Let me start off by saying I have been on a year long mission to learn about reef tank lighting the best I can... Actually this has been a 7 year mission, but only in the last year have I been able to afford the testing equipment to run tests myself.
So.. over the last 12 months, reading every scientific paper I can, emailing Sanjay with questions, talking with people like Tyree at frag shows about his experience and knowledge on the topic, and buying my very own PAR meter and Spectrometer to run my own tests.
Quick breakdown of what I have learned:
Lux = meaningless for corals
WPG (Watts per Gallon) = meaningless for lighting in general (T5HOs really blew that one out of the water so to speak)
PAR (Photosynthetic Active Radiation) = meaningless unless you have up to date spectral graphs of the bulbs your testing.
PUR (Photosynthetic Usable Radiation) = Better than PAR, but no scientific information on specific corals to determine exact PUR values, and each coral is different.
Kelvin = Not even sure about this one, what is written on the bulb/description and what I see visually are definately 2 different things, ie, my Phoenix 14k is MUCH bluer than my Radium 20k. If Kelvin meant anything at all, the Radium would be much more blue than the Phoenix, but it isn't.. and for that reason, the Phoenix having more blue, I have concluded for my setup it is hands down the best "coral bulb" on the market for growth/photosynthesis. The Radium has a higher PAR reading than the Phoenix, but I can without hesitation and with witnesses say the Phoenix gives me just about 2x the growth on my SPS corals as the Radium does. So again, PAR=Meaningless without knowing the spectral output of the bulb.
Spectrum = The output wavelengths of the bulbs means everything for this hobby.
DLI (Daily Light Index) = Great to figure out how long to run your lights each day when combined with specral output knowledge and PAR readings.
So.. first thing. PAR. PAR meters measure the entire wavelength of 400-700nm. Higher wavelengths in the Orange/Red area, the photons hit the PAR sensor more often than Blue. Think of a wave.. higher wavelengths = shorter distance between wave peaks, meaning more waves hit the sensor. The meter will then take ALL the photons that hit the sensor and give you an average # of them all, which is the # you see on the screen. Here is the problem. If the Red spectrum is hitting the sensor at a rate of 3 photons to every 1 blue photon and then you get the average of that, it doesn't tell you how good your lighting actually is. Corals in nature rely primarily on the 400-500nm spectrum for the majority of its photosynthesis (higher wavelengths do not penetrate water nearly as much and lower wavelengths, this is why the ocean is blue once you get down to a certain depth.) Higher wavelengths play minor roles in photosynthesis but bigger roles in pigmentation/coloration of the coral. Simply running a tank all in the 400-500nm spectrum will give you good growth and healthy corals, but they probably will not look very good to the eye, and your tank will not look as pleasing to you either.
Here is where PAR really fails. Take a blue bulb, like T5HO Actinics, measure the PAR. On my 22" bulbs I get about 150PAR per actinic bulb, and I know the output is entirely in the "good light" area for my corals. Great thing about blue/actinics are they do not lose their intensity much with depth. 150 at the surface, 135 @ 20" underwater. Now, if you could get one, put a pure Red bulb in place of the actinic and do the same measurements. 300 PAR at the surface.... but 40 PAR @ 20", huge drop off, not to mention the entire light it is putting out will do almost zero for your corals. They will definately not grow under red lighting. People often think of photosynthesis as a one shoe fits all type thing and it simply isn't true. Photosynthesis in plants on land can utilize the Red spectrum, and have a much broader range of light it can use for photosynthesis, where as corals, living under water, are not exposed to the higher wavelengths much at all. Sure, low tides can expose some, and I have nothing to back this up but I think when corals are exposed to the air and emmit their slime protection they hinder their ability for photosynthesis because they are more concerned about staying alive and not being killed by the elements/UV light. I don't know if it a complete stoppage or a severe slowdown, but corals are obviously not in the "growing mood" when out of the water. This last part is simply speculation though and if someone can show me proof to the contrary I would be delighted to read it.
So anyway, I can sit here and write several books on the topic, but I think this is more than enough to scare most people away as it is.. so anyone still left reading and interested, I would love to hear your thoughts, insights, links to studies (I have read every article on Advanced Aquarist 100x over now, and they have done some AMAZING work in the lighting studies area).
