Key takeaways
- PAR measured at the substrate is the only intensity figure that predicts growth, and it can be 80 percent below the number printed on the fixture.
- Kelvin decides how the tank looks to you and has very little effect on plant growth anywhere between 5000K and 10000K.
- Six to eight hours suits low tech tanks and eight to ten suits injected tanks, with anything past ten hours mostly feeding algae.
- Watts per gallon stopped working the moment LEDs broke the link between power draw and usable light output.
- Pick the plants first, then buy a fixture that delivers their PAR band at your measured water depth.
A customer dropped off a 120 litre (32 gallon) planted tank last spring with a complaint that had nothing to do with broken equipment. The fixture worked, the timer worked, and the carpet was still rotting from the base up.
A meter answered it in about ninety seconds. Just under the surface the light read close to 180 PAR. At the substrate, 50 cm down through slightly tannin stained water, it read 26.
That gap is the whole subject. Most lighting arguments in this hobby are two people quoting a number measured in one place at someone whose plants live somewhere else.
Three numbers describe a light, and only three
For our purposes a fixture is fully described by how much usable light it delivers, what colour that light appears, and how many hours it runs. PAR, Kelvin, photoperiod.
Everything else printed on the box is marketing or engineering trivia. Lumens weight brightness to human eye sensitivity, which peaks in green, and green is the waveband plants reflect rather than absorb.
Buy on PAR at your substrate depth, pick Kelvin for how you want the tank to look, and set photoperiod last. With those three settled, a struggling planted tank becomes a nutrient or carbon problem instead of a lighting one.
PAR is the only intensity figure worth arguing about
PAR is photosynthetically active radiation, the band between roughly 400 and 700 nanometres that plants can use for photosynthesis. It is counted in micromoles of photons per square metre per second.
The number means nothing without a location attached. A fixture advertised at 150 PAR is quoting a reading taken at a distance the manufacturer chose, often 15 cm below the lens in open air.
Your plants get whatever survives the air gap, the surface, and 40 to 60 cm of water. Where those readings come from and how far they can be trusted is worked through in the numbers plants actually respond to at leaf height.
What happens to light on the way down
Intensity drops with distance in air, then drops again to absorption and scattering in water. Tannins from driftwood, fine suspended particles, and the green tint of a bloom all steepen the loss.
The figures below are typical of a mid range LED bar over reasonably clear water. Optics vary a great deal between fixtures, so read the shape of the curve rather than the exact percentages.
| Depth below fixture | Share of rated PAR | What it supports |
|---|---|---|
| 15 cm | 100 percent | High light, carbon injection required |
| 30 cm | 45 to 60 percent | Medium light stem plants |
| 45 cm | 25 to 35 percent | Low to medium |
| 60 cm | 12 to 20 percent | Low light species only |
Doubling the distance does not halve the light, it roughly quarters it. A 5 cm change in hanging height can move a tank out of one plant category and into another.
Kelvin changes the room, not the growth
Colour temperature in Kelvin describes what a light looks like, not how strong it is. A 5000K lamp reads warm and yellow, 6500K reads neutral daylight, 10000K reads crisp and faintly blue.
Plants grow under all of them. Chlorophyll absorbs hardest in the blue around 430 nm and the red around 660 nm, and every white LED in the 5000K to 10000K band carries enough of both to run photosynthesis at a useful rate.
What Kelvin decides is whether your cardinal tetras look red or grey, and whether green leaves read as fresh or olive. That trade is picked apart in what each colour temperature does to fish and leaf colour.
Photoperiod is a lever, not a setting
Six to eight hours suits most low tech tanks. Eight to ten is normal on a tank with injected carbon dioxide, strong light, and a fertiliser routine that keeps pace.
Longer is not better. Past about ten hours you are mostly feeding whatever grows fastest, and in a tank without injection that is rarely your plants. Hour by hour figures for each tank type are set out in the daily run times that suit each light level.
If the glass greens within four or five days of a clean, shorten the day before changing anything else. Thirty minutes off the end is a real adjustment and it costs nothing to test.
Why watts per gallon stopped meaning anything
The old rule of 2 to 3 watts per gallon came from an era of T12 and T8 fluorescent tubes, where every tube of a given length turned electricity into light at roughly the same rate.
LEDs broke that relationship. A 30 watt LED bar with tight optics over a shallow tank can put more usable light on the substrate than a 90 watt fluorescent hood over the same footprint.
