Wikipedia:Reference desk/Archives/Science/2022 August 17

= August 17 =

Red-green colorblind.


Which green is for red-green colorblind? I have a feeling it is the light model, rather than the paint model. So "dark green" is more differentiated from the red. 67.165.185.178 (talk) 09:27, 17 August 2022 (UTC). Edit: and someone feel free to shrink the images. 67.165.185.178 (talk) 09:28, 17 August 2022 (UTC).
 * I shrunk your images. -- Jayron 32 10:48, 17 August 2022 (UTC)
 * It's not a specific shade, rather its a range of color families that get confused. See Color blindness for more information.  The human eye has three different kind of colored light receptor cells known as cone cells, and each is sensitive to a different range of colors, roughly corresponding to wavelengths of light around the red, green, and blue ranges respectively.  In people with red-green colorblindness, their "red" and "green" receptors essentially respond to the same wavelengths of light in the same way, so the entire range of colors that could be distinguished differently by those two receptors all look the same to them.  -- Jayron 32 10:52, 17 August 2022 (UTC)
 * I always thought it was by wavelength of light, and so, 555 nm of green for example. So I wonder what does pink look to a red-green colorblind person. 67.165.185.178 (talk) 12:07, 17 August 2022 (UTC).
 * 555 nm is perceived as a shade of green, but there are other ways you can perceive green, for example as a mixture of other wavelengths, none of which is in the range of 555 nm, but which stimulate your three cone cells in such a way that your brain can't tell the difference, you will still perceive that as green. Color vision is much more complex than you seem to be thinking of it.  The answer is that people who are red-green color blind perceive colors like, say pink and sea-green to be indistinguishable from each other.  -- Jayron 32 12:53, 17 August 2022 (UTC)
 * The color blindness article talks about inability to tell a red apple from a green one, and has a section on red-green color blindness, but that section only describes the inability to tell the neutral point (between the two cones) from white: it does not describe a mechanism for inability to tell spectral red from spectral green, and logically two different types of cone cell (blue and red, anyway) should suffice to distinguish all spectral colors. I don't get it. Card Zero  (talk) 16:06, 17 August 2022 (UTC)
 * The point is, you don't see spectral colors either. You see stimulations of the cone cells in the retina of your eyes, and each cone cell responds to both intensity and wavelength of the incoming light, and your brain does a complex bit of processing to determine what color you perceive.  Even with normal color vision, you cannot distinguish between a pure wavelength of light that you interpret as a specific color (say, for argument's sake, "yellow" at 580 nm) and a mixture of wavelengths that with absolutely no contribution at all from 580 nm light, but which stimulates the three kinds of cones in the same way that 580 nm light does.  You just see "yellow" in both cases.  That's why the RGB color model works in things like TV displays.  A series of closely spaced dots of only three very specific wavelengths is enough to make you see all of the other colors as well.  The display on your phone, TV, computer monitor, produces essentially no light of wavelength 580 nm and yet, you can see yellow on that display just fine.  Because all you see is the stimulation of your cone cells, as processed by your brain.  The way to think of color blindness in this way is that the brain of colorblind people is still expecting inputs from three different kinds of cone cells, but two of them are sending the same signals.  So, anything that in a non-colorblind person would stimulate the red cone cells differently than it would the green cone cells, in a colorblind person would be stimulating both kinds of cone cells the same amount, all the time, regardless of what the incoming light is.  As a result, their brains interpret as the same color two colors that a non-colorblind person would interpret as different.  It doesn't really matter what is happening at the blue cone, the brain expects three signals and is still processing the input like three signals coming in.  So it interprets every possible combination of inputs on the red and green cones as some shade of yellow, which is what "equal amounts of red and green stimulation" tends to mean in a non-colorblind person.  And that isn't strictly correct; since color is a qualia, there's real problems with associating the color one person sees, in an absolute sense, with what another person sees.  But you get the idea.  -- Jayron 32 16:25, 17 August 2022 (UTC)
 * Singular: quale. --Lambiam 17:35, 17 August 2022 (UTC)
 * Some parameters:
 * This question-hijack is only about spectral colors.
 * Let's suppose the person's short and long wavelength cones work, but the green ones aren't sending a signal.
 * So spectral red should stimulate the red cones a lot, and not the other type(s), which is just what red ordinarily does. Then spectral green should stimulate the red cones somewhat and the blue cones somewhat less, and yet not stimulate the green cones at all (since they're missing), which I guess is an impossible color, but why would it register the same as red? Now I'd better read that article too. Maybe later. Card Zero  (talk) 16:38, 17 August 2022 (UTC)
 * It's not that the green ones don't exist, or don't send a signal (at least in normal red-green color-blindness, perhaps there is some disorder where that happens; I've not heard of it), it's that the green ones and the red ones exist and send signals perfectly fine, and they are differentiated by your brain, but the two respond to light in the same way. I'm not really sure what would happen in your situation, because it's not a thing as far as I am aware.  (also, I've greatly simplified the process of red-green colorblindness, aka Daltonism, it's a bit more complex, but this gets the spirit of it).  -- Jayron 32 17:35, 17 August 2022 (UTC)


