You can say one thing for outdoor sodium lamps and the ugly yellow light they generate: The light doesn't suppress the generation of melatonin in humans and animals.
The same can't be said, however, for the light generated by white LEDs. At least so say researchers Fabio Falchi, Pierantonio Cinzano, Christopher D. Elvidge, David M. Keith, and Abraham Haim. Writing in a recent edition of the Journal of Environmental Management, the team describes their research examining to what degree different kinds of light bulbs suppress melatonin production and lead to light pollution at night.
Researchers first calculated the wavelength and energy output of bulbs generally used for outdoor lighting. They then compared that information with existing research regarding melatonin suppression to determine how much each type of bulb would suppress melatonin production. It turns out that the metal halide bulb, which gives off a white light and is often used for stadium lighting, suppresses melatonin at a rate more than three times greater than the high-pressure sodium (HPS) bulb. White-light LEDs suppress melatonin at a rate more than five times higher than HPS bulb. Researchers also note the road surface materials, either asphalt or concrete, reflect less short wavelength radiation, implying that lamps emitting more long-wavelength radiation such as HPS lamps will reflect more from roads.
But lamps such as white LEDs that emit more in shorter wavelengths reflect less from road surfaces. This lowers the visual benefits of these lights, researchers say. "Accumulated evidence of the demonstrated negative effects of light at night may well outweigh the positive ones," the researchers warn. "Moreover, most of the positive effects used to justify the huge expenses to build, maintain and power external lighting are based on anecdotal indications or poor statistical analysis. Even for the road safety effect there is a lack of studies using randomized controlled trials," they continue. They also warn that a migration from HPS lamps to blue-rich white LEDs could boost artificial night sky brightness by 2.5 to 5 times as perceived by the dark-adapted human eye. "Such large increases, combined with the usual growth of installed flux, may produce a tenfold increase of the scotopic sky brightness in the next ten years or so," they say. They expect the same order of magnitude increase on the melatonin suppression action spectrum.
But there is a way out of this dilemma. LEDs could be tuned and produced with very different spectra, they note, so it is advisable that industry research be pushed toward the production of less polluting warm LEDs, with no blue emissions.
You can find the full article here: http://www.sciencedirect.com/science/article/pii/S030147971100226X