This notion keeps popping up on CPF, and it's not true.
Yellow fog lights are a marketing gimmick, made semi-plausible by a misunderstanding of Rayleigh scattering. The fact is that fog particles are much too large to cause the strongly wavelength-dependent scattering that comes from air molecules (over very long distances) or hydrocarbon haze.
And anyway, you cannot make a beam penetrate better by putting a filter in front of it (which is what standard "fog" lights are). A filter merely removes some photons from the beam; the remaining photons continue onward in the same manner as if the filtered-out photons had remained.
The reason this might be true however, in the case of very bluish lights is that under very dark conditions, our eyes are a lot more sensitive to blue back-scattered light, than they are to say green, amber, or red back-scattered light. So the glare from a cool white light in fog is more irritating, even if more energy is not actually being reflected back. This is a very maginal effect in my experience. I believe 90% of the reputation of LEDs sucking for throw in the fog has to do with the fact that they have very bright spill, and weak main beams in general, while most incans concentrate most their lumens into the hotpsot. Comparing cool white LEDs with aspheric lenses, to similarly-bright incans in deep reflections (EG similar lumen output AND similar beam profile) the difference in fog-penetration is minor. A concentrating beam with no spill, held off-axis from your eyes (So backscatter does not hit you as much) is best, regardless of color temperature.
Another benefit for warm vs cool for fog penetration is the following:
Outdoors most object being viewed do not reflect strongly in the blue portion of the spectrum -- Let's say for an extreme case of argument you shine a blue LED, through fog, at a tree.
I shine an amber LED at the same tree. The fog reflects back an equal proportion from both lights, but the
tree reflects back very little blue. So the ratio of light reflected from the tree, compared to light reflected from the fog is a lot worse for the blue, meaning the tree is "washed out" by the fog which is still very bright (water reflects all colors equally)
The same argument can be extended to using warm white flashlights outdoors, versus cool white. Materials like trees, grass, and dirt tend to reflect a lot more in the green to red range and very little in blue. I find that lights in about the 4000k range look the most
accurate as far as color rendition goes, compared to the same scene viewed under daylight. This is due to the sensitivity shift of our eyes.
This makes sense if you consider that full moonlight is naturally around 4000k, that this is where our "low ambient light" white balance would work best. Consider that ehen the moon and sun are visible simultaneously, the moon actually looks brown -- but when there is a full moon out, the moon looks almost bluish white. It is our perception that is changing.
Incandescent lamps however at 3000k give me somewhat more contrast -- the red dirt or tree branches tend to "pop out" a lot more compared to the green grass surrounding them. This effect is similar to wearing brown tinted sunglasses during the daytime -- color accuracy is actually worse, but contrast for particular sorts of things is enhanced.
Outight "bug light" colored amber lighting, in my experience, is actually worse than using incan. This is because there really isn't any advantage from the effects I'm describing, but the overall color rendering is a lot worse to begin with.
In this case, you are viewing the tree with the blue light will see a very dim tree, which is being masked by bright blue backscatter (which the eyes are disportionately
more sensitive to in low concentrations). In my case the ratio of amber light reflected off the tree, to light reflected off the fog is much greater.
