Pressure with trend

[Holli]

The pressure value alone says nearly nothing. Important for the upcoming weather is the pressure trend.

I know that in the METAR information there is no trend information, but users with their own PWS can set a value for the trend in the last 3 or 6 hours. So it would be nice to have an ascending or descending arrow behind the pressure value if a trend value is given.

[daddylito]

Hi Holli!

I find this to be a good idea!

But if there is only 3 variants of the value (up-arrow, down-array and no arrow) then
arrows should only be used when pressure is pretty rapidly changing!
But there is also the possibility to have more than 3 variants of the value!

I guess it is possible for YoWindow to collect and save pressure values from the METAR,
so when YoWindow have been running a while, a trend for the METAR would be possible to display.

For the forecast data it´s also possible to compare the pressure between different
time intervals in order to find any trends.

Perhaps it should at least be 4 variants of the value, where no arrow represents that
YoWindow lacks data to determine any pressure changes. The fourth value could for instance be
an arrow pointing at right.

[gemini06720]

Indeed trend arrows (or other symbols) should be used, not only for the pressure but also for the temperature.

And, as already mentioned, there should be 4 arrows/symbols:
- one up arrow/symbol
- one down arrow/symbol
- one horizontal bar for when the trend is fixed
- one symbol to indicate the lack of trend information

[Holli]

And, as already mentioned, there should be 4 arrows/symbols:
- one up arrow/symbol
- one down arrow/symbol
- one horizontal bar for when the trend is fixed

Flash is not delimited to fixed symbols. The angle of the arrow can symbolize the speed of the increasing or decreasing. So one arrow is enough.

- one symbol to indicate the lack of trend information

That's easy: no information, no arrow.

[daddylito]

Hi Holly!

Fascinating idea that the angle of the arrow should represent
the speed of the pressure change.

I like it!

But, my guess is that Pasha would like to have a
definition of how to determine the angle.
A pretty rapidly change would be perhaps 60 degrees.
Extreme changes and close to extremes are 90 degrees.
You shouldn´t exceed 90 degrees! (Or?)

[Holli]

You shouldn´t exceed 90 degrees! (Or?)

No, 90 deegrees must not be exceeded. The best way would be using a tangens function, but I don't know if this is possible in the programming language Pawel is using.

2nd best is to use a limit of maybe 6hPa/3h and all values between are proportional to this. That means 3hPa/3h will give a raising or falling arrow of 45 deegrees, 1hPa/3h an arrow with 15 deegrees.

If the limit is set to 9hPa/3h the arrow would be at 30 and 10 deegrees. I think, this limit must be enough. If the pressure is raising or falling more rapid, it doesn't matter to the weather. It will be as stormy as it can be.

[daddylito]

Hi Holli!

If I understand you right, I find the second version most suitable.
A non proportional handling would certainly confuse many! (i.e. tangens)

And it´s also nice if one easily can see a significant pressure change,
even if it´s not extreme! (The second seems better than third)

But please convince me that tangens are right except for the
reason that it can handle all situations, without limits!

[par]

Guys, I really like the idea of picking the pressure trend.
I even want to add this feature as soon as I can.

Though I need your help before I start coding this.

What is the right interval to compare the pressure between now and the past?
1 hour?
3 hours?
6 hours?

Consider this scenario.
11:00 - 1010 hPa
12:00 - 1015 hPa
13:00 - 1020 hPa
14:00 - 1015 hPa
15:00 - 1010 hPa (Now)

So what's the pressure trend in this case?


Holli, yes, I can use argtangent function. It is available in Acion Script - YoWindow is written in it, and in all programming languages I know about.
But I like the variant with the limited value.

[tph06784]

The National Weather Service defines pressure trend as[quote]Pressure Change
The net difference between the barometric pressure at the beginning and ending of a specified interval of time, usually the three hour period preceding an observation./quote]

[Holli]

What is the right interval to compare the pressure between now and the past?
1 hour?
3 hours?
6 hours?

I don't know if there is a "right" interval. I think 3h is a reasonable interval, but I don't know if every PWS software is able to give this value. If it is fixed to 6 or 12 hours, it would be best to set the interval by an attribute.

Consider this scenario.
11:00 - 1010 hPa
12:00 - 1015 hPa
13:00 - 1020 hPa
14:00 - 1015 hPa
15:00 - 1010 hPa (Now)

This is a normal scenario when a high or low passes by. For the trend it doesn't matter. The trend is always the difference from the begin to the end of the interval.

So what's the pressure trend in this case?

For 3h -5hPa = -1.66hPa/h. It will be quite windy

Holli, yes, I can use argtangent function. It is available in Acion Script - YoWindow is written in it, and in all programming languages I know about.

Then I would use atan (hPa/h). Small diffences are shown clearly, but very great diffences never reach 90°.

But I like the variant with the limited value.

Then I would set the limit to 3hPa/h, and all values between proportional to this. But as said above, this is the 2nd best solution.