Measurement of atmospheric pressure

[MetricRocketeer]

Hi TRF colleagues,

First, I will get to the point. During a hurricane, the National Weather Service (NWS) uses an efficient and intuitive metric measurement of atmospheric pressure — millibars. After the hurricane passes, the NWS reverts to using the inefficient and nonintuitive foot-pound measurement of inches of mercury. Why?

As my TRF username suggests, I support using the metric system, including for measuring weather conditions.

Unfortunately the National Weather Service does not report weather conditions using metric units. Doing so would go a long way towards metrication.

Interestingly, however, the NWS does use one metric unit when it reports atmospheric pressure during a hurricane. Namely, it uses millibars (abbreviated mbar). Incidentally, one millibar exactly equals one hectopascal (hPA), which equals one-tenth of a kilopascal (kPa). Internationally, hectopascals and kilopascals are used to report atmospheric pressure. The International Bureau of Weights and Measures (BIPM, from the French wording) — the international agency charged with administering the metric system on a day-to-day basis — fully approves using hectopascals or kilopascals to measure atmospheric pressure. Nevertheless, since 1 mbar exactly equals 1 hPa, we can consider the millibar to be an approved metric unit.

Here is why the millibar is efficient and intuitive. Normal atmospheric pressure at sea level measures 1013.25 mbar (this is a very close approximation). This is not exactly 1000 mbar, of course, but it is so close that you can get a quick sense of atmospheric pressure. If the pressure is less than 1000 mbar, we have a low-pressure situation. If the pressure is greater than 1000 mbar, we have a high-pressure situation. Beautiful and lovely!

And here is why inches of mercury (inHg) is inefficient and nonintuitive. Normal atmospheric pressure at sea level measures 29.92 inHg, rounding to two decimal places. Wait! What? Who can remember this? What a ridiculous and useless unit of measurement.

But the NWS, following the stubborn and foolish American refusal to adopt the metric system, uses inches of mercury — except during a hurricane. During a hurricane, the NWS uses millibars. Good for them. But once the hurricane passes, the NWS reverts to inches of mercury. Why, pray tell?

Since the United States steadfastly refuses to adopt the metric system and instead persists in using the foot-pound system, the NWS’s decision to use a foot-pound measurement for atmospheric pressure makes sense. The NWS is consistent in its measurement backwardness. Asking the NWS to use the metric system would be tantamount to asking it to part the Red Sea. Amazingly, however, during a hurricane the NWS does use the metric system to measure atmospheric pressure, which makes comprehending the strength of the hurricane so much easier.

So, then, I ask the question. Since the NWS already makes this concession to modernity — during a hurricane it measures atmospheric pressure using the metric system — why does it revert to the foot-pound system once the hurricane passes?

Does anyone know?

Thank you for thinking about this.

Stanley

 

[Charles_McG]

Torr!

 

[Steve Shannon]

Most of us are capable of moving between the different systems. It just doesn’t seem to matter.

 

[Marc_G]

That is odd!

Then again, my car gets 40 rods to the hogshead, and that's the way I like it!

(personally, as a scientist, I'm pro-metric, but have adapted to many backward tendencies around me!)

 

[cls]

send a copy your good letter to NWS.

 

[Funkworks]

Does anyone know?

I think it's just a matter of tradition.

To understand the point of view of those who don't want to change, ask yourself if you'd accept a minute having 100 seconds (instead of 60). We could adapt to decimal time, but no one seems ready to make that apparently logical step.

 

[heada]

What if they use metric all the time and then convert and report whatever is most appropriate for that situation

 

[Marc_G]

I think it's just a matter of tradition.

To understand the point of view of those who don't want to change, ask yourself if you'd accept a minute having 100 seconds (instead of 60). We could adapt to decimal time, but no one seems ready to make that apparently logical step.

Lunch at 5 o'clock. Love it!

 

[FredA]

A slight aside:
A week or so back, one of the hurricanes set a new pressure record - I don't remember the specifics, but not important.
One of our local TV news reporters changed up the units a bit while reading the prompter.
Instead of giving units of milli-bar the newscaster call out mega-bytes.

