The number that matters is not a property of the gas. It is a property of your tank. Carbon dioxide lowers blood pH and reduces how much oxygen the blood can carry, so losses happen while the measured oxygen level still looks normal. The same 30 ppm is unremarkable in a well agitated tank and a real risk in a still, heavily stocked one.
Target bands
There is no single correct target, and any source that gives you one without asking about your tank is guessing on your behalf. Three working bands cover most planted setups:
| Tank | Target | Why |
|---|---|---|
| Low tech | 10 to 20 ppm | Modest light and slow growth do not need more, and the margin to trouble stays wide |
| Standard planted | 20 to 30 ppm | The usual compromise between plant demand and livestock safety |
| High tech | 30 to 35 ppm | Only with strong surface movement and attention paid daily |
Where risk actually rises
Two thresholds are worth holding in mind, and neither is a cliff edge:
- About 40 ppm. Risk climbs noticeably from here in an ordinary tank.
- About 60 ppm. Treat this as acute regardless of how the tank is set up.
These are deliberately not described as lethal figures. Published acute values sit far higher, in the range of 91 to 140 mg/L for bluegill and 235 to 306 for fathead minnow. The real hazard in an aquarium is chronic stress plus the oxygen coupling, not sudden suffocation at 40 ppm, and a source that claims otherwise loses credibility with anyone who has run a tank properly.
What moves the threshold
Three conditions lower the point at which a concentration becomes risky, and they stack:
| Condition | Effect | Reason |
|---|---|---|
| No surface agitation | about 8 ppm lower | Nothing is replenishing oxygen at the surface |
| Heavily stocked | about 5 ppm lower | More demand on the same dissolved oxygen |
| Shrimp or snails present | about 5 ppm lower | Invertebrates show distress before the fish do |
Run all three and a tank that would tolerate 40 ppm is in trouble in the twenties. This is also the reason more surface movement lets experienced keepers push injection further: the limit was never the gas on its own.
Fish hovering at the surface or gasping is a late signal, not an early one. Increase surface agitation immediately, cut injection, and do a partial water change if the behaviour does not settle. Act on the animals rather than on any calculated figure.
The night is the dangerous half
After lights out plants stop consuming carbon dioxide and start producing it, while dissolved oxygen falls to its daily minimum. That combination is the worst point of the 24 hour cycle. Shut the gas off an hour before lights out and run an air stone through the dark period.
The daily pH swing is not the problem
A pH swing of a full unit over the photoperiod is normal and expected in an injected tank, and warnings about it mislead more than they help. What deserves attention is the absolute low point against the tolerance of your species, and whether carbonate hardness is stable. An unstable KH is the signal worth reacting to.
Judge the number in the context of your tank
The app takes surface movement, stocking and invertebrates into account before it calls a value safe, on target or risky, and it flags the night aeration case rather than leaving it to a footnote.
