Author Topic: Strange temperature data. from Gladstonefamily  (Read 934 times)

0 Members and 1 Guest are viewing this topic.

Offline OldAlaskaGuy

  • Alaska, The Last American Frontier
  • Forecaster
  • *****
  • Posts: 399
  • Living in the "Last American Frontier"
Strange temperature data. from Gladstonefamily
« on: April 07, 2020, 07:37:35 PM »
Strange temperature data. Here are my data quality charts from my CWOP. Note the temperature graph. This appears to be an aberration. This does not show on any of my other dashboards. Any ideas? Also, the pressure, should it be closer to the redline? Is that the expected pressure. Is a calibration in order? I have calibrated it in reference to my local airport but it is a few hundred feet higher.
https://weather.gladstonefamily.net/cgi-bin/wxqchart.pl?site=F7065
« Last Edit: April 07, 2020, 07:51:54 PM by Weatheroger »

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: Strange temperature data. from Gladstonefamily
« Reply #1 on: April 07, 2020, 09:18:20 PM »
What you see in Gladstonefamily.net should be taken lightly. It isn't the best running analysis site. You know your data. You should have other places to look at your data and see if any anomaly is occurring elsewhere or not. Sometimes strange things show up on Gladstonefamily.net and it isn't yours stations fault.

Also don't take the analysis red line as gospel. It is only as good as your neighbor stations....and even then the site glitches on its own even when neighbors are good.

Ignore the wind analysis as it hasn't been working for almost a year. Who knows when or if it will get fixed.

I recommend that you consider MesoWest for better trusted analysis.

That said your barometer does not seem to be calibrated for Absolute pressure. There are two parts to the barometer. There is Absolute and there is Relative pressure. Although your Relative pressure may seem right to you on your console and on other online services when you upload to CWOP you need to also be certain that your Absolute pressure is correct for CWOP. Most people ignore their Absolute pressure. When you upload to CWOP the software (AmbientCWOP.com in your case) uses your Absolute pressure and applies a formula to that value to come up with your Altimeter pressure. If your Absolute value is incorrect then you see the bad results you are experiencing. It doesn't matter that your Relative pressure is correct....because that isn't used for CWOP uploads. Sure you may wonder why doesn't it use that value which you have already set? The answer is because Sea Level Pressure is different than Altimeter pressure (though they are close). CWOP is one of the few sites that requests Altimeter rather than Sea Level Pressure be sent to them. Your weather station does not have a way to calculate Altimeter....this must be done by the CWOP software (AmbientCWOP.com in your case). To calculate this Altimeter value AmbientCWOP needs to use Absolute pressure not Relative pressure. That is why Relative pressure is ignored.

So how do you set your Absolute pressure and what does Absolute pressure mean? Well there is no other direct source that you can use to directly calibrate your Absolute pressure, so it needs to be done by way of reasoning what this value represents. Absolute pressure is the raw pressure of your barometric sensor at your given elevation. The trouble is that your station hardware doesn't have a way to easily give it your elevation. It needs to know your elevation in order for the Absolute pressure to be correct.

There is a 3 way relationship between 3 values....Absolute Pressure, Relative Pressure and your Elevation. If you know any two values then you can deduce via a calculation what the missing last value is. Therefore what is true for all of us is that we know two things....we know what Relative pressure is by way of looking up the local METAR and we know our elevation by way of map survey topo maps. With these two values we can calculate what your Absolute pressure should be at any given moment. But once you set it, it is done and you never need to adjust it again....because your elevation doesn't change...you are permanently at that location...at a given elevation.

So what you have to do is understand the relationship between these 3 things and then use the method that I have posted (in other linked thread below) in order to properly calibrate both your Altimeter and your Relative pressure. Basically to sum all of this up....the difference between your Relative pressure and your Absolute pressure is indicative of your elevation....this difference is static...it never changes. If you properly dial in both Absolute and Relative pressure then you basically have informed your station what your elevation is, even though the station isn't really asking you for your elevation.

It really is a shame that Fine Offset clone stations are so complicated when it comes to this simple matter. If they could have only changed the interface to ask for your elevation and then they did the math and figured out what your Absolute pressure was based on that elevation this would be a simple matter for anyone. But with the services that Fine Offset upload to they only use Relative pressure and so this doesn't affect most users. But here you are attempting to upload to CWOP and if you want to do this properly then you have to go through this added effort.