 

[dr wogz]

A slight aside:
A week or so back, one of the hurricanes set a new pressure record - I don't remember the specifics, but not important.
One of our local TV news reporters changed up the units a bit while reading the prompter.
Instead of giving units of milli-bar the newscaster call out mega-bytes.

"they don't have to be smart, but they better look good doing it.."

 

[mh9162013]

Metric versus imperial: I use whatever gives me round numbers to work with.

 

[MetricRocketeer]

send a copy your good letter to NWS.

Hi @cls,

Thank you for the suggestion. I have now done that.

I sent an e-mail with my message to [email protected]. I hope that someone with some authority reads it.

Stanley

 

[Peartree]

So, your contention is that if the normal pressure is almost, but not quite, 1000, and that higher than 1000 is "high" and lower than 1000 is "low" then the system is beautiful and elegant.

But if normal pressure is almost, but not quite, 30, and that higher than 30 is "high" and lower than 30 is "low" then the system is ugly and inelegant?

Sure. Okay.

 

[Funkworks]

Lots of metrication resources in here:

https://www.nist.gov/pml/owm/metric-si/us-metrication
And metric week is very soon:

https://usma.org/national-metric-week

 

[lakeroadster]

https://www.brightcellars.com/blog/how-much-is-a-buttload-of-wine/

 

[Tyeeking]

And metric week is very soon:

https://usma.org/national-metric-week

Weird that metric week would have 7 days.

 

[cerving]

All "standard" units are based on purely arbitrary criteria, so honestly one is as good as another. They've done some fancy things over the years to make them more repeatable (i.e. basing the the meter on a number of wavelengths of Cesium), but the fact still remains that somebody once upon a time declared "THIS is a unit."

 

[MetricRocketeer]

So, your contention is that if the normal pressure is almost, but not quite, 1000, and that higher than 1000 is "high" and lower than 1000 is "low" then the system is beautiful and elegant.

But if normal pressure is almost, but not quite, 30, and that higher than 30 is "high" and lower than 30 is "low" then the system is ugly and inelegant?

Sure. Okay.

Hi @Peartree,

Thank you for your reply.

If normal atmospheric pressure were exactly 1000 mbar, then the system would be fully beautiful and elegant. Unfortunately, normal atmospheric pressure is only approximately 1000 mbar. Nevertheless, comparing against a base of 1000 is mathematically much easier than against 30. For example, 3.5% greater than 1000 is quickly calculable to 1035. And 3.5% less than 1000 is quickly calculable to 965. But 3.5% greater than or less than 30 requires more arithmetic.

Stanley

 

[SecondRow]

Hi @Peartree,

Thank you for your reply.

If normal atmospheric pressure were exactly 1000 mbar, then the system would be fully beautiful and elegant. Unfortunately, normal atmospheric pressure is only approximately 1000 mbar. Nevertheless, comparing against a base of 1000 is mathematically much easier than against 30. For example, 3.5% greater than 1000 is quickly calculable to 1035. And 3.5% less than 1000 is quickly calculable to 965. But 3.5% greater than or less than 30 requires more arithmetic.

Stanley

You’re willing to accept an error of 1.3% using your metric approximation, rather than the 0.27% error of the imperial approximation? No thanks, in this case I’ll take the superior approximation, and keep my error to almost 1/5 of the inefficient metric one.

Oh, and 3.5% of 30 is 1.05. C’mon, you don’t even need a calculator for that. Who can’t multiply by 30? But even if you did need a calculator, nearly everyone of us has one in our pocket. Ease of calculation should no longer be the standard for how to choose a measurement system. Rather, it should be based on what works better for the measurement you are trying to make.

For instance, Fahrenheit is vastly superior to Celsius when trying to measure body temp or ambient air temp, because it has a larger range, and can be more precise without resorting to more significant digits. But if you want to boil or freeze water, at 1 atmosphere of pressure only of course, and the water is pure and has no salts in it, then…well…Celsius is the way to go! Never mind that no other substance has melting and boiling points that easy to remember. We should definitely choose our everyday temperature system based on the phase change points of water at standard atmospheric pressure.

