What is 1.025 specific gravity in ppt?
On a refractometer calibrated with a 35 ppt standard, 1.025 is 33.2 ppt. On a glass hydrometer marked 60/60 °F floating in 78 °F tank water, the same 1.025 is 36.1 ppt — three parts per thousand saltier, and past the top of the reef window. That single fact is behind most of the "is 1.024 or 1.026 right?" threads: specific gravity is a comparison of two densities, the sample's and pure water's, and both depend on temperature. Say what the instrument compares to and the confusion disappears. The calculator above does that; the numbers come from the UNESCO 1983 equation of state for seawater, not from the "(SG − 1) × 1,340" shortcut, which would call 1.025 33.5 ppt and be a third of a ppt out before temperature even enters.
A refractometer with working ATC reads the sample as if it were at 25 °C and compares it with water at 25 °C — the "25/25" scale. Natural seawater, 35 ppt, is 1.0264 there. A glass hydrometer reads the density of the water it's actually floating in, against a scale drawn for water at its calibration temperature — usually 60 °F. Float it in 78 °F water and 35 ppt reads 1.0242: same salt, lower number, because warm water is less dense. Neither instrument is wrong. "1.026" and "1.024" for the same tank are both right — for their instrument. What's wrong is comparing them without saying so.
Salinity chart: ppt to specific gravity, by instrument
| ppt (psu) | Refractometer 25/25 | Hydrometer 60/60, water at 60 °F | Hydrometer 60/60, water at 78 °F | mS/cm at 25 °C | kg/m³ at 78 °F |
|---|---|---|---|---|---|
| 28 | 1.0211 | 1.0215 | 1.0189 | 43.5 | 1,017.9 |
| 30 | 1.0226 | 1.0230 | 1.0204 | 46.2 | 1,019.4 |
| 31 | 1.0233 | 1.0238 | 1.0212 | 47.6 | 1,020.2 |
| 32 | 1.0241 | 1.0245 | 1.0219 | 49 | 1,020.9 |
| 33 | 1.0249 | 1.0253 | 1.0227 | 50.4 | 1,021.7 |
| 34 | 1.0256 | 1.0261 | 1.0234 | 51.7 | 1,022.4 |
| 35 | 1.0264 | 1.0269 | 1.0242 | 53.1 | 1,023.2 |
| 36 | 1.0271 | 1.0276 | 1.0249 | 54.4 | 1,023.9 |
| 37 | 1.0279 | 1.0284 | 1.0257 | 55.7 | 1,024.7 |
| 38 | 1.0286 | 1.0292 | 1.0265 | 57.1 | 1,025.4 |
| 40 | 1.0302 | 1.0307 | 1.0280 | 59.7 | 1,027 |
UNESCO 1983 equation of state and PSS-78, computed by this page's engine. Natural seawater is 35 psu. Reef target 34–36 ppt. Refractometer column assumes calibration with a 35 ppt standard and working ATC.
Why does my hydrometer read lower than my refractometer?
Because the water is warm. Here is what a 60/60 glass hydrometer and a 77/77 one read for exactly 35 ppt as the tank warms up:
| Tank water | 60/60 hydrometer reads | 77/77 hydrometer reads | Refractometer reads |
|---|---|---|---|
| 68 °F / 20 °C | 1.0258 | 1.0278 | 1.0264 |
| 72 °F / 22 °C | 1.0252 | 1.0272 | 1.0264 |
| 75 °F / 24 °C | 1.0247 | 1.0267 | 1.0264 |
| 78 °F / 26 °C | 1.0242 | 1.0262 | 1.0264 |
| 80 °F / 27 °C | 1.0238 | 1.0259 | 1.0264 |
| 82 °F / 28 °C | 1.0235 | 1.0255 | 1.0264 |
| 84 °F / 29 °C | 1.0231 | 1.0251 | 1.0264 |
All rows are the same water — 35 psu. Computed by this page's engine.
So a hobbyist with a 60/60 hydrometer who was told to "keep it at 1.025" and does so at 78 °F is actually running 36.1 ppt. One told to keep it at 1.024 is at 34.8 — about right. Both the "1.024" and "1.026" camps online are usually correct for their own instrument. If you use a glass hydrometer, either use the calculator's hydrometer mode with your temperature every time, or work in the corrected number once and note what your hydrometer should read for your tank at your temperature (1.0242 for 35 ppt at 78 °F on a 60/60). Swing-arm hydrometers add their own error on top and aren't worth correcting.
Most hobby refractometers are built for brine, not seawater: their scale assumes a sodium-chloride solution, and seawater's ion mix refracts a little differently. Zeroing on RO makes 0 read 0 and leaves the seawater end reading about 1–1.5 ppt high. Calibrating with a 35 ppt seawater standard pins the scale where you actually use it. The calculator's "calibrated with RO" option applies the typical correction; a $10 bottle of standard removes the need for it. And ATC only compensates for the temperature of the prism — it does nothing for a wrong calibration.
Conductivity: what a probe or a controller measures
Controllers and lab meters don't measure density at all; they measure how well the water conducts electricity, in mS/cm, and convert to salinity with the Practical Salinity Scale (PSS-78) — which is why the unit on those displays is "psu," practical salinity units, and why psu and ppt are the same number for hobby purposes. Conductivity rises about 2% per °C, so a probe reading must be temperature-compensated: 35 psu is 42.9 mS/cm at 15 °C and 53.1 at 25 °C. The calculator takes the raw reading and the water temperature and does the compensation itself; if your controller already shows psu, its probe is calibrated (or not) against a conductivity standard, and this page can only tell you what it should read.
What salinity should a reef tank be?
Natural seawater is 35 psu (1.0264 on a properly calibrated refractometer), and the reef consensus is 34–36 ppt — the corals came from 35, and the case for running lower is habit from the fish-only era. Stability matters more than the exact figure: an auto-top-off that holds the level holds the salinity, and the daily swing from evaporation in a tank without one is bigger than the gap between 34 and 36. Density, for anyone who needs it (aquaculture, dosing by weight, the weight calculator): 35 psu at 78 °F is 1,023.2 kg/m³, about 2.6% denser than fresh water at the same temperature. When you need to change the salinity, the salt-mix and salinity-adjustment tools take it from here.