I like the fact that the LED street lights out there typically waste less light by directing more of it downwards, and less skywards.

That comes out to $137.50 per light installed. That seems low specially for an LED light bulb.

Volume discount I guess. Or maybe 2.2 mil is just for the hardware, and the instalation comes from the regular maintenance budget. Maybe the LED modules they bought fir in the old lamposts?

"Because Anchorage has 85 days a year with less than 8 hours of daylight, any benefit over the tradition lighting architecture are compounded. "

Uh, pretty much every location on Earth recieves 6 months of light, 6 months of dark a year, its just a question of how it's distributed. Those lights will be off the entire months of June and July.

Jeromy's right. The extra sun in summer should just about balance the reduced sun in winter.

It evens out over the year, but it makes a difference in peak demand during the dark time, which is what ultimately matters to power plants, no?

Wait a minute, wait a minute.

I hear that LEDs for the house are still not as low power as CFLs quite yet. Also, I know that street lighting is obscenely efficient already, at least the newer more amber/orange/pink/soft/sulfur shaded lights. The newer lights take longer to warm up than the old ones, but are more efficient (why else would they be so prevalent).

So, are these LED lamps really better? Maybe they aren't replacing the new more economic softer lights but instead the wasteful painful old white ones that stupid landscape architects put in too many places. That would be better.

(As you can see, I hate the old white streetlights compared to the new. The new ones don't blow out my night vision. They are gentle like old gas lamps. And I assume they must use less power.)

The maintenance issue alone made the switch to LEDs practical on streetlights years ago, I'm not surprised that white LEDs would make sense for streetlights.

I am all for daylight balanced street lighting (especially lighting that is focused downwards and not glaring in pedestrian's or driver's faces). I walk at night often (especially during the long winter nights here in Scotland) and it is much easier on the eyes (and psychologically calming) to have daylight balance rather than the funky spectrum that sodium vapour emits.

I wonder also, LEDs are dimm-able without changing the temperature of the light. At some locations, electric street lighting was once switched depending on use (if you were walking down the street, you'd switch it on). Would it be feasible to install motion detectors to some of these lights so that they would dim down to a certain percentage that still illuminated the street, then up to full when someone approached? That would be even more efficient, no?

There is more to these LED decisions than people have room to explain. No one has said empirically why they are being used.

No one is saying that LEDs are more efficient than fluorescent streetlamps yet, and everyone agrees they are less efficient than High Intensity Discharge (HID) lamps.

(BTW, the current white streetlights which I said I HATE above, are HID with Mercury Vapor or Metal Hallide. The new prevalent (except where designers don't care about money, climate change, people's night vision, or light pollution) amber lamps are HID with High Pressure Sodium. The amber HID HPS are 10% more lumens/W than Metal Hallide.)

People are saying that LEDs point light more directly. Well, I think the city planners must have found a way of measuring that (or they'd have no way to decide on LEDs), and I wish one of the journalists would touch on that. I wish the person above who mentions the maintenance issues would make a quantitative comparison as well.

my sources:
http://www.eskimo.com/~jrterry/lamps.html
http://www.full-spectrum-lighting.com/durotest/Annual%20Cost%20of%20Lighting.htm
http://members.misty.com/don/lede.html

I must agree with Jason about the possibility of switching LEDs on and off to save money/energy. The HIDs switch on and off very slowly, so that's not a possibility for them.

As for the color, I guess I have to admit it's down to taste. But the astronomers are on my side, saying the lights with more limited spectrum (not white) are more easily filtered out. And I can add that the sodium lamps are similar in color to the old gas lamps, which light people seem to think are cool. Not the open flame ones that ignorantly nostalgic overly rich people like (I don't think piped-gas open flames never existed, folks, certainly not natural gas!), but the elegant lamps that attempted to make gas much more efficient.

I am all for daylight balanced street lighting (especially lighting that is focused downwards and not glaring in pedestrian's or driver's faces). I walk at night often (especially during the long winter nights here in Scotland) and it is much easier on the eyes (and psychologically calming) to have daylight balance rather than the funky spectrum that sodium vapour emits.

I wonder also, LEDs are dimm-able without changing the temperature of the light. At some locations, electric street lighting was once switched depending on use (if you were walking down the street, you'd switch it on). Would it be feasible to install motion detectors to some of these lights so that they would dim down to a certain percentage that still illuminated the street, then up to full when someone approached? That would be even more efficient, no?

Here's a good article about it in the Anchorage Daily News. Apparently the project is expected to cost $5m. Perhaps the $2.2 was just the initial funding? Also, the LED street lights are expected to last 20 years, whereas the sodium ones have a life expectancy of 3-4 years. That's a lot less maintenance, an 80% percent reduction. Imagine only having to mow the lawn every five or six weeks instead of every week.

LEDs use MORE power than the usual street lights so no, these will NOT save electrical costs. They are white light so its nice, but nothing touches the yellow street lights in terms of power use!

The cost issue is that it the estimated costs for replacing burned out bulbs are so high for cities that even using more power comes out cheaper if you have the bulb last years longer.

see http://en.wikipedia.org/wiki/Luminous_efficacy

@John B
Thanks for pointng out luminous efficacy - I haven't seen any economical studies of maintenance and replacemenet costs of low pressure sodium lamps compared to LEDs. This particular installation seems to me as a typical greenwash fueled asset-stripping of the city budget.
Anchorage could retrofit current streetlights with sodium lamps and save the rest of budget for maintenance electricians (which with LED streetlights get fired and become unemployed I suppose).

While HPS having a better luminous output to start with, that degrades quite quickly (like with CFLs) and even if not broken after 4 years it still takes the same energy, without the light. Horticultural HPS are often replaced every year because of that. And don't forget the losses in the ballast a HID has to be driven with, they get hot, and hot means loss. Those devices are simple inductors, while LEDs mostly use switching power-supplys which can easily reach over 90% efficiency.
LEDs usually don't break and still have 70% of light output after 50.000 hours. In addition the whole light is emitted downwards, unlike the HID which has to use a lossy reflector that degrades over time, too. Certainly I forgot some of the pros and cons of each, but it shows that a lot of factors are to be taken in to consideration.

Hmm, aren't these Luxim bulbs better than 16,000 LEDs?

Luxim 250w bulb @ Gizmodo