As Dodgeville (and many other towns and cities) are planning to replace their streetlights with LED luminaires, it is imperative that we use LEDs with a CCT (correlated color temperature) of 3000 K or less (Jin et al. 2015). This is a “warm” white light (similar to incandescent) rather than the “cold” blue-rich light often seen with LEDs. Outdoor LED luminaires often come in at least three “flavors”: 3000K, 4000K, and 5000K. For example, American Electric Lighting’s Autobahn Series. 5000K luminaires provide the bluest light, and should be avoided at all costs. Of these three, 3000K would be best, and if 2700K is offered, use that.
Why does this matter? On June 14, 2016, the American Medical Association issued guidance on this subject.
Incidentally, for your residential lighting needs, a good local source for LED bulbs that are not blue-rich is Madison Lighting. They have many LED bulbs in both 3000 K and 2700 K. I use 2700K bulbs exclusively in my home, and the warm white light they provide is an excellent replacement for incandescent and compact fluorescent bulbs. Never purchase LED lighting without knowing the color temperature of the lights.
If you’re skeptical that the color temperature of LEDs is an important issue, I suggest you purchase a 2700K bulb and a 4000K or 5000K bulb with the same output lumens and compare them in your home. I believe that you will much prefer the 2700K lighting. If 2700K lighting is best for your home, then why should it not be best for outdoor lighting as well?
Besides, most streetlighting is currently high pressure sodium (HPS), which is inherently non-blue-rich. You will find that 2700K LED lights offers better color rendering than HPS without the need to go to even bluer lights.
If you have ever been irritated at night by an oncoming vehicle with those awful “blue” headlights, you’ve experienced firsthand why blue-rich light in our nighttime environment must be minimized.
Why are 4000K and 5000K LED lights so prevalent? They are easier and cheaper to manufacture, but with increased demand of 2700K and 3000K LED lights, economies of scale will reduce their cost, which today are generally slightly higher than blue-rich LEDs.
Now, a bit more about why blue light at night can be detrimental to human health, and the primary reason why the AMA issued a guidance on this subject.
In addition to image-forming rods and cones, there exist non-image-forming retinal cells in the human eye called intrinsically photosensitive retinal ganglion cells (ipRGCs) that help regulate our circadian rhythms. Studies have shown that blue light is far more disruptive to our circadian rhythms than redder light (Lockley et al. 2003).
Now, on to the environment. Using a clever technique that compared sky brightness at several locations on several nights both with and without snow cover, Fabio Falchi (Falchi 2011) determined that at least 60% of light going up into the night sky is direct waste lighting, and 40% or less is reflected light. This is as good an argument as any that we still have a long way to go towards using only full-cutoff luminaires that do not produce any direct uplight. Blue light scatters much more in the night sky than red light, and this is due to Rayleigh scattering which tells us that the amount of scattering is proportional to the inverse of the wavelength of light to the fourth power, σs ∝ 1 / λ4. This also explains why the daytime sky is blue.
Bluer wavelengths of light thus increase artificial sky glow to a much greater extent than redder wavelengths do. Not only is an increase in blue light bad for astronomy, but its impact on the natural world is likely to be adverse as well.
Falchi recommends a total ban of wavelengths shorter than 540 nm for nighttime lighting, both outdoor and indoor. He goes on to say that, at the very least, no more light shortward of 540 nm should be allowed than that currently emitted by high pressure sodium lamps, lumen for lumen.
Falchi, F. 2011, MNRAS, 412, 33
Falchi, F. 2016, The World Atlas of Light Pollution, p. 44
Jin, H., Jin, S., Chen, L., et al. 2015, IEEE Photonics Journal. 7(6), 1-9
Lockley, S. W., et al. 2003, J Clin Endocrinol Metab. 88(9), 4502–5