Identify it by texture, not color — the confusions that matter (black beard vs staghorn, cyanobacteria vs diatoms) look similar in a photo and behave completely differently in your fingers. Then be skeptical about causes: a few are reliable (cyanobacteria means low flow, diatoms mean a new tank), some are plausible, and several of the most-repeated ones are folklore. The levers that work are almost always the same regardless of type: less light, healthier plants, more flow, physical removal.
Identify it by what it does, not what it looks like
Every algae guide identifies by photograph, which is the wrong tool for the job. Photographs are fine for the easy cases and useless for the two mistakes people actually make — confusing black beard with staghorn, and confusing cyanobacteria with diatoms. Those pairs look alike on a screen. They feel completely different in the tank:
| What you're seeing | The physical test | Where it starts | What it is |
|---|---|---|---|
| Brown dust | Wipes off with a fingertip, clouds the water, returns in days | Glass and substrate of a tank under 3 months old | Diatoms |
| Slimy sheet, blue-green to reddish | Peels off in one piece and smells earthy or musty | Substrate surface, dead spots, where flow doesn't reach | Cyanobacteria — a bacterium, not an alga |
| Hard green dots | Won't wipe; needs a blade or a credit card | Glass and the oldest, slowest leaves | Green spot algae |
| Fine green film | Wipes off easily, then returns on a roughly 3-week cycle | Glass, evenly, in an otherwise healthy tank | Green dust algae |
| Green threads | Pulls away in strands; winds onto a toothbrush | Plant edges, hardscape, filter outlets | Hair or thread algae |
| Dark tufts, black to grey | Tough and anchored — won't wipe, won't pull cleanly | Leaf edges, wood, equipment in the strongest flow | Black beard algae (a red alga) |
| Grey-green branching tufts | Also tough, but branched like a tiny antler rather than bushy | Same places as BBA, often alongside it | Staghorn (same family as BBA) |
| The water itself is green | Nothing to touch — it's suspended | Whole tank, worst in sunlight | Green water — see cloudy water |
The two tests that settle most arguments: does it peel off as a sheet (cyanobacteria does, nothing else does), and does it resist a fingertip (black beard and staghorn do, soft green growth does not).
It is a photosynthetic bacterium — the same group that oxygenated the atmosphere — which is why it forms sheets rather than filaments, why it can fix its own nitrogen and so thrives in tanks with nutrients driven too low, and why an antibiotic clears it when nothing touches real algae. "Blue-green algae" is a 200-year-old misnomer that still misleads people into treating it like the green stuff.
What the type actually tells you
Here is where the guides diverge from each other and, quietly, from the evidence. Most assign one confident cause per type. The honest position is that some of those attributions are well supported, some are reasonable inference, and some have been copied between pages for twenty years without anyone checking. Graded:
- Brown diatoms in a tank under three months old. A maturation phase, not a fault. They fade as the tank stabilizes and something starts eating them.
- Cyanobacteria. Low flow plus accumulated organics, almost every time. Find the dead spot — under the substrate surface, behind the rock, the corner the powerhead does not reach.
- Any algae after a light left on 10+ hours, or in direct sun. Light duration and intensity are the most reliable lever in the hobby, and the easiest to get wrong.
- Black beard algae and unstable CO₂. Very widely observed in injected tanks, and the mechanism — stressed plants stop competing — is reasonable. Weaker evidence in tanks with no CO₂ at all, where BBA still appears.
- Green spot algae and low phosphate. Consistently reported by planted keepers, and it is one of the few cases where the fix is to raise a nutrient rather than cut one. Worth trying before anything drastic.
- "Hair algae means excess iron". Repeated everywhere with nothing behind it. Iron is rarely the limiting nutrient in a tank that has any fish in it, and cutting iron mostly starves the plants you were relying on to compete.
- "Diatoms mean silicates in your tap water". Diatoms do use silica, but they appear in new tanks on RO water too. The age of the tank predicts them far better than your water report does.
- "Balance your nitrate-to-phosphate ratio". A unit error stacked on a category error — see below.
The nutrient-ratio trap
The most pseudo-precise piece of algae advice in the hobby is that you should balance your nitrate against your phosphate — usually quoted as the Redfield ratio, 16:1. It is worth taking apart, because it fails in three separate ways and the failures stack.
First, the units. Redfield's 16:1 is a ratio of atoms. Quoting it at a test kit compares moles to milligrams, and correcting that changes the number twice over:
| Version | Basis | Ratio | |
|---|---|---|---|
| As Redfield stated it | atoms of N per atom of P | 16 : 1 | The original, from plankton tissue chemistry. |
| Converted to mass | mg of N per mg of P | 7.24 : 1 | Right conversion, still not what a kit reads. |
| In test-kit units | ppm NO₃ per ppm PO₄ | 10.45 : 1 | The only version you could compare a reading to. |
Computed by this site's engine from the molar masses. Your kit reads whole nitrate and phosphate ions, which carry oxygen as well as the nitrogen and phosphorus — which is why all three numbers are in circulation and why people quoting them at each other rarely mean the same thing.
Second, the arithmetic gives the game away. Apply the corrected ratio to an ordinary planted tank at 10 ppm nitrate and it asks for 0.96 ppm of phosphate — a level every reef chart in existence treats as a disaster and most planted keepers would find alarming. When a theory applied correctly tells you to do something the whole hobby considers wrong, the theory has not transferred.
