Author Topic: Quality Checking on GladStoneFamily Analysis Wrong? - (MADIS is working on it!)  (Read 2783 times)

0 Members and 1 Guest are viewing this topic.

Offline danoh

  • AmbientCWOP.com
  • Contributor
  • ***
  • Posts: 103
    • AmbientCWOP
I don't know if Philip even maintains this anymore, but I'm sure someone here is smarter than I am.

It seems the analysis reference for barometer is way off for my station. It jumps sometimes +/- 3 mb

It's been like this for months.

My baro is calibrated and matches the local civilian airport and local military base, but gladstone site's analysis jumps all over the place causing spatial consistency errors.  If I adjust my baro to make gladsone happy, it's way higher than official stations nearby.

Should I just ignore that the quality check on gladstone is telling me, and go with what I know - or is there something else going on that I haven't figured out yet?

Reference links:
https://weather.gladstonefamily.net/site/F4363?lang=en
http://www.findu.com/cgi-bin/wxpage.cgi?call=FW4363
https://www.aviationweather.gov/metar/data?ids=kvld&format=raw&hours=0&taf=off&layout=on
https://www.aviationweather.gov/metar/data?ids=kvad&format=raw&date=&hours=0


« Last Edit: June 27, 2019, 09:50:10 AM by danoh »
Ambient Weather station CWOP connector: AmbientCWOP.com

Offline lightmaster

  • MB-HWS Template Dev
  • Forecaster
  • *****
  • Posts: 324
    • Baxley,GA Weather
Re: Quality Checking on GladStoneFamily Analysis Wrong?
« Reply #1 on: June 27, 2019, 09:39:04 AM »
Mine does that to, the analysis making impossible swings up and down. I've talked with MADIS support and seems to be an issue on their end. Last I heard the guy working on it is waiting to start a 30 day test of some new code before he can put it out live to correct some issues with the analysis.

Long story short, if you know your station is good, then don't worry about the analysis line right now. MADIS is working in a fix for it.

Offline danoh

  • AmbientCWOP.com
  • Contributor
  • ***
  • Posts: 103
    • AmbientCWOP
Re: Quality Checking on GladStoneFamily Analysis Wrong?
« Reply #2 on: June 27, 2019, 09:40:06 AM »
That's good news.  Thanks!
Ambient Weather station CWOP connector: AmbientCWOP.com

Offline lightmaster

  • MB-HWS Template Dev
  • Forecaster
  • *****
  • Posts: 324
    • Baxley,GA Weather
Re: Quality Checking on GladStoneFamily Analysis Wrong?
« Reply #3 on: June 27, 2019, 09:41:59 AM »
I thought your graph looked eerily similar to mine, before realizing you're from Valdosta, like 2 hours from my place. Check my graphs in the CWOP link in my sig.

Offline danoh

  • AmbientCWOP.com
  • Contributor
  • ***
  • Posts: 103
    • AmbientCWOP
Re: Quality Checking on GladStoneFamily Analysis Wrong?
« Reply #4 on: June 27, 2019, 09:45:13 AM »
I was just through Appling the other day  [tup]
Ambient Weather station CWOP connector: AmbientCWOP.com

Offline lightmaster

  • MB-HWS Template Dev
  • Forecaster
  • *****
  • Posts: 324
    • Baxley,GA Weather
Re: Quality Checking on GladStoneFamily Analysis Wrong?
« Reply #5 on: June 27, 2019, 09:47:43 AM »
And I was at Wild Adventures last Saturday, lol.

Offline danoh

  • AmbientCWOP.com
  • Contributor
  • ***
  • Posts: 103
    • AmbientCWOP
Re: Quality Checking on GladStoneFamily Analysis Wrong?
« Reply #6 on: June 27, 2019, 09:48:41 AM »
got your taste of tequila  :lol:
Ambient Weather station CWOP connector: AmbientCWOP.com

Offline BKS97

  • Senior Contributor
  • ****
  • Posts: 173
    • Sartell East Weather
Re: Quality Checking on GladStoneFamily Analysis Wrong?
« Reply #7 on: June 27, 2019, 10:30:18 AM »
Long story short, if you know your station is good, then don't worry about the analysis line right now. MADIS is working in a fix for it.

I quit sending to CWOP because wildly-inflated barometer readings resulted in a red X on the Gladstone page.  My RainWise MK-111 tracks closely to the airport altimeter readings in my area day and day out, and other networks like PWSweather, WU, AWEKAS report the readings accurately.  So I am glad to learn there may be a problem that MADIS is trying to fix, because I would like to send my data to CWOP.

