As some of you may know, I have a strange, difficult-to-explain interest in weather forecasting. It predated my brief experience, auditioning to announce predictions on the local news (described here) and had something to do with my leaky roof…but that's not the whole story. Anyway, the Internet affords me — and you — an interesting means of getting perhaps the most accurate weather forecast that is humanly possible…moreso than listening to any one TV weatherperson or dialing the forecast on the phone or reading any newspaper.
There are several outfits that predict the weather, starting with the National Weather Service and continuing through private companies like AccuWeather and The Weather Channel. Consulting all of them is like getting opinions from a number of doctors instead of just one. If they all agree, you can have great (though not absolute) trust in the verdict. If they don't, you can get some idea of the parameters. That is to say that if one says there's a 20% chance of showers and another says 40% and another says 60%, you can figure on 40% but with a lower level of confidence than if all said 40%.
More useful though are the sites where the forecasters explain to other forecasters, how they arrived at their forecasts, how much confidence they have in them, when they think they may have to modify them, etc. As far as I know, only the NWS forecasters supply this and perhaps not all of them. The one for Los Angeles can be read at this link and since it changes several times a day, it may have changed by the time you click over there. Right at this moment though, it reads, in part…
BY WEDNESDAY, ALL BETS ARE OFF AS THE ETA AND CANADIAN MODELS HAVE MOVED TO A MORE PROGRESSIVE PATTERN WITH THE TROF WHILE THE AVN, MRF, AND ECMWF CONTINUING TO INDICATE MORE OF REX BLOCK PATTERN WITH THE LOW CUTTING OFF WELL OFFSHORE AND EVENTUALLY MOVING WELL SOUTH INTO BAJA. SUCH A HUGE DIFFERENCE IN THE HANDLING OF THIS PATTERN IS RESULTING IN A VERY WISHY WASHY FORECAST AT THIS TIME. MY GUT FEELING IS THAT THE LOW WILL BEHAVE MUCH LIKE THE AVN SUGGESTS AND LEAVE US HIGH, DRY, AND MILD. BUT I FEEL LIKE I CAN'T COMPLETELY IGNORE THE ETA/CANADIAN SOLUTION. EVEN THE 18Z MESO ETA CONTINUES THIS MORE PROGRESSIVE SOLUTION. SO HAVE DECIDED TO ADD A 20 PERCENT POP FOR WEDNESDAY IN ALL ZONES WITH THE CAVEAT THAT CONFIDENCE IN ANY ONE MODEL SOLUTION IS EXTREMELY LOW. CONCEIVABLY SPOTTY SHOWERS COULD ARRIVE ALONG THE CENTRAL COAST IN THE PRE-DAWN HOURS WEDNESDAY BUT WILL JUST INDICATE POPS FOR THE DAYTIME PERIOD THERE. IN ALL LIKELIHOOD REALITY WILL BE SOME COMBINATION OF THE CURRENT SOLUTIONS, BUT THE RESULTING WEATHER COULD BE JUST ABOUT ANYTHING. IN FACT, IF THE ETA WAS RIGHT ON TARGET THUNDERSTORMS WOULD NOT BE OUT OF THE QUESTION WEDNESDAY.
THE GOOD NEWS IS THAT ALL SOLUTIONS AGREE ON A DRY AND WARMER PATTERN FOR THURSDAY THROUGH SUNDAY SO THE ONLY REAL QUESTION MARK IS WEDNESDAY.
Some of that may seem unduly technical but all you really have to know is that they use various computer models, and that the gent who wrote this is trying to decide which one is giving the more accurate projection. There's a low approaching the coast and it may dip down and move inland over Baja, or it may come straight on in through Southern California. He's using the different computer projections the way I suggest you use the different services' forecasts — going by the majority opinion but allowing for any dissenting voices. (You usually find, as a given weather-maker nears, diverging forecasts slowly merging. That is, NWS will say a storm is coming on Thursday evening; AccuWeather will say Saturday morning and, as the front grows nearer, both will amend and inch over to mid-day Friday.)
I don't know about you but I find this kind of thing enormously helpful. The forecast I can read in the paper or on any public site right this minute will tell me that Thursday and Sunday will be dry and warmer but they won't tell me that there seems to be great confidence of that. (And not only do all the NWS models point that way but so do the AccuWeather and Weather Channel projections.)
The NWS public forecast tells me there's a 20% chance of showers on Wednesday but it doesn't tell me they're saying that because, essentially, they're hedging their bets. Most of their computer models think the storm will come nowhere near us but one says it will hit L.A. It would be quite a different thing if all indicators were saying that the storm would definitely come our way but there would only be a 20% chance of it yielding precipitation.
If you'd like to "triangulate" this way and read the various weather services' opinions, you can get your local NWS forecast at http://www.nws.noaa.gov and with that as a starting point, you may be able to find a "discussion" page for your local forecast. The AccuWeather forecast may be obtained at www.accuweather.com and the Weather Channel forecast is over at www.weather.com. There are several other services on the 'net that offer predictions — some, arrived at via still other computer models; others, disguised versions of one of these — but I tend to think that the three I recommend should do it for anyone. My experience has been that, on occasion, all are wrong…but that by consulting them all, you get a much more accurate forecast than by consulting any one.