Author Topic: Anybody explain this in simple terms?  (Read 1053 times)

0 Members and 1 Guest are viewing this topic.

Offline funsutton

  • Contributor
  • ***
  • Posts: 140
    • Weather Underground
Anybody explain this in simple terms?
« on: June 06, 2020, 07:58:07 PM »
My wind vector apparently had some problems yesterday resulting in a somewhat-failure I don't quite understand. I'm not worried about it, per se, because there isn't much I can do about my situation. Just wondering if someone could translate this...

 [ You are not allowed to view attachments ]

Offline funsutton

  • Contributor
  • ***
  • Posts: 140
    • Weather Underground
Re: Anybody explain this in simple terms?
« Reply #1 on: June 06, 2020, 08:46:42 PM »
I noticed it just got updated to this if it helps.. I assume the 'analysis' is where it's expected to be and the blue line is where it was today...

 [ You are not allowed to view attachments ]

Offline funsutton

  • Contributor
  • ***
  • Posts: 140
    • Weather Underground
Re: Anybody explain this in simple terms?
« Reply #2 on: June 22, 2020, 11:11:34 PM »
While I was just looking at my Gladstone page, I noticed it had the local KGSO METAR station in those near me.

https://weather.gladstonefamily.net/site/KGSO

So I clicked it and found that Gladstone fails it on wind as well. It's a freaking METAR station and it still gets two thumbs down on wind:

 [ You are not allowed to view attachments ]

If a METAR station can't pass Gladstone, then I'm pretty sure there's no hope for me ever passing on wind. (Btw I've had two thumbs down almost my entire short MADIS history)

Offline davidmc36

  • He who dies with the most toys wins!
  • Forecaster
  • *****
  • Posts: 1246
  • FN25ie61jw
    • MorewoodW34
Re: Anybody explain this in simple terms?
« Reply #3 on: June 23, 2020, 03:52:46 AM »
I saw that on mine and others, just like you see. I figure their "ideal" is average of all reports.

Any single station's average is likely not in step with the overall average. Depending on just how the weather pattern goes and your location within it even nearby stations could see very different average wind direction..

That's what my analysis was when I saw that. I figured it was useless to worry about.

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: Anybody explain this in simple terms?
« Reply #4 on: June 23, 2020, 09:49:16 AM »
Wind analysis on Gladstonefamily.net used to work about a year ago and it worked for years before that. This failed QC analysis is a relatively new thing with Gladstonefamily unfortunately. Something changed in MADIS and Gladstonefamily was not updated to interpret the new way. Unfortunately Gladstonefamily has been sort of abandoned. It is just limping along. It is what it is.

CWOP is not one thing. It exists because a lot of pieces came together with different people and organizations. Just take a look at the complexity of all the different websites you need to know about to get going with it. You have your sign up and general information at wxqa.com, then there is Findu.com to see your telemetry, then there is the APRS network working behind the scenes, and then there is NOAA for finalizing of the registration, and then there is MADIS for ingestion, then there is Gladstonefamily.net for QC but they are only one of the QC sites available as there are others like MesoWest and the NWS MADIS map. That is a lot and I skipped over differentiation of ham APRS servers vs CWOP servers and other stuff. As you can see CWOP only exists as a collaborative of a lot of people and organizations. Some of these pieces unfortunately have not been maintained to have current information nor have they been maintained for general improvement of the look and feel and modernization of the service. Lots of people just ignore Gladstonefamily.net even though they continue to upload to CWOP. Even in years before when wind analysis worked better, there was still contention about the analysis of all metrics and how one bad neighbor affected a good station. That is still true today. Therefore QC analysis was never perfect. But wind analysis today is definitely useless and broken and as I said it wasn't like this before.

The big takeaway is that CWOP isn't one thing. It isn't like WU or PWSweather or AWEKAS or any other online service where one body (person/organization) controls the entire operation. Take it or leave it. It is what it is. There is nobody to complain to.
« Last Edit: June 23, 2020, 09:52:21 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 miraculon

  • Sunrise Side Weather
  • Forecaster
  • *****
  • Posts: 4107
  • KE8DAF
    • Sunrise Side Weather in Rogers City MI USA
Re: Anybody explain this in simple terms?
« Reply #5 on: June 23, 2020, 11:41:46 AM »
According to the Quality Spread Distribution Page, I have 90-100mph wind speed errors. Note the "target" rectangle near the origin.



