Forecasts at high elevation / in mountains seem inconsistent
I really like Windy, I've been using it for a while but I believe there are inconsistencies with how some information is presented. It is confusing and not user friendly. Let me explain:
I am currently providing daily forecasts for a group doing an expedition on Mt Logan in Yukon (almost 6,000m). Models available are : ECMWF, GFS, ICON, METEOBLUE, NAM.
You can see Mt Logan forecasts here:
For example, when you look at the forecasts at the summit (5959m), using the Basic or Meteogram views, each forecast (except METEOBLUE) shows temperatures in the -10C/-15C range:
Only METEOBLUE seems to provide the elevation corrected forecast showing temperatures in the -25C/-30C range:
I am currently using the Airgram to have more details on how high the clouds are and the expected temperatures and wind speed at the elevation I need, and you can see for example that at 470hPa, roughly the pressure at the summit elevation, the temperatures are in the -30C range. You can also get this information using the map temperature and wind layers and increasing the elevation with the cursor to the desired pressure/elevation.
So I have to juggle with lots of stuff to try prepare the best forecast I can.
There may be a good reason for doing this but why not presenting the information in the Basic and Meteogram views consistently with the Airgram information (or Map layers). This is my main point, to be clear.
Furthermore, because of this, I also don't know if the Windgust data that is only displayed on the Basic and Meteog. views is ok or completely wrong and if I should only stick with the Airgram wind data...
While Meteoblue seems to be providing ok temperatures with elevation change, it lacks the Meteogram and Airgram displays presenting the cloud cover at different elevation. And its windgust data seems always under by quite a bit compared to ECMWF's wind forecasts using the Airgram. So I don't really trust either.
This post is a bit longer than I intended, I appologize, but I hope it will explain clearly what my problems are. It currently makes my life harder and I am not super confident with the information I am providing.
I also forgot to mention that I use the Sounding tool to get the hopefully correct information at the desired elevation:
But again why is not available on the Basic and Meteogram views. And how to get the windgusts speed (if it's even possible) at the elevation I want (assuming that the information provided on the Basic and Meteogram views is not accurate - is it?)
vsinceac | Premium last edited by vsinceac
I'm not sure global models would be the best choice to correctly forecast weather for such a very specific place. I think better would be to try more specialized forecast sites like this one, or even better to ask Canadian Weather Office for dedicated forecast.
Weather models are not very good at giving a direct temperature forecast in mountains. In particular global models but also the regional ones. This is due to their horizontal resolution and mainly to the coarse orography model that they include in their computation.
As example GFS shows these altitudes in its orography model.
This model cannot ‘’see’’ Mount Logan at its real altitude. Limited area models like NAM have generally a finer orography model. But that is not enough and in mountains the only way is to consider the temperature at different levels in free atmosphere as you do with Airgram.
In the Sounding you have posted you see that the ECMWF model considers Mount Logan at 3028m (model altitude) instead of 5858m. It cannot give the right temperature !
It would be useful that Windy shows the model altitude in Meteogram.
Meteoblue model called METEOBLUE AI has been developed in Switzerland to better integrate the altitude in mountain weather forecast. This is probably the reason why you get a better information with this model.
Thanks. Yes I've been looking at it too. It is fairly user friendly, especially the ability to have conditions at different elevations. I would like to understand what model they are using as input data for their algorithm. I have also been looking at Environment Canada, using Spotwx.com that is a very useful tool as well. I asked if Environment Canada models could be added to Windy as Windy's visualisation tools is hard to beat in my opinion. I still find the ECMWF model to be remarkably good in most situations though. Thanks again for your comment.
Thanks for your comment. Yes I can see the orography and model resolutions as being a limitation indeed.
I still think that if I clicked on a map at 5000m for example I would like Windy to give me on all its views (basic, meteogram, Airgram) the conditions at this elevation that can be found using the Airgram view. Right now the basic view displays the elevation of the picked location (in my example it would show 5000m on the right hand side), but as you suggested, it may display forecast values of lower elevation (your point on the Sounding graph).
As far as I can tell, the Sounding tool matches the data from Meteogram and Airgram, so the data is there. (However I don't know how the Wind Gusts are handled and if they are available within the data provided by these models at any elevation).
At the end I can still get the information I want, but I think it could be made simpler, and furthermore more consistent. I would expect the different views/modes (basic, airgram, etc) to provide consistent information, but this is not the case currently. Well, at least not for these higher elevations, I haven't really investigated yet for lower mountains.
So I think the case is still open and could be added to the backlog... :)
As far as I know wind gusts are only available at 10 m high.
For ECMWF model, see § 1.1.5
Forecasters use to estimate the wind gusts with the averaged wind at 950 hPa (approx. 500m above surface). In high mountains that is more difficult to say.
@idefix37 Good article, thanks. And thanks for getting back to me quickly. Cheers