Water-change math explained: why 2 × 50% ≠ 100%
Here is a question that catches almost everyone: if you do two 50% water changes, how much nitrate have you removed? The intuitive answer — 100% — is wrong. The first change removes half the nitrate. The second removes half of what is left. You finish at 25% of where you started. This one piece of arithmetic explains most water-change confusion, and it is worth internalizing before you design a schedule.
Key takeaways
- Each change multiplies the remaining nitrate by (1 − p). Percentages compound; they never add.
- 2 × 50% → 25% of the original remains (75% removed). 4 × 25% → 31.6% remains (68.4% removed).
- Many small changes move more total water but remove less nitrate than fewer large ones.
- With nitrate in your tap water, dilution floors out at the tap value — more changes stop helping.
The model, exactly
Drain a fraction p of a well-mixed tank and refill with nitrate-free water. New concentration = old × (1 − p). Repeat n times and the multiplier is (1 − p)n. That superscript is where intuition breaks: 0.5² is 0.25, so two half-tanks-out leave a quarter. Ten 10% changes leave 0.910 ≈ 34.9% — you carried roughly one full tank of new water over a fortnight and still kept a third of the nitrate. Meanwhile three 30% changes (0.7³ ≈ 34.3%) shift almost the same fraction with three buckets instead of ten.
The general version with tap nitrate Ctap: Cn = C0·(1−p)n + Ctap·(1 − (1−p)n). Check it by hand once and you will never trust an additive intuition again: start at 80 ppm, 50% changes, tap at 10 ppm. Change one: 80×0.5 + 10×0.5 = 45. Change two: 45×0.5 + 10×0.5 = 27.5 — and no number of further halvings gets you below 10.
Worked comparisons
| Programme | Water moved | Nitrate remaining (0 ppm tap) |
|---|---|---|
| 1 × 50% | 50% | 50% |
| 2 × 50% | 100% (over two sessions) | 25% |
| 4 × 25% (weekly, a month) | 100% | 31.6% |
| 10 × 10% | 100% | 34.9% |
| 3 × 30% | 90% | 34.3% |
The pattern: for the same total water moved, fewer-and-bigger beats more-and-smaller on nitrate. That is not an argument against small frequent changes — stability, fish stress and convenience all matter — but it is the reason "I change a bit every day" can coexist with "my nitrate never drops". Our water change calculator runs this table for your own numbers, including the one-shot percentage needed to hit a target.
What the guides actually recommend
Aquarium Co-Op's water-change article states the arithmetic plainly — "a 30% water change reduces nitrates by 30%" — and their nitrate guidance makes 50 ppm the trigger for a 50% change, repeated every four days if needed, until nitrate is at most 25 ppm. That is exactly the compound-removal logic above: one 50% change from 50 ppm lands at 25 ppm, tap permitting. In a real stocked tank, nitrate also climbs continuously between changes as the end product of the nitrogen cycle, so the honest framing is a sawtooth you steer, not a level you erase.
Why "I'll just do 100%" isn't the answer
The algebra tempts you toward one total change, but water changes are a stress event, not just a dilution: temperature and chemistry shock fish, and you never want to strip a tank to zero. Large changes are fine — even 80–90% is routine for some keepers with matched, dechlorinated water — but the point of the math is precision, not maximalism. Know your real tank volume first (volume calculator), your tap nitrate second, and your target third. Then the schedule designs itself.
Sources
- Aquarium Co-Op — How often do you have to change water?: "30% removes 30%" framing (checked 2026-11-16).
- Aquarium Co-Op — Is nitrate good or bad?: 50 ppm trigger, 50% changes, ≤25 ppm target (checked 2026-11-16).
- Dilution algebra: geometric series, Cn = C0·(1−p)n + Ctap·(1−(1−p)n).
Heuristic, not gospel
The model assumes perfect mixing between changes and ignores ongoing nitrate production, plant uptake and substrate release. It is exact for what it models — always validate against your own test kit, and treat fish behaviour as the override on any schedule the math produces.