Wattage now tells you what the fixture costs to run and very little else. The history of that rule and what replaced it is covered in why the old wattage formula no longer predicts anything.
Spread is the number nobody prints
Two fixtures can read identical PAR directly under the centre of the lens and behave completely differently at the front corners. Narrow optics concentrate output into a hotspot and leave the ends of a tank in shade.
On a 90 cm tank I expect the corners to land somewhere near 60 to 70 percent of the centre reading. Below roughly half, plants at the ends stretch toward the middle and carpets thin out against the front glass.
A single narrow pendant over a long tank is the usual cause. Two smaller bars spaced along the length beat one bright unit parked in the middle almost every time.
Choosing a fixture by tank depth
Depth is the first question, ahead of volume. A 60 cm cube and a 60 cm long shallow tank hold similar water and need completely different fixtures.
| Water depth | Fixture class | Realistic goal |
|---|---|---|
| Under 30 cm | Budget LED bar, around $25 to $50 | Carpets without much effort |
| 30 to 45 cm | Mid range planted LED, around $120 to $250 | Medium light, injection optional |
| 45 to 60 cm | High output bar or pendant, around $250 to $450 | High light, injection not optional |
| Over 60 cm | Pendant on a hanging kit | Low light species, or a much larger budget |
Deep tanks punish every mistake twice. Anything over 60 cm from rim to substrate needs the fixture chosen before the plants are bought, not after.
When lighting goes wrong
Three failures account for most of what reaches my bench. Too much light over too long a day, a fixture mounted at the wrong height, and an LED channel that quietly stopped working.
Green spot algae on the glass and black brush algae on slow growing leaves both point at light outrunning carbon and nutrients. That chain is traced properly in the algae that begins as a photoperiod problem.
A dead channel is sneakier. If a tank that ran clean for two years starts leaning green or blue while the plants pale, photograph it and compare against an old picture of the same scape.
Never run a planted fixture at full output on day one of a new tank. Start at six hours and roughly 50 percent, then climb over four to six weeks while the filter and the plants establish.
The 120 litre tank from the opening got a hanging kit, and the fixture came down from 45 cm above the rim to 18 cm. Substrate PAR went from 26 to 58, the photoperiod dropped from eleven hours to eight, and the carpet that had been melting rooted properly inside three weeks.
The order to spend money in
Choose the plants first, then buy light to suit them. Picking a fixture and hoping something will grow under it is how people end up with three dead carpets and a drawer full of fertiliser.
Species by light demand, with the PAR each one wants at substrate level, are laid out in the demand table for common aquarium species. Match that against the depth table above and the shopping decision makes itself.
- Measure water depth from rim to substrate in centimetres
- Decide your hardest plant, not your easiest
- Buy a fixture that hits that plant’s PAR at that depth
- Start at six hours and climb in 30 minute steps
- Add carbon dioxide before adding light if growth stalls
A quantum meter such as the Apogee MQ-510, an underwater sensor running somewhere around $400 to $500, ends these arguments permanently. Most hobbyists borrow one from a club rather than buy, and a single afternoon of readings usually lasts the life of the tank.
Apogee publish their sensor specifications openly at Apogee Instruments, and Tropica list a light requirement against every species they sell at Tropica. Between those two references and a tape measure, most lighting questions answer themselves.
Lighting is the one part of a planted tank you cannot correct later with a water change. Get the figure at the substrate right and everything downstream becomes easier.
Questions readers ask about this
Watch the plants rather than the fixture. Stems with long bare gaps between leaves and pale new growth mean the substrate is receiving under about 25 PAR. Compact growth with algae on older leaves means the opposite. A borrowed quantum meter settles it in an afternoon, and most aquarium clubs own one.
It will grow plants, and it is a poor idea. Household and workshop fixtures are not sealed against splash or condensation, which is a genuine shock risk over an open tank. They also lack ramping and colour control, and their output is aimed at human vision rather than at chlorophyll absorption peaks.
For a tank 30 to 45 cm deep, somewhere between 15 and 30 cm above the rim is a sensible working range. Higher gives wider coverage and lower substrate PAR, closer gives a bright centre and dark corners. A hanging kit at around $20 to $40 makes this adjustable rather than permanent.
Only up to the point where the cheap one delivers the PAR your plants need. What money buys past that is control, spread, colour rendering, and build quality. A $40 bar over a 25 cm deep tank can outperform a $300 pendant hung badly over a 60 cm one.