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
I would not stop sending to CWOP because of issues with Gladstone. They are separate sites maintained by different people/organizations. There are numerous benefits to send to CWOP. I know many CWOP users that don't bother with Gladstone ever. Gladstone is just pulling the data and showing you pretty graphs and some analysis from MADIS...so this is a separate and extra thing...it isn't integral. Lots of people find Gladstone less than useful. I find that Gladstone only works well when your area has more quality CWOP neighbors. If you quit and your neighbor quits then you mutually hurt each other. Stay onboard and then see what happens a few months or a year down the line, or encourage a friend or family to join you with their station.

There are other ways to check your QC for CWOP data. You can go to Mesowest and find links to both their QC and to Synoptic Labs:
https://mesowest.utah.edu/cgi-bin/droman/regress_database_top.cgi?stn=F3708
https://synopticlabs.org/demos/network-qc/?display_interval=3600&status=active&showemptystations=0&recent=2880&radius=F3708,25

Or you can just look at the Mesowest tabular data for the QC column:
 [ You are not allowed to view attachments ]

You can go to MADIS Surface Map and find your station and see the QC result codes:
https://madis-data.ncep.noaa.gov/MadisSurface/

Windy.com is another place where your data shows up automatically and you can see how your data compares there to the different forecast models. This is nice as many people have struggled with uploading directly to Windy.com and by having a CWOP upload they don't have to do anything extra (although there are some benefits to direct to Windy.com upload).

Lastly consider that by not sending to CWOP then you are missing out on your data showing up at all these great resources:
https://www.wxforum.net/index.php?topic=35089.0
« Last Edit: June 27, 2019, 12:16:04 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 danoh

  • AmbientCWOP.com
  • Contributor
  • ***
  • Posts: 103
    • AmbientCWOP
Re: Quality Checking on GladStoneFamily Analysis Wrong?
« Reply #9 on: June 27, 2019, 11:20:38 AM »
Long story short, if you know your station is good, then don't worry about the analysis line right now. MADIS is working in a fix for it.

I quit sending to CWOP because wildly-inflated barometer readings resulted in a red X on the Gladstone page.  My RainWise MK-111 tracks closely to the airport altimeter readings in my area day and day out, and other networks like PWSweather, WU, AWEKAS report the readings accurately.  So I am glad to learn there may be a problem that MADIS is trying to fix, because I would like to send my data to CWOP.

Not quite sure why you stopped sending to CWOP because of GladStone, they're not mutually inclusive, but ok.
Ambient Weather station CWOP connector: AmbientCWOP.com

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
The bottom line is this. If you know your stuff is working and properly calibrated then don't be concerned. You should never chase analysis baselines. The analysis is just there as a quick reference. When I see my analysis errors I don't think my stuff has gone haywire. Rather I think that one of several things have gone wrong and then it it up to you to chase the culprit. The problem could be a neighbor or the problem could be just MADIS screwing up as it does sometimes (especially in less dense areas).

Even AWEKAS is not perfect for analysis. It seems with AWEKAS you just get an average from your neighbors. All it takes in one bad neighbor to throw things off. Recently I noticed that my AWEKAS humidity comparison was really off. I realized that my station was not at fault. Then I checked out all my AWEKAS neighbors and sure enough I found the problem station. I contacted this neighbor and let them know. With AWEKAS you need to act or they shut you down for whatever sensor fails. My AWEKAS neighbor told me that they were out of town till next month. I told them if they could remote in to shut it down preemptively. They did not have such remote capability set up. So sure enough AWEKAS within a day or two contacted them and warned them. Then a few more days later they shut the sensor down "permanently." With AWEKAS what they call permanently is really not permanently if you fix it and contact them to turn it back on. As soon as AWEKAS shut them down my humidity sensor came back to looking good with the remaining neighbors.

 [ You are not allowed to view attachments ]

I've done the same thing with other CWOP neighbors and helped them out when their stuff goes bad.
« Last Edit: June 27, 2019, 02:38:32 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 W Thomas

  • Welcome to my area!
  • Forecaster
  • *****
  • Posts: 1643
  • In Loving Memory Of Hooker The Weather Dog !
    • Smyth-Grayson Weather
I think the QC Analysis must still be seriously out of calibration. But I have made several changes to my station's location after moving and I just need to refresh my memory and set some things straight.

First of all I am using a Davis VP2 normally aspirated station. I also have a VUE console in the family room strictly for monitoring.
My data that goes to my WX software is fed from the Davis VP console.
I'm using Weather Display for all my translation and uploads.
I have the box ticked in the WD setup that indicates it uses Altimeter data for CWOP.

 My console is displaying just what the BARO sensor sees.  According to Davis ( just confirmed today), That is the correct setup. Looks terribly low in comparison to ly closest META site which is relatively close to my elevation.

