When we brought out the D-D H2Ocean Pro+ reef salt a few years ago, we were adamant that we wanted to print the exact levels and tolerances for the main important elements that we would expect to find in our salt on every bucket; information that should be readily available to allow an informed choice for every good reef salt.

Of course once we gave people levels to test against we received a steady stream of emails and threads from individuals who were measuring readings outside these guaranteed parameters. The main reason that we found for this discrepancy, other than general mixing and test kit errors, was due to low specific gravity, i.e. not enough salt in their water.

Further investigation brought to light a number of anomalies and misunderstandings regarding the whole issue of S.G. and S.G. measurement and has prompted me to write this article which tries to illustrate some potential problems when using specific gravity as a unit of measurement without a full understanding of what is actually being measured.

Most people, if asked, would probably tell you that their reef aquarium is running at an S.G. (Specific Gravity) of 1.025. Those with greater understanding may go into further detail and tell you that the S.G. is 1.025 at 25C, but where does this figure come from, and is it the best way to measure or express the amount of salt in your aquarium, is it even the correct level to be running at?

The salinity of the world’s oceans is not constant and varies substantially from one location to the next with concentrations of just 10-15ppt in the Baltic Sea to around 40ppt in the Red Sea.

But stop a minute!! Why are we talking about Salinity now instead of S.G. and what is the difference??


Salinity is a true measurement of the concentration of salt in the ocean and is calculated as the total weight of ‘dry’ salt dissolved in a total of 1000 weight units of water or parts per thousand, (ppt).   Salinity is a pure weight per weight measurement and as weight is not influenced by temperature then neither is salinity.

The generally accepted standard salinity for natural seawater, (NSW), is 35ppt.

It is not possible to measure the salinity of the ocean or your aquarium directly unless you take a precise weight of seawater and evaporate all of the H20 from it then measure the weight of the residue, (mass solids analysis). Obviously this is not a convenient technique for the average hobbyist, so we use other methods to determine the salinity indirectly.


Specific Gravity, or relative density, is expressed as the ratio of the density of seawater relative to the density of the same volume of pure water. This of course can be quantified quite easily, however as the density of a liquid is calculated as the weight of a unit volume of that liquid and because liquids expand with temperature, then the volume that we measure and subsequently the weight of that volume will also change. We should always therefore qualify the S.G. by quoting it at a specific measured temperature, e.g. 1.025 at 25C.

Many people may have noticed on their hydrometers that there is a temperature quoted on the side, normally in small writing. This is the temperature at which the hydrometer was designed/calibrated to measure the water sample at –

SIMPLES - to use the words of a well known Russian meerkat - All we need to do is to ensure that our sample water is at that temperature and the specific gravity will be exactly right.

NOT QUITE TRUE - At a simple level this is correct however as you get further into the subject you will find that the S.G. reading that you obtain may be more specific to the equipment you have chosen to use to measure it rather than to the actual salt concentration in your tank.

I suspect, rightly or wrongly, that the specific gravity levels that we as hobbyists use in our aquariums may be historical from general oceanographic studies and it is not immediately clear when or where the 1.025 at 25C standard originates from.

If you trawl the internet and scientific literature you will find that the specific gravity measurements and tables for seawater which you come across were first developed long ago by shipping companies and oceanographic bodies like the NOAA. These charts were used to allow a simple way of calculating the changes in seawater density across the various oceans which affects how much cargo a ship can carry on a journey from one destination to another. The higher the density or salt content the more cargo the ship can hold without sinking.

If you go back to the original definition of specific gravity you will see that we calculate the S.G. as the density of a sample of SEAWATER at a specific temperature divided by the density of PURE WATER at a specific temperature. It is with density of the pure water that the problem or potential for error occurs.

Unfortunately there are a number of different temperature standards commonly used in oceanography and so printed tables for PURE WATER density can be quoted at 4C, 60F (15.56C), 20C or 25C depending on their intended application. Each temperature standard will result in a different pure water density figure which will give a different S.G. when we use that density in our calculation.

