Jump to content

  • Member Statistics

    17,512
    Total Members
    7,904
    Most Online
    12bet1 net
    Newest Member
    12bet1 net
    Joined

Models Have Us On Northern Edge Of Snow 3/4/16

bluewave

By bluewave,
in New York City Metro

Recommended Posts

  • Replies 1.3k
  • Created
  • Last Reply
Brian5671

Brian5671

Posted
Posted

Correct. A higher probability relative to climatology does not provide guarantees. Moreover, a cluster of the guidance now shows only weak blocking.

 

Having said that, it's still premature to assume that March won't have any measurable snow. But the probability of a much snowier than normal March is declining.

And one has to wonder if we even get that.   Blocking in the long range has turned out to be largely false.  Our one good block came from a record Kara sea block as bluewave pointed out and even that only led to a 5 day -NAO and then it flipped right back to positive.

Link to comment
Share on other sites
KeithB

KeithB

Posted
Posted

I don't know the weather models in great detail, but I do know how they operate, since I've done computational fluid dynamics modeling for small chemical reactors, which essentially utilizes the same heat, mass and momentum transfer equations and these also include phase changes and chemical reactions, so the composition is changing over time.  For both systems, the models start with defining an initial state (initial conditions in the reactor or the global "reactor" for meteorology), certain boundary conditions, and the best physical approximation of the fundamental governing equations, which allow one to move forward in time, predicting what the next "state" looks like and then repeating the calculations for subsequent points in time.  In both cases, the actual systems are so complex that "analytical solutions" do not exist (i.e., the various differential equations can't be "solved" per se), so a collection of usually non-linear numerical approximations are made for how the variables involved change over time, which is why so much computer horsepower is required.  

 

So, given that the initial conditions, globally in three dimensions at very fine grid spacing are not known well at all, and given the complexity of the equations involved, combined with the fact that weather is, by nature, inherently chaotic, meaning minor errors in initial conditions or forward moving calculations in time can lead to major errors in the various future state model outcomes (the butterfly effect as Lorenz called it, wherein a butterfly flapping its wings a certain way, in theory leads to a hurricane thousands of miles away well into the future), the question isn't, "why are these models so bad?" but really, how amazing is it that these models are this good?"  It's hard to predict the future state of mixing in a chemical reactor I can hold in my hand and yet people are complaining that we can't predict the future state of a global system which still isn't that well understood on many levels?  I wish I could convey better how unbelievably difficult it must be to even have any accuracy at all with these weather models.

. Thanks for the info. My question still remains, but I'll word it differently: weren't the model outputs in the 120-144 hour range better in the previous 3-4 years, than they have been this year? I don't recall this much flip flopping, and to this mile distance flip flopping the last few winters.
Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Guest
This topic is now closed to further replies.
Go to topic listing

  • Recently Browsing   0 members

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