An intriguing, anecdotal finding was recently reported by some news outlets that the implementation of blue-colored streetlights has reduced both crime and suicides:

Glasgow, Scotland, introduced blue street lighting to improve the city’s landscape in 2000. Afterward, the number of crimes in areas illuminated in blue noticeably decreased.
The Nara, Japan, prefectural police set up blue street lights in the prefecture in 2005, and found the number of crimes decreased by about 9 percent in blue-illuminated neighborhoods. Many other areas nationwide have followed suit.
Keihin Electric Express Railway Co. changed the color of eight lights on the ends of platforms at Gumyoji Station in Yokohama, Japan, in February.

Since the railway company introduced the new blue lights, they’ve had no new suicide attempts.
This effect may be attributed to a few possible reasons (some of which are mentioned in the comments section of the article):

• The light color is new and unusual, causing people to act more cautiously in the area (as a person is unsure what to expect in the unusually-lit area).
• Blue is a light color almost universally associated with a police presence, suggesting it is an area of stricter law enforcement.
• Blue may be a more pleasant illuminating color to most people, as opposed to yellow, orange or red (according to some research, such as Lewinski, 1938).

In fact, the article quotes from a professor at the end, noting it may just be an “unusualness effect:”

Prof. Tsuneo Suzuki at Keio University said: “There are a number of pieces of data to prove blue has a calming effect upon people. However, it’s an unusual color for lighting, so people may just feel like avoiding standing out by committing crimes or suicide under such unusual illumination. It’s a little risky to believe that the color of lighting can prevent anything.”

There is a lot of research into the psychology of color, but not as much has looked into the color of blue illumination itself (as opposed to the color of an object or wall). But some research looking into short wavelength light (blue) has demonstrated that it is a potentially effective treatment for seasonal affective disorder (a seasonal type of depression; see for instance, Glickman, et al., 2006), and helps to reduce the stress response in fish (it hasn’t been yet tested on humans).
If this finding is robust and the behavior change associated with it is still prevalent a few years from now (when everyone has become accustomed to the new light color), it would be an interesting finding. A simple, inexpensive change might be effective in helping reduce at least one method of suicide (and reduce crime to boot).