Understanding probability forecasts

[Frances Boardman]

I am really enjoying using your site. In particular i find it useful when you give a probability of precipitation or snow. Looking at the 10 day forecast for Hemel Hempstead there is a 45% precipation and a 95% chance of snow on 7th January. How can the chance of snow be higher than the chance of precipitation? Surely snow is a form of precipation and there never could be a situation where there was snow and no precipitation.

[Paul]

Hi Frances, welcome to the community

The precipitation and snow risks are seperate, with the snow risk on the basis that if there is precipitation, that's the percentage risk of it falling as snow. If there's a zero risk of precip though, the risk of now will be zero too. 

Hope that helps!

Paul

[Frances Boardman]

Interesting. So the risk of snow on the 7th January is 95% of 45 or 43%. 

[Chris.R]

As far as I understand it, if there was precipitation falling, of which there is a 45% chance this could happen, then there is a 95% chance of this precipitation falling as snow and therefor a 5% chance it will not fall as snow.

To put it another way: it is slightly more likely that no precipitation will fall at all, but if it does fall then it is almost certain to fall as snow rather than anything else.  Hope this helps :-)

 I see Paul has already answered. Yes, 95% of 45%. 

Edited December 29, 2016 by Chris.R