Salinity — measurement, units and calibration

Salinity — measurement, units and calibration

Salinity is the single parameter in a reef aquarium that most affects every other measurement. ICP laboratory reference values are calibrated to 35 ppt. Alkalinity, calcium and magnesium consumption rates are calculated for a specific salinity range. Ozone reacts differently at different salinities. Temperature affects density and thereby the salinity reading.

Incorrect salinity does not just deviate from the target — it distorts every other measurement, because all reference values are tied to it. Before other parameters can be meaningfully interpreted, salinity must be known with certainty.

1. Three units — one truth

Salinity is expressed in the hobby in three different ways, each measuring different physical properties but all referring to the same thing.

ppt (parts per thousand, ‰) is a mass ratio: grams of salt per kilogram of solution. Natural seawater salinity is 35 ppt — that means 35 grams of salt in 965 grams of water, giving a total weight of 1,000 grams. ppt and PSU (Practical Salinity Units) are practically synonymous for hobbyist purposes: both refer to the same target value of 35.

Specific gravity (SG) is the ratio of water density to the density of pure water at the same temperature. Here lies a problem: density is temperature-dependent. SG 1.025 does not mean the same thing at 25 °C and 10 °C. Older swing-arm hydrometers are often calibrated to 15 °C, which is why they give a wrong reading in a tank at 25–26 °C. The reference temperature in most hobbyist salinity tables is 25 °C — this always needs attention.

Conductivity (mS/cm) is what electrical conductivity meters actually measure. At 35 ppt in seawater, conductivity is 53 mS/cm at 25 °C. Conductivity meters have temperature compensation capability — they calculate salinity from conductivity and temperature, giving a reliable reading regardless of tank temperature as long as calibration has been done correctly.

Conversion table at 25 °C:

An important practical consequence: if the tank is kept at 25 °C and salinity is 35 ppt, SG is 1.0264 — not 1.025, which is often seen in hobby literature. SG 1.025 corresponds to approximately 34 ppt at 25 °C. The difference is not large, but it is systematic, and it explains why many tanks inadvertently run slightly below their target.

2. Measurement methods

Optical refractometer

Optical refractometers are the most common hobbyist salinity meter and a cost-effective choice. The device measures the refractive index of light — that is what it actually measures, even though the scale shows ppt or SG.

Strengths: inexpensive, no batteries required, fast for field use, long-lasting.

Limitations: requires the correct calibration standard (not just RO/DI water — more on this in section 3), the sample must be taken at tank temperature or allowed to equilibrate, and there is subjectivity in reading the prism line.

A critical detail: a refractometer “calibrated for seawater” does not mean it is calibrated correctly. It means the scale has been converted correctly — but if the device has an offset error in calibration, all readings are systematically wrong. See section 3.

Brix-scale refractometers are designed for the sugar industry — they must not be used for seawater without a conversion table. The same applies to clinical urine and protein refractometers whose scale reads ppt or SG for protein solution, not seawater.

Digital refractometer

Digital refractometers (e.g. Milwaukee MA887) work on the same principle as optical ones but give a numerical reading instead of a prism line. They reduce reader subjectivity and are easier to use in poor lighting.

Calibration still needs to be done correctly — the same RO/DI myth applies to digital ones too. A good digital refractometer is reliable and usable for hobbyists.

Conductivity meter

Conductivity meters are the most accurate and repeatable method for salinity measurement — they measure a physically different quantity from refractometers and are not as sensitive to calibration error. When conductivity is 53 mS/cm and temperature is 25 °C, salinity is 35 ppt — this is a physically precise correlation.

Limitation: a quality conductivity meter is more expensive than an optical refractometer. Cheap devices’ electrodes wear quickly. The probe must be stored moist.

The salinity sensor in Profilux, Neptune Apex and GHL controllers is a conductivity meter — the reading these give is generally reliable as long as calibration has been done with a standard solution.

Hydrometer (swing-arm)

The traditional swing-arm hydrometer measures density by buoyancy. It is inexpensive and simple, but its accuracy is the weakest of all: air bubbles interfere with measurement, temperature correction is inadequate, and mechanical wear affects the result over time. Randy Holmes-Farley demonstrates with test data that swing-arm hydrometers give systematically low readings — often 1–3 ppt below actual. A swing-arm hydrometer is acceptable as a backup tool or in an emergency, but is not suitable for precise long-term monitoring.

