The water cycle: evaporation, water changes, and internal flux
Most hobbyists think of tank water as a fluid you "replace." It is not. The water in your tank is continuously moving — molecules evaporate skyward, minerals stay behind, the filter shuttles the column, plants transpire, fish gills extract gases, and the substrate slowly exchanges with the column above. What governs the long-term chemistry of your tank is not how much water you swap, but the ratio of rates at each node of this hidden circulation. This guide computes those rates explicitly.
1. Evaporation: the silent concentrator
A 60 L open-top tank with an above-water light fixture in a 22 °C, 50%-humidity room loses 0.5–1.5 L per day to evaporation. Over a year, that is one to two tank volumes — but only pure water leaves. Minerals, fertilizers, accumulated salts, and chloride all remain.
Consequences:
- GH and KH drift upward. Topping off with tap water (which already contains minerals) without periodic large water changes can double the hardness in three months.
- Trace salts accumulate. This is the mechanism behind sudden Caridina shrimp deaths in long-running tanks that "haven't changed water in six months." TDS quietly climbs past 600 μS/cm.
- Fertilizer concentration creeps up. The Estimative Index method (Barr, 2005) requires weekly 50% water changes specifically to reset this concentration curve.
Top-off must use RO or distilled water, never untreated tap water. Tap brings in minerals; using it to replace pure evaporation is equivalent to adding hardness to the tank.
2. Water changes: a physical reset
A water change has four functions, all distinct:
| Function | Mechanism |
|---|---|
| Dilute nitrate | NO₃⁻ accumulation rate = nitrogen input ÷ tank volume; dilution is the only physical export (plant uptake is slow) |
| Replenish minerals | Ca²⁺, Mg²⁺, HCO₃⁻ consumed by plants and lost to evaporation are restored from tap |
| Remove allelopathic compounds | Fish skin secretes chemical-signaling molecules that suppress growth at high concentration; water changes flush them |
| Reset pH drift | In long-running acidic-substrate tanks, KH depletion leads to ongoing pH decline; new water restores buffer |
Change volume: aquaculture literature converges on 20–50% per week; tropical ornamental practice settles around 25–30% (Walstad, 2013). Larger changes (>70%) at once shock the steady state and re-suspend substrate organics. Smaller changes (<10%) cannot keep pace with accumulation; nitrate creeps up.
Temperature matching: a difference >2 °C between tank and replacement water triggers a measurable stress response in fish (Bowden, Fish & Shellfish Immunology, 2008). Tropica's operational guideline is ±1 °C.
3. Surface gas exchange: the invisible breathing
The water surface is the primary gas-exchange interface. Lewis & Whitman's two-film theory (1924) gives the rate:
J = K × A × (C_sat − C_water)
- J = flux (mg/h)
- K = mass transfer coefficient (function of temperature and surface turbulence)
- A = interface area
- C_sat = saturation concentration
- C_water = actual concentration
Practical implications:
- A protein/oil film reduces K by 50% or more. Use a surface skimmer or a paper towel pull.
- Surface turbulence increases K by refreshing the boundary layer.
- At equal volume, a wide-shallow tank exchanges gas 2–3× faster than a tall-narrow column of the same volume. This is why "Arowana columns" require strong mechanical aeration while shallow Iwagumi layouts can run low-disturbance.
4. Internal circulation: filter, flow field, substrate
4.1 Filter loop
A canister or HOB removes water from the tank, passes it through media, and returns it. The target turnover rate (TR):
- General community tanks: 4–6× tank volume per hour
- High-stocking tanks: 6–10×
- Planted tanks with CO₂: 5–8× (balancing diffuser residence time)
- Slow-flow species tanks (betta, paradise fish): 3–5×
Too-low TR: ammonia and nitrite peaks don't reach biofilter media fast enough; fish receive prolonged exposure. Too-high TR: plant uprooting, substrate scouring, and reduced diffuser dissolution efficiency.
4.2 In-tank flow field
A return pipe is not just "where water comes back" — it shapes the entire flow geometry of the tank. Tropica's design convention recommends a lily pipe with a return aimed along the front glass, producing a single-loop circulation pattern. When the whole water column completes one large loop every 8–15 minutes, nutrients and gases distribute uniformly and dead zones (where algae prefer to settle) disappear.
4.3 Substrate flux
Substrate is not just decorative. Porous active soils (ADA Aquasoil, Tropica Soil, Fluval Stratum) sustain a slow exchange of water through capillary action and plant-root respiration. Diana Walstad used isotopic tracers in her monograph to show that most plant nitrogen uptake comes from substrate ammonium, not from the water column. This is why a bare-bottom tank with floating plants has a measurably lower nitrogen removal rate than the same plants rooted in active substrate.
Substrates more than 10 cm deep, or undisturbed for years, develop anaerobic sulfide zones (black, sulphide-smelling layers). Disturbing them releases hydrogen sulfide, which is acutely toxic. Routine gentle stirring during replanting or maintenance prevents accumulation.
5. Transpiration: the forgotten output
Submerged plants lose almost no water through their leaves. Emergent and floating plants do — through stomata in their surface tissue. Floating-leaf plants (duckweed, water lettuce, frogbit) can transpire at 50–80% the rate of terrestrial plants (Lai et al., Hydrobiologia, 2013). A surface fully covered in duckweed adds roughly 30% to total tank water loss compared to a bare-surface tank.
6. A worked water-budget example (60 L tank)
| Flow term | Daily volume (L) | Notes |
|---|---|---|
| Evaporation | 0.8 | 22 °C, 50% humidity, half-covered top |
| Water changes | 2.6 | 30% weekly (18 L/7 days), pro-rated |
| Filter throughput | 7,200 | 5× turnover, 24 h |
| Duckweed transpiration | 0.2 | 30% surface coverage |
| Net outflow (evap + transpiration) | 1.0 L/day | Replace with RO water |
7. Operational recommendations
- Buy a TDS meter (~$5–15). Measure both tank water and tap water weekly; the difference tells you the accumulation rate. A weekly difference >50 μS/cm indicates water changes are too small or too infrequent.
- Use a partial lid to reduce evaporation by ~70%, leaving a small gap for gas exchange. Light fixtures may need a heat sink.
- Top off with RO water, not tap. Approximately $0.03/litre at scale.
- Vacuum substrate organics every 2–4 weeks to prevent decomposition products from entering the anaerobic layer.
References
- Lewis, W. K., & Whitman, W. G. (1924). Principles of gas absorption. Industrial & Engineering Chemistry, 16(12), 1215–1220.
- Walstad, D. (2013). Ecology of the Planted Aquarium (3rd ed.). Echinodorus Publishing.
- Bowden, T. J. (2008). Modulation of the immune system of fish by their environment. Fish & Shellfish Immunology, 25(4), 373–383.
- Lai, W. L., Wang, S. Q., Peng, C. L., & Chen, Z. H. (2013). Root features related to plant growth and nutrient removal of 35 wetland plants. Hydrobiologia, 716(1), 33–44.
- Barr, T. (2005). The Estimative Index of Dosing. The Barr Report, online publication.
- Boyd, C. E. (1998). Water Quality in Pond Aquaculture. Auburn University.