Here's a little behind the scenes of what is going through a nws head tonight...
Forecasting potential extreme events ("black swan" events as some call them) is always a psychological as well as scientific challenge. Yesterday I hemmed and hawed about mentioning blizzard conditions for Long Island for the upcoming storm and went with it in the end, the signal was too much to ignore. Now I'm confronted by the latest shorter term models, which refuse to back down from extreme snowfall amounts from Sunday afternoon into Monday morning. How much do I accept those amounts?? I think those models are seeing what the European model still isn't, and that's the potential for heavy thundersnow with rates of 2-4 inches per hour for several hours, preceded and followed by additional snow at lesser rates. The European never gets hourly snowfall rates above an inch per hour, and with an intense storm like this that makes no sense. But if I bought the extreme numbers verbatim I'd have to forecast at least 2 feet of snow for nearly the entire NYC Tri-State area, Long Island, and all of NJ, and locally as much as 30-40 inches for parts of the Jersey shore up into western and central Long Island. Nope that's too extreme. I can be conservative and say as much as 20-25 inches for the former and 25-30 inches for the latter, there is still time to zero in on the right amounts. If the extreme numbers don't back down with the daytime Sunday model guidance and I can still provide a rationale for why they might happen, then I may have to keep trending toward them. For now time to get some sleep