Pressure trends and solar tide compensation

[adrianj]

A search did not turn up anything about this.

Does the pressure trend number take account of the daily solar atmospheric tide pressure variation ? Typically it amounts to about 2 HpA over 6 hours, so about 0.3 Hpa per hour. Not much, but it does influence what you see when tracking a slow oncoming front. For instance the high tidal pressure here happens about 9:30 am, so before this I see the pressure increasing, and after it decreases. But this has nothing to do with the weather.

I saw a couple of references to taking the trend over either 1 or 3 hours, but nothing about actually compensating for the tidal changes.

Any comments ?

[steve]

Cumulus doesn't do anything clever with the pressure trend - it just uses the pressure readings supplied by your weather station.

[adrianj]

Thanks Steve,

Thats what I thought from the displayed effects. I might play with some pressure trend analysis and run some Fourier transforms, to see if I can 'improve' the figures.

Great software ! And really easy to get going with the WH3081 station I bought.

I had an Acurite station before this, and while the hardware was good, their logging and reporting software was abysmal, never did get it going properly, after a month or so of fooling around.

[freddie]

A search did not turn up anything about this.

Does the pressure trend number take account of the daily solar atmospheric tide pressure variation ? Typically it amounts to about 2 HpA over 6 hours, so about 0.3 Hpa per hour. Not much, but it does influence what you see when tracking a slow oncoming front. For instance the high tidal pressure here happens about 9:30 am, so before this I see the pressure increasing, and after it decreases. But this has nothing to do with the weather.

I saw a couple of references to taking the trend over either 1 or 3 hours, but nothing about actually compensating for the tidal changes.

Any comments ?

Tidal changes are heat-related and, as such, are most pronounced in the Tropics. In other latitudes tidal changes can quite often be seen in the summer half of the year, but are easily dwarfed by pressure changes forced by atmospheric motion.
Over most of the globe the commonly used pressure trend time period is three hours - this period of time is not random, it is directly related to dynamical metoerological theory. To cancel out the effect of tidal changes in regions where they are dominant (i.e the Tropics) the pressure trend time period in these regions is 24 hours. This is the practice applied by professional meteorologists.

Maybe Steve could offer a user-defined pressure trend time period to cater for users who live in/near the Tropics? If so, then I would assume this would be offered for Cumulus MX but not Cumulus 1.

[adrianj]

Hmm, I dont doubt that the meteorologists know what they are doing, but if the pressure changes are temperature related, how come I see two peaks per 24 hours.

I always thought ( from one of the first number crunching exercises I did way back in Uni days ) that the pressure change was due to gravitational effects, mostly from the sum, as distinct from the sea tides, which are mostly from the moon. But still that means there will be two tides per day, just as I see.

See attached file for the last few days.

[freddie]

We sure do know what we're doing
This Wikipedia article tells you a lot about atmospheric tides. I have linked directly to the bit about why the period of the heating-driven pressure oscillation is not 24 hours.
https://en.wikipedia.org/wiki/Atmospher ... eric_tides

[adrianj]

Thanks freddie. I got to the Wiki too. Should probably have gone there first. Still find it odd that there should be such a big 12 hour component, but I am an electronics engineer, not a meteorologist.

Having read the relevant bit, I still dont 'get it'.

Even if the heating is roughly square wave - and I can see why that is reasonable - there should be little or no 2nd harmonic content, and plenty of first ( 24 hour ) harmonic, and the odd harmonics at 1/3, 1/5 etc of that.

Delving a bit further into ancient theory, I can see that if the atmosphere has a 'natural' period of 12 hours, then forcing it every 24 hours may still produce 12 hour harmonics. I am happy.

[jdc]

That's not your central heating, is it?

[adrianj]

Er, no. Dont need central heating here, unlike your local temperature !. See attached for temps for the same period
The pressure sensor is inside, of course, so it only sees about +/- 1 deg change here. Insulation does that. You can see on the last day that I left the windows and doors open, so we get a bit higher temp inside.