Various sources of weather information have got it all wrong. Let me list the ways…
Your crappy weather icons. What do they mean? It’s fine to have a quick visual reference, but only if either there is an easily accessible glossary/guide or if the icons are blindingly obvious. Neither of these are in evidence.
Percentage. What does it mean? 20% chance that there will be rain/snow at some point during the day? Or it will rain/snow 20% of the day?
What is your purpose? Figure it out.
If you are just like an astrology weather service, for people who just want to know what it’s going to look like outside their window then be that.
If you are going to be a data conveyor, with time tables and inches and so forth, be that.
If you are going to be a weather helper, letting us know what is appropriate to wear or plan, if we can drive, if things will be closed, then be that.
But don’t be some half assed useless hodge podge of all three unless you are going to do it right.
Most of the online weather sites I’m used to visiting have stopped presenting – or hidden so deeply I cannot find it – any estimate of “inches of snow.” The regular TV network news still says “3 to 5 inches predicted,” but going online, it’s been very hard to find that.
Since that’s far and away the most interesting and important weather news of all, to working people or people in school, it’s the primary reason I ever look up the weather.
A recent foray onto Weather.com and Weatherbug, drilling down to local reports by zip code for each day and the coming week, did not reveal any “inches” prediction at all, making it impossible to verify the rumors being floated around the office. I finally found a “winter storm warning” link on that first site that brought up a dtailed discussion of the coming snow for my location, including the “dew point,” that did not include any mention of inches predicted, amount of snowfall, or what have you.
And add “wind chill” to the list of weather things people care about that’s no longer (easily) available. Yesterday it was 17 degrees in my town, and there was a prominent graphic showing current wind direction, but no discussion at all of “wind chill.” I know it’s a made-up estimate, but it’s still useful for comparison to previous made-up estimated wind chill I’ve experienced. Wind direction? I’m not trying to decide whether to pull up my anchor and sail for Spain; I just want to know if it’s cold enough I need to wear long underwear.
This seems intentional. I mean, they’ve got to know people are more interested in the anticipated snowfall than the isobar map; more interested in the wind chill than the “Trending [Twitter] keywords in [your area].” I already know everyone else is talking about the snow – why don’t the weather people want to tell me about it?
Am I to be reduced to nodding and telling co-workers, “How 'bout that dew point, though?”
Nate Silver talks about this in his recent book The Signal and the Noise. He says that, while all TV weather forecasters get their information from the National Weather Service, they deviate from the NWS forecast, highlighting any possibility of rain/bad weather. The Weather Channel does it some, while local news can be wildly off from the NWS forecast.
Why? Because of two things:
Rain helps ratings. Tell your viewers in the tease prior to commercials that the rest of the week is sunny and they won’t stay tuned. Tell your viewers that there is rain in the future and they’ll stay tuned to find out that on Friday there’s a 20% chance of rain, but sunny the rest of the time. And if you can tweak your forecast to say that there’s a 40% chance of rain on Friday and 20% on the other days… who is going to check? Who cares?
Viewer reactions to wrong forecasts. Few people complain when you forecast rain and it doesn’t happen. Far more people complain when you forecast sun and it rains. So highlight the possibility of rain and drive ratings up while minimizing complaints. It’s a win-win. (Except for accuracy, that is.)
Nate does mention that with better modeling and vast increases in computing power, the NWS is one of the few instances where modeling chaotic systems has dramatically improved over the past 20 years. Compare that to economists who completely missed or ignored the housing bubble, even though the data showed that one was forming.
Go to weather.gov. Type in your city, state in the box in the top-left of the screen - that will give you the 7-day forecast.
For the various periods, it will tell you the amount of snow that is predicted.
At Weather.com, type in your city+state or zip code (I tried this with zip 83840 which is to have snow tonight according to the NWS.) Click on “Today”, and the amount of snow will be listed. You may have to expand the list to see the amount of snow for tonight, but it will be there.
I recommend this as well. I can’t say if the NWS is going to be more accurate (although at least for me, it blows weather.com waaaay out of the water), but there is as much or as little information as you want there. Definitely you will find predicted rain/snow amounts, broken down into how much to expect during the day and overnight.
You can also find a graph with the hourly predicted temperature/dew point, precipitation chances, severe weather chances, wind speed information, and a lot more. You can look ahead several days on this graph to get an idea of what might be happening at a specific time on a particular day.
If you want to go in-depth, take a look at the forecast discussion. The forecasters talk about the major atmospheric features driving the weather and why they think things will happen the way they predict. Lots of arcane acronyms and abbreviations in there, but I still find it interesting and informative.
The National Weather Service does a great job, and weather.gov is a tremendous resource. Joe Bob says check it out!
Yes. National Weather Service. That’s the only one I use. Dig around that site enough, and you’ll find everything the OP asks about, and everyone else in this thread mentioned wanting.
The only forecast I’ve found more accurate and useful than that is one of the three local TV channels, because they have a damn good meteorologist who knows the local area and can tweak NWS into something with better local accuracy for the various places in the station’s coverage area. But even he isn’t all that much better.
I guess “proficiency in Weatherscript” isn’t a job requirement for their webmaster.
Weather.com lost me when they replaced their formerly fully functional and useful website with yet another shitty, cluttered, eye-gouging Web 2.0 collection of superfluous visual trash. But hey, if you want a weather website that lets you “like” or “dislike” the current temperature (God, how I wish I were kidding), they’re the one for you!
The Weather Channel iPhone app is still okay, I suppose because the screen is too small to pack it with all the garbage they’d like to.
I use the Firefox plugin, Forecast Fox, from Accuweather. It still shows percipitation in inches. It also has a handy popup radar screen, which I find more useful, personally. I find what I really want to know is how much longer a weather system is going to park itself on top of me and what’s headed my way. There’s a version for Chrome, too.
What area? No they are saying (and on the sites I use it is broken down by hour) at 8am there is a 20% chance it will be raining and a 80% chance it won’t based on previously observed conditions. Not that there will be rain in 20% of the state or county or whatever.
I forget which site it was, but I recently saw one that allowed me to register my opinion about the then-current snowstorm. I had a choice of “Love!” or “Ugh!” Sadly, there was no “Barf” option.