Please see this other thread for my calibration method to properly set both Absolute and Relative pressure:
https://www.wxforum.net/index.php?topic=39088.0

PS - I hear this all the time...people say how their local airport (METAR) is at a different elevation than them. Well that fact does not matter. The whole purpose of Sea Level Pressure equivalence is how meteorologists are able to compare different locations even when they are at different elevations. That is the whole purpose of Sea Level Pressure...it equalizes every location as if everyone was at the same elevation...it basically reports a pressure that is what it would be if you were at sea level. If everyone then compares their sea level pressures then you have equalized everyone and taken elevation out of the picture. I've simplified this concept to get the bigger educational benefit through. There really is a lot more to this...but you have to start somewhere. Bottom line is that it doesn't matter that your local airport (METAR) is at a different elevation. It also doesn't matter how far that METAR is....that does require a bit of extra knowledge to get things right and it deals with understanding isobars. If you want to learn about this advanced method then read that other thread and after you get those basic concepts down you'll see that in there is another link to the isobar post for taking things to the next level.


« Last Edit: April 07, 2020, 09:53:30 PM by galfert »
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline OldAlaskaGuy

  • Alaska, The Last American Frontier
  • Forecaster
  • *****
  • Posts: 399
  • Living in the "Last American Frontier"
Re: Strange temperature data. from Gladstonefamily
« Reply #2 on: April 08, 2020, 04:57:12 PM »
Here is what I came up with. Station elevation is 864.2 M, difference between 1013.25 (sea level) - 915.46 (from calculator) = 97.79hPa = 2.88 in Hg In the GW1000B calibration page where does this value go rel offset or abs offset. I currently have the offset at 2.84 in Hg. Current conditions at KOGO Metar at the time of calculation were 29.93 inHg = 1013.5 in Hg
Thanking you in advance for your help.

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: Strange temperature data. from Gladstonefamily
« Reply #3 on: April 08, 2020, 05:29:16 PM »
On the GW1000 the 97.79 hPa or 2.89 inHg gets entered into Rel Offset.  You forgot to round and just dropped significant digits on this conversion from hPa to inHg...it is 2.89 and not 2.88.

But I would really recommend you don't use inHg units. In WS View go to Menu / Unit Settings / and change Pressure to use hPa units. Do so just for calibration purposes....later you can switch back to inHg...it is better to enter the offset as hPa units as that is what your device uses internally and hPa with one decimal place is more precise than inHg with 2 decimal places. It just is.

Then review how far off your Relative pressure shows on the Live Data dashboard. Then determine by how much you need to lower or increase. However much that is that you need to change it by is by how much you then adjust the Abs Offset. The reason is because you affect Relative by actually changing the Absolute.
« Last Edit: April 08, 2020, 05:32:20 PM by galfert »
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline OldAlaskaGuy

  • Alaska, The Last American Frontier
  • Forecaster
  • *****
  • Posts: 399
  • Living in the "Last American Frontier"
Re: Strange temperature data. from Gladstonefamily
« Reply #4 on: April 08, 2020, 07:15:53 PM »
Almost there. On live data ( changed to hPa) Relative pressure now reads 1014.9 hPa, KOGO METAR now reads 1013.2 hPa (29.92 inHg) 1.7 hPa difference. Now, do I enter -1.7 hPa in the Abs offset?
« Last Edit: April 08, 2020, 07:33:10 PM by Weatheroger »

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: Strange temperature data. from Gladstonefamily
« Reply #5 on: April 08, 2020, 08:22:23 PM »
No. You are getting ahead of yourself. We don't need to worry about the METAR yet.

There are two steps.
Step 1. - Dial in the difference between Relative and Absolute by using your elevation. We don't care about the METAR just yet. You are still at this first step. This difference is your Rel Offset on the GW1000. For your elevation you enter 97.8 hPa into Rel Offset. Then you never ever ever touch Rel Offset. Because 97.8 hPa is indicative of your elevation. Your elevation doesn't change...neither should your Rel Offset because they mean the same thing.  That is unless you move your console between floors and change its elevation, because the barometric sensor is in your console (it is inside of the GW1000).

Step 2. After you have dialed in your Rel Offset (97.8 hPa) based on your elevation. Then you can compare your Relative to the METAR and see how much of an adjustment you need to make. BUT the adjustment is not done to the Relative...it is done the the Absolute...not where the Absolute then is made to match the METAR but what is actually being done is to look at by how much your Relative reading needs to go up or down and then that is by how much you adjust Absolute. Because however much you adjust Absolute by will also be reflected in your Relative.
« Last Edit: April 08, 2020, 08:36:45 PM by galfert »
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline OldAlaskaGuy

  • Alaska, The Last American Frontier
  • Forecaster
  • *****
  • Posts: 399
  • Living in the "Last American Frontier"
Re: Strange temperature data. from Gladstonefamily
« Reply #6 on: April 08, 2020, 08:36:10 PM »
Sorry to confuse you. I have entered 97.8 into Rel Offset and saved. And the difference in the METAR is in my previous post. So the ABS offset of -1.7 goes in the ABs offset?

« Last Edit: April 08, 2020, 08:38:54 PM by Weatheroger »

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: Strange temperature data. from Gladstonefamily
« Reply #7 on: April 08, 2020, 08:44:18 PM »
Okay then...we are ready for Step 2....

