I understand the desire for a rule to plug things into to determine light intensity, and why the wpg thing is so popular. Most people don't have access to a PAR meter and just want an easy way to get into the right ballpark with their lighting.

But, from what I can see, one of the major drawbacks of any such rule that is using gallons as one of the components of the formula, is that I don't think the volume of a tank really has all that much to do with the light intensity in that tank. Yes, it makes some amount of sense that a bigger tank would need more light. But wouldn't the actual dimensions of the tank be a much more accurate way of generalizing how much light a particular tank would need?

For instance a standard 10 gallon tank has a surface area of 200 square inches, a ratio of 20 square inches per gallon. A 50 gallon has 648, only about 13 square inches per gallon. So wouldn't it make sense that the 10 gallon would require substantially more light intensity per gallon to achieve the same effect as the 50 gallon? So just for purposes of this illustration, lets say we have 3 wpg over the 50, and we're trying to get the same over the 10g.... we would need 154% of that based on surface area, so we would actually need 4.62 watts per gallon.

However it's true that that doesn't take into account the depth of the tank. The 10 gallon is 12" deep and the 50, 19". There should be some type of calculation (which I'm not smart enough to figure out) that would account for the additional light needed to have the same intensity at 19" as you would otherwise at 12"..... or for this illustration since we're solving for the 10 gallon, the same intensity at 12" as we do at 19" in the 50g. Add or subtract this intensity from our previous calculation based on surface area, and maybe a more accurate watts per gallon would result.

Then the other issue is that you have a greater amount of surface area IN GENERAL in a smaller tank PER GALLON than you do in a larger tank. Light shining unimpeded in a tank through the water would seem to me to be much more powerful than light that bounces off glass, substrate, etc. and ends up getting either partially absorbed or lost. So maybe rather than just taking the surface area itself into account, the total of all the walls, substrate, and surface area should be calculated.

Anyway, all of this does not remove all the variables involved with the type of lighting and reflectors...but it seems to me that with some math, a formula or two should be able to be used to determine quite a bit more closely the lighting intensity in a particular tank, far more accurately than wpg could ever do, without the use of a PAR meter.