AccuWeather: You cannot be serious (new 45-day forecasts)
41 replies, posted
[url]http://www.washingtonpost.com/blogs/capital-weather-gang/wp/2013/08/06/accuweather-you-cannot-be-serious-new-45-day-forecasts/[/url]
[quote=Washington Post (under new ownership!)]Accuweather calls its new 45-day weather forecast launched Monday “revolutionary.” I call it a joke.
By and large, weather forecasts beyond 7-10 days in the future simply do not have “skill.” That is, they provide no more accurate information than the historical average conditions in that time horizon.
AccuWeather insists it is innovating and pushing the envelope by providing these forecasts. In its press release, it boasts of being a leader in long-range forecasting.[/quote]
Alternate title: "AccuWeather introduces "45-day" forecasts, Washington Post's Capital Weather Gang (under new ownership!) calls them out on their shit"
They can barley get it right when its tomorrow let alone 45 days!
I got pretty suspicious when AccuWeather told me the heat index was 127 degrees.
I love accuweather, it's really accurate for me, but this is just bullshit, there is no way that they can predict 45 days into the future considering that stuff can actually happen during this time, billions upon billions of variables.
Hell you might as well consult a fortune teller.
so many variables go into predicting this, we need better models and a hell lot more computing power to get an accurate 45 day forecast
They can get a monkey to throw shit at a spreadsheet on a wall to decide the forecast at 45 days and it will probably be more accurate
[QUOTE=Liem;41737727]They can get a monkey to throw shit at a spreadsheet on a wall to decide the forecast at 45 days and it will probably be more accurate[/QUOTE]
Hmm, I don't think this is true
[QUOTE=Xmeagol;41737707]so many variables go into predicting this, we need better models and a hell lot more computing power to get an accurate 45 day forecast[/QUOTE]
Computer models and computer power can never, ever predict the weather that far. Its 100% unpredictable.
EDIT:
I am talking about predicting 45 days into the future. It can predict pretty good for a couple weeks but not that far.
I stick to the weather channel, accuweather always has my area 4-10 degrees hotter than it really is
[QUOTE=mark6789;41737738]Computer models and computer power can never, ever predict the weather that far. Its 100% unpredictable.[/QUOTE]
It [I]is[/I] predictable, it's just very hard.
The reason it's hard is because you would need to know the exact position and velocity of every air molecule in the atmosphere. Ever single one.
That's 5.25x10^21 molecules, or 5,250,000,000,000,000,000,000
So yeah, you would need a lot of computers.
[editline]6th August 2013[/editline]
[QUOTE=Medevilae;41737833]Weather Channel? AccuWeather?
Get on my level scrubs
[t]http://www.st.nmfs.noaa.gov/nauplius/media/images/logo_noaa.gif[/t]
[URL="http://www.noaa.gov/"]Weather in the raw[/URL][/QUOTE]
Pfft.
[url=http://www.wunderground.com/Aviation/?query=KJFK]Real men use METAR[/url]
[QUOTE=mark6789;41737646]They can barley get it right when its tomorrow let alone 45 days![/QUOTE]
I see this sentiment a lot, and in my experience it's totally unwarranted.
I never use Accuweather or applications/websites associating with it.
Generally I use the BOM's local weather stations for an average temperature in the region.
well the old farmer's almanac has a trend of producing long-range weather predictions as much as a year and a half in advance which is supposed to be based on a "secret formula" based on sunspots and other shit.
apparently they're reasonably accurate, but it could be all just dumb pseudoscientific guesses
Yeah no. Just to get the weather right for the next day a full scan of where the air particles are being brought and the likes- is required, and then a simulation occurs. You can't simulate how the particles will behave 45 days from now, there are just too many factors that lead to countless different possibilities.
what they do is just publish the most likely by percentage
of course the NOAA also has a generalized outlook which is supposed to have predictions [url=http://www.cpc.ncep.noaa.gov/]90 days in advance, too.[/url]
[QUOTE=Sgt-NiallR;41737955]I see this sentiment a lot, and in my experience it's totally unwarranted.[/QUOTE]
Depends where you live. I live in the Northeast USA and there are honestly times when they're wrong about weather 1-2 hours in the future, let alone the next day. I don't necessarily blame them; predicting the weather accurately is an extremely difficult task after all. That in mind, given how inaccurate the standard "7 day forecast" tends to be, we should be focusing on improving the accuracy of the predictions long before we start predicting even further into the future.
[QUOTE=Medevilae;41737833]Weather Channel? AccuWeather?
Get on my level scrubs
[t]http://www.st.nmfs.noaa.gov/nauplius/media/images/logo_noaa.gif[/t]
[URL="http://www.noaa.gov/"]Weather in the raw[/URL][/QUOTE]
Raw weather data is fine if you're the size of a city but you'll always get more relevant forecasts when that raw data is interpreted by local meteorologists
This message brought to you by the Weather Network TM
Mesoscale discussions are all I need.
The Met Office in the UK sort of does forecasts out to 30 days, but the region of 15-30 days is always incredibly vague and talks about general patterns. Or sometimes just hedges it's bets and gives a forecast that can pretty much cover anything:
[QUOTE]The broadly unsettled weather type expected across the UK during mid-August is likely to continue into the early part of this forecast period. Thereafter, during late August and early September, no strong signals currently exist as to most likely dominant weather types. As such, most regions can expect to have fine and at least pleasantly warm weather at times, southern and eastern UK likely to see the most prolonged fine spells. Equally some bouts of unsettled weather are also expected, perhaps most likely to the north and west.[/QUOTE]
Are there any weather companies that present the possible error in their forecasting along with their actual forecast?
Like, "Wednesday - between 15 and 18 degrees at noon, between 10 and 12 degrees at midnight"
Imagine one day when we can perfectly simulate the weather of Earth and predict the weather perfectly for virtually any date?
This is a prediction, not a fact. They know it will be inaccurate, but it can still be helpful.
What does it take to be a weather forecaster, I'm looking for a job and I'm good at making up bullshit on the spot
[QUOTE=fruxodaily;41738851]What does it take to be a weather forecaster, I'm looking for a job and I'm good at making up bullshit on the spot[/QUOTE]
You need a BS degree.
[QUOTE=mark6789;41737738]Computer models and computer power can never, ever predict the weather that far. Its 100% unpredictable.
EDIT:
I am talking about predicting 45 days into the future. It can predict pretty good for a couple weeks but not that far.[/QUOTE]
You would be surprised
Do I have to get the newspaper and spray bottles out, Amazon?
[QUOTE=Mr._N;41738947]You need a BS degree.[/QUOTE]
[url=http://en.wikipedia.org/wiki/BS_degree]he's right, actually[/url]
To accurately 100% perfectly predict the weather over any amount of time you would need to know the exact movement of every single particle of air on the planet.
Sorry, you need to Log In to post a reply to this thread.