Third, it is unmeasurable at hobby precision anyway. That same tank at 0.05 ppm phosphate is running a ratio of 200:1. Move the phosphate one step up the color card to 0.10 and it becomes 100:1 — the number halves on a difference you cannot reliably read. You would be tuning noise.
Redfield measured the composition of plankton tissue — what the organisms are made of, averaged across the ocean. It was never a statement about what the surrounding water should contain, and later work found the ratio varies widely between species and conditions. Using it to set targets for tank water is not a mis-application of a good rule; it is applying a description of algae to the thing algae live in.
The levers that work on nearly all of it
The corollary of the grading above is freeing: because type predicts cause only loosely, the response is mostly the same whatever you have. In rough order of how reliably they work:
- Shorten the photoperiod. Six to eight hours is plenty for most tanks, and a timer makes it free. Block direct sunlight entirely — no photoperiod control survives a sunny window.
- Get the plants growing. Healthy fast growers outcompete algae for the same light and nutrients, and they are the only fix that keeps working after you stop paying attention. Struggling plants are the most common root cause of an algae problem in a planted tank.
- Physically remove it every week. Unglamorous and effective — a toothbrush for hair algae, a blade for green spot, siphon out cyanobacteria sheets. You are removing biomass faster than it grows, which is the same principle that beats any bloom.
- Fix the flow. Specifically for cyanobacteria, and it is close to a complete cure on its own — the dead spot is the problem, not the water chemistry.
- Keep nutrients in range, not at zero. Under 20 ppm nitrate for a planted tank, phosphate detectable. Starving the tank starves the plants first.
- Wait, if the tank is new. See below.
New tank algae is a phase, not a fault
Most algae panic happens in the first three months, and most of it resolves without intervention. A new tank has surfaces with nothing established on them, plants still converting to submersed growth, and a filter still finding its population — so the fastest colonizers get there first. Brown diatoms almost always appear and almost always leave. Green dust algae often cycles on and off for a couple of months.
Doing something drastic here — cutting nutrients, tearing down, blacking out repeatedly — usually extends the phase, because it keeps knocking back the plants that would eventually win. Keep the photoperiod short, wipe the glass, do the normal water changes, and let the tank mature.
Spot treatment, honestly
Spot-dosing hydrogen peroxide or a liquid carbon product directly onto black beard algae works — it is the standard planted-tank answer for the types nothing eats. The technique is to turn the pumps off, apply it directly to the affected leaf or hardscape with a syringe, leave it for a few minutes, then restart the flow. Treated BBA turns pink or white over the following days and is then grazed or falls away.
Two honest caveats. We don't publish doses for this — tolerances vary by species, shrimp and fish are considerably less tolerant than plants, and an overdose in a small tank is a serious event rather than a setback. And it treats what you can see, not why it grew: spot treatment without a light or plant-health change buys a few weeks. Use it to reclaim a plant while you fix the cause, not instead of fixing it.
What you'll be told, and what's true
- "Balance your nitrate to phosphate ratio." Three problems: the famous number is atomic and your kit is not, the corrected version demands 0.96 ppm phosphate at 10 ppm nitrate, and the ratio moves by half on one step of a color card. It also describes algae tissue, not water.
- "Cut the nutrients and starve it out." Starves the plants first, because they adapt more slowly than algae does. Cyanobacteria in particular thrives where nitrate has been driven low, since it can fix nitrogen from the air and its competitors cannot.
- "Blue-green algae is algae, treat it like algae." It is a bacterium. That is why it peels off in sheets, why it appears in low-nutrient tanks, and why an antibiotic clears it in days when nothing you would use on real algae touches it. The name is the problem.
- "Hair algae means too much iron." Repeated everywhere, supported by nothing. Iron is rarely limiting in a stocked tank, and cutting it mostly weakens the plants you need competing. Look at the photoperiod first.
Common questions
Is algae actually bad for my tank?
A film of it on the back glass and the hardscape is a sign of a functioning tank, and a completely algae-free aquarium usually means something is being suppressed hard — either the lighting is too low for the plants or the nutrients are. What is worth acting on is algae that is smothering plants, coating the front glass faster than you can clean it, or spreading week on week. Steady and cosmetic is not a problem to solve.
Will more plants stop algae?
Genuinely yes, and it is the most durable fix available — but only if the plants are actually growing. Fast growers like hornwort, water sprite and floating plants compete for exactly the light and nutrients the algae wants, and they are far better at it. A tank full of struggling plants is worse than a tank with none, because you have added competitors that lost.
Should I cut nutrients to starve the algae out?
Rarely, and it is the most common way people make things worse. Driving nitrate and phosphate toward zero starves the plants first — they are slower to adapt than algae — and you end up with a tank that has no competition left in it. Keep nitrate in the normal band (under 20 ppm for a planted tank), keep phosphate detectable, and take the light away instead.
Do algae eaters fix an algae problem?
They manage it; they do not fix it. Amano shrimp, nerite snails, otocinclus and Siamese algae eaters each specialize — nerites are excellent on green spot, amanos on soft filamentous growth, and almost nothing eats black beard or cyanobacteria. Add them because you want the animals, size them to the tank, and treat the grazing as a bonus rather than a treatment.