The evils of interior fluorescent lighting

Ok, so maybe “evils” is a little too strong. There’s no doubt that fluorescent light fixtures consume less power per emitted unit of perceived light.

But, look at this:

Not exactly the same distribution, right? Look at those nasty spikes on the short end of the visible spectrum. Jesus, it’s no wonder that things look cold and dead under bad fluorescent light. It’s not remotely an approximation of anyone’s concept of “daylight,” one version (D65) of which is plotted below (note: the x-axis scales are not equivalent)!

(Sorry there’s no scale on the y -axis of the first plot, but it’s not my graph. Anyone with a better source for fluorescent light spectra are welcomed to submit them to me; in particular, I’d really love to see a comparison of the “daylight” ~5000k bulbs vs. the cheaper fluorescent bulbs. I’m not convinced that they make things look any better–instead, it seems to me that it’s just less red!)

Update: Here we go…lots of ugly, ugly lamp spectra! http://ledmuseum.home.att.net/spectra7.htm

So…this power output distribution, taken in concert with the photopic (”light-adapted,” the black plot) and scotopic (”dark-adapted,” the green plot) response of the human retina:

…produces…

References and Selected Readings:

A quick aside: Randy Kardon, M.D., Ph.D., of the University of Iowa has published research regarding a preferred response to those precise wavelengths by a subpopulation of melanopsin-expressing (and photosensitive) ganglion cells within the retina, a population of cells previously believed to lack photoreceptive capabilities. See the abstract (and the paper, if you have access) here: J Neuroophthalmol. 2007 Sep;27(3):195-204.

See these articles for more information: