Light requirements

The discussion on light requirements is tending toward opinions based
on "What works for <insert your name here>". This is all well and
good, but doesn't provide much basis for determining "what will work
for me" based on "what works for you".  There are quite a few

1) Your tank size
2) Your water conditions
3) Your maintenance regimen
4) Your specific plants
5) Your aquascaping particulars
6) Your criteria for "success"

The simplified "watts per gallon" criteria may be useful for beginners
to show them that the 15 watt bulb that comes with their 29 gallon
tank is insufficient, but doesn't really help that much.  A more
pseudo-scientific method is sorely needed that brings together the
known parameters. Let me try to start a discussion by rambling a bit.
Feel free to comment.

Success criteria

A "successful" plant tank can mean almost anything.  Newbies may be
satisfied if the plants don't die right away.  Casual aquarists might
be happy of the plants stay green and they don't have to weed the
tanks every month and don't need to mortgage the house to buy
fertilizers. Enthusiasts want actual growth and propagation so they
can impress their friends and win HAP points.  Fanatics like me want
massive growth and aren't happy unless the plants are supersaturating 
the tank with O2 and can be harvested for sale at plant auctions to
pay for the high tech products being used.

The oft-maligned Kevin Osborne is perfectly happy with his smattering
of non-demanding plants.  I would be bored to tears. 

I wonder if we could develop some objective terms to describe our
successes?  Somehow I think the results obtained with a gro-lux and
cool white bulb and a bag of breath are different than results
obtained with obscenely expensive triton and ultra trilux bulbs and
automated CO2 injection.  Both techniques are satisfying to their
proponents but need to be quantified somehow for someone trying to
decide which route to go.   

Light requirements for plants

Different plants require different amounts of light energy for growth. 
Growth is limited by the amount of carbohydrates that are created
during photosynthesis.  Photosynthesis is limited by the amount of
light energy available (required intensity and spectrum) and the
amount of nutrients and CO2 available.  At a certain threshold of
energy, no photosynthesis will occur. Above that threshold, more and
more photosynthesis occurs until an upper threshold is reached.  See
the discussion in "Dynamic Aquaria" for more details.  

There is precious little information about how much energy various
plants need. The only quantitative data I've found is in "The Complete
Book of Aquarium Plants" (Allgayer and Teton, translated from French). 
The following table is from that book (roughly treanslated):

                Amount of lighting in lumens/m^2 (lux)
Plant genera    500 1000  1500 2000 2500 3000 3500 4000 4500 5000
surface plants      |------------------------------------------->
Myriophyllum        |--------------------------X----|
Bacopa            |--------X---------|     
Hottonia             |-------X--------------------------------|
Ludwigia             |-------X---------------------------|
Aponogeton           |--------X----------------------|
Ceratopteris         |---X-----------------------------|
Nomaphila            |--------X---------------------|
Rotala               |--------X------------------|
Vallisneria        |-------X--------------|
Echinodorus      |--X-------------|
Sagittaria        |-X---------|
Cryptocoryne     |-X--------|
Marsilea         |-X--------|
Micorsorium      |-X--------|

"X" indicates the optimum for each genera.

Depending on the plants in the aquarium, you can decide how much light
intensity you need *at the plant*.  For a tank holding Ludwigia,
Rotala and Nomaphila, the plants would need over 1000 lux at the
leaves.  Foreground plants like E. tenellus would need around 600 at
the leaves.  The upper leaves will get plenty of light so you need to
consider the needs of the lower leaves. Once you know the intensity
required at the leaf, you can figure out how much you need at the

Light loses intensity very quickly in water.  What we need is a chart
that shows how much is lost so we can determine how much light is
needed at the surface to produce the required amount at the depth we
desire. The Allgayer book has such a chart but it was apparently
garbled in the translation and is useless. There are aquascaping
specific considerations, also.  Do tall plants overshadow forground
plants?  Are emersed leaves blocking the light?  Also, light is
reflected from the aquarium sides so that plants near the glass get
more light than plants in the middle.  Perhaps such a chart could take
"typical conditions" into account.  Perhaps actual measurements with a
luxmeter could be taken. 

If you know how much light is needed at the surface, you can determine
approximately how many lumens you need in the hood.  If all the lumens
from the bulbs could be focused onto the water surface, the lux at the
surface is (total lumens / area of surface in meters^2).  Derating
factors such as reflector efficiency and hood configuration could be

You would also need to take in account the reduction in lumens over
time as the bulbs age.  If you designed your light system for just the
right amount of lumens based on the initial lumen rating, it would be
suboptimal a short time after startup and would never be right unless
all the bulbs were changed at once.  So a factor considering something
like 25% new bulbs, 25% almost new bulbs, 25% getting old bulbs and
25% old bulbs would need to generated. 
Hmmph, maybe "2 watts per gallon" isn't so bad after all ;-).

As for light type, there is no clear data that indicates whether
blue-red or full spectrum light is "better" for growth.  I prefer the
appearance of full spectrum light.  The intensity of bulbs is
expressed in "lumens" which is weighted towards humnan sensitivities.
There are no commerical ratings geared towards plants so we are stuck
with lumens.  The output in lumens of a gro-lux bulb looks poorer than
a full-sepctrum bulb even though the "useful" light energy *may* be
the same (I don't know for sure).  But anything we come up with will
have to deal with lumens with perhaps some factor for the bulb type.