Water really is blue! Who knew?

December 22, 2004 7:26 AM

Teaching bad science is not something only creationist wingnuts do. The redoubtable Bill Beaty sets us straight. (thanks, Laen)
posted by flabdablet (35 comments total)

This reminds me of the wonderful, if sometimes cranky, "Bad Science" site.
posted by TonyRobots at 7:32 AM on December 22, 2004

flabdablet this was a good post without the "creationist wingnut" crack.

many of the misconceptions about physics are simply that: misconceptions or misunderstandings. and that's what makes it bad science. creationism isn't bad science, it isn't science at all.
posted by three blind mice at 7:49 AM on December 22, 2004

Great post, flabdablet, I was chagrined to realize that I believed a couple of these. And the creationist crack was entirely appropriate in this context.
posted by LarryC at 8:02 AM on December 22, 2004

btw: creationism isn't "bad science." It's not science at all.
posted by bshort at 8:02 AM on December 22, 2004

Sorry, I'm gonna have to fight Beaty on this one:

Some authors state that bodies of water are blue because the water reflects the sky. But wouldn't this only make the shiny surface-reflections look blue? And doesn't water remain blue on cloudy days? Exactly. There's no mystery here; water looks blue because water *is* blue. It's not the sky that creates the color.

If you're several feet below the surface of a lake, looking up through it on a cloudy day, I imagine it still looks blue. But if you're above the surface of the lake, looking at it, it indeed looks gray on an overcast day, and looks black at night. So I'm sticking with the "reflects the sky" explanation, as long as you're above the surface, and not below it looking through the water.
posted by DevilsAdvocate at 8:06 AM on December 22, 2004

And let me draw everyone's attention to my boyfriend (shhh...don't tell Mr. S) Ben Goldacre's fantastic "Bad Science" column in the Grauniad.
posted by Sidhedevil at 8:08 AM on December 22, 2004

water looks blue because water *is* blue. It's not the sky that creates the color.

Then wouldn't a boat that's just come out of the water be covered with lots of blue drops of water instead of clear ones?
posted by Cyrano at 8:13 AM on December 22, 2004

And, of course, water doesn't always look blue. In the ocean, it sometimes looks green; sometimes it looks gray; and sometimes it looks almost white.
posted by Sidhedevil at 8:20 AM on December 22, 2004

Cyrano, directly from the site:

In fact, this is exactly how it works: water is nearly clear, but it's very very slightly blue. A small amount of water is too thin, so a small amount looks clear rather than blue. But look through thirty feet of water, and you'll see a strong color. Gaze into a hundred feet of deep pure mountain lake water, and you'll see exactly what color the water actually has. Yet if you scoop a canteen full of that lake water, it will look totally clear.
posted by Khalad at 8:25 AM on December 22, 2004

Scientology is very bad science.
(I hope they don't sue me for saying that.)
posted by nofundy at 8:30 AM on December 22, 2004

Great site. These kinds of things always get me interested in science for a little bit and then I remember: fiction is much wierder.
posted by OmieWise at 8:33 AM on December 22, 2004

weirder.
posted by OmieWise at 8:34 AM on December 22, 2004

Yah! Slashdotting! Don't worry, eskimo.com shrugs off high transients.

Thanks for the high traffic everyone!

And Merry X-MASS.
posted by billb at 8:49 AM on December 22, 2004

"So I'm sticking with the "reflects the sky" explanation, as long as you're above the surface, and not below it looking through the water."

I guess I need to find references to link to all those articles. The "blue water" one came from a long discussion among physicists on PHYS-L. They posted absorbtion spectra of ultra-pure water, sea water, etc. Yes, water really is blue. This is best observed from an aircraft flying over tropical beaches with extremely white sand. The ocean looks just like various thicknesses of deep-blue glass: light blue along its thin edges, and fading smoothly to blue-black in the depths.


Beware of this psychology effect:
"When a man finds a conclusion agreeable, he accepts it without argument,
but when he finds it disagreeable, he will bring against it all the
forces of logic and reason." -Thucydides
posted by billb at 8:56 AM on December 22, 2004

Hi, Bill! Great to have you here discussing your site.

Maybe I wasn't clear--I am not questioning that water is blue, or your explanation of why water is blue, when you look through it, as a diver looking up through the lake, or someone in an airplane looking through the water at the sand below the surface.

What I am saying is that I don't believe this explanation applies to most people's most common experience with a body of water--standing on or near the shore (eye level just a few feet above the surface of the lake) and looking out across the expanse of the lake (not the tiny part of the lake right by them, where they can look through the water and see the lake bottom.)

