The days of unreliable forecasts could be over - as the Met Office has started using a computer programme so powerful it can foretell the weather across every mile of the country. 

Read more: http://www.dailymail.co.uk/news/article-2755203/New-Met-Office-supercomputer-end-forecasting-gaffes-predicting-localised-cloudbursts-black-ice-areas-previously-small-pinpoint.html#ixzz3DMOBdU74 
Follow us: @MailOnline on Twitter | DailyMail on Facebook

What a load of tosh. Yes no doubt the new computer will aid the prediction of weather somewhat but I totally disagree that the Met Office got the flooding risk wrong last Winter. It was well forecast down here (weeks) in advance and was only catastrophic because of the sheer volume of water and nowhere for it to go due to the fact that River dredging had not been done for years meaning it's not the fault of an incorrect forecast from the Met. Of course if you talking about seasonal predictions from months ahead then this computer will probably be as unreliable as the last as the weather has a habit of kicking us in the teeth at times two days ahead let alone months. If I hadn't of looked closer I would of thought it was taken from the Express. In any event I think the column can be disregarded again as it is just another reason to have a dig at the Met rather than a revolutionary breakthrough in forecasting accuracy.

Did someone mention tosh and weather

This is the Met Office's own article about this new software, ENDGAME

http://www.metoffice.gov.uk/barometer/science/2014-08/ENDGame

It already wins first prize for me in the category of worst reverse engineered acronym ever

"Even Newer Dynamics for General Atmospheric Modelling of the Environment"

but it is a close race between this and the even higher resolution successor to be launched in a few years, Gung Ho

"Globally Uniform Next Generation Highly Optimized"

The ENDGame article states this will be used "across all weather and climate timescales" implying that they will be resurrecting some sort of long range (seasonal ?) forecasting that they dumped after the 2009 Barbecue Summer

I'm afraid my belief is that this sort of forecasting is impossible with mathematical models however high def they may be. IMO the weather is too chaotic to predict much more than 7-10 days ahead

You only need to look at the LRF synoptics produced by CFS models etc to see how it turns on it's head every couple of days as the output draws a map of the noise generated by the models and not much else

You obviously don't work in NMR (nuclear magnetic resonance) then - my own favourite is called INADEQUATE. This comes from Incredible Natural Abundance DoublE QUAntum Transfer Experiment. NMR is full of acronyms, most of which seem to have been concocted to fit a 'cool' acronym, but this one is superb, and fittingly is also acurate as the technique is mostly useless except in highly unusual situations... Inadequate indeed!