EXAMPLE: If we calculate the specific gravity for a sample of 35ppt seawater using a standard density temperature of 25C for the seawater sample but vary the density temperature used for the pure water, we can see how this affects the specific gravity result?

Specific Gravity = (?s(T)/?0(Tx)) X 1000

?s is the density of 35ppt seawater at temperature T, in this case 25C = 1.023343

?0 is the density of pure water at a range of commonly used temperature standards (Tx).


?0 (15.56C) = 0.9990166       SG = 1.023343/0.9990166              = 1.02435

?0 (20.0C) = 0.998203            SG = 1.023343/0.998203                = 1.02519

?0 (25.0C) = 0.997047958     SG = 1.023343/0.997047958         = 1.02637

We can see from the above that by using a pure water temperature standard of 20C that the S.G. of 35ppt NSW does in fact approximate to 1.025 which is perhaps where the figure comes from however you can also see the potential range that becomes available if another standard temperature is used.

All of these readings are for the same sample of 35ppt seawater at the same sample temperature of 25C. To measure the amount of salt in our aquarium by way of an S.G. reading, we must therefore understand which pure water temperature standard and which sample water temperature standard was originally used when calibrating the hydrometer or refractometer which is not constant between all equipment and manufacturers.

The Tropic Marin High Precision Hydrometer for example quotes 25C/25C on the instructions which means that it is calibrated using the density of both the sea water and pure water at 25C. With this instrument you should therefore be looking to measure an S.G. of 1.0264 and not 1.025 for the same 35ppt water sample.

This brings us back full circle to the point that the salinity scale which we talked about at the start is perhaps a much better way of expressing the salt concentration of your aquarium as we do not have the same potential for variation however most glass hydrometers do not even show this scale.


The main drive behind this article was to enlighten people on the potential variation between different specific gravity measurements, and to show individuals why they often find varying concentrations of major elements such as calcium and magnesium in their tanks compared with what they ‘expect to find’ or have been ‘told that they should find’ in their salt mix.

If as a consequence of measurement or calibration technique you do end up with a low salt concentration in your tank or salt mix, a salinity of only 33ppt for example. You should then equally expect that the concentrations of ‘all’ elements will be low as a consequence compared with the target concentrations found in natural seawater.

If the normal calcium level of NSW at 35ppt is 420ppm then at 33ppt salinity this will only be 396ppm and if the aquarist is not aware of his low salinity and adds more calcium to raise the level to 420ppm then they will be putting this particular element out of balance with the rest of the salt whilst ignoring other elements that may be just as important for the proper biological function of the organisms we keep. What they should really do is to add more ‘salt’ to raise the salinity, and bring ‘all’ of the levels up together.

D-D quote figures on the concentrations of the main elements in our H2Ocean salt at 35.5ppt which is an average of the readings found in the coral seas. With our new found understanding of the relationship between salinity and specific gravity we can expect that the 35.5ppt standard can correctly equate to an S.G. anywhere between 1.0247 and 1.0267 depending on the hydrometer used.


We discussed earlier that as aquarists we do not measure salinity or specific gravity directly and instead measure another parameter which has a relationship with the salt concentration such as refractive index in the case of a refractometer, buoyancy in the case of a hydrometer and conductivity if measured electrically.

Unfortunately all of these alternative parameters ‘are’ affected by temperature and so even when measuring salinity, which is not temperature dependant, we must still make an allowance because of the indirect test method.


There are various glass and plastic hydrometers available to aquarists on the market which work on the principle of buoyancy, where the denser the liquid in which it is immersed, the higher it will float.


 The more expensive glass hydrometers are very accurate but time consuming to use as they require a clear sided vessel in which to float and for the water to be at the exact calibration temperature. They are extremely delicate and easy to break and generally only display an S.G. scale which means you need a complete understanding of both the sample calibration temperature required, and the pure water calibration standard used to be able to relate the reading back to a particular salinity.