Under these conditions, the lake appears blue on a clear day, gray on an overcast day, and black at night. Therefore I'm concluding that in viewing a body of water this way on a clear day, it's blue because it reflects the sky.

I reject (under these conditions) your explanation, because it conflicts with my personal experience that a lake appears gray on an overcast day.
posted by DevilsAdvocate at 9:18 AM on December 22, 2004

"I reject (under these conditions) your explanation, "

It's not my explanation. As I said, it's from the physics community, and their assertion is that the misconception of "water is blue because the sky is blue" is common and widespread, and could be usefully attacked by physics instructors in the classroom in order to stimulate thinking. It's also common knowledge among physics teachers that long-held misconceptions are frequently impossible to shift.

"because it conflicts with my personal experience that a lake appears gray on an overcast day."

Ok, but the entry is not about the color of the reflections, it's about the color of the water itself. Do brown muddy rivers appear blue on a bright clear day? Yes, their sky-reflections do, but only when viewed at a glancing angle, and it doesn't lead us to think that muddy water is blue water.

My personal experience is unusual, so the "blue sky" misconception was easy for me to question. I saw a swimming-pool reactor at Cornell on a tour. It was 30ft deep if I recall. The notable thing was that, when they turned on the lights, the water was brilliant blue in the depths, but utterly colorless on the surface. (And when they turned off the lights, there was no Cerenkov glow, since the reactor was shut down.) It was not D2O or anything, it was just well-filtered water.

Also, I grew up on Guam. My personal experience is not that of mainlanders. Guam has white beaches, eye-searingly blue water, and if you've ever seen a postcard of Guam (or Hawaii,) you'll notice that it's not the blue surface reflections you're viewing.

And yes, during overcast days the brilliantly blue pacific water was far less blue. But so were all other colors in the environment.

This "grey water" effect wasn't discussed on PHYS-L, but I suspect that it's an unusual effect worthy of discussion all on its own. Not "why is water blue," but instead "why does bright direct sunlight reveal bright colors in the environment, while overcast diffuse illumination greatly affects the perceived colors?" Hmmm. Maybe the light from an overcast day is not just dim. Maybe it's blue...
posted by billb at 9:39 AM on December 22, 2004

fiction is much wierder.

Only superficially.
posted by rushmc at 9:58 AM on December 22, 2004

I think this and this are the ones that remained unchallenged for the longest time for me. It really is alarming What They Teach In These Schools (tm).

Bill, I've been wandering around your site for far too long now, and I'm off to sleep before it All Happens Again :-)

Thanks again for an outstanding body of excellent work.
posted by flabdablet at 10:05 AM on December 22, 2004

MetaFilter is blue because it reflects the sky this guy.
posted by weapons-grade pandemonium at 10:08 AM on December 22, 2004

Ok, but the entry is not about the color of the reflections, it's about the color of the water itself.

Well, now we're getting into semantics, but I would still argue that the original statement you're arguing against ("LAKES AND OCEANS ARE BLUE BECAUSE THEY REFLECT THE BLUE SKY") would be understood by the average person to mean the blue that it appears when viewing at a glancing angle, which is most people's most common experience in viewing lakes and oceans.

If you doubt it, I propose the following experiment to determine whether that constitutes standard usage. Take a number of people to a lakeshore on a clear day. Ask them, "What color is the lake?" Count how many respond "it's blue," and how many respond "I can't tell, because the color of the lake is obscured by the reflection of the sky."

Ok, but the entry is not about the color of the reflections, it's about the color of the water itself.

No, actually the original statement you're trying to argue against is not about the color of the water, but the color of the lake or the ocean. (Again, the statement at the top of your page you are attempting to refute is "LAKES AND OCEANS ARE BLUE BECAUSE THEY REFLECT THE BLUE SKY," not "WATER IS BLUE BECAUSE IT REFLECTS THE BLUE SKY.") When a person views a lake at a glancing angle on a clear day, what they see appears blue. In common, everyday English, this is expressed as "the lake is blue." Any explanation, intended for laypeople, which begins "A lake is blue because..." needs to either take this common usage into account, or else qualify right up front that when it says "A lake is blue..." it does not mean the blue that is observed when standing on the lakeshore on a clear day.

And yes, during overcast days the brilliantly blue pacific water was far less blue. But so were all other colors in the environment.

Sure, grass seems less green on a cloudy day than on a bright one. But my experience (including living for two years just a hundred feet away from a decently sized lake, not to mention random vacations to lakes of all sizes, and yes, even oceans) is not that a body of water is "less blue" on an overcast day; it's not blue at all.

