are the stations that collect data proprietary? And do different people have different information? Or do they use different type of analysis to come up with a degree or two of difference/different ideas about precipitation.
Edited to add an example:
I use the Wewow app and it lets you choose from different weather services and shows forecasts for each of them. For 11 PM tonight here’s a sample is what’s forecasted (c = degrees celsius, pop= probability of precipitation).
World weather online 10c 75pop
AccuWeather 12c and 48pop
Aerisb13c and 0pop
Thanks!
Editing again to say I’m in Ontario, Canada
In: 3
Can you give an example? In the US, at least, basically everyone reporting on the weather gets it from teh same place – the national weather service. I don’t think I’ve really ever seen contradicting reports from different places because the underlying data is all the same.
The simple answer is that weather is really complicated and the data is extremely noisy, devices that are a few meters apart can measure slightly different results and lead to different predictions. Even if they were all the same the predictions aren’t an exact science (they’re very good but obviously not perfect) so a little change in measurement can lead to a bigger change in prediction. That said they keep getting better and noise reduction is constantly improving so the predictions are never too far off
To put it simply: chaos.
In the sense that weather is really complicated and effectively impossible to actually predict, just estimate. That’s why, for example, you always get a report of the CHANCE it will rain, not a “yes” or “no”.
Even two services working with the same data can come to different conclusions depending on the simulations they run. And the same simulation can give different results if you slightly change the initial data.
Of course it’s possible for a service to be just flat out wrong if they have bad data and crappy simulations, but its difficult to check even that.
I find that they don’t vary very much.
But sometimes there is a band of bad weather coming though, and one service will predict it’s arrival 6 hours earlier than another. Or that it will pass 50km further north or south.
So if you look at the prediction for one town at one time it could breed very different. But if you look at the bigger picture you are just seeing a small difference in the predictive model.
It depends heavily on what they consider the forecast area and forecast duration. One service’s definition of Ontario may be larger or smaller than another’s. I think this is the case for your example, as the temps aren’t drastically different but the precipitation chance is.
Precipitation chance is really the amount of a given area experiencing rain over a given interval.
A storm covering an entire area for an entire interval is 100% chance of rain. A small storm that only covers a tiny strip but moves across the entire area in the interval is also 100%. The same storm that only moves in halfway is a 50% chance.
One service may consider a 5 km^2 area over 1 hour having a 50% chance, while another may consider a 10 km^2 area over 30 minutes having a 30% chance even though they’re talking about the exact same storm centered on the exact same location.
Are you referring to the measurement of current conditions or to forecasts?
There are numerous forecasting models that meteorologists can choose from, so their answers will differ slightly.
Latest Answers