Good discussion. My biggest concern is the same one the NWS employees union has, i.e., that centralizing is just the first step to major job reductions, despite what the politicians are saying. At the end of the day, I think, as proposed, the new system could work with most folks at regional offices, but including subteams for each local area, just that they're housed at a regional HQ, and still including some local staff for data collection and communications with local authorities.
And having a lot less mets to make good local forecasts (whether in NJ or some regional center) would be a disservice to public safety, as these guys generally do a great job and generally (not always, of course, with last year's "blizzard" being the worst example in years) do better than most private sector forecasts. I'm not 100% convinced having forecasters physically local to an area is that important, however.
Despite 4Real's comments about the NWS vs. the NHC, the NHC did better with storm surge forecasts than either the NYC or Philly offices of the NWS, as per the attached from Bryan Norcross, from TWC, who is one of the best in the business. The NWS had low storm surge predictions on Friday and Saturday and didn't catch up to the more dire NHC predictions until Sunday, one day before landfall.
http://www.wunderground.com/blog/bnorcross/unraveling-the-sandy-stormsurge-forecast
The point Norcross makes is a good one: the local NWS guys are good at local forecasts, but the NHC guys are the storm surge experts (they're the ones who run the models and study tropical systems full time). Way too many people were comparing it to Irene, which was a horrible comparison, but by far the most negligent folks were in the NYC emergency management team, who completely blew it on Sandy, by providing horrible information.
And the NHC/NWS decision to not issue hurricane watches/warnings, while technically "correct" (given that the system was morphing into an extratropical one), was bad from a public safety perspective - they should've scared the crap out of people given the potential impact, rather than worrying about being "correct." It's likely that more people would've evacuated and there would've been less loss of life and property.
Having said all that, Szatkowski and the NWS-Philly clearly did a much better job of communicating with emergency officials and getting evacuations going than they did in NY. Szatkowski was out in front, leading the charge for evacuations (even if their surge forecasts were too low - they were still high enough to demand action), while Bloomberg and the NYC folks were minimizing the risks right up to 48 hours before landfall. See the link below.
http://www.climatecentral.org/news/...y-showed-limits-of-an-accurate-forecast-16648
To me, the bottom line is that I simply don't want to see a reduction in meteorologists paid to protect the public and I'd like to see our government spending more money on improving the models and ensuring we have enough satellites (and balloons and weather flights into storms) to provide the data we need to feed the models to make better forecasts. There's no way we should still be behind the Europeans on this (with the models).
