vortex95

See attachment It is said that for ML models, the equations of motion and thermodynamics are not satisfied, so that is going to lead to unrealistic output and egregious errors at times. Not that physics-based models are immune to these same problems, but we often know why they occur due to the limitations of simulating the atmosphere best, as one example. For ML models, does that same apply? That is, would we be able to detect why they are wrong for a given forecast? How far do ML models not satisfying equations go? For instance, Navier-Stokes? The paper states: "New forecast methods based on statistical estimation, including neural networks and ML, are neither designed nor constrained to yield dynamically coherent, physically consistent evolutions of the atmosphere." The above would appear to be a huge problem. The paper also discusses ML models lacking in resolution detail, and give a more broad overview. That's fine within itself, but modeling has come so far, getting the broad strokes right as to what is going on, say synoptically, is no longer an issue. What matters most now are the details and fine-tuning on a more local level and shorter time frames in weather forecasting. That's what the public/partners want and demand. How vulnerable are ML models to chaos theory? More or less than physics-based? Another item suggested in the paper, it seems that ML models can not exist or do well/improve without physics-based models. This would ask the question, how much of a statistical database of weather history/analogs do we need for an ML model to perform better? bams-BAMS-D-25-0214.1.pdf

Snow weenies rejoice! GFS V17 goes from 13 to 9 km for horizontal resolution in Oct. Scott counting the days now for the first day of meteorological winter! LOL. pns26-29_Science_for_GFSv17.pdf

One last chance this season for it to look "VIOLENTLY BEAUTIFUL" out there!

SE MA OES in later April? "Unprecedented!" CoastalWx will be out during overnight shaving off every tenth of an inch for max total snow! NORLUN apparent. I like the -31 C at 500 and -10 at 850 and the axis of the 500 trough right overhead.

Instant PT?

From my contact in Woburn, 5 days w/ thunder this month now. 5 TS days is pretty good. That's what normal in July this area. CoastalWx still MEH though I bet!

I'll save CoastalWx the trouble -- WCWGTH??

4/18/1996 27 inches of snow over the last three days from an upper level low pressure system raised the snow cover on top of Mount Mansfield, Vermont to an impressive late season 135 inches.

Try that week in third week of May 2006, "endless Nor'easter" capped off by the bow echo and tornado at Hampton Falls NH later.

Peterborough NH Thu evening. Photo by Thomas Bensenhaver.

Or winter equivalent "DIAMOND DUST AT 30,000 FT!"

Scott needs another 6/1/11 wicked CG supercell show like he had when living in Dorchester. Right, biggest svr wx day in years and he still has a Bruins playoff party at his house, and of course he loses power! I know, another "PC jolt" he got from the +CG strike some years ago!

Reports from ME/NH/VT absolutely spectacular LTG show for April this evening, and still going strong Downeast ME as I type this! Find me an event in April w/ more CGs in several hours in New England! I checked the mid-level lapse rates plots on the NAM, nothing special, but that is 700-500. If you look at 750-600 on the soundings, they were as high as 7.0 C/km. so there you go. MUCAPE ahead of the SQLN were as high 1500, most of it elevated. 0-6 km shear 50-60 kt. Even CoastalWx *has* to be impressed! He recalls the crazy LTG shows that overnight in Feb 2016 RI/SE MA? Basically a ring of fire in April this week. Strong ridge SEUS and direct feed flow straight from Plains to put an EML into NEUS. BDF?, no problem!

Had nice white VIL core on Radarscope for two scans. Usually, that is enough for a SVR.