Glass spindle hydrometer — a high-quality alternative

A traditional quality glass spindle hydrometer is a different device from a cheap swing-arm plastic hydrometer. The physical measurement principle is the same — buoyancy — but the glass spindle’s accuracy is in a completely different class: hand-blown glass, high resolution and industrial calibration make it a reliable instrument that requires no recalibration.

The Tropic Marin High Precision Hydrometer is the best-known quality glass spindle among hobbyists. Specifications:

• Measurement range: SG 1.021–1.031
• Resolution: 0.0001
• Maximum measurement error: ±0.001 (25 °C / 77 °F)
• Calibration temperature: 25 °C
• Calibration: done at the factory, does not require renewal
• Construction: hand-blown glass, measuring cylinder included
• The measuring cylinder has a scale marking where the SG range optimal for reef use is highlighted as a visual reference

Usage principle: the spindle is lowered into the measuring cylinder or directly into a sufficiently deep, flow-free section of the sump. The spindle floats freely and the reading is taken at the waterline. The sample temperature should be as close to 25 °C as possible for an accurate result — temperature change affects density and thereby the spindle level. The manufacturer provides a temperature correction table for measurements at other temperatures.

The Tropic Marin device also serves as an excellent reference measuring device for calibrating other instruments — it is a physical standard whose accuracy does not depend on reagents or electronics.

Glass spindle hydrometers of equivalent quality are also manufactured by other manufacturers for laboratory and industrial use. The key criteria are that the device is calibrated to 25 °C, resolution is 0.0001 or better, and maximum measurement error is documented. Cheap imported glass spindles without documented specifications must not be equated with quality laboratory-grade instruments.

3. The freshwater calibration myth

This is one of the most common misconceptions in reef keeping: “Calibrate the refractometer to zero with RO/DI water and you’re ready.”

This is incorrect — and the error can be significant.

A refractometer measures the refractive index of light. The refractive index of pure water is 1.33300. The refractive index of 35 ppt seawater is 1.33940. The difference between these is what the refractometer measures.

The problem: many refractometers sold to hobbyists are factory calibrated so that they show the correct value only over a specific refractive index range or at a specific temperature. If the device has an offset error — meaning the prism line is systematically wrong — RO/DI calibration corrects the zero, but does not correct the error in the seawater range. A refractometer’s error is typically linear: if it shows zero correctly but shows 35 ppt as 32 ppt, the error is the same across the entire scale.

Holmes-Farley demonstrates this with examples: a refractometer calibrated with RO/DI water can give readings up to 3–4 ppt below actual. This means a hobbyist believes they are keeping the tank at 35 ppt, while in reality salinity is 31–32 ppt.

The correct calibration process:

1. First calibrate with RO/DI water (ensures zero is zero)
2. Verify calibration with a seawater standard of known salinity (35 ppt)
3. If the device shows the wrong value on the standard, adjust with the calibration screw

Calibrating with RO/DI alone skips step 2 — and that is precisely the step where an offset error would be revealed.

4. Calibration standards

Commercial standards

Commercial calibration standards are the simplest way to verify refractometer accuracy. An important note: not all commercial standards are suitable for refractometers, even if they are intended for salinity measurement.

Example: the American Marine Pinpoint 53 mS/cm calibration solution has a conductivity of 53 mS/cm — which corresponds to the conductivity of 35 ppt seawater. But its refractive index does not necessarily correspond to seawater, because KCl solution (used for PSU definition) and seawater are not optically identical. The refractive index of a 53 mS/cm KCl solution is approximately 1.3371, which corresponds to less than 26 ppt seawater — not 35 ppt.

A conductivity standard must therefore not be used for refractometer calibration unless the manufacturer specifically states that the solution has been calibrated for both conductivity and refractive index.

Valid for refractometers:

• Standards explicitly manufactured as seawater (with seawater ion ratios, not just NaCl or KCl)
• Products whose refractive index value (RI) is stated — 35 ppt seawater = RI 1.33940

DIY standard

Holmes-Farley has calculated that a sodium chloride solution with a concentration of 3.65% by weight optically matches 35 ppt seawater. The solution is prepared as follows:

3.65 g NaCl + 96.35 g RO/DI water = 100 g of a standard matching 35 ppt seawater

Scaled up: 36.5 g NaCl + 963.5 g RO/DI water.

The salt must be pure sodium chloride — not iodised table salt or sea salt with other additives. A purely NaCl-based food-grade salt (e.g. pickling salt) works. A scale accurate to 0.1 g is sufficient.