 * I was following the article, where it says Deutan (6% of males): Lacking, or possessing anomalous M-opsins for medium-wavelength sensitive cone cells. I figured "lacking" would be the simplest case. But if the green-sensitive cells are responding to red, then spectral green should still look unlike spectral red, because it would stimulate the blue cells a bit. Or ... maybe the responses are narrow enough that relative closeness to blue is imperceptible. That must be it. Card Zero  (talk) 17:57, 17 August 2022 (UTC)
 * You could very well be right then... I'm getting in over my head, and am backing out in favor of your clearly greater expertise here. -- Jayron 32 18:06, 17 August 2022 (UTC)
 * Just trying to make reality conform to the explanations I've somehow got hold of. Or the other way round. I don't know about expertise, we're all just making finger-puppets in a cave, or whatever Plato said. Card Zero  (talk) 18:20, 17 August 2022 (UTC)
 * If we go all math-y, each cell cone type detects the integral of the object’s radiation over all wavelengths weighted by the cone spectral sensitivity.
 * It is therefore true that two different types of cone cell [suffice] to distinguish all spectral colors (as long as the ratio of spectral sensitivities of the two cones varies along the whole spectrum, and subject to signal-to-noise limitations at very low sensitivities). [EDIT: the signal-to-noise stuff is likely the cause of most difficulties, see below.] But it does not prevent a "red" apple (which is not monochromatic-red) to have the same few weighted integrals as a "green" apple. In fact, it is fairly clear that as the eye sees in a few discrete colors (two or three), most of the information of a continuous spectrum is lost, and therefore many objects with random continuous spectra will appear the same to the discrete-transform of the eyes. (Of course, that’s with the math hat on; with the biology hat on, "random continuous spectra" do not occur in nature, and a small number of well-chosen discrete colors probably gives 99.9+% of the environmental information needed to thrive in most ecosystems, and avoids the need to synthetize hundreds of different photosensitive receptors.)
 * I find it hard to believe that someone who can link to spectral colors missed this; sorry if I misunderstood the source of your confusion. Tigraan Click here for my talk page ("private" contact) 16:32, 17 August 2022 (UTC)
 * Oh, good! Non-spectral apples, that did cross my mind, but the main thing is, spectral red and green would look different? So fully-saturated red and green on a computer screen would in fact be distinct to a red-green color-blind person. Which sounds intuitively like the purest example of what they should not be able to distinguish, so the counter-intuitive reality is interesting. Card Zero  (talk) 16:45, 17 August 2022 (UTC)
 * I don't believe so. LED traffic signals output essentially monochromatic light, and the red and green can be confused by someone with red-green colorblindness.  This discussion thread between people with colorblindness note tricks they use, from location of the bulbs, to senses of brightness or saturation to tell them apart, but they are not able to distinguish them by hue.  -- Jayron 32 17:44, 17 August 2022 (UTC)
 * Hmm, I do have to amend my answer above somewhat - it does not contain anything wrong but it’s certainly misleading and too math-centric.
 * Let us assume deuteranomaly, the most common form of color blindness, where the individual lacks functional M-cones ("green" receptors). If you look at the graph of cone sensitivity vs. wavelength posted above, you can see that above ~550nm, the S-cones ("blue" receptors) have essentially zero sensitivity. (Perception of light is highly non-linear etc. etc. so I am not sure what the actual threshold is, but the point is that L-cones see nothing beyond a certain threshold.) Therefore, in that area, a deuteranomalous person "sees" a single color channel (from cones at least). And then, the L-cone response is non-monotonous, so two points of the curve that have the same height are perceived as the same color; eyeballing the curve, this happens at 550nm (clear green) vs. 590nm (orange-ish red).
 * So while mathematically, two colors channels are enough to distinguish any two spectral colors, physically, the S-channel does not work in a large range (because of perception threshold / signal-to-noise / etc. issues), thus deuteranomalous humans can only rely on one channel in that range, which causes them to mix up certain colors. However, there is no real reason that "pure" colors should be harder to distinguish than "mixed" colors (without making assumptions of the spectrum of the "mixed" colors). Tigraan Click here for my talk page ("private" contact) 08:57, 18 August 2022 (UTC)