Lastly, to prove the point about PAR, there are several ways a person can do this on their own even without a meter. If you have the ability to setup a test tank, and can get Red MH or T5 bulbs, try this one. Put in a blue bulb (or several depending on your light setup) and put a test coral in the tank and observe for 3 days. Then take the blue out and replace with Red and run for another 3 days (if you can, usually at the end of day 2 the coral is going south quickly). If you can use a PAR meter you will see the Red bulb is putting out 2-3x the PAR as the blue bulb (makes sense, 2-3x the photons hit the sensor in red light vs blue light), yet the coral is dying, where as with the blue bulb they live. Another test I did.. I have a 6x T5HO light on my 29G. I put all 10k bulbs in there, PAR was crazy, over 1000 at the surface, but only about 200 at the bottom. Obviously there is still blue in 10k bulbs, just not as much as a pure blue bulb. Result, corals (Hyacinth birdsnest was my test) was dead right between week 3-4 of the test. I then replaced all the 10K bulbs with ATI Blue+ bulbs and ran the same test with a new frag of the same coral. PAR was around 700 at the surface and 400+ at the bottom. Much more penetration happening. Result, coral lived and actually grew after 4 weeks and kepts its coloration when placed next to the mother colony in my display tank. I encourage everyone to run their own tests if this is a topic that interest them.
I always say "Read everything you can about the topic your interested in, and use the information the best you can, but do not rely on the information unless you have independently confirmed it yourself. I live by that line of thinking and would expect no less from others reading what I write. I am not the end all be all expert on this subject by any means and will never claim to be, but I have spent over $5000 on test equipment and bulbs just for this topic in the last 12 months so it is one that I have great interest in. I also stayed at a Holiday Inn last night.. j/k
Anyone... by all means, challenge me, tell me I am wrong about something if I am, but if you do PLEASE post links showing why and how I am wrong so I can revise/rerun tests to get more accurate data.
So.. over the last 12 months, reading every scientific paper I can, emailing Sanjay with questions, talking with people like Tyree at frag shows about his experience and knowledge on the topic, and buying my very own PAR meter and Spectrometer to run my own tests.
Quick breakdown of what I have learned:
Lux = meaningless for corals
WPG (Watts per Gallon) = meaningless for lighting in general (T5HOs really blew that one out of the water so to speak)
PAR (Photosynthetic Active Radiation) = meaningless unless you have up to date spectral graphs of the bulbs your testing.
PUR (Photosynthetic Usable Radiation) = Better than PAR, but no scientific information on specific corals to determine exact PUR values, and each coral is different.
Kelvin = Not even sure about this one, what is written on the bulb/description and what I see visually are definately 2 different things, ie, my Phoenix 14k is MUCH bluer than my Radium 20k. If Kelvin meant anything at all, the Radium would be much more blue than the Phoenix, but it isn't.. and for that reason, the Phoenix having more blue, I have concluded for my setup it is hands down the best "coral bulb" on the market for growth/photosynthesis. The Radium has a higher PAR reading than the Phoenix, but I can without hesitation and with witnesses say the Phoenix gives me just about 2x the growth on my SPS corals as the Radium does. So again, PAR=Meaningless without knowing the spectral output of the bulb.
Spectrum = The output wavelengths of the bulbs means everything for this hobby.
DLI (Daily Light Index) = Great to figure out how long to run your lights each day when combined with specral output knowledge and PAR readings.
So.. first thing. PAR. PAR meters measure the entire wavelength of 400-700nm. Higher wavelengths in the Orange/Red area, the photons hit the PAR sensor more often than Blue. Think of a wave.. higher wavelengths = shorter distance between wave peaks, meaning more waves hit the sensor. The meter will then take ALL the photons that hit the sensor and give you an average # of them all, which is the # you see on the screen. Here is the problem. If the Red spectrum is hitting the sensor at a rate of 3 photons to every 1 blue photon and then you get the average of that, it doesn't tell you how good your lighting actually is. Corals in nature rely primarily on the 400-500nm spectrum for the majority of its photosynthesis (higher wavelengths do not penetrate water nearly as much and lower wavelengths, this is why the ocean is blue once you get down to a certain depth.) Higher wavelengths play minor roles in photosynthesis but bigger roles in pigmentation/coloration of the coral. Simply running a tank all in the 400-500nm spectrum will give you good growth and healthy corals, but they probably will not look very good to the eye, and your tank will not look as pleasing to you either.