I'm not convinced my QC checks are OK  but I just want to make sure my system is benchmarked as close to perfect as I can get it. I understand I believe my location to be in somewhat of a microclimate also.

The main thing I need to know is that I have CWOP data configured correctly in WD or is there an outboard program that will handle this data better?  The pressure that my closest METAR displays in quite different than what I am showing and I'm guessing it displays Altimeter at the METAR site.   It's been quite a while since I have tried to iron out discrepancies like this and my age and other factors make me doubt my calculations LOL!

Any ideas and or methods are appreciated!


WT

https://www.smythweather.net
CW8217 CWOP


     Best Regards
     Wayne

CWOP CW8217
KVAWHITE22 Wunderground   Davis VUE &  Davis Vantage Pro 2  /   Dedicated Server
GR Level 3 ,Level 2 AE Radars  Weather Display 10.37P  Mid Atlantic Weather Network Member
SkyWarn & Spotter Network 6092

Offline CW2274

  • Forecaster
  • *****
  • Posts: 6731
    • Conditions @ CW2274 West Tucson-Painted Hills Ranch
I have the box ticked in the WD setup that indicates it uses Altimeter data for CWOP.
Since you're using your VP2 console through WD for online, the first thing I'd do (if you haven't already) is see what your Vue console shows for the altimeter setting since that is exclusive to it. Just make sure you have your proper altitude MSL dialed in and that the Vue console is selected to REDUCED BAR ALT SETTING.

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
The main thing I need to know is that I have CWOP data configured correctly in WD or is there an outboard program that will handle this data better?  The pressure that my closest METAR displays in quite different than what I am showing and I'm guessing it displays Altimeter at the METAR site.   It's been quite a while since I have tried to iron out discrepancies like this and my age and other factors make me doubt my calculations LOL!

Any ideas and or methods are appreciated!

Your recent calibration is right on the money. If I was to nit pick you are still 0.2 hPa too high. I used MesoWest to determine this. But what you have is good enough. Then again I may be wrong and only because right now isobars are not perfectly straight between you and METAR. To make the 0.2 hPa correction you'd have to swtich to hPA if you are currently using inHg on your display. That is because with the VP2 console you can only correct down to 0.01 inHg but you can correct to 0.1 hPa which is more precise. The VP2 console does display inHg down to 0.001 but you can't adjust at that level.

 [ You are not allowed to view attachments ]
Very nice correction!

Yes ignore the red analysis line, as it is wrong usually. The reason it is wrong for many is because of your poorly calibrated neighbors. The key thing is to compare yourself to local METAR as I've done here with KGEV. It is also important to calibrate with distant METAR when isobar lines are the same. They are currently in a loop pattern and they aren't straight. So better to double check when they are straight isobar lines between you and the METAR.
« Last Edit: October 01, 2019, 08:37:19 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 W Thomas

  • Welcome to my area!
  • Forecaster
  • *****
  • Posts: 1643
  • In Loving Memory Of Hooker The Weather Dog !
    • Smyth-Grayson Weather
Thanks for the reassurance!  I just want my data to be as close to actual as possible. I thought my setup was correct but thought it worthwhile to make sure :)

Thanks again for all the information


     Best Regards
     Wayne

CWOP CW8217
KVAWHITE22 Wunderground   Davis VUE &  Davis Vantage Pro 2  /   Dedicated Server
GR Level 3 ,Level 2 AE Radars  Weather Display 10.37P  Mid Atlantic Weather Network Member
SkyWarn & Spotter Network 6092

Offline CW2274

  • Forecaster
  • *****
  • Posts: 6731
    • Conditions @ CW2274 West Tucson-Painted Hills Ranch
The main thing I need to know is that I have CWOP data configured correctly in WD or is there an outboard program that will handle this data better?  The pressure that my closest METAR displays in quite different than what I am showing and I'm guessing it displays Altimeter at the METAR site.   It's been quite a while since I have tried to iron out discrepancies like this and my age and other factors make me doubt my calculations LOL!

Any ideas and or methods are appreciated!
The VP2 console does display inHg down to 0.001 but you can't adjust at that level.
No, the console itself does not display to 0.001", it's resolution is only to 0.01". However, software such as WL does resolve to 0.001" and does adjust to it. Besides, forget the isobars and use the equipment one has and use the VUE. Simple.

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
So you are saying that with Weatherlink software it will talk to the console and adjust it? or are you saying that Weatherlink adjusts down to 0.001 inHg and then that is carried forward as data sent from logger but does not make its way back to the console? I'm trying to understand which way it is going.

