Outside Temperature Change

[Altocumulus]

This ought to be a logical trend showing on localhost displays : Dashboard and Graphs. Legend says C/hr.....

This morning I had a positive graph showing negative trend?

0754 : 0.9C
0854 : 3.0C Steady rise : Trend -0.7C/hr
0914 : 3.1C Trend 0.5C/hr

Wiki suggests temptrend is over the last 3 hourly period which, at 0945, gives a more realistic value at 0.6C/3hr......

Am I missing something obvious?

Geoff

[Altocumulus]

At 10:44

3.0C has a trend of 0.7C/hr which is neither a one hour change (0.0) or a three hour change (+2.1).

[mcrossley]

A 0.7C/hr trend = 2.1C/3hr ???

[KarlS]

@Altocumulus: see also the short discussion @ viewtopic.php?f=40&t=19550

[Altocumulus]

Thanks Both.

[Altocumulus]

Thanks Both.



Yes, but, A 0.7C/hr trend = 2.1C/3hr ???

It was warming on a negative trend (I missed the "-" on the 0.7C)

[Altocumulus]

p.s. Why use an hourly pressure change in the gauges and tables when Cumulus returns a 3 hourly value?

WMO standard is for a 3 hourly trend.

It is possible to also use the standard 9 pressure trend descriptions, especially on the gauges?

[mcrossley]

p.s. Why use an hourly pressure change in the gauges and tables when Cumulus returns a 3 hourly value?

WMO standard is for a 3 hourly trend.

MX does use a 3 hour change value everywhere?

It is possible to also use the standard 9 pressure trend descriptions, especially on the gauges?

What trend descriptions are these, and for which country? CMX uses the phrases adopted by the UK Met Office - minus the "now" as CMX does not do the comparison against the previous rate.

You can always set your own trend descriptions in MX using strings.ini

[Altocumulus]

Localhost display.

Dashboard hPa/hr
Hover - over pressure gauge, trend is hPa/hr
Current data table trend hPa/hr

Why take the 3 hour trend and change to a one hour?

9 pressure tendency descriptions over 3 hours: codes 0 to 8 in the synop, used by all Countries.

[mcrossley]

Ah, you mean why report the three hour change as an hourly rate.

The short answer - because it's always been like that in Cumulus!

Longer answer is I do not know the history, but is probably because I think most people would expect a rate to be in units/hour, not in units/3*hours.
If you use units/3*hours you might as well just report the absolute change rather than the rate of change.

[sfws]

Your subject title says "outside temperature change", yet you seem now to be focussed on "pressure change". EDIT: Ideally you, or administrators, should change subject on first post, as the subject there is what is used to identify what topic is about for other readers.

For original post on Cumulus pressure trend text, based on three hour change see viewtopic.php?p=14832#p14832, it has been repeated in many topics ever since as many people have asked about that.
For the pressure change text from Davis stations, based on three hour change, see viewtopic.php?p=83961#p83961

Steve Loft, see his past posts at both viewtopic.php?p=54837#p54837 and viewtopic.php?p=72748#p72748 knew that officially the pressure change over 3 hours was reported. It is known he was implementing the rain rate (where he expresses change between two successive rain counts as an hourly rate) at the same time, so it is likely as Mark says that he assumed people wanted a rate in "unit/hour". EDIT: Or because Cumulus was written in his "leisure" time basically for Steve's own use, (although made available to others), it may be because as Steve was not a professional meteorologist he implemented what suited him!

Steve used to host this support forum, but he EDIT: lost some old posts, when he changed to a new version of the software running his support forum. Therefore there is no documentation still existing from the time when Steve introduced pressure change rate EDIT: if he did once explain why he selected rate rather than absolute change. It is however recorded in several places that for temperature, Steve selected a rate expressed over one hour based on three hour change because that was how he had done it for pressure. There was at least one enhancement request (all those were lost by Steve Loft) for more alternative periods for pressure, and temperature, these were never implemented before he retired from Cumulus, apart from that he later added web tag <#TempChangeLastHour> for actual change over one hour for temperature.

Mark EDIT: has already added web tag <#PressChangeLast3Hours> for actual absolute pressure change over 3 hours (mentioned at end of his recent post) - see viewtopic.php?f=40&t=19328 topic. So although you don't see that either in the supplied MX (dashboard) interface, or in the default web site, you can choose on your own web page to have it displayed.

Note: Edits above added after this post was quoted in full in later post by originator of topic, because the reply there suggests parts of my original text were misunderstood.

[mcrossley]

@sfws Thanks for the update and old post links

[Altocumulus]

Ah, you mean why report the three hour change as an hourly rate.

The short answer - because it's always been like that in Cumulus!

Longer answer is I do not know the history, but is probably because I think most people would expect a rate to be in units/hour, not in units/3*hours.
If you use units/3*hours you might as well just report the absolute change rather than the rate of change.

Chuckles!

I accede that most people would expect a rate to be in units per hour - but traditionally pressure has been reported and analysed as per 3 hours. Every practising meteorologist around the world uses this as a standard, with the 9 types of change one steady and 4 each of rising/falling. But water beneath the bridge.

Could it be implemented through CMX for those of us who would like to use it?

[Altocumulus]

Your subject title says "outside temperature change", yet you seem now to be focussed on "pressure change".

For original post on Cumulus pressure trend text, based on three hour change see viewtopic.php?p=14832#p14832, it has been repeated in many topics ever since as many people have asked about that.
For the pressure change text from Davis stations, based on three hour change, see viewtopic.php?p=83961#p83961

Steve Loft, see his past posts at both viewtopic.php?p=54837#p54837 and viewtopic.php?p=72748#p72748 knew that officially the pressure change over 3 hours was reported. It is known he was implementing the rain rate (where he expresses change between two successive rain counts as an hourly rate) at the same time, so it is likely as Mark says that he assumed people wanted a rate in "unit/hour".

Steve used to host this support forum, but he did not keep old posts, when he changed to a new version of the software running his support forum. Therefore there is no documentation still existing from the time when Steve introduced pressure change rate. It is however recorded in several places that for temperature, Steve selected a rate expressed over one hour based on three hour change because that was how he had done it for pressure. There was at least one enhancement request (all those were lost by Steve Loft) for more alternative periods for pressure, and temperature, these were never implemented before he retired from Cumulus, apart from that he later added web tag <#TempChangeLastHour> for actual change over one hour for temperature.

Mark added web tag <#PressChangeLast3Hours> for actual absolute pressure change over 3 hours (mentioned at end of his recent post) - see viewtopic.php?f=40&t=19328 topic. So although you don't see that either in the supplied MX (dashboard) interface, or in the default web site, you can choose on your own web page to have it displayed.

The thread moved onto pressure as part of the discussion regarding 1 hour or 3 hour reporting of elements. As a meteorologist I never saw change of temperature reported in this way, but pressure in 3 hours is a standard measure; I was 'just' querying why, if C1 reported 3 hourly the table and gauges show 1 hour - seems odd. All weather charts show at the time of observation, "instant" elements except the pressure trend.

But water beneath the bridge....

[mcrossley]

I was 'just' querying why, if C1 reported 3 hourly

As far as I know C1 used the same as MX, the three hour change expressed as a rate per hour?

Capture.jpg