Here is where PAR really fails. Take a blue bulb, like T5HO Actinics, measure the PAR. On my 22" bulbs I get about 150PAR per actinic bulb, and I know the output is entirely in the "good light" area for my corals. Great thing about blue/actinics are they do not lose their intensity much with depth. 150 at the surface, 135 @ 20" underwater. Now, if you could get one, put a pure Red bulb in place of the actinic and do the same measurements. 300 PAR at the surface.... but 40 PAR @ 20", huge drop off, not to mention the entire light it is putting out will do almost zero for your corals. They will definately not grow under red lighting. People often think of photosynthesis as a one shoe fits all type thing and it simply isn't true. Photosynthesis in plants on land can utilize the Red spectrum, and have a much broader range of light it can use for photosynthesis, where as corals, living under water, are not exposed to the higher wavelengths much at all. Sure, low tides can expose some, and I have nothing to back this up but I think when corals are exposed to the air and emmit their slime protection they hinder their ability for photosynthesis because they are more concerned about staying alive and not being killed by the elements/UV light. I don't know if it a complete stoppage or a severe slowdown, but corals are obviously not in the "growing mood" when out of the water. This last part is simply speculation though and if someone can show me proof to the contrary I would be delighted to read it.
So anyway, I can sit here and write several books on the topic, but I think this is more than enough to scare most people away as it is.. so anyone still left reading and interested, I would love to hear your thoughts, insights, links to studies (I have read every article on Advanced Aquarist 100x over now, and they have done some AMAZING work in the lighting studies area).
Lastly, to prove the point about PAR, there are several ways a person can do this on their own even without a meter. If you have the ability to setup a test tank, and can get Red MH or T5 bulbs, try this one. Put in a blue bulb (or several depending on your light setup) and put a test coral in the tank and observe for 3 days. Then take the blue out and replace with Red and run for another 3 days (if you can, usually at the end of day 2 the coral is going south quickly). If you can use a PAR meter you will see the Red bulb is putting out 2-3x the PAR as the blue bulb (makes sense, 2-3x the photons hit the sensor in red light vs blue light), yet the coral is dying, where as with the blue bulb they live. Another test I did.. I have a 6x T5HO light on my 29G. I put all 10k bulbs in there, PAR was crazy, over 1000 at the surface, but only about 200 at the bottom. Obviously there is still blue in 10k bulbs, just not as much as a pure blue bulb. Result, corals (Hyacinth birdsnest was my test) was dead right between week 3-4 of the test. I then replaced all the 10K bulbs with ATI Blue+ bulbs and ran the same test with a new frag of the same coral. PAR was around 700 at the surface and 400+ at the bottom. Much more penetration happening. Result, coral lived and actually grew after 4 weeks and kepts its coloration when placed next to the mother colony in my display tank. I encourage everyone to run their own tests if this is a topic that interest them.
I always say "Read everything you can about the topic your interested in, and use the information the best you can, but do not rely on the information unless you have independently confirmed it yourself. I live by that line of thinking and would expect no less from others reading what I write. I am not the end all be all expert on this subject by any means and will never claim to be, but I have spent over $5000 on test equipment and bulbs just for this topic in the last 12 months so it is one that I have great interest in. I also stayed at a Holiday Inn last night.. j/k
Anyone... by all means, challenge me, tell me I am wrong about something if I am, but if you do PLEASE post links showing why and how I am wrong so I can revise/rerun tests to get more accurate data.
I will post more pictures from over the year, side by side with dates to show growth under a 10k bulb vs the other bulbs. I will agree that my tests are in no way setup to be "scientific" like some of the test scenerios people have suggested but I do plan on doing just those types of tests in the near future so I can answer all these great questions people are asking. 
