UKC

Assessment and selection of the weather forecast

New Topic
This topic has been archived, and won't accept reply postings.
 Jake80 04 Apr 2019

How do we approach this? The situation is very different if you want to climb in the European Alps and when you are going to remote part of the Andes or Tajikistan.

I have read an article in the past explaining how different weather forecast providers use different approaches and techniques to provide forecast for specific regions and altitudes. It mentioned that some were not accurate enough (grid density etc) and therefore high risk to use really. Surprisingly one of my 'favourite' websites was there as not too good! 

I'm not an expert but if someone is it would be good to understand how to select source of the information to be successful and safe. I'd appreciate advice or even link to some reputable sources.

Thanks 

 alexm198 10 Apr 2019
In reply to Jake80:

I'm afraid I'm not an expert, but I thought I'd bump this as it's a good question and I'd also be interested in any advice. 

Do you have a link to the article you mention?

I commonly use MeteoBlue's Multimodel view to get a wider spread of data: https://www.meteoblue.com/en/weather/forecast/multimodel/chamonix_france_30... but I'm not sure how reputable it is. Mainly it's good for convincing yourself that even though 9 out of 10 forecasts says it's going to piss it down, the one that's forecasting blazing sunshine makes it still worth a try

 jgc 10 May 2019
In reply to Jake80:
Weather in the mountains is complicated.  Resolution is very important (grid density as you say, check our post in that respect: https://meteoexploration.com/wordpress/index.php/2019/04/11/model-resolutio... ).  But also initial conditions and number of weather stations in the direction of the coming weather, which is useful to initialise models. That's why forecasts for Eastern Europe are usually more accurate than for Western Europe, as there are hardly any weather or radiosounding stations in the Atlantic, but there are plenty of them in Western Europe.
Model ensembles (for example https://meteoexploration.com/forecasts/Monte-Rosa/?module=ensembles ) gives you an idea of how reliable the forecasts are, sometimes the dispersion is such, that forecasts are just a guess, and then local knowledge is crucial. A local, experienced forecaster knows the behaviour of weather systems when they interact with local topography and no freely available model goes to that level of detail.

All weather (and climate) models are of limited resolution and had inherent limitations in simulating the chaotic behaviour of weather, although they do a really good job many times.

In a summary, check resolution, vertical levels and how they interpolate to the altitude of the peaks, which model and model initial conditions they use, and don't trust anything beyond five days (treat everything beyond three days with suspicion).

 tehmarks 10 May 2019
In reply to Jake80:

Scotland: check a selection of weather forecasts, pick the most optimistic, and invariably go and have a fine day on this classic wee route: No match for climb id:349250

Alps: check a selection of weather forecasts, pick the most pessimistic, and invariably waste a perfect blue sky day sat in the valley when the weather doesn't materialise.

Post edited at 22:07
 Webster 10 May 2019
In reply to Jake80:

the important thing is to look at multiple different forecasts and look for common themes, i.e. they are all agreeing that a cold front is coming from the NW, but disagreeing on timing etc... one model will be more accurate on one occasion and another on a different occasion. the important thing is not to take specific predictions too literally and look at the bigger picture.

 Dave Cundy 10 May 2019
In reply to Jake80:

I'd offer two views on the problem.

As an aeronautical engineer, the weather models will always be limited by a) grid resolution, b) limited knowledge of todays condition (i.e. spot points in a 3D grid) and c) while fundamental aspects of fluid dynamics are well understood, the the modelling of mixing is not.  So different organisations will have their own preferred mixing models, none of them will be reliable everywhere and all of the time.

As a former paraglider pilot, those who got the most air time were prepared to go out on slightly crappy forecasts, knowing that 30% of those days would actually be flyable.  The rest of us moaned more, flew less but spent fewer fruitless hours in the car.

So in summary, no forecast organisation will be perfect.  So go out climbing but be prepared to change objective, or to go home, if the weather continues to look crap.  And don't count the wasted days, they're not wasted, they are the price you pay for sneaking a cheeky route on the marginal days when you mates stayed at home and moaned

Post edited at 23:43

New Topic
This topic has been archived, and won't accept reply postings.
Loading Notifications...