2015 Severe Weather Season Coming to Life

[Nate Weikal]

For a while, the US was in a weather pattern that favored bringing cold arctic air down the eastern 2/3 of this country, keeping the weather in those regions rather tranquil. Well, the bear is finally waking up from hibernation. That aforementioned pattern right now is breaking down, allowing warmer, moist air to start flowing into the central US. On top of that, the models are starting to hint at several impressive looking troughs sliding through the central plans at the end of this month. It's still very far off though for anything to be set in stone.

The models are still conflicting on everything from the tilt, to the available moisture, cape and temperatures. However, I have a feeling that this is pointing to a very active April and possibly May. I mention this because as soon as that winter pattern dissolves, according to the models, it seems we'll be bombarded with these systems. Do you agree that we'll have an active next couple of months due to this scenario?

 

[Adam Lucio]

Not necessarily, a pattern that is TOO progressive with system after system sometimes means gulf sweeping cold fronts and no chance for the real good juice to make it inland. I also hate to see such variability among the models. They have a climatology bias built into them and when I see such spaztastic model behavior that far out it gives me even lower confidence. At least there is a sign of hope though, and something to watch. It sure beats seeing 20 degree air dumping into the midwest.

 

[Marcus Diaz]

For future reference, we should keep discussion about the upcoming season, including predictions, in this thread:

https://stormtrack.org/threads/thoughts-about-the-upcoming-2015-chase-season.27843/unread

 

[Brian G]

Just curious...which models have climatological bias corrections built into their simulations? Other than post processed products like MOS and using climatological values for some parameterization scheme inputs (like SSTs) I've not seen any mention in NWP literature of a climatological bias correction mechanism built into the solver part of a model like the GFS. It's likely that I just missed it though. Can you guys point me to some literature? I'd be very interested in delving deeper into the topic.

 

[Justin Wilcox]

For future reference, we should keep discussion about the upcoming season, including predictions, in this thread:

https://stormtrack.org/threads/thoughts-about-the-upcoming-2015-chase-season.27843/unread

Snitch

 

[Marcus Diaz]

I will thumbs down your ass so fast...

 

[Justin Wilcox]

I will thumbs down your ass so fast...

Eew!

 

[Marcus Diaz]

I mean that in whatever context you can think of.

 

[Nate Weikal]

Not necessarily, a pattern that is TOO progressive with system after system sometimes means gulf sweeping cold fronts and no chance for the real good juice to make it inland. I also hate to see such variability among the models. They have a climatology bias built into them and when I see such spaztastic model behavior that far out it gives me even lower confidence. At least there is a sign of hope though, and something to watch. It sure beats seeing 20 degree air dumping into the midwest.

Sorry for the late reply. (working 90+ hours per week really exhausts you). I guess that could be too much of a good thing. I do, however, want to point out that the GFS has been really consistent with bringing in a negative tilt trough across the central US around April 1. Again, too far off to guarantee anything but this one looks more promising.

 

[Jeff Duda]

They have a climatology bias built into them...

This isn't really true. Model designers do not insert code to try to force the model to a certain atmospheric state at a certain forecast range. Rather, every model has an attractor, which we call a bias because it tends not to be exactly where the atmosphere winds up in the long run, towards which the model will move in a general sense as time moves forward and IC (and LBC for limited area models) errors build up and saturate. This attractor, or bias, is related to the internal structure of the model. Obviously it is the goal of every model to have zero bias, but none of the numerical schemes used to solve the PDEs that make up the bulk of most models are truly complete unless they contain an infinite amount of terms (which they don't). They are consistent and stable, however, but the nonlinear nature of the atmosphere and our limited compute technology masks the behavior that you would see in a long-term integration of a PDE or ODE to a very distant time.

 

[Jeff Duda]

Just curious...which models have climatological bias corrections built into their simulations? Other than post processed products like MOS and using climatological values for some parameterization scheme inputs (like SSTs) I've not seen any mention in NWP literature of a climatological bias correction mechanism built into the solver part of a model like the GFS. It's likely that I just missed it though. Can you guys point me to some literature? I'd be very interested in delving deeper into the topic.

Yes, every operational modeling center includes some statistical post-processing to the output model fields (i.e., it does not act during the model integration as any part of the dynamics or solver as you said). The statistical post-processing is done to remove bias and to improve reliability of probabilistic forecasts like precipitation. However, that is all done in a diagnostic sense - it does not alter the model fields, just the end product that the user sees.

 

[Stephen Szulborski]

I think there could be a period of interest around April 1st, the long range models are picking up on an abnormally strong upper level jet that crosses the Pacific and comes ashore just before the 1st. It has been a while since we saw a strong upper level jet running completely across the entire Pacific this time of year. Usually when that happens big things occur.