The cheaper plastic swing arm hydrometers are easy to use and robust but are not always intended to give an absolutely accurate reading and not normally used by more experienced reef keepers housing delicate and expensive stock. There can be issues with these units giving false readings over time due to bubbles or deposits attaching to the swing arm or pivot point which change the buoyancy and therefore the reading obtained. Swing arm hydrometers normally show a salinity scale and an S.G. scale but calibration temperatures may vary from one manufacturer to another.


Refractometers work on the principle that changes in salt concentration affect the refractive index of light and therefore light entering the instrument is deflected by varying amounts onto a graduated scale. They are a convenient, fast and easy method of keeping check on the salinity in your aquarium or when mixing up new batches for water changes.

The majority of refractometers in the hobby are manufactured and calibrated for use with saltwater and not seawater and intended for measuring sodium chloride, (brine) concentrations. Although sodium chloride is the main constituent of NSW, it varies in refractive index due to the lack of other ions such as calcium and magnesium which are found in seawater. The result of these additional elements is that the refractive index changes so that a 35ppt seawater solution approximates to the same refractive index as that of 36.8ppt brine. Whilst this may seem a small difference we should remember that it does have a significant impact on the true concentration of the dissolved elements in our aquariums.

When using a conventional ‘saltwater’ or ‘brine’ refractometer for the reef aquarium we must therefore look to achieve a reading of 36.8ppt if our intended salinity is 35ppt or 37.3ppt if our intended salinity is 35.5ppt as quoted for the H2Ocean salt.


To use a refractometer properly we must first understand the correct calibration procedure. If you look down the lens of most good refractometers you will see 20/20 printed on the screen. This means that it was originally calibrated with both pure water and sample water temperatures set at 20C and so in order to recalibrate the instrument properly you must use the same temperature for your calibration sample. What many people do not understand however is that it is the temperature of the instrument that should be at 20C and not the liquid as the few drops of sample water contains so little heat that it soon equilibrates to the same temperature as the refractometer body.

So with your refractometer at 20C, which is close to normal room temperature, you add a sample of distilled or RO water to the screen and use the adjustment screw to line up the blue marker with the zero salinity mark. This can now be used to measure the sample salinity accurately but only whilst the refractometer is still at 20C. Remember that even if your aquarium is running at say 27C (80.6F), the refractometer will always read as if it were at 20C. 

If you have an auto temperature compensating model (ATC) you must still calibrate the unit at the calibration temperature of 20C however this version, once correctly set, will auto adjust for environments where the instrument warms or cools away from this temperature. Normally the range is 10-35C and is achieved by a small bimetallic strip in the body of the refractometer which responds to the change and moves the graduated scale accordingly. For this reason copper bodied refractometers are better than the lower cost plastic ones as they conduct the ambient temperature changes faster.

True seawater refractometers have never really existed in the hobby at an affordable price until now but as part of the development of our salt and salt related products, D-D have brought out a new model of refractometer which is specifically calibrated to the salinity of NSW thus removing the requirement to apply an offset to obtain the true reading for your aquarium.

These new models are easily identified as seawater refractometers on the viewing screen and have been produced to show the correct S.G. of 1.0266 for NSW at a salinity of 35ppt and a sample/pure water density of 20C and a reduced range of salinity from 0 -40 rather than 0 – 100.


Hydrometers – understand the calibration temperature requirements, and take note that S.G. is not a constant number across all measuring equipment and reference sources.

Refractometers – understand the calibration requirements and apply an offset to measure the true seawater salinity if using a saltwater/brine model. Understand that the S.G. scale is not a constant number across all measuring equipment and reference sources.

Remember that the most important thing for your aquarium is consistency and stability. This applies to both the salinity and the relative concentrations of all elements within the salt and so any changes in salinity, salt brand or parameters should be made gradually.

Top up evaporated loss regularly or fit an auto top up system to reduce swings in salinity to a minimum.