Does the sea need rain?
Cloud-seeding made me think of this. Suppose there was some technological means (with no side-effects, for the purpose of this question) by which people diverted all rain to fall on land. Would this be ecologically harmful to the sea somehow? I suppose there'd be more stuff washed into the sea, leading to algae blooms and maybe salinity increase. Can't think of any other drawbacks. Card Zero (talk) 17:30, 17 August 2022 (UTC)


 * I'll leave it to those versed in geography, hydrography, climatology, etc. to answer this in depth, but in the most naive estimation (using simple available numbers) we can take the 1.386 billion km^3 of total water (I believe this includes vapor) on Earth, 1.338 billion km^3 (96.5%) being in oceans; and the 505,000 km^3 of annual water falling as precipitation, 398,000 km^3 (about 80%) over the oceans; and calculate that that if none of the rain that fell on land returned to the oceans in a balancing cycle, the oceans would be depleted in at least 13,000 years (much longer, maybe 2x-10x longer or so, because it's actually a differential equation even in a simple estimation as the rate of evaporation from the oceans and thus their contribution to rainfall decreases as they are depleted, as does the total area over which rain will conceivably fall into it, again being completely naive with this calculation), which to me sounds extremely fast, but the water cycle is enormous. The average annual drop in sea level is surprisingly tricky to measure (and varies by region) based on water volume alone, but there's probably a simple estimation method somewhere. SamuelRiv (talk) 17:58, 17 August 2022 (UTC)
 * I wasn't expecting the water to pile up on land and stay there! Though this literal interpretation is pleasing. Card Zero  (talk) 18:05, 17 August 2022 (UTC)
 * Ohhhhhh! So this isn't the "I demonstrate basic arithmetic using a grade-school understanding of Earth science" Reference Desk! SamuelRiv (talk) 18:09, 17 August 2022 (UTC)
 * Rain water does not contain mineral salts, unlike the river water that flows into the seas. The influx of mineral salts would about triple. Since sea water is already so much more saltier than river water, by a factor of about 300, while the yearly influx of river water is less than 1/2500th part of the ocean water volume, I guess it will take centuries before the effect in increased salinity exceeds 1‰. --Lambiam 18:19, 17 August 2022 (UTC)
 * But while the flow of river water increases, the concentration of dissolved materials in this river water could decrease. The concentration of suspended solids obviously increases after heavy rain, but I expect the total amount of nitrates coming from farmland to remain the same. That's set by what the pigs and chickens drop. PiusImpavidus (talk) 10:47, 18 August 2022 (UTC)


 * "High levels of rainfall are commonly observed over the tropical regions of Earth's oceans, impacting physical processes that influence weather and climate from the microscale to the basin scale."
 * "Rain alters the physics and carbon chemistry at the ocean surface to increase the amount of CO2 taken up by the ocean."
 * "...rain and wind effects combine nonlinearly to enhance air-water gas exchange. "
 * "Rainfall over the sea modifies the molecular boundary layers of the upper ocean through a variety of different effects. These cover the freshwater flux stabilizing the near-surface layer, additional heat flux established due to rain versus surface temperature differences, modification of physical parameters by temperature and salinity changes, enhancement of the surface roughness, damping of short gravity waves, surface mixing by rain, and transfer of additional momentum from air to sea."


 * The above (thanks User:Fiveby) is the kind of thing I had a hunch must be the case: that the life in the ocean needs regular watering for some reason. Putting CO2 into the water makes it more acidic and is generally thought of as a bad thing, but presumably the splashing rain also helps oxygenate the water, which must matter a lot for life near the surface. The last link hints at a layer of reduced salinity and moderate temperature as well, but I don't know if that constitutes an environment that maybe plankton depends on. I'll also just note that "gravity waves" are not the same as gravitational waves, because that took me a moment. Card Zero  (talk) 13:50, 18 August 2022 (UTC)
 * IIRC, there are some animals that returned to the sea, but still prefer to drink fresh water, like sea snakes. After heavy rain, the fresh rainwater doesn't mix immediately with the denser seawater, so they can drink relatively fresh water from the top few millimetres. PiusImpavidus (talk) 10:47, 18 August 2022 (UTC)