Jump to content
  • Member Statistics

    17,509
    Total Members
    7,904
    Most Online
    joxey
    Newest Member
    joxey
    Joined

Jan 15-16 Winter Storm


Brick Tamland
 Share

Recommended Posts

26 minutes ago, calculus1 said:

 

If you guys think I am arguing against using ensembles in forecasting, I am not.  I am arguing against displaying a map that gives precision probabilities.  If WRAL derives their “probabilities” from ensembles, I still disagree with their language.  If they said “14 out of 50 ensemble members indicated 3 inches or more of snow for Clinton” or even “28% of ensemble members indicated 3 inches or more of snow for Clinton”, I would have no issue with those statements.  Those are indisputable facts based off model output.  But equating such fractions of ensemble members to probabilities is a misuse of statistics, in my opinion.

The probability of a success (getting three inches of snow or more, in this case) is equal to the number of successes divided by the number of possible outcomes (failures + successes).  The ensemble members do not represent all the possible outcomes; they only represent approximately 50 of the infinite number of possible outcomes.  Thus, we can’t really say there is a 28% probability of something happening based off ensemble members.  In fact, I would argue there’s really no way to calculate such a probability, because it’s impossible to account for all the possible outcomes that arise from tweaking just one minuscule atmospheric condition somewhere over the entire globe.

It’s the language that I take issue with, not the use of ensembles.  If they were to change their title to “Percent of Ensemble Members that Predict 3 Inches of Snow”, I would see that as much more transparent.  As it is, I think it’s a misleading graphic.

My head exploded, but I agree with you.  I don't know how you could give a realistic probability of snowfall occurring at a given location 5 days out.  If you had 10 possible outcomes and knew only 3 led to snowfall but didn't know which of the 10 would occur but did know that each had an equal chance of occurring, then it would be easy to conclude that there is a 30% chance of snow.  That's unrealistic though.

On the other hand you have to develop messaging that is framed within a forecast that the general public can consume and understand.  It's probably always going to have a subjective quality to it because you can't quantify the entire possibility set or the likelihood of each possibility for any future event (or non-event) on any given date.  Idk

  • Thanks 1
Link to comment
Share on other sites

Just now, BornAgain13 said:

Very sharp cut off to the snow... shows around 7" here , with around 20" for @Buddy1987. Goes from heavy snow to a sleet fest.

If this were the final solution I’d bet 10$ it would be tainted for sure even up here. I will take what I can get. These last two storms royally screwed me. I need at least a “decent” one

Link to comment
Share on other sites

1 minute ago, wncsnow said:

Welp.. maybe that will help with any further big trends 

Yes to be exact there were 21 dropsondes. Message issued by the Senior Duty Meteorologist out of College Park MD. I am salivating to see what GEFS looks like. It’s been a war now brewing between ops and ensembles. Canadian up next 

  • Like 2
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...