and out of focus.

Photographing falling snowflakes is probably pretty tricky, but some of them are just awful.
eg.
http://www.inscc.utah.edu/~tgarrett/Snowflakes/Gallery/2013.01.27_06.25.55.5_flake_17839_cam_2.png
http://www.inscc.utah.edu/~tgarrett/Snowflakes/Gallery/2012.02.18_22.43.52_flake_1406_cam_2.png
http://www.inscc.utah.edu/~tgarrett/Snowflakes/Gallery/2012.11.09_06.14.16_flake_107_cam_0.png
http://www.inscc.utah.edu/~tgarrett/Snowflakes/Gallery/2013.02.22_16.06.08.425688_flake_16605_cam_1.png

Re:Image quality

[Unipuma]

I thought this one to be interesting:
http://www.inscc.utah.edu/~tgarrett/Snowflakes/Gallery/2012.11.08_17.53.16_flake_1_cam_0.png

(Freudian photography by a researcher hoping for a new grant to rain down on them?)

Re:Image quality

[Anonymous Coward]

Sure - but the point was just to get enough, good enough, images to do 3D reconstruction in software. They can afford a very low rate of perfect pictures; it's not like getting a lot of exposures was especially hard or expensive.

Wow!

[Anonymous Coward]

The shape diversity of snowflakes truly is amazing!
http://www.inscc.utah.edu/~tgarrett/Snowflakes/Gallery/2012.11.08_17.53.16_flake_1_cam_0.png

Shortly after posting the results...

[TheSkepticalOptimist]

Someone cried out "Oh I have wasted my life...!"

SCIENCE B****!

[ProfessorKaos64]

Enough said.