 

[Hobie1dog]

You’re willing to accept an error of 1.3% using your metric approximation, rather than the 0.27% error of the imperial approximation? No thanks, in this case I’ll take the superior approximation, and keep my error to almost 1/5 of the inefficient metric one.

Oh, and 3.5% of 30 is 1.05. C’mon, you don’t even need a calculator for that. Who can’t multiply by 30? But even if you did need a calculator, nearly everyone of us has one in our pocket. Ease of calculation should no longer be the standard for how to choose a measurement system. Rather, it should be based on what works better for the measurement you are trying to make.

For instance, Fahrenheit is vastly superior to Celsius when trying to measure body temp or ambient air temp, because it has a larger range, and can be more precise without resorting to more significant digits. But if you want to boil or freeze water, at 1 atmosphere of pressure only of course, and the water is pure and has no salts in it, then…well…Celsius is the way to go! Never mind that no other substance has melting and boiling points that easy to remember. We should definitely choose our everyday temperature system based on the phase change points of water at standard atmospheric pressure.

ok, I take note that this guy knows his stuff.

 

[jadebox]

I am a big supporter of using standard units ... where appropriate. For example, when doing woodworking, using feet and inches tends to be better because 12 has more factors than 10. Half of 12 is 6. A third of 12 is 4. And a fourth of 12 is 3.

As far as units for pressure when talking about the weather, I don't have a preference. I tend to be concerned more about relative measurements - is the pressure rising (good) or falling (bad).

I have worked with the military for years and almost everything they do uses metric units. The notable exception is aircraft altitude. I have heard multiple reasons for using feet. The most common is that it is a legacy thing. But, that doesn't make sense because all of the other things using metric used to be measured in imperial units. More likely it is because of the higher precision you get using integer feet than integer meters.

 

[KennB]

The first company I worked for out of college made vacuum pumps and compressors. In the test department, there was a mercury barometer. Very easy to read and, because we were in the United States, we used the Imperial system with 29.92 inches of mercury as standard pressure at sea level (a very high tide would flood part of our parking lot so we knew sea level). I believe it had a dual scale showing ~760mm of mercury as standard.

I don't recall ever needing to calculate a percentage of atmospheric pressure but some people may. We did use a rule of thumb that for every 1000 feet of altitude, you lose 1 inch of mercury pressure. This came in handy when I was visiting our office in Denver. A facilities person at a hospital was wondering why his vacuum pump that was rated at 25" of vacuum was only reaching 20" on the gauge. I had to remind him that the rating was at sea level and could be expressed as 5" absolute pressure. With only 25" of atmospheric pressure available in Denver, the best the pump could pull down to would be 20".

For the general public, units don't matter that much. If they know what's "normal" they can grasp the meaning of numbers that are higher or lower. Magnitude generally doesn't register unless it's extreme.

 

[bobby_hamill]

I have several ways of measuring pressure and in different units

1= Belfort Aneroid Barometer reading Millibars & Inches Of Mercury 2 places behind the decimal

2= Belfort Microbarograph land version 5-800A and marine version 5-800A-M1 they also read in
Millibars & Inches Of Mercury 2 places behind the decimal

3= Princo 453X Fortin ( mercury ) Barometer that is good up to 12,000 feet and can measure 3 places behind
the decimal in Inches Of Mercury or Millimeters Of Mercury
The Fortin Barometer also has a chart to account for the temperature of the mercury itself

 

[Funkworks]

I am a big supporter of using standard units ... where appropriate. For example, when doing woodworking, using feet and inches tends to be better because 12 has more factors than 10. Half of 12 is 6. A third of 12 is 4. And a fourth of 12 is 3.
...
I have worked with the military for years and almost everything they do uses metric units. The notable exception is aircraft altitude. I have heard multiple reasons for using feet.