Maybe the light from an overcast day is not just dim. Maybe it's blue...

I observe that other blue objects (even, say, a mural that uses sky blue for its background color) still appear blue on overcast days.

It's not my explanation.

Sorry if I was unclear--I meant "your explantion" as in "an explanation offered by you," not "an explanation which you were the first person to conceive of."
posted by DevilsAdvocate at 10:36 AM on December 22, 2004

billb, devilsadvocate: I think you're talking past each other. If someone is asking the question "why does the sea look blue?" then the answer has got to depend on the way the glancing reflection colors the light. You could argue that the glancing reflection colors the light in accordance with the bulk transmissive color of the water; billb's argument implcitly relies on this idea. But, AFAIK, in general that kind of grazing reflection doesn't color the light very much¹, so the answer's likely to be "because the sea reflects the sky". Other evidence in support of this is that the sea is gray on an overcast day, blood-red below a sunset, and so forth.

If you're asking "what color is water?" then you're asking a very, very different question. My experience, living in Seattle, is that seawater is green; pull up a bucket of it to look and it's distinctly green, presumably due to floating phytoörganisms. Pure water in the mountains is blue: I'd always assumed this was because of fine suspended particles, but it could be inherent in the water. But this is beside the point unless you want to argue that the bulk transmissive color of the water is what determines the color of the glancing reflection.

¹ Thought experiment: How much water does the light interact with when it glances off the surface? A few wavelengths deep, I'd guess. What color is that thickness of water? Clear.

On preview: what devilsadvocate said.
posted by hattifattener at 10:50 AM on December 22, 2004

Great web site. If you're interested in similar kinds of stuff with regard to history, I recommend James Loewen's Lies My Teacher Told Me. He's a great speaker too, if you get the chance.
posted by grouse at 11:01 AM on December 22, 2004

Because sound travels much more slowly than light, people understand sound much better than color. If your neighbour’s loud music were bouncing off your house, you would never think that your house was the source of the sound. That’s because you hear the time lag from the source. But paint your house blue, and you think the blue comes from your house, or the paint on your house. Nope. The blue comes from the sun (or some other light source). It’s white sunlight echoing off your house with some of the low frequencies removed. The house is in fact orange, and that’s why you don’t see orange—it stops at the house. Your eyes, like your ears, are passive—they don’t somehow travel over to the house to pick up the color. When you understand that air, clouds, water, and other substances don’t possess color, but merely selectively echo sunlight, things are easier to understand. Black is the absence of light—the visual equivalent of silence. If you take a red object into a completely dark room, do you really think the red is still there but you just can’t see it? If the orchestra stops playing, is the sound still there but you just can’t hear it?
posted by weapons-grade pandemonium at 11:15 AM on December 22, 2004

Popular opinion to the contrary, only a small fraction of the sunlight striking the surface is reflected, unless the angle of incidence is very close to the horizon.
Even when the sun is only 10 degrees above the horizon, 2/3 of the sunlight enters the lake, and only 1/3 is reflected.
Blue light is strongly scattered by water, but red is only weakly scattered.
Because blue light is very weakly absorbed but strongly scattered by pure water, very deep unproductive lakes (i.e. nearly free of plankton or other particles) appear blue under natural light.
-Light in Natural Water

"Light from the sun and sky passes through the surface into the water. Not all the radiance is transmitted through the surface, a small percentage is reflected back up. The magnitude of the diffuse transmission coefficient for incident light is dependent on the sun zenith angle, atmospheric conditions, the refractive index of the water and the sea state. Within the water some of the downwelling light is scattered back up by the water and suspended particulates and passes back through the air-water interface where a proportion is reflected back down (again depending on the incidence angle and the seawater refractive index)."
-Oil-Water Contrast for Surface Oil

PS: Bill, I've been going to your site on and off for about 6 years now, thanks!
posted by nTeleKy at 11:51 AM on December 22, 2004

What color is a single molecule of H2O in sunlight?
posted by rushmc at 12:43 PM on December 22, 2004

"Blue light is strongly scattered by water, but red is only weakly scattered."

LOL! This one above is also a misconception, but it's at the physics-undergrad level, so I didn't put it on my page.

If the blue of water was caused by scattering of short wavelengths (Rayleigh scattering, Tyndall effect), then a view through a thick layer of water would appear red/orange.

