Stop using this blanket statement.

It is technically wrong; it is practically wrong; and it is often used as a misleading mental model for firefighters.

GPM vs BTU make little sense

GPM (LPM) is flow, which in a firefighting context could be the amount of water flown per time unit. It is Gallons Per Minute (Liters Per Minute or LPM in SI-units).

A BTU (Joules or J in SI-units) is an amount of energy. In a firefighting context it could be the potential chemical fuel in a chair. But in this example, I will use it as the total amount of energy being released from the fire.

But knowing the total amount of BTU’s (J) released by the fire does not give you a time, how fast or slow that energy is being released. That total energy could be released in seconds, or years without more information.

The statement GPM vs BTU make little sense for firefighters as GPM is time dependant and BTU’s are not.

GPM vs BTU’s/min make sense

BTU’s/min (may be converted to watts in SI-units) would instead give us energy per time unit. Another word for BTU’s/min would be power. It could be cooling power or heating power, both possibly measured in BTU’s/min. Or we can use the term Heat Release Rate HRR if we only talk about heating. That is a very useful value for firefighters to understand.

GPM vs BTU’s/min (usually converted to Kg/s vs watts) would make technical sense. Some amount of water per minute (cooling power) is required to absorb some energy being released per minute (heating power). With these values we would know if the firefighters cooling power is higher compared to the fires heating power. Basically, if the firefighters are winning or not.

Practically wrong and misleading

For this article we will imagine a bedroom fire in a house with a HRR of 200k BTU’s/min (3.5 MW). That energy is transferred from the fire in the form of heat.

The blanket statement GPM vs BTU’s/min is often used to imply that you must have more then 200k BTU’s/min of cooling power (water flow) to suppress the HRR of that imaginary fire. But that is not true and used like that the statement is practically wrong and misleading.

Firefighters need to cool

Imagine that we want to suppress the fire by doing an interior attack through the front door, down the hallway and into the bedroom.

When the firefighters move down the hallway they need to cool the smoke. They need to absorb some of the energy coming from the fire with their water, to prevent the smoke from heating up the firefighters too much. And they need to cool the surfaces of the hot building materials. Cooling surfaces lowers radiating heat and prevents possible offgasing.

The closer the firefighters get to the primary fire, the more energy they must be able to absorb. Eventually the firefighters reach the bedroom and get line of sight to put water on the fuel surfaces and stop the fire.

The building provides cooling

The fire in the bedroom is producing energy which is transferred as heat. Some of that energy is lost when the fire heats the fuel surfaces to make them offgas more. The fire is also heating all the cold air being pushed in by gravity. And surfaces not able to offgas are also being heated. It means that a lot of the energy created by the fire is already absorbed in the materials.

But some of the energy from the fire follows the smoke (fire plume) out into the hallway. The smoke will transfer a lot of its energy to the hallway materials making them hotter, and the smoke colder.

When the smoke finally leaves the open front door to the house, its potential to heat things has dramatically been reduced. The building has already absorbed most of the energy produced by the fire.

GPM does not have to be higher than the total HRR

If the firefighters want to cool the smoke and the materials inside the front door, they do not need 200k BTU’s/min (3.5 MW) of cooling power. They may only need 5 % of that. The rest of the energy from the fire has already been absorbed by the building before the smoke reaches the front door.

When the firefighters reach the hallway, they might need 10 % of the fires total HRR. And when they reach the primary fire compartment, they might need 50 % of the fires HRR.

The rest of the needed cooling power is already supplied by the building. And we do not have to cool and absorb all the energy stored in the building materials to achieve suppression.

I just made up those numbers. The actual numbers vary heavily depending on the situation. But the needed cooling power (GPM or LPM) is not going to be as high as the total amount of energy being produced (BTU’s/min or W) by the fire, not even inside the fire compartment.

GPM have to be higher than the local HRR

Locally, the HRR varies inside the house. Some heat may come from potential combustion in that local space, for instance inside the front door. And some heat may come from hot things (like building materials, furniture and the smoke) radiating heat.

