Author Topic: bizarre values in 10-day forecast  (Read 824 times)

0 Members and 1 Guest are viewing this topic.

Offline danxx

  • Contributor
  • ***
  • Posts: 108
bizarre values in 10-day forecast
« on: December 25, 2018, 11:55:41 AM »
So in at least one 10-day forecast, the table data often starts out with a bizarre value that sets the scale for the whole plot, making it practically useless.
Gee WU, dontcha know that DP=-45F is a little screwy? Anyone know why this happens? See below.

 [ You are not allowed to view attachments ]

Offline Jasiu

  • Forecaster
  • *****
  • Posts: 949
    • LexMAWeather
Re: bizarre values in 10-day forecast
« Reply #1 on: December 25, 2018, 02:21:15 PM »
What is the weather station listed above the forecast? Anything from the left of the "now" line should be historical data from that station - at least that's what I see. (include more of the screen shot...)

I have a similar situation with wind speed because my location is a bit isolated and the winds are always below the forecasted speeds.
https://lexmaweather.info
On Mastodon: @LexMAWeather@toot.community

Offline danxx

  • Contributor
  • ***
  • Posts: 108
Re: bizarre values in 10-day forecast
« Reply #2 on: December 25, 2018, 02:51:46 PM »
It is from Lamar Middle School KTXAUSTI318. https://www.wunderground.com/forecast/us/tx/austin/KTXAUSTI318

A few hours later, it's a little different, but same problem. Looking into the history of that station, there are no bizarre values. I see this problem often with this station, and I don't recall ever seeing it with other stations. This really makes the 10-day plots halfway useless. Is this a station issue? If so, I may just abandon this station for my 10-days.

 [ You are not allowed to view attachments ]
« Last Edit: March 11, 2019, 10:36:25 AM by danxx »

Offline Maverick3837

  • Member
  • *
  • Posts: 13
Re: bizarre values in 10-day forecast
« Reply #3 on: December 25, 2018, 02:58:56 PM »
Mine has done this as well, quite a bit here lately. We're getting ready for a big snow storm tomorrow, Southern Colorado, and it was really showing bizzare values for the last three days, up until the storm was about 24-36 hours away. It's been much more stable today, but I do occasionally catch a glimpse of it throwing out some goofy numbers. lol 

Plus, it's not just my station doing it. The surrounding stations 10 day forecast will also show weird readings when it happens.
« Last Edit: December 25, 2018, 03:02:00 PM by Maverick3837 »

Offline danxx

  • Contributor
  • ***
  • Posts: 108
Re: bizarre values in 10-day forecast
« Reply #4 on: December 25, 2018, 03:03:06 PM »
Well, the funny thing is that looking at the station history, it doesn't seem to be throwing funny numbers at WU. You would think the stuff to the left of the "now" line would be from the station history, but that's not what seems to be happening. Looks to me like WU is making bizarre stuff up out of whole cloth for the history part of the 10-day. Usually everything looks right, but a few times a week, it looks like this.

Offline Maverick3837

  • Member
  • *
  • Posts: 13
Re: bizarre values in 10-day forecast
« Reply #5 on: December 25, 2018, 03:09:40 PM »
It's gotta be some type of problem on WU's end, I agree. My station, and surrounding stations history are not adding up either when it comes to what is showing up, and what is actually being reported from the stations themselves.

Offline danxx

  • Contributor
  • ***
  • Posts: 108
Re: bizarre values in 10-day forecast
« Reply #6 on: December 25, 2018, 03:16:35 PM »
I have sent a comment to WU Feedback. Hopefully they can work this out.

Offline danxx

  • Contributor
  • ***
  • Posts: 108
Re: bizarre values in 10-day forecast
« Reply #7 on: January 08, 2019, 07:40:53 PM »
So here it is, two weeks later. Nothing done, no response. 10-day displays are showing DP=-55F for this site. I assume WU has no staff anymore, correct?

Offline danxx

  • Contributor
  • ***
  • Posts: 108
Re: bizarre values in 10-day forecast
« Reply #8 on: March 11, 2019, 09:34:42 AM »
OK, now the problem has infected ALL the WU 10-day forecasts I look at, from all around the country. WU has decided that all dew points before the present time shall equal precisely zero. That means that the 10-day plots that include dew points automatically comes up on a squashed scale. This is nuts. For example,
« Last Edit: March 11, 2019, 10:36:54 AM by danxx »

Offline danxx

  • Contributor
  • ***
  • Posts: 108
Re: bizarre values in 10-day forecast
« Reply #9 on: March 24, 2019, 07:41:52 PM »
Just FYI, I believe WU has now responded to this issue and says that a formal ticket has been created. We'll see.
http://www.wxforum.net/index.php?topic=33651.msg374031#msg374031 (reply #1739)

 

anything