Instead of 12 inches, a metric person would use 30 cm (a metric ruler), which can be cut in 2, 3, 5, and 6.

In aviation however, there's the issue that airspace near control towers is split like floors of a building: 1000ft, 1500ft, 2000ft and so on. By having different planes fly at different levels, they stay safe from each other. Pilots are trained to keep that in mind, and instruments are built to display those numbers. Changing to metric would require all pilots and instruments to change to metric at the same time. The last thing you want is pilot flying at 1000 feet instead of 1000 meters. Confusion must be avoided.

As for pressure, I learned all the units at the same time, so I'd pick metric, but I've had to use imperial units for work because of some equipment we had, so I'm quite indifferent in practice.

 

[MetricRocketeer]

Hi TRF colleagues,

I appreciate all the helpful and thoughtful comments that people have offered.

I would like to take a different tack for a moment and ask this about measuring atmospheric pressure.

Would it make more sense just to measure pressure in atmospheres? This is a genuine question. I am not a meteorologist, and I am curious as to what people think.

Here is my reasoning. Whatever units we use serve as just an index, right? Just to stick with millibars or inches of mercury for the moment. Let's say that the pressure were 985 mbar, a value which I saw while tracking Hurricane Ian. To two digits past the decimal place, that would be 29.09 inHg.

But both units mean that that the pressure is 97.2% of an atmosphere. So perhaps weather services around the world should report that the atmospheric pressure equals 97.2 atm (or whatever precision you want, perhaps just 97 atm).

Would this make sense?

Stanley

 

[Tyeeking]

Would this make sense?

It depends. Who is the intended audience?

 

[MetricRocketeer]

It depends. Who is the intended audience?

Anyone who wants to know the weather conditions.

I mean, why is the barometric pressure shown in the daily weather report? I am asking whether reporting the barometric pressure in atmospheres would be better than reporting it in inches of mercury or in millibars.

 

[Funkworks]

Would it make more sense just to measure pressure in atmospheres?

I think we have to distinguish between how it's measured, and how it's reported.

"Reporting" a pressure as a percentage of the atmosphere's (atm) seems like a great idea. I'd go for it. However, a pressure cannot be "measured" as a percentage of the atmosphere (atm). An atmosphere (atm) is defined as 101325 Pa. That is the unit defined in terms of the kg, s, J, and N of the metric system. An atmosphere (atm) can then be calculated from the Pa value.

The metric system has a few base units (kg, s, J, N, and a few more) from which all others (Pa, Watt, volt, etc) are derived.

 

[MetricRocketeer]

I think we have to distinguish between how it's measured, and how it's reported.

"Reporting" a pressure as a percentage of the atmosphere's (atm) seems like a great idea. I'd go for it. However, a pressure cannot be "measured" as a percentage of the atmosphere (atm). An atmosphere (atm) is defined as 101.325 kPa. That's the unit defined in terms of the kg, s, J, and N of the metric system.

The metric system has a few base units (kg, s, J, N, and a few more) from which all others (bar, Watt, volt, etc) are derived.

Hi Funkworks,

Good for you. I agree with what you just said. Thank you for the nice explanation.

Just to be clear then, I suggest reporting barometric pressure in atmospheres. I think I suggest that. I am not a meteorologist, so I would like someone to please confirm that the suggestion makes sense.

Stanley

 

[Funkworks]

Hi Funkworks,

Good for you. I agree with what you just said. Thank you for the nice explanation.

Just to be clear then, I suggest reporting barometric pressure in atmospheres. I think I suggest that. I am not a meteorologist, so I would like someone to please confirm that the suggestion makes sense.

Stanley

I think it makes sense. So here's how it would work:

- The real reference has to be 0 (no force or pressure, as in space).
- From 0 Pa, you measure atmospheric pressure over many years and find a average value in Pa. What they found is that 101325 Pa is a good average pressure to pick as a reference, and they defined 1 atm to be just that: 1 atm = 101325 Pa.
- From that 1 atm meteorological reference, you can report percentages of 1 atm, daily.