The GPM (or more correct the cooling power) would only need to be higher then the actual HRR in that specific local place. If the cooling power if higher, the temperature will start to go down and the firefighters are winning.

But that is not how HRR is measured in research. Researchers usually only look at the total amount of HRR for the whole fire.

And when a blanket statement like GPM vs BTU/min is used, it is misleading when compared to the total HRR of the entire fire.

Cooling through the inlet is easier

Imagine the same bedroom fire, but we remove the room around the items on fire. It is just a bed and other items burning on a parking lot.

Let us still imagine that the total HRR is 200k BTU/min (3.5 MW). But most of the energy produced by the fire is now lost through convection and radiation into the air around and above it.

If the firefighters were to attack this fire they would not need to cool at all during the advance. There is no smoke or hot surfaces to cool down. The needed cooling power is 0 % of the total HRR during the advance.

To stop the fire the firefighters need to put water on the hot fuel surfaces that are offgasing. As most of the heat goes up into the air the water stream can reach the hot surfaces without having to pass through, and possible absorb, energy coming from the fire. The water only needs to cool down the fuel surfaces and cool some radiant heat coming back from the flames above.

Suppressing this 200k BTU’s/min (3.5 MW) fire would only require a fraction of the water (cooling power). The water does not have to absorb all the total HRR from the fire. In this case, most energy is absorbed and lost to the air around the fire. The water only need to be able to cool the local HRR above that fuel surface.

When cooling into the inlet of the fire, the cooling power needed is dramatically reduced.

Cooling through the smoke is easier

Imagine the same bedroom items on fire again, but this time they are positioned in the end of a long hallway. So now we have a room around the burning items that absorb the energy from the fire.

Imaging that our firefighters are in the other end of that hallway. They have line of sight to the fire and the hot surfaces. But between the firefighters and the hot fuel surfaces we now have the hot smoke with lots of energy.

The firefighters could use a straight stream (as compact as possible) through the hallway towards the hot surfaces. If they aim well, the stream will cut through most of the energy in the smoke as the stream does not absorb much energy in that form. The stream reaches the fuel surfaces and stops the fire, without having to cool the smoke and materials in the hallway.

Again we only need a fraction of the total HRR produced by the fire. Because we do not need to absorb all the energy coming from the fire to achieve suppression.

Usually there are twists and turns that prevents line of sight between the firefighters and the hot fuel surfaces. We most often need to cool smoke and materials during the advance. But when line of sight exist it can greatly reduce the needed cooling power for suppression.

Cooling from behind the smoke is easier

Sometimes we also have the option to insert water directly into the primary fire compartment with a piercing nozzle. The water is sprayed from behind the smoke, directly to the fuel surfaces. The water does not have to absorb all the energy in the smoke to get there.

This very sneaky fire attack is just one of the reasons why a piercing nozzle may not even require 1 % of cooling power compared to the total HRR of the fire. The understanding of water as steam explains a lot of the rest, but that is a topic for another article.

Too simple model for firefighters

I think the blanket statement GPM vs BTU’s/min as a mental model is too simplified to be useful for firefighters. Water flow is just one fairly simple factor of understanding cooling and fire suppression.

Where the cooling power is needed, how to get it there and in which form are some examples of much harder problems to solve. But if we want great firefighters we need to make them understand it.

The right flow is critical

Some might read this article as advocating for lower flows and that would be partially true. You should not flow more then you need. Time is our enemy and higher flows are most often heavier and slower. Or might deplete the available water too soon. And higher flows might also bring unnecessary property damages. The right flow is the goal.

Not understanding how water is used effectively and efficiently is bad regardless of the amount we have available and can use. There is always going to be an upper limit and we want to make sure we rarely meet it by being efficient with the water all the time.

But at potential rescues, and for firefighter safety, an interior attack should probably use as much water as we can to reduce temperatures as fast as we can. Going way past the fires local HRR in that space.

But an interior attack might not be the best option, or a lower flow could make it faster and thus better. It is complicated and we should encourage firefighters to understand it and strive for excellence.

Latest blogs


A fires total heat release rate is not a good value for how much water flow that is needed for suppression, and simplified statements are misleading.