Water is ACTUALLY BLUE. Just take a look at:

http://www.dartmouth.edu/~etrnsfer/water.htm
posted by billb at 1:07 PM on December 22, 2004

rushmc - It's obviously squant. Of course, you'll need SquantView^TM to view it in your browser.
posted by nTeleKy at 1:10 PM on December 22, 2004

What color is a single molecule of H2O in sunlight?

It is colorless, because it is smaller than the wavelength of light. That's why a soap bubble becomes clear and colorless when it gets really thin, just before it breaks. The iridescence, produced by interference of light waves, disappears.
posted by weapons-grade pandemonium at 1:13 PM on December 22, 2004

BTW, Bill, I don't mean to be overly negative by focusing on the one entry. Overall, it's a great site; some of those misconceptions I already knew were misconceptions; some I knew were false but didn't realize that they were common misconceptions; and for some I had my own misconceptions corrected. I especially like the one on the scientific method. If I were teaching children about the scientific method, I'd point out that "experiment" should really be "observation" (experiment is one very important type of observation, but not the only type), and also I'd replace "hypothesis" with "at least two hypotheses." If you only have one hypothesis, that's your only option, so why test it? You need at least two hypotheses, which make disparate predictions in some cases, before you can hope to design an experiment (or at least conceive of an observation) which would distinguish between them. And sometimes the simple logical negation of the first hypothesis can serve as the second (but even then it ought to be spelled out), but sometimes it can't.
posted by DevilsAdvocate at 1:18 PM on December 22, 2004

The "how wings create lift" entry is clearly wrong, and thus all of the other entries that so triumphantly declare the truth are now suspect to me. Too bad, it was fun at first.

Note: I have a degree in aerospace engineering.
posted by intermod at 8:48 PM on December 22, 2004

On second reading, I'm going to retract the statement that it's "wrong" and just say it's bad science. Ha!
posted by intermod at 8:51 PM on December 22, 2004

I was wondering about that one, intermod. Can you give us a layman's explanation or link?
posted by weapons-grade pandemonium at 10:30 PM on December 22, 2004

>On second reading, I'm going to retract the statement that it's "wrong" and just say it's bad science. Ha!


Heh. I'd think that you'd want to see the mistake corrected, and so you'd explain in detail WHAT IT IS.

Unfortunately much of the email attracted by the site is similar: people are moved to bad-mouth certain concepts, but then they find it impossible to explain what's wrong with them. I suspect that the problem revolves around "That's Not How All Textbooks Always Explain It." Yeah, but that's a failure of textbooks, since "If you don't have two or three ways to explain a piece of physics, then you don't really understand it." (?-Feynman?)

Also, the "wings and lift" section grew far beyond that one small entry. The airfoil misconception is part of standard pilot licenscing exams, which led to flamewars of no small extent. A pair of physics people published a popular article about the misconception, leading to quite a bit of news coverage. Follow the link to

http://amasci.com/wing/airfoil.html for lots more stuff.
posted by billb at 3:19 AM on December 23, 2004

CORRECT: THE SKY IS BLUE BECAUSE AIR IS BLUE.

Hey, I guessed right. =P Self-link. (I wonder who got there first; his pages don't seem to have dates.)
posted by Tlogmer at 4:00 AM on December 23, 2004

> BTW, Bill, I don't mean to be overly negative by focusing on the one entry...

Don't worry... I *enjoy* the intense discussion. It really helps give me a kick in the pants, thinking-wise.

For example, I finally see a big semantic problem: the phrase "the color of the ocean" has two interpretations. I was using that phrase in the same sense as "the color of a brightly-colored automobile." What color is a red car? It's RED, even if we view it from a glancing angle across it's shiny hood during a sunny blue-sky day. Or in other words... if we're talking about the "color of an object," then we are required to view it from an angle where surface-reflections aren't an issue. If all we can see is the surface reflections, then we don't know the color "of the car."

But "the color of the ocean" has a second interpretation: the actual color that hits your retina. With bright fluffy clouds on the horizon, the ocean "is white." During sunset, the ocean "is red."

Children think that the color of the sky somehow infects the ocean water, making it blue. The misconception isn't about surface reflection colors. The misconception could be stated more clearly I guess... more like this: "Water is not a blue substance, instead it's colorless, and if water ever looks blue, it's because it somehow got its color from the sky above it."

Here's a great photo that demonstrates that a thick layer of ocean water is not colorless. All we need is some submerged white surfaces, such as beach sand:
NASA photo: Bahamas

Of course this applies only to warm seawater. Warm ocean water is extremely transparent. The cause was finally discovered in the 1980s: it's because the enormous population of viruses kill off nearly all the single-celled life.
posted by billb at 1:17 PM on December 23, 2004