Slap some Bear fat on the south side of an Oak tree, if it runs down, there will be weather.

Dial Twister wrote:

Microclimates.

The next town over from us (a mere 10 km away) often has very different weather than do we.

Imagine trying to forecast for an audience than can span across the entirety of Southern Ontario, from Windsor to the Quebec line, and north to cottage country and beyond.

You're right. So my argument is don't promise what you can't say is going to happen. Do a 5-day, maybe, instead of a 7-day or something within the realm of probability. I've said it before here but outside of a day or two ahead, I simply don't believe them anymore. 

RadioActive wrote:

You're right. So my argument is don't promise what you can't say is going to happen. Do a 5-day, maybe, instead of a 7-day or something within the realm of probability. I've said it before here but outside of a day or two ahead, I simply don't believe them anymore. 

Perhaps they should run a scrolling banner during the forecast...

'Possibly Fake Forecast'

😁

paterson1 wrote:

First off, the tone of this thread is all wrong.  Nobody is promising anything with weather, especially in a 7 day forecast.  It is called a forecast for that reason.  In fact they are not promising anything two or three days ahead either.

So please get over this "promise" thing which has never existed with weather forecasts. Stations that have a weather guarantee usually limit it to the temperature, and even then give themselves a few degrees leeway.   

Also the tone on here seems to be that the forecasters are giving out false information. Saying it is false is implying it is also intentional.  Huh?   Weather systems change all the time.  What was showing 5 days ago can change quickly and with not a lot of notice.
                                                                                                                                                         
 As a few others above have mentioned, the weather can be very different a few blocks from where you live.  Once after work I went to the grocery store which is five blocks away from my house.  Torrential downpour when I was leaving the store.  When I got home my neighbourhood was bone dry and we never received any rain the rest of the evening.   How do you predict that? 

Nowhere in my post did I ever use the word "promise." Nor did I ever say it was intentional. That's your interpretation of a "tone" that doesn't exist in this kind of communication. But you obviously wanted it to say that, so that's how you saw it. 

My point was really pretty simple - assuming they get most of their information from Environment Canada, how does one station wind up with sunshine next week, while the other says it will rain? Of course, weather is very localized and there can be different patterns in different places. But so diametrically opposed? I would love to know how such a huge difference can happen, even in a "best guess" scenario. 

In any event, I've learned not to trust the 7-day forecasts. They're almost never right (except, for some reason, when they're predicting awful weather like a snowstorm, and then it generally happens!) beyond the first two days or so. But sometimes I need to plan for a future event (which I am next Monday morning) and an accurate forecast would be helpful. It just feels impossible to really find one. 

I think the thing to remember is that there are more than just one data set and data models in use.  There are more than 10 different models in use and each one uses its own dataset.  Note all services use Environment Canada’s dataset.   Some use their own dataset’s from sensors that they only have access to! 

A great example is Environment Canada yesterday issued a severe thunderstorm warning for Bobcaygeon yesterday yet both Accuweather and another service didn’t issue a warning for the same area.  It didn’t rain at all in Bobcaygeon yet Fenelon Falls got dumped on!   In fact it was sunny almost the entire time the warning was issued.

   I remember when I was working that there were many choices as to what dataset and models  available and the costs were all very different.   Not all used Environment Canada’s data, either.

It all comes down to what service and dataset that the individual station had access to. 

Don’t blame the on air people.   It’s how the different services available use their respective data.

RadioActive wrote:

Dial Twister wrote:

Microclimates.

The next town over from us (a mere 10 km away) often has very different weather than do we.

Imagine trying to forecast for an audience than can span across the entirety of Southern Ontario, from Windsor to the Quebec line, and north to cottage country and beyond.

You're right. So my argument is don't promise what you can't say is going to happen. Do a 5-day, maybe, instead of a 7-day or something within the realm of probability. I've said it before here but outside of a day or two ahead, I simply don't believe them anymore. 

You have used the word promise many times when talking about the weather.  See above. I have brought this up to a few times before. You seem to agree, but next time you post or comment about the weather there always is something like "don't promise what you can't deliver.." 

Also I wasn't saying that it was all your comments, but the tone of the thread which includes more than just what you wrote about. 

So because stations get most of the information from Environment Canada all the forecasts should be the same?  I think weather forecasters and meteorologists should  be given some room for interpretation.  Even with dramatic weather like a hurricane we see the differences between the Canadian, US and various European models, and the differences can be quite different.