[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: light attenuation
On Mon, 17 Jul 2000, Wright Huntley wrote:
> > Nevertheless, I contend the *difference* in substrate-level available light
> > between an 18"-deep tank and a 24"-deep tank is essentially insignificant.
To which Ivo Busko replied:
> Sorry if I misunderstood, but you mean the article is wrong ? According
> to the "pure H2O" curve, which is the one relevant for this purpose, a
> 16" layer of pure water should absob 15-20 % of the ligth at the 600-700
> nm band, about 5% at the visible range and nothing at the blue.
Adey and Loveland (1991) gives values for the attenuation of natural PAR
(so, not specific to any particular wavelength) in various natural water
bodies. I've taken the liberty of converting the values from 1/meter to
percent per foot:
Oceanic Water
Sargasso Sea 0.9%
Gulf Stream off Bahamas 2.4
Pacific Ocean, 100 km off Mexico 3.4
Coastal and estuarine waters
Bjornafiord, Norway 4.6%
Gulf and California 5.2
Tasman Sea, Australian coast 5.5
Port Hacking estuary, Australia 11.3
Clyde R. estuary, Australia 21.6
Inland waters
North America
Crater Lake, Oregon 1.8%
L. Ontario, Canada 4.6 - 18
San Vicente reservoir, California 20
Lake Minnetonka, Minnesota 21 - 85
McConaughy reservoir, Nebraska 49
Yankee Hill, reservoir, Nebraska 76
Pawnee reservoir, Nebraska 88
Europe
Esthwaite Water, England 24 - 49%
Loch Croispol, Scotland 45
Loch Uanagan, Scotland 72
Sea of Galilea 15
Sea of Galilea (Peridinium bloom) 101
Africa
Lake Simbi, Kenya 91 - 380%
Lake Tanganyika 4.9
Australia
a) southern tablelands
Corin Dam 27%
Lake Ginninderra 45
Burrinjuck Dam 50
L. Burley Griffin 86
L. George 460
b) Northern Territory (Magela Creek bilabongs)
Mudginberri 38%
Leichardt 51
Georgetown 260
c) Tasmania (lakes)
Perry 6.4%
Ladies Tarn 13
Risdon Brook 18
Barrington 34
Gordon 66
Values of a few percent per foot can be used the way they read. Higher
numbers are a little problem. Math-impaired may want to skip down a ways
to where the English starts again.
The coefficients are for exponential attenuation so the coefficient always
overstates the amount of attenuation you would actually get as light
passed through a foot of water.
The equation is:
I/I0 = exp(-Kd*D/100)
Kd is the attenuation coefficient given above
D is the depth in feet
I0 is the incident light intensity
I is the light intensity at depth D.
<English here I think>
I think that means that light attenuation over short distances in pure
water would be insignificant -- less than 1% of the light per foot and
less than 1.3% in 16 inches. The amount of attenuation in an aquarium
will be pretty variable but like Wright, I expect that our tanks are
usually on the clear side of natural water. Let's say, something like 5%
per foot on a good day. In the article I circulated last fall, Davis and
Brinson (1980) cited a value that converts to 6% per foot for clear lakes
like Lake Tahoe. At 6%, the light penetrating to 24" inches would be
almost as bright (97% as bright) as the light at 18". There's leeway for
that number to be a little higher and much lower. Also, reflection and
refraction in a tank would effect the vertical distribution of light.
While this is all very nice I think it neglects the most important factor,
which is the plants themselves. The plants adsorb red and blue light and
reflect green light, so the light in a heavily planted tank often looks
pretty green. If you have a deep tank or you let the plants close in at
the surface then the light can get *really* green. Our eyes are sensitive
to green light, so it still looks only a little darker than the unfiltered
light, but to plants that green light is nearly useless.
Roger Miller
Adey, Walter H. and Karen Loveland, 1991. Dynamic Aquaria. Academic
Press, San Diego, Ca. 643 pp.
Davis, Graham J. and Mark M. Brinson, 1980. Responses of vascular plant
communities to environmental change. US Fish and Wildlife Service
FWS/OBS-79/33.