I don't own a Davis but I've worked on several of my neighbor's Davis VP2 and I've not been able to calibrate better than 0.01 inHg.  But if I switch to hPa then yes I can calibrate down to 0.1 hPa and that is more precise. You want the console to be as precise as possible. Further corrections that are possible down the line with software if they don't make it back to the console although nice for reporting services purposes do not help the console.....unless you are saying that it does get back to the console.

Please clarify.
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
...oh and I don't know now exactly where (I can't recall at the moment) but I have seen 0.001 inHg resolution on the VP2 console somewhere. I think it had to do with the historical stuff.
« Last Edit: October 01, 2019, 10:26:27 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 CW2274

  • Forecaster
  • *****
  • Posts: 6731
    • Conditions @ CW2274 West Tucson-Painted Hills Ranch
I've should have been more clear. Both Vue and VP2 consoles calculate pressure to the foot, which is 0.001". However, neither console displays that resolution, only to 10 feet, which is 0.01". WL software displays the resolution that the consoles are capable of, to the foot, and can be adjusted accordingly. It's precision is just as good as hPa.

Offline CW2274

  • Forecaster
  • *****
  • Posts: 6731
    • Conditions @ CW2274 West Tucson-Painted Hills Ranch
...oh and a don't know now exactly where (I can't recall at the moment) but I have seen 0.001 inHg resolution on the VP2 console somewhere. I think it had to do with the historical stuff.
If so, I've never seen it.

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
I've should have been more clear. Both Vue and VP2 consoles calculate pressure to the foot, which is 0.001". However, neither console displays that resolution, only to 10 feet, which is 0.01". WL software displays the resolution that the consoles are capable of, to the foot, and can be adjusted accordingly. It's precision is just as good as hPa.

There are two adjustments to make with the VP2 console. First you dial in your elevation...to the foot. If that gives you 0.001 inHg resolution great. But if the raw sensor's station pressure is still not right then you have to go to that other adjustment. And there you can only adjust up or down by 0.01 inHg.  Therefore I say...switch to hPa and you can have greater precision. When you are done, then switch back to inHg if that is what you want displayed.

If Weatherlink software has adjustments beyond what the console sends out great....but not necessary if you dial in the console just right as I've described.
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline CW2274

  • Forecaster
  • *****
  • Posts: 6731
    • Conditions @ CW2274 West Tucson-Painted Hills Ranch
I've should have been more clear. Both Vue and VP2 consoles calculate pressure to the foot, which is 0.001". However, neither console displays that resolution, only to 10 feet, which is 0.01". WL software displays the resolution that the consoles are capable of, to the foot, and can be adjusted accordingly. It's precision is just as good as hPa.

There are two adjustments to make with the VP2 console. First you dial in your elevation...to the foot. If that gives you 0.001 inHg resolution great.   Therefore I say...switch to hPa and you can have greater precision. When you are done, then switch back to inHg if that is what you want displayed.

As I've stated, you can not dial in pressure to the foot on the console with inHg. That being said, I never use other than inHg, so if one's using the console alone, then hPa would be the better resolution.

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Agreed on what is and isn't possible on the console.

Thanks!


Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline CW2274

  • Forecaster
  • *****
  • Posts: 6731
    • Conditions @ CW2274 West Tucson-Painted Hills Ranch
What confuses me, if W Thomas wants to use the altimeter, why bother with all the other in-between crap and just use his Vue console capability as intended?  :???:

Offline W Thomas

  • Welcome to my area!
  • Forecaster
  • *****
  • Posts: 1643
  • In Loving Memory Of Hooker The Weather Dog !
    • Smyth-Grayson Weather
It's not so much that I want to use the altimeter measurement but supposedly that is what CWOP looks for and not an actual SLP.
I personally would be fine using whichever mode NWS uses.  I don't think I can ever achieve a thumbs up all the way around here due to being in a hollow. Things change at a totally different rate here LOL!

FWIW my VUE console doesn't have the data logger. I used it and honestly had great results until I got the VP2 and at that point, Davis sent me a newer data logger for the VP2 so I have just used it to download data. Actually, some of the functionality the VUE has should be incorporated in the VP in my opinion.  I like having the VUE as a reference that shows exactly what the sensor reads so I can switch between measurements
« Last Edit: October 01, 2019, 11:05:30 PM by W Thomas »


     Best Regards
     Wayne

CWOP CW8217
KVAWHITE22 Wunderground   Davis VUE &  Davis Vantage Pro 2  /   Dedicated Server
GR Level 3 ,Level 2 AE Radars  Weather Display 10.37P  Mid Atlantic Weather Network Member
SkyWarn & Spotter Network 6092