As time passed weather models started showing unprecedented and unparalleled consistency that a big time coastal storm capable of dropping historic snow amounts will in fact track up the coast between January 21st and 22nd. Exactly a week ago the EURO, GFS, CMC, NAVGEM, JMA, UKMET models were all showing that. However, as the week went on and we were within 3-5 days from potential impact, the aforementioned models trended south with the storm track which kept the crippling impacts contained to the Mid-Atlantic region. They also pushed back the start date from the 22nd to the 23rd which is also not a great sign when tracking east coast Nor'easters because it gives the pattern a chance to break. Suddenly the extraordinary consistency we once saw was gone and the NYC Metro area was no longer in the cross-hairs of this beast. Weather forecast hobbyist such as myself did not completely buy into this trend. It was in my January 17th Mo Mo (Monday Morning) blog that I highlighted the upper level pattern and possible scenario's. As the week grew older it became clear the global models were suffering from convective feedback while the mesoscale models kept a surface low closer to the coast drawing plenty of moisture northward to bring a historic snowfall. The extreme upper level dynamics and complex set-up were too much for global models to handle. It was not until Thursday night or early Friday that I had full confidence a Godzilla was coming.
Here are a couple of graphics from the January 17th Mo Mo. It was clear at the time a big east coast snowstorm was possible because the ingredients were all there...a "recipe for Godzilla" one could say. It was just a matter of determining how they all come together.
I then put together 3 scenario's this storm could take. While this storm did not play out exactly as described above, it was close. The 500mb low actually followed a track similar to what's outlined. The surface low developed off the coast of NC and tracked due north tucking and stalling into the Delmarva before heading east-northeast.
The 500mb low, Sub-Tropical Jet Stream, and above normal Atlantic sea surface temperatures played critical roles in getting the surface low to track as far north as it did and - most importantly - pivoting snow bands into the NYC Metro area. Reality is the global models DID agree with the mesoscale models where the surface low will track but the difference was the mescoscale models tucked and stalled the low near Delmarva while the globals tried to transfer energy to a second low well off the coast. The STJ is enhanced from the strong El Nino conditions over the Tropical Pacific. You can see in the first image how far west the STJ extends. The storm was being fed energy from the Tropical Pacific AND Tropical Atlantic (second image). Mild Caribbean air was being drawn into the core of the surface low which developed a lot of convection on the east side of the trough. This is partly the reason why global modeled kept "jumping" the surface low from Delmarva to 150-200 miles east into the Atlantic. With the 500mb low south-southwest of us, it does not make sense to see a surface low so far off the coast. Instead, it should remain near the coast as bands of heavy precipitation rotate into the area.
Here is how the GFS (top) and EURO (bottom) models handled the storm 12-24 hours before impact. Both models agreed the crippling snowfall accumulations will be contained from Washington D.C. - Philly Metro - and southern NJ. Notice what they do with the surface low pressure. It's off the coast of VA/DE but a couple of frames later they re-develop it well off the coast southeast of the Benchmark. They are taking it where all the enhanced convection from the Atlantic feed is.
Meanwhile, the NAM at the same time was doing the complete opposite. Despite all the convection on the east side of the trough the model still kept the surface low close to the coast while pivoting prolific snow bands into the area. The NAM had support from the SREFS and RPM models - two other mesoscale models. The RGEM did not jump on board until the 12z Friday run. The track of the H5 low being further north than where the global models had it immensely helped with its almost perfect verification. Syoptically it made sense. The warm SST's in the Gulf of Mexico and raging STJ does not support an H5 low digging into the southeast as much as the EURO showed. Also, the western ridge was oriented SW to NE more than N to S and began propagating east due to a s/w crashing into the west coast. In the end, the NAM and other mesoscale models saw these features the best and they not only won the battle but also the war against global models.
The above map is observed and reported snow totals. The bottom map is my final snow prediction map. My map did very well in some areas but also poor for others. The difficulty was trying to forecast where the cut-offs will be. In the back of my mind I felt the NAM had the right idea but may have been too bullish with how far N&W it was showing the snow bands get considering the surface low is way down off the Delmarva! The enhanced dynamics from the northern and southern jets allowed these bands to get as far north as they did. Some of those bands dropped 3-4" of snow an HOUR. Given the complexity of the situation I will give myself a B-. Any other event probably a C.
Here is an animation and radarscope image of some of the banding that took place. The bulls-eye ended up being from east-central PA to the NYC Metro area. Some towns saw snow totals exceeding 30 inches with drifts as high as 6-8 feet from the wind.
Here are some photos from my home. Cranford, NJ - 30 inches.
Everyone will have their own opinion on where this storm ranks in their personal list of memorable snowstorms. For me this definitely ranks #1. I do not remember the Blizzard of 96' and Boxing Day 2010 happened overnight in a 10-12 hour span effecting a smaller area compared to this storm. I come out of every winter a stronger and more capable forecaster because of the people I share this passion with. There are great minds and passionate enthusiast's out there who make this hobby worth the time I put into it day in and day out. I come out of every winter learning something new and gaining a better understanding winter weather forecasting. The Blizzard of 2016 will forever be coined the Roidzilla of 2016. I look forward to tracking the next one with you all.
Time to dig out.
Best,
Francesco Paparatto
______________________________________________________
Where schools closed and anything else?
ReplyDelete