This is the final boss of explanations on what specific gravity actually is, spoiler, its a lot of things and nothing

Don't skim this one. Read the entire paragraph twice!

The density of water for a specific gravity calculation is a contentious subject. The most technically correct answer is that every specific gravity measurement should be listed with the temperature of the fluid under measurement and the temperature of the water that was used as a reference. See "The Wise Words of an ASME Trusted Source" below. However, I have yet to see anyone use specific gravity while specifying both of these temperatures.

It is quite common to see something like this:

Specific Gravity of Jet A @70°F = 0.80722

However, I have never seen anyone specify the following:

Specific Gravity of Jet A @70°F (water temp at 39.164°F) = 0.80722

This means that people need to make assumptions about the temperature of the water that was used as a comparison point. In different industries and at different companies, different standards exist.

• Option 1: Assume the density of water is water at its most dense. Water is the most dense at 3.98°C (39.164°F). This corresponds to a density of 1.9403 slugs/ft³ (999.973 kg/m³). This is the most common method.
• Option 2: Assume the density of water is water at room temperature. Some industries/companies say room temp is 60°F (15.55°C), some say 66°F (18.88°C), and others say 70°F. This is the second most common method. Most companies say room temp is 60°F.
• Option 3: Assume the density of water is the density at the same temperature as the liquid under test. I think this makes the most intuitive sense, although it is the least convenient because you need to have the density of water handy (see the Density of Water Calculator and the Specific Gravity Calculator). This is the method used by Mettler Toledo.

Bonus:

It is common to see a specific gravity equation written where the density of water is 1000 kg/m³. However, the most dense that water can be (under standard atmospheric pressure) is 999.973 kg/m³. I understand that this rounds up to 1000 kg/m³, but I think it is sloppy to round this because it implies that the density of water is 1000.00000 kg/m³.

The following is the official trusted fluids textbook of ASME:

Mark's Standard Handbook for Mechanical Engineers, 12th Edition:

"The specific gravity (sp. gr.) of a substance is a dimensionless ratio of the density of a fluid to that of a reference fluid. Water is used as the reference fluid for solids and liquids, and air is used for gases. Since the density of liquids changes with temperature for a precise definition of specific gravity, the temperature of the fluid and the reference fluid should be stated, for example, 60/60°F, where the upper temperature pertains to the liquid and the lower to water. If no temperatures are stated, reference is made to water at its maximum density, which occurs at 3.98°C (39.164°F) and atmospheric pressure. The maximum density of water is 1.9403 slugs/ft³ (999.973 kg/m³). For gases, it is common practice to use the ratio of the molecular weight of the gas to that of air (28.9644), thus eliminating the necessity of stating the pressure and temperature for ideal gases."

What is the point of a "convenient" dimensionless ratio if every time the number comes up you have to question exactly how to convert it back to density? I think specific gravity is overused. Absolute density units are much less ambiguous. Specific gravity is a muddled, confusing mess of a unit. Let's obsolete this unit and start over.

At the temperature extremes, does specific gravity even make any sense?

• The specific gravity of ethanol at -20°F is 0.807. Water at this temperature would be frozen.
• The specific gravity of MIL-PRF-23699 at 300°F is 0.88. Water at this temperature would be boiling.

I would argue that it makes no sense to compare the density of a liquid to the density of an ice cube or to superheated steam.

Special thanks to Camren for encouraging me to add this clarification and doing the background research. Additional thanks for being half of KasperCalc's web traffic.