I also get the L3 QC check failure. I am not sure it is worth anything.

It may be related to this issue captured in this thread: https://www.wxforum.net/index.php?topic=37810 ("supersonic" wind errors..)


Greg H.


Blitzortung Stations #706 and #1682
CoCoRaHS: MI-PI-1
CWOP: CW4114 and KE8DAF-13
WU: KMIROGER7
Amateur Radio Callsign: KE8DAF

Offline Liamdog4

  • Member
  • *
  • Posts: 40
Re: Anybody explain this in simple terms?
« Reply #6 on: June 23, 2020, 08:47:10 PM »
Seems somebody must be updating info on the Gladstone site. I recently seen "neighboring Stations" not showing for a day or so and now they are showing different sites. A lot of local airports now.
Vantage Pro2 Plus Wireless, 24 hr Aspirated, Soil Moisture Station, Leaf Moisture, 2 Extra Temperature Stations

Offline funsutton

  • Contributor
  • ***
  • Posts: 140
    • Weather Underground
Re: Anybody explain this in simple terms?
« Reply #7 on: June 23, 2020, 10:06:02 PM »
Wind analysis on Gladstonefamily.net used to work about a year ago and it worked for years before that. This failed QC analysis is a relatively new thing with Gladstonefamily unfortunately. Something changed in MADIS and Gladstonefamily was not updated to interpret the new way. Unfortunately Gladstonefamily has been sort of abandoned. It is just limping along. It is what it is.

CWOP is not one thing. It exists because a lot of pieces came together with different people and organizations. Just take a look at the complexity of all the different websites you need to know about to get going with it. You have your sign up and general information at wxqa.com, then there is Findu.com to see your telemetry, then there is the APRS network working behind the scenes, and then there is NOAA for finalizing of the registration, and then there is MADIS for ingestion, then there is Gladstonefamily.net for QC but they are only one of the QC sites available as there are others like MesoWest and the NWS MADIS map. That is a lot and I skipped over differentiation of ham APRS servers vs CWOP servers and other stuff. As you can see CWOP only exists as a collaborative of a lot of people and organizations. Some of these pieces unfortunately have not been maintained to have current information nor have they been maintained for general improvement of the look and feel and modernization of the service. Lots of people just ignore Gladstonefamily.net even though they continue to upload to CWOP. Even in years before when wind analysis worked better, there was still contention about the analysis of all metrics and how one bad neighbor affected a good station. That is still true today. Therefore QC analysis was never perfect. But wind analysis today is definitely useless and broken and as I said it wasn't like this before.

The big takeaway is that CWOP isn't one thing. It isn't like WU or PWSweather or AWEKAS or any other online service where one body (person/organization) controls the entire operation. Take it or leave it. It is what it is. There is nobody to complain to.

Great explanation galfert. Thanks.

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: Anybody explain this in simple terms?
« Reply #8 on: June 23, 2020, 11:16:21 PM »
Seems somebody must be updating info on the Gladstone site. I recently seen "neighboring Stations" not showing for a day or so and now they are showing different sites. A lot of local airports now.

Yep that happens totally randomly. Sometimes the airports are there, then they disappear and then they reappear. It's like the WU gold stars. You just never know and it happens randomly as systems and databases decide to sync up or not. Nobody is at the helm.
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline Storm017

  • Senior Contributor
  • ****
  • Posts: 172
Re: Anybody explain this in simple terms?
« Reply #9 on: June 23, 2020, 11:20:37 PM »
Thanks galfert.  Our local office is looking for more PWS to register for a CWOP ID.

