safely lowering salinity


New member
Hello all,

I just found out that my refractometer is way out of calibration, so I need to start lowering my salinity. I'm looking for some information about what type of duration this should be done in safely. How slowly, and by how much can I lower the salinity at a time.

The Grav/Sal is up to 1.030/40 and I would prefer it to be at or near 1.025/35

So far I have removed 20g of water and replaced 15 with a lower salinity mix. I figure this achieved my water change for the week, started lowering the salinity (hopefully not too quickly) and created space for RO to go in over the course of today and maybe tomorrow.

How long do I need to wait after adding new water before I can get an accurate reading on Grav/Sal? And how long do I have to wait between changes?

Tank is a 90g, with 35g sump, probably close to 100g of water volume.

Thanks in advance!
If your current salinity is at 1.030 and you want to reduce it down to 1.025 with a total water volume of 100 gallons, you will need to remove around 17 gallons of water from your tank, dump it and replace with rodi water. You don't want to do this all at once. Perhaps replacing 2.5 gallons with rodi water per day would be a good target (you can take half out in the AM and the other half out in the PM to be safer. This should take about 1 week to reduce your salinity where you want it.

You just need to give your system enough time to thoroughly mix before taking a salinity reading, perhaps 1-2 hours should be fine with good water circulation. ;)
You should be calibrating your refractometer with a proper seawater standard using a salinity calibration fluid 53.0mS/1.026SG & not pure water. ;)


New member
WOW... 17g, I had no idea it would take that much to change it! Just curious, why are you recommending not to use pure water to calibrate?
I have taken the easier to understand part of Randy's article regarding proper refractometer calibration, if you want more info it is in this article. ;)

Refractometers and Salinity Measurement

From it:

"Despite the fact that many refractometers sold to aquarists recommend calibration in pure water, such a calibration alone will not ensure accuracy for the reasons described above. So my recommendation for calibration is as follows:

1. First calibrate the refractometer in pure freshwater. This can be distilled water, RO (reverse osmosis) water, RO/DI water, bottled water and even tap water with reasonably low TDS (total dissolved solids). Calibrating with tap water that has a TDS value of 350 ppm introduces only about a 1% error in salinity, causing readings in seawater to read a bit low. So 35 ppt seawater (specific gravity = 1.0264) will read to be about 34.7 ppt, and will show a specific gravity of about 1.0261.

This calibration should ordinarily be carried out at room temperature using an ATC refractometer. The directions with some ATC refractometers insist that the calibration be carried out at a specific temperature, but I've never understood how that could matter and I would not worry about it. If the refractometer is not an ATC refractometer, then careful temperature control or correction is necessary, and such corrections are beyond the scope of this article.

Calibration is usually performed by putting the freshwater on the refractometer, letting it sit for at least 30 seconds so it comes to the same temperature as the refractometer, and adjusting the calibration screw until it reads a value appropriate for freshwater (e.g., refractive index = 1.3330, salinity = 0 ppt, specific gravity = 1.0000). Normally, this step is a quick and easy procedure, and may often be all that is required IF the refractometer has been verified to have passed the second calibration step below at least once. This is an offset calibration, as described above.

2. The second step in calibration should be performed at least once before relying on a refractometer to accurately measure the salinity of a reef aquarium. This step involves testing it in a solution matching the refractive index of 35 ppt seawater (or some similar solution near the range of measurement). Remember to let it sit for at least 30 seconds so it comes to the same temperature as the refractometer. Suitable commercial and do-it-yourself standards were described earlier in this article. Using one of them, place a drop onto the refractometer and read the value. If it reads approximately 35 ppt, or a specific gravity of 1.0264, or a refractive index of 1.33940, then the refractometer is properly calibrated and is set to go.

If it does not read correctly, and is off by an amount that is significant relative to your salinity precision requirements, then you need to recalibrate it using this second fluid. I suggest that a salinity error of ± 1 ppt or a specific gravity error of ± 0.0075 is allowable. If the refractometer is off significantly, and you used a do-it-yourself standard made with crude techniques such as Coke bottles, a good next step might be to buy a commercial standard.

To correct errors using these seawater standards, simply adjust the calibration screw on the refractometer until it reads the correct value for the standard (35 ppt, or a specific gravity of 1.0264, or a refractive index of 1.33940). This type of slope calibration makes the refractometer suitable to read solutions whose salinity is close to seawater's. After such a calibration, refractometers may not read freshwater correctly.

Again, despite the claims in the directions of some refractometers to have the standard at a particular temperature, when calibrating an ATC refractometer with this seawater standard, I'd just use it at room temperature.

If you are using a refractometer for hyposalinity, such as when treating a sick fish, I'd just use one calibrated in freshwater, because that is closer in salinity than seawater to the hyposaline solution usually used (say, specific gravity = 1.009). A new standard for hyposalinity can also be made by mixing one part 35 ppt seawater and two parts freshwater, but that is probably overkill."
FWIW, this is the forumla that Randy has provided for determine salinity in the past. This statement refers to a system with a total of 160 gallons of total water volume, with a salinity of 1.0260.

"The 35 ppt part is easy: it is just proportional.

So the final salinity would be:

[140 x (35) + 20 x (0)]/(140+20) = 30.6 ppt

Using sg, you can just drop the leading 1.0, so:

[140 x (26) + 20 x (0)]/(140+20) = 22.75 ---> 1.02275

Randy Holmes-Farley
Club 65535

Current Tank Info: 120 mixed reef"


So the final calculation for your situation would be:

(83 x 30) + (17 x 0)/(83+17) = 25 ---------> 1.025

Please check my calculations. ;)
At simpler forumla for figuring reducing salinity when replacing rodi water with zero salinity would be:

(current tank salinity) x N / (Total water volume) = (your wanted final salinity)

where N = (Total remaing tank water volume after replacing approrate amount with pure water)

Total water volume minus N = the amount of reef tank water you need to replace. ;)
Last edited: