Table of Contents
The Parameter That Kills Quietly
In a recirculating aquaculture system, ammonia is the parameter that ends operations. Fish excrete it continuously as a metabolic byproduct, and because RAS packs stock at roughly ten times pond density, it accumulates fast.
The toxic form is unionized ammonia (NH₃), not the ammonium ion. Gill damage and reduced growth start at low concentrations — guidelines commonly put the chronic threshold for NH₃-N in the 0.02 mg/L range — and acute mortality follows at higher levels. The number that matters on any given afternoon depends on pH and temperature, which is exactly why a periodic test is a weak instrument.
The Problem With Once-a-Day Testing
Traditional RAS ammonia monitoring means grab samples, colorimetric kits, or a lab run. All three share a flaw: ammonia doesn’t hold still.
Feeding events push ammonia up within two to six hours as fish metabolise protein. Biofilter performance shifts with temperature and organic load. In a facility feeding four or five times a day, TAN traces a saw-tooth curve. A single daily sample catches one point on that curve — often the baseline before feeding, sometimes the peak, and you can’t tell which without more data.
What happens between tests is where the losses occur. A biofilter that has quietly lost nitrification capacity shows up as higher post-feeding peaks for days before fish show symptoms. An operator sampling at 09:00 every morning will see “normal” right up until it isn’t.
Shanghai ChiMay’s ammonia nitrogen sensor measures continuously and needs no reagents, which removes both the blind spot and the consumable-supply problem that comes with colorimetric kits at remote sites.
What Continuous Data Changes
Reading the Feeding Cycle
With a continuous trace, the feeding cycle becomes visible instead of inferred. That supports three practical judgements.
Biofilter capacity. If post-feeding peaks stay elevated for hours rather than returning to baseline, the biofilter is undersized or its nitrification rate has dropped.
Feeding schedule. Peak height and duration tell you whether frequency or portion size is the variable to change — a decision most sites currently make by feel.
Biofilter degradation. A slow rise in peak height over a week or two is the earliest available signal. Media cleaning or backwashing at that point is maintenance; two weeks later it’s an emergency.
Ammonia, Oxygen and Aeration Are One Control Loop
Ammonia and dissolved oxygen can’t be managed separately. Stressed fish consume more oxygen. The biofilter converting ammonia to nitrate consumes oxygen too — nitrification is an aerobic process with a substantial oxygen demand.
Run both parameters continuously through Modbus RTU/TCP and the automation can do what a threshold alarm can’t:
- raise aeration when ammonia climbs, serving fish respiration and nitrification at once
- open a water-exchange valve when ammonia crosses a setpoint
- adjust feed recommendations against the system’s current ammonia-processing capacity
Sensors that feed your AI water model, not just your dashboard. Once ammonia and DO are on the same time base, the data supports models that anticipate a peak rather than react to it.
How the Sensor Works
Shanghai ChiMay’s ammonia nitrogen sensor uses an ion-selective electrode with automatic temperature and pH compensation. It measures ammonium directly and calculates total ammonia nitrogen from the concurrent pH and temperature readings.
That compensation isn’t a convenience feature. It’s the difference between a number and a usable number.
| Parameter | Specification |
|---|---|
| Measurement range | 0-100 mg/L NH₃-N (configurable) |
| Accuracy | ±5% of reading or ±0.1 mg/L |
| Response time (T90) | Under 60 seconds |
| Operating temperature | 0-50 °C |
| Output | Modbus RTU, 4-20 mA |
| Protection | IP68 submersible |
| Calibration interval | 3-6 months, depending on water quality |
Data output is configurable from 30 seconds to 5 minutes per cycle. Reagent-free operation removes the shelf-life, cold-chain and shipping constraints that make kit-based testing awkward at coastal sites far from a lab supply chain.
Temperature, pH and Why Compensation Matters
The share of total ammonia present as NH₃ rises steeply with pH and, less dramatically, with temperature. At pH 8.0 the unionized fraction is roughly 4% at 20 °C and around 7-8% at 30 °C. Push pH to 8.5 and it roughly triples again.
The practical consequence: a TAN reading that is safe at 15 °C in March can be dangerous at 28 °C in July, with the same sensor showing the same TAN. Monitoring ammonia without pH and temperature context produces a number that looks reassuring and isn’t.
The sensor reports the toxic NH₃ fraction alongside TAN, so seasonal temperature swings don’t quietly change what your alarm threshold means.
Watching the Biofilter Directly
The biofilter is where nitrogen is actually dealt with: nitrifiers oxidise ammonia to nitrite (still toxic) and then to nitrate (much less so). The bacteria are sensitive to DO, temperature, pH and organic loading, which is why the biofilter is usually the first thing to go wrong.
Placing sensors at the biofilter inlet and outlet gives a direct performance measure — the drop across the bed is the nitrification rate. When that drop narrows over days, you have warning before fish health is affected, and you have a documented reason to schedule media cleaning or a pH adjustment.
Reagent-free operation matters most here. A large RAS facility monitoring four or five points consumes a lot of test kits; an ISE-based sensor consumes none. IP68 sealing and a 3-6 month calibration interval fit the maintenance rhythm of a commercial farm.
A Deployment Pattern for Shrimp RAS
A land-based shrimp facility producing around 200 tonnes a year typically monitors four points:
| Monitoring point | Purpose | Response |
|---|---|---|
| Culture tank outlet | Exposure to stock | Emergency aeration and water exchange |
| Biofilter inlet | Loading | Track production against feeding schedule |
| Biofilter outlet | Nitrification efficiency | Detect performance decline |
| Sump / recirculation | System-wide level | Validate overall removal |
Each sensor reports via Modbus RTU to a controller managing blowers, oxygen injection and exchange valves. Add DO, pH and temperature and you have the full picture the control logic needs.
Market Context
Mordor Intelligence identifies aquaculture as the fastest-growing end-use in the water and wastewater sensor market, at 8.44% CAGR through 2031, in a market moving from USD 6.76 billion in 2026 to USD 8.88 billion by 2031 at 5.61% CAGR overall. Inside that segment, dissolved oxygen is the fastest-growing parameter at 8.61% CAGR — and ammonia monitoring is its operational partner, since the two drive the same control decisions.
The structural driver is straightforward: as land-based RAS replaces open-net pens to address environmental constraints, the number of facilities that depend on continuous water quality data keeps climbing.
Supply and Support
Shanghai ChiMay holds 5-8 working day standard lead times on ammonia nitrogen sensors from China-based production, with low-MOQ terms for operators building out monitoring point by point. Calibration standards and replacement sensor caps ship through the same channel, so maintenance intervals don’t stretch because a consumable is on back order.
Tier-1 performance, without Tier-1 lead time.
References
- Mordor Intelligence, “Water and Wastewater Sensors — Market Share Analysis, Industry Trends & Statistics, Growth Forecasts (2026–2031),” July 2026. https://www.mordorintelligence.com/industry-reports/water-and-wastewater-sensors-market
- Emerson et al., “Aqueous ammonia equilibrium calculations: effect of pH and temperature,” Journal of the Fisheries Research Board of Canada, 1975 — standard basis for the NH₃/NH₄⁺ fraction tables used in aquaculture.
Author: Shanghai ChiMay Aquaculture Applications Team.