https://www.weather.gov/bgm/hydrologyCWOP

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: Anybody explain this in simple terms?
« Reply #10 on: June 23, 2020, 11:33:47 PM »
Storm17,
Excellent, I'm still a big advocate for decent weather station stewards to join CWOP. I know the data is very important to a lot of private and public organizations. I just wish that NOAA would somehow do more than the minimal effort so that the sign up and general information and QC worked better and seemed more modern.
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline Liamdog4

  • Member
  • *
  • Posts: 40
Re: Anybody explain this in simple terms?
« Reply #11 on: June 24, 2020, 07:31:52 PM »
I agree Galfert, with better QC on stations that make proper reporting, or trying to report proper data and having stations reporting incorrect data making the good ones look bad from what I understand. If you want to report your station data, then be responsible for it and try to make it correct as possible. I have seen some local stations in my area not showing up for a couple days. I know my temperature is a little low compared to my local airport report but I am a few hundred feet higher than they are and not sure if I should change it to make it closer to them. Galfert since you seem to know alot, why does bad stations  data affect what other good stations report? Is it like they are averaging station data together or something like that? I'm just trying to understand.
Vantage Pro2 Plus Wireless, 24 hr Aspirated, Soil Moisture Station, Leaf Moisture, 2 Extra Temperature Stations

Offline CW2274

  • Forecaster
  • *****
  • Posts: 6731
    • Conditions @ CW2274 West Tucson-Painted Hills Ranch
Re: Anybody explain this in simple terms?
« Reply #12 on: June 24, 2020, 08:38:28 PM »
If I've said it once here, I've said it a million times...forget QC, at least til you understand it's very real limitations. QC is completely dependent on your neighbors data. If ALL were to the same standard, things would be much more representative of actual conditions, but usually, it's not. Some locales may have "good" data to compare with, however, most likely do not as a whole. Pay attention to your own PWS..and know that it's "where it should be". You'll save yourself a lot of meaningless, worthless headaches.

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: Anybody explain this in simple terms?
« Reply #13 on: June 24, 2020, 09:14:40 PM »
why does bad stations  data affect what other good stations report? Is it like they are averaging station data together or something like that?

Other station's data doesn't affect what your station reports. Every station stands alone. But they all feed into (get ingested) by MADIS and then analysis data comes out of that. Then sites like Gladstonefamily.net try to interpret this analysis and basically flag your station if not within the norm of what is expected based on the data of other stations. If you ignore the analysis data then no harm comes of your station and your data. This analysis data could be right or it could be wrong. It is like a consensus. But If you have enough neighbors with bad data then you have to realize that they are going to be affecting more than just your analysis data, as the data ingested by MADIS is used to produce weather forecast data. If you are in a sparse area then it becomes a he said she said situation and you both lose. Typically analysis data does better when there are more densely populated stations, unless you are unlucky to have many neighbors neglect their stations and their bad data. But this analysis data as it pertains to your station should be secondary to your own understanding of weather and your hardware. You should be able to discern if your station data is bad. And you should be using other means to determine this. You can use the QC not from just Gladstonefamily but also from MesoWest and MADIS Surface Map. And you can also report to AWEKAS and use some neighbor comparison there too. But you can also just look at your data and sometimes just instinctively know that something is wrong and you should be doing maintenance on your hardware and physically inspecting and cleaning it. And you can also invest in secondary equipment to make some comparative analysis of your own.

But you can do something about it. You can contact your neighbors and inform them nicely that they should take a look at their stations and perhaps consider some maintenance. I monitor my data, and my analysis data and I monitor stations within my 20 or so mile radius. And I contact them when I feel something needs their attention. I've successfully been able to police my area and get neighbors to fix or take down and stop reporting to CWOP until they fix the problem. It is the right thing to do. I tell them to continue to upload to WU where anything goes, but to respect CWOP and respect their neighbors and the multitude of organizations that depend on good data.

If you'd like to contact your neighbors just look them up using the wxqa.com shortform lookup:
http://wxqa.com/shortform.html

Everyone's email is right there. You'll need to do a little deciphering and figure out that some funky symbols in the email address are actually the @ sign. Just remove the [$20At} or [ At%7D and replace that with the @ sign.



« Last Edit: June 24, 2020, 09:20:32 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


 

anything