View Full Version : The accuracy of weather forecasts?
02-05-2009, 07:41 AM
Has anyone at the observatory studied or attempted to quantify the predicability of weather on Mt Washington? Maybe, for example by measuring how accurate forecasts have been relative to anywhere else?
02-05-2009, 08:04 AM
We have had a few interns study our accuracy in relation to forecast models (MOS) for our station as well as compared to NWS Grey, ME's higher summits forecast for Mount Washington. I forget the exact number and being off summit this week, I can't just look them up. For the summits forecast compared to models, I do remember that for the 12 hour period we were about 70%, overnight we were closer to 80% and for the next day, we were in the 70% range again. Compared to the NWS forecast, we were better than them about 80% of the time. But to compare our forecast to NWS is unfair because we forecast for Mt Washington itself and have hikers assume that if it is -20F with 100 mph on the summit, it will probably be less than that lower where as NWS calls higher summits anything over 3500 ft in western ME and most of NH. Quite a larger forecast area to try an pin down. And to compare us to other locations is difficult for a few reasons. One, there aren't many summit stations. Two, most of the ones that do exist don't forecast. Third, the summit locations don't match the set up for the summit (ie, higher, lower, in less/more storm tracks, further north/south/east/west, etc.). And lastly, again with location, location is everything. For instance, it is easier to forecast for San Diego than for Mt Washington. But to compare those two is like comparing a head of lettuce to an apple.
The thing to keep in mind is that the faster computers get and the longer the climate record, the better our forecast become. Plus, experience is everything. But also, our forecasting record isn't that long. Although people think we have been forecasting since the get go, a printed forecast has only really been in existance since I believe the 90's (someone can correct me if they know for sure, I get this number from our archive data). And older forecasts vary as a uniform wording and format has only been around since I arrived. But it is like I always tell our visitors: "Our forecasts are a tool, not a word set in stone. It is up to you to gather as much information as possible, including current conditions while out, and make an informed decision from that. I am not going to say whether or not to hike in 20 below with winds over 100 mph. If you can do it, fine. Would I do it? No. Will I set out to rescue you in those conditions? Probably not since we are volunter Search and Rescue (as is state park). We usually only rescue if it will not cause us to be put in harms way. So, use the forecasts, trail conditions/reports, etc and decide what is safe for you. And always remind yourself, it is safer to head down than up almost all the time."
02-05-2009, 08:09 AM
Good question -- I imagine that Mt. Washington's forecasts are done by the most qualified, top meterologists....yet accuracy can never be 100%.
Also, as an aside...one must remember that the forecast for Mt. Washington does not apply to all the summits....there's a difference in wind speed and temps on 4000 foot mountains vs. 5000 footers vs. Washington. Whenever I plan a hike, I look at a variety of forecasts for a variety of altitudes, and I keep in mind which way the wind will be blowing, cloud cover, etc.
02-05-2009, 10:54 AM
Considering the general accuracy of weather forecasting the job that the OBS crew does with this is amazing in my estimation. The slogan for Mt Washington should perhaps be changed from "Home of the world's worst weather" to "Home of the world's worst and most unpredictable weather - and the best meteorologists". :)
02-05-2009, 05:55 PM
The forecasts need to be geared toward the user. For hiking a temp spread of ten degrees is precise enough so is a wind range of forty mph. There is no need to forecast to the nearest degree.
Powered by vBulletin® Version 4.2.0 Copyright © 2014 vBulletin Solutions, Inc. All rights reserved.