This standard is excellent for checking a refractometer’s offset error. It is not perfect — seawater is not pure NaCl — but from an optical refractive index perspective, the difference is less than 0.1 ppt, which is insignificant at hobbyist level.

5. Temperature correction

Refractometers are generally calibrated to 20 °C. A reef aquarium runs at 25–26 °C. The temperature difference cannot be compensated by an “ATC” function alone (Automatic Temperature Compensation) — ATC only compensates for the change in light angle entering the device’s prism, not the change in the sample’s viscosity or refractive index with temperature.

Practical correction: take a sample from the tank and allow it to equilibrate to room temperature (or the device’s calibration temperature) before measurement. Alternatively, use a calibration standard you know is 35 ppt and verify the device regularly at this temperature.

Conductivity meters compensate for temperature automatically through calculation — they are more reliable than refractometers in this respect.

6. Target range and stability

Natural seawater salinity on tropical reefs is typically 34–36 ppt depending on location. The Red Sea is higher (~41 ppt), coastal areas during the rainy season are lower. Open ocean surface water at tropical latitudes is close to 35 ppt.

Riuttareef recommendation: 35 ppt (SG 1.0264, 25 °C), acceptable daily variation 34.5–35.5 ppt.

More important than the absolute value is stability. Salinity changes primarily through evaporation — evaporation raises salinity, because water evaporates but salt stays. Tank size and openness, lighting heat and ambient air humidity affect evaporation rates. A typical reef tank evaporates 0.5–2% of its water volume per day.

ATO (Automatic Top-Off) is the practical solution for maintaining stable salinity. ATO replenishes evaporated water with RO/DI water automatically, controlled by an optical sensor or float. Without ATO, salinity rises continuously through evaporation — particularly in warm months or under intense lighting.

Manual top-off is possible but requires a daily routine. Top-off is always done with RO/DI water — not saltwater — because evaporation removes water, not salt.

7. Cycling-phase SG and raising to target

The Hovanec protocol cycle is performed at low salinity, SG 1.015 (~20 ppt). Nitrification bacteria activity at this salinity is more than 40% higher than at 35 ppt, and cycling speeds up significantly.

Salinity is raised to target — 35 ppt / SG 1.0264 — only after cycling is complete: both ammonia and nitrite show zero on measurement. Salinity is not raised during the cycling phase.

The raise is done gradually over 1–2 days by adding salt mix or pre-mixed saltwater in small increments. A rapid jump from low to high salinity stresses the bacterial population.

8. Finnish water and salinity measurement

Finnish tap water is extremely soft and nearly ion-free. This does not mean it can be used in a reef aquarium as-is — it means its problems are different from hard water: instead of removing calcium and magnesium (as with hard water), you must remove phosphates, nitrates, chloramines and organic compounds.

RO/DI equipment is mandatory, and the quality of water it produces must be verified with a TDS meter. Below 5 ppm TDS is sufficient — below 1 ppm is the goal. A high TDS reading after RO/DI indicates worn membranes or depleted DI resin.

Calibration of salinity instruments is particularly important precisely because in Finland hobbyists mix the tank from entirely synthetic components — RO/DI water + synthetic salt. In this process, salinity is the only reference point for which there is no “correct” comparison value without a calibrated meter.

References

Peer-reviewed research

• Millero, F. J. (1978). Freezing point of seawater. Eighth Report of the Joint Panel on Oceanographic Tables and Standards. UNESCO Technical Papers in Marine Science, 28, 29–35.
• Millero, F. J. & Poisson, A. (1981). International one-atmosphere equation of state of seawater. Deep-Sea Research, 28A(6), 625–629.

Hobbyist literature and brand documentation

• Holmes-Farley, R. (2010). Refractometers and Salinity Measurement. REEF EDITION. Reef2Reef.com
• Holmes-Farley, R. (2010). Reef Aquarium Salinity: DIY Calibration Standards. Reef2Reef.com
• Hovanec, T. A. & Coshland, J. L. (2022). The Ultimate Cycle. CORAL Magazine, January/February 2022, 46–57.

Books and textbooks

• Delbeek, J. C. & Sprung, J. (1994). The Reef Aquarium, Vol. I. Ricordea Publishing, Miami.
• Borneman, E. H. (2001). Aquarium Corals: Selection, Husbandry, and Natural History. Microcosm, Charlotte.