How do you calculate oxygen requirements in aquaculture?

In RAS aquaculture, you calculate oxygen requirements by turning every major oxygen “sink” into kg O₂ per hour, then sizing your oxygenation/aeration equipment so its actual oxygen transfer rate (AOTR) exceeds that demand with a safety margin.

A practical sizing equation is:

Required O₂ supply (kg/h) = (Fish O₂ use + Biofilter nitrification O₂ use + Other system O₂ use) × Safety factor
Then: choose oxygenation so AOTR (kg/h) ≥ Required O₂ supply (kg/h)

For RAS, two sinks dominate:

1. Fish respiration (changes with biomass, temperature, activity, and feeding).
2. Nitrification in the biofilter: a widely used design stoichiometry is 4.57 g O₂ per g NH₄-N oxidized (i.e., per gram of “ammonium-N” converted to nitrate).

RAS is unforgiving. Water volume is small. Biomass is dense. If oxygen delivery falls behind demand, the system does not give you much time.

Most oxygen calculations fail for one simple reason: people calculate fish demand, but they forget the rest. In RAS, the biofilter is a constant oxygen consumer. Feeding creates oxygen spikes. Water chemistry and temperature reduce oxygen transfer. And “rated” oxygen transfer from vendor brochures is often not what you get on site.

Define the target: DO setpoint and the “oxygen budget” view

The goal is not “high DO.” The goal is stable DO above your species and system needs, with enough reserve to handle feeding peaks and power glitches.

General guidance used across aquaculture is often >5–6 mg/L DO for many cultured species, with higher needs for coldwater fish like trout.

Those numbers are not magic. They are a risk boundary. In RAS, you also need DO high enough to keep the biofilter happy. Nitrifying bacteria are oxygen-limited under low DO.

Think in two layers:

• Concentration layer (mg/L DO): what the probes read.
• Mass-flow layer (kg O₂/h): what your equipment must add to offset consumption.

Sizing oxygen supply is a mass-flow problem.

Convert your RAS into kg O₂ per hour

Start with a simple budget:

Total oxygen demand (TOD)

TOD = Fish respiration demand

• Nitrification demand (biofilter)
• Other microbial/oxidation demand (solids, degassing units, sump, etc.)
• Operational margin

You do not need perfection. You need a number that keeps you safe.

Fish respiration demand: the baseline you can’t ignore

Step 3.1 — Calculate biomass

Biomass (kg) = number of animals × average weight (kg)

Example: 30,000 fish × 0.2 kg = 6,000 kg biomass.

Step 3.2 — Choose a specific oxygen consumption rate

You will see oxygen consumption reported many ways (mg O₂/kg/h, mg O₂/kg/min). Use mg O₂ per kg fish per hour and convert to kg/h.

A simple working approach:

Fish O₂ demand (kg/h) = Biomass (kg) × OCR (mg/kg/h) ÷ 1,000,000

The hard part is OCR. It changes with:

• Temperature (warmer water drives higher metabolic demand).
• Species and size (small fish often consume more per kg).
• Activity and stress (grading, crowding, poor CO₂ stripping).
• Feeding (this matters more than many people think).

Step 3.3 — Add the feeding “oxygen spike” (SDA)

After a meal, fish oxygen consumption rises due to the cost of digestion and assimilation. This is commonly discussed as specific dynamic action (SDA). Controlled studies show that oxygen consumption can peak after feeding, and that feeding can take a large share of aerobic capacity.

For oxygen sizing, you do not need to model SDA physiology in detail. You need to recognize the pattern:

• Your highest risk window is often after feeding, especially in warm water.
• If you only size to “average” respiration, you will see DO sag after meals.

Practical rule: treat feeding periods as a separate operating mode and size oxygen for that mode (or add an automatic control strategy that ramps oxygenation up during and after feeding).

Biofilter nitrification demand: the hidden oxygen bill in RAS

If you run a proper RAS, nitrification is almost always present.

A standard stoichiometric design value is:

O₂ required for nitrification ≈ 4.57 g O₂ per g NH₄–N oxidized 1

This value is widely used in environmental engineering design manuals and is a solid planning number.

Step 4.1 — Estimate your daily TAN production or removal

You can estimate TAN load from:

• measured TAN at biofilter inlet/outlet and flow
• feed rate and feed protein (common in practice, but variable)
• historical facility data

If you have real measurements, use them. If you do not, start with a conservative estimate and refine once you collect TAN data.

Step 4.2 — Convert TAN oxidation to oxygen demand

Nitrification O₂ (kg/day) = TAN oxidized (kg NH₄–N/day) × 4.57 (kg O₂/kg NH₄–N) 

Then convert to kg/h: kg/h = kg/day ÷ 24

This biofilter oxygen demand is often steady. That is good and bad. Good because it is predictable. Bad because it never turns off.

Oxygen transfer: why “SOTR” is not what your fish get

Once you have demand (kg/h), you must translate it into equipment.