Look at Abs Offset and whatever it says there subtract 6.8 hPa from it. This will be just an initial adjustment. We may need to revisit and fine tune later. Any subsequent fine tuning will only be done changing Abs Offset.  We never touch Rel Offset ever again as that is your elevation.

If you want to know where I got 6.8 hPa from, I looked at your current CWOP data and saw by how much it differed from KGOO.

Your data shows 1019.8 hPa (at this moment)
http://www.findu.com/cgi-bin/raw.cgi?call=FW7065

KGOO METAR shows 1013 hPa (at this moment)
https://aviationweather.gov/metar/data?ids=KGOO&format=decoded&date=&hours=0

The difference is 6.8 hPa between you and the local METAR. So you are that much too high. You fix that by reducing your Abs Offset by that much than what it currently shows. In other words you don't enter in -6.8 hPa into Abs Offset unless it was currently showing 0. For example if your Abs Offset now before adjusting shows 5.4 hPa then you substract 6.8 hPa from that and that make you new Abs Offset -1.4 hPa.
« Last Edit: April 08, 2020, 08:59:13 PM by galfert »
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline OldAlaskaGuy

  • Alaska, The Last American Frontier
  • Forecaster
  • *****
  • Posts: 399
  • Living in the "Last American Frontier"
Re: Strange temperature data. from Gladstonefamily
« Reply #8 on: April 08, 2020, 09:52:26 PM »
All data entered. I will wait for a while then review. Again, thank you. It is very confusing for the novice. I do thank you again for your time and patience.

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: Strange temperature data. from Gladstonefamily
« Reply #9 on: April 09, 2020, 06:08:16 AM »
Using WS View for GW1000 (hPa):
What are the following current offsets
Abs Offset ____
Rel Offset ____

At the top of the hour (any hour) take the following readings and let me know what these are:

Using WS View Live Data view for GW1000 (hPa):
Absolute pressure ____
Relative pressure ____

http://www.findu.com/cgi-bin/wxpage.cgi?call=FW7065
Pressure at top of this page for FW7065 in mb ____

https://aviationweather.gov/metar/data?ids=KGOO&format=decoded&date=&hours=0
Altimeter pressure for KGOO in mb ____
« Last Edit: April 09, 2020, 06:15:21 AM by galfert »
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline OldAlaskaGuy

  • Alaska, The Last American Frontier
  • Forecaster
  • *****
  • Posts: 399
  • Living in the "Last American Frontier"
Re: Strange temperature data. from Gladstonefamily
« Reply #10 on: April 09, 2020, 10:03:59 AM »
Using WS View for GW1000 (hPa):
What are the following current offsets
Abs Offset  -6.8 hPa
Rel Offset   97.8 hPa

At the top of the hour (any hour) take the following readings and let me know what these are:

Using WS View Live Data view for GW1000 (hPa):
Absolute pressure  912.4 hPa
Relative pressure 1010.1 hPa

http://www.findu.com/cgi-bin/wxpage.cgi?call=FW7065
Pressure at top of this page for FW7065 in mb 1015.0 mb

https://aviationweather.gov/metar/data?ids=KGOO&format=decoded&date=&hours=0
Altimeter pressure for KGOO in mb 1015.7 mb

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: Strange temperature data. from Gladstonefamily
« Reply #11 on: April 09, 2020, 10:10:23 AM »
Okay we are getting there. A little fine tuning required....

Change Abs Offset to -6.1 hPa.
This will add +0.7 hPa to current Abs Offset.
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline OldAlaskaGuy

  • Alaska, The Last American Frontier
  • Forecaster
  • *****
  • Posts: 399
  • Living in the "Last American Frontier"
Re: Strange temperature data. from Gladstonefamily
« Reply #12 on: April 09, 2020, 10:22:08 AM »
Done. I assume that this will take a few days to sort out through the QC monitoring. Thank you. It is interesting to watch the pressure waves passing. Thank you again.

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: Strange temperature data. from Gladstonefamily
« Reply #13 on: April 09, 2020, 10:41:51 AM »
Yes, it will take some time for QC to catch up.

For high elevations such as you have this is a much more difficult task. I'll keep an eye on it and see if further fine tuning is required. But do realize that it won't be perfect.
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline OldAlaskaGuy

  • Alaska, The Last American Frontier
  • Forecaster
  • *****
  • Posts: 399
  • Living in the "Last American Frontier"
Re: Strange temperature data. from Gladstonefamily
« Reply #14 on: April 09, 2020, 10:51:32 AM »
Yes, it will take some time for QC to catch up.

For high elevations such as you have this is a much more difficult task. I'll keep an eye on it and see if further fine tuning is required. But do realize that it won't be perfect.
Thanks. Since this stay home policy I have taken a renewed interest in data quality. It is not the best station but I am now thinking of upgrading to a better one and possibly setting up a wether page of my own. Thank you and stay safe.

 

anything