Windy Community
    • Unread
    • Categories
    • Groups
    • Go to windy.com
    • Register
    • Login

    Weather Models Wildly Inaccurate

    Scheduled Pinned Locked Moved Your Feedback and Suggestions
    8 Posts 3 Posters 1.9k Views 1 Watching
    Loading More Posts
    • Oldest to Newest
    • Newest to Oldest
    • Most Votes
    Reply
    • Reply as topic
    Log in to reply
    This topic has been deleted. Only users with topic management privileges can see it.
    • E Offline
      edbevan
      last edited by

      I'm currently off Norway, where all the models are suggesting that it is blowing 14-15 knots from the East.
      It is actually blowing 35-40 knots, and has been for the last 6 hours.
      We have been watching weather forecasts closely.for the past few weeks, and they have been consistently 20 knots below what were actually experiencing.
      How can we tell the forecasters that they are so wrong, we have lost all faith?

      idefix37I 1 Reply Last reply Reply Quote 0
      • idefix37I Offline
        idefix37 Sailor Moderator @edbevan
        last edited by idefix37

        @edbevan
        It is not forecasters who provide weather prediction in Windy, but weather models.
        Numerical weather prediction (NWP) involves huge computation programs carried out by the major weather Centers. So, unfortunately, this is no chance that your observations can be used to modify and improve these weather models.
        Did you try all models available in your area?

        E 1 Reply Last reply Reply Quote 2
        • E Offline
          edbevan @idefix37
          last edited by

          @idefix37 hi, yes, I know its models not forecasters. I have checked all the models available in my location and they are all suggesting lower than that which we experience.
          The ship I am on is also paying for a 'personalised' forecast, which is also consistently lower than experienced.

          T idefix37I 2 Replies Last reply Reply Quote 0
          • T Offline
            tuxcs_ @edbevan | Premium
            last edited by

            @edbevan I'm not sure if this helps but if your main priority is accuracy you can check out yr.no which I believe is the met.no's official weather app for Norway. Hopefully the model they use (MetCoOp EPS) can be added but for now that should be more accurate I believe.

            idefix37I 1 Reply Last reply Reply Quote 0
            • idefix37I Offline
              idefix37 Sailor Moderator @edbevan
              last edited by idefix37

              @edbevan
              It is strange that you find such discrepancy between model forecasts and observations. For the wind or other parameters it is quite easy to explain discrepancies on rough terrain, specially in mountains. But at sea models are quite accurate except special conditions (in summer, convective conditions, thermal breezes…).
              For instance here we had very recently the strong storm named Ciáran in Brittany and Britain (12 Beaufort winds). This storm has been perfectly predicted by all models.
              At which latitude off Norway are you experiencing thes discrepancies?

              E 1 Reply Last reply Reply Quote 0
              • E Offline
                edbevan @idefix37
                last edited by

                @idefix37 we are off Southern Norway, south of Larvik, heading towards Oslofjord. Noticed the UKV model is more accurate. The most accurate aspect of the forecasts has been the trend, but 20 knots more than predicted.

                idefix37I 1 Reply Last reply Reply Quote 0
                • idefix37I Offline
                  idefix37 Sailor Moderator @edbevan
                  last edited by idefix37

                  @edbevan
                  According to your location description you seem not so far from the coast. The ECMWF has a 9 km resolution, which means that the wind forecast can be lower if you are close to the coast. In this case a higher resolution model like UKV with a 4 km resolution in this area gives better values.

                  00EC612E-9D22-46F9-BE8E-E95DFF3114EF.jpeg
                  w/customized colour scale

                  1 Reply Last reply Reply Quote 0
                  • idefix37I Offline
                    idefix37 Sailor Moderator @tuxcs_
                    last edited by idefix37

                    @tuxcs_
                    The AROME - MetCoOp with a 2.5 km resolution would surely give a more granular prediction.
                    https://journals.ametsoc.org/view/journals/wefo/32/2/waf-d-16-0099_1.xml

                    1 Reply Last reply Reply Quote 0
                    • First post
                      Last post
                    Windy Community  |  Powered by excellent NodeBB
                    Terms of Use     Privacy Policy     Windy.com