Most oxygenation and aeration devices are sold with a clean-water performance number such as SOTR (Standard Oxygen Transfer Rate). SOTR is a standardized way to compare devices under defined conditions.

Your RAS is not standard. Real water has:

• different temperature
• different salinity
• different barometric pressure (altitude, weather)
• surfactants and organics that reduce transfer efficiency

Older aquaculture research on aerators discusses correction factors for real culture water versus clean water. 
Engineering guidance in aeration design also uses correction factors (alpha, temperature, pressure) to adjust standard transfer to actual transfer. 

So you should think like this:

AOTR = SOTR × (site and water correction factors)

You do not need to publish a perfect correction model in your blog post. But you should teach readers one habit:

Size oxygen delivery using AOTR, not brochure SOTR.

A worked example (simplified but realistic)

This is a template example. Replace the numbers with your data.

System

• RAS fish biomass: 10,000 kg
• Water temperature: warm (feeding active)
• Estimated fish OCR during non-feeding: 300 mg O₂/kg/h (placeholder)
• Feeding mode OCR multiplier: 1.5× (to represent SDA peak)
• TAN oxidized by biofilter: 2.0 kg NH₄–N/day
• Safety factor: 1.25 (25%)

Step A — Fish oxygen demand (baseline)

Fish O₂ = 10,000 kg × 300 mg/kg/h ÷ 1,000,000
Fish O₂ = 3.0 kg O₂/h

Step B — Fish oxygen demand (feeding peak)

Fish O₂ peak = 3.0 × 1.5 = 4.5 kg O₂/h

Step C — Nitrification oxygen demand

Biofilter O₂ = 2.0 kg NH₄–N/day × 4.57 = 9.14 kg O₂/day 
Per hour: 9.14 ÷ 24 = 0.38 kg O₂/h

Step D — Add a “other” bucket (solids + unknowns)

If you have no data, add a conservative placeholder, say 0.3 kg O₂/h (you will refine later).

Step E — Total demand

Normal TOD = 3.0 + 0.38 + 0.3 = 3.68 kg O₂/h
Feeding TOD = 4.5 + 0.38 + 0.3 = 5.18 kg O₂/h

Step F — Apply safety factor

Required supply (feeding mode) = 5.18 × 1.25 = 6.48 kg O₂/h

Now you have an equipment target: your oxygenation system must reliably deliver ≥ 6.5 kg O₂/h AOTR during the worst expected operating window.

That number tells you whether you need:

• pure oxygen injection (cone, LHO, oxygenation tower)
• heavy-duty diffused aeration
• backup oxygen banks for outages
• control logic tied to DO probes

Monitoring and control: how to stop guessing

In RAS, oxygen is not only a design problem. It is a control problem.

A simple but strong setup:

• DO sensors in culture tanks and after oxygenation
• alarms for low DO and sensor failure
• automatic oxygen valve control or blower VFD tied to DO
• feeding schedule linked to oxygen ramp-up

Even basic extension guidance emphasizes frequent DO monitoring, and RAS generally needs closer attention than ponds. 

If you want to write a blog post that ranks, this is where you beat thin content. Most pages stop at “keep DO above X.” You show how to keep it stable during feeding, nitrification, and daily operations.

Common mistakes that break oxygen calculations in RAS

Mistake 1: Using only fish biomass and ignoring nitrification

Biofilters can consume a meaningful share of oxygen, and they consume it continuously. The 4.57 g O₂ per g NH₄–N rule gives you a clean way to budget it. 

Mistake 2: Sizing to average load, not peak load

Feeding pushes oxygen demand up. SDA is real and observed in fish oxygen uptake after meals. 
Your oxygen system must survive the peak.

Mistake 3: Treating SOTR as real performance

Standard ratings are useful for comparison, but actual transfer depends on temperature, pressure, and water quality. Use correction thinking and keep margin. 

Mistake 4: Forgetting what DO probes don’t tell you

DO can look fine in one tank and bad in another. Poor mixing, short-circuiting flow, or sensor placement can hide a problem. Always pair DO readings with flow understanding.

A simple checklist you can paste into your SOP

1. Calculate biomass (kg).
2. Estimate fish OCR for normal and feeding periods.
3. Estimate TAN oxidized (kg NH₄–N/day).
4. Compute nitrification oxygen using 4.57×. 
5. Add “other” oxygen uses (solids, unknowns).
6. Pick a safety factor (start 1.2–1.3 if you lack data).
7. Size oxygenation so AOTR ≥ demand in feeding mode.
8. Validate with real DO logs. Update OCR and TAN estimates monthly.

FAQ

Does nitrification stop if DO is low?
Nitrification slows under low oxygen. Guidance notes DO needs to be maintained to sustain nitrification. 

Is 5 mg/L DO enough for every RAS?
No. Many warmwater systems target around that range, but species and management matter. Coldwater fish often need higher DO for best performance. 

Can you “over-oxygenate”?
You can create gas supersaturation problems if you inject oxygen without proper degassing and control. This is more of a system design and monitoring issue than a reason to run low DO.