AK, I got that impression too and it explains why post 7 day forecasts are very poor and why skill in forecasting isn't much better than 20 years ago. They might improve forecasts for smaller areas (eg I'm in a weather bubble that isn't exactly the same as even a few miles to each side due to the way clouds react to the hills in Wales and the Pennines) but it doesn't make that much difference.

The justification for each bigger computer is that they'll be better at climate but I don't see those predictions getting better. I want to see them recreate the last 10,000 years. They've got values for all the big players.

It s my understanding that UK weather forecasts are not based upon computer models but upon comparisons with former weather conditions. Occasionally a weather forecaster will admit on air that their forecasts for their long range (4days ahead) are less accurate because they could not find a unique previous match, especially relating to the speed with which weather fronts might progress across the country. As records accumulate, these instances will become less frequent.

If you want to see just how poor the computer based weather forecasts are try this. Take a screen shot of the next 24,48,hrs etc and then compare them with the live radar. It reveals just how rubbish they are especially in the longer term. This also illustrates what futile rubbish the climate models are. Any scientist should acknowledge predicting a multivariable multicoupled chaotic system is a nonsense!

Feb 17, 2020 at 9:57 AM AK
In the early 1980s I was working in New Hampshire. I was amazed by the accuracy of the weather forecasts made 2-3 days previously. In hindsight, the weather systems had been tracked and measured on land for a few thousand miles before they arrived.

The last 2 weekends have seen storms over the UK, forecast warnings were broadcast 5+(?) days previously, as they had been tracked across America and then via satellite across the Atlantic.

For rainfall now, will it start or stop, I use Rainfall Radar.

I am happy to accept and acknowledge the improvements in weather forecasting. The Met Office had to accept that its seasonal weather forecasts were not fit for purpose. I hope they can improve their 10-14 day weather forecasts, rather than waste too much effort on 10-100 years.

Re the Met Office's new computer......much needed so they can be wrong more accurately!

Inside the Commission there is an “epic battle between those who want change and those who don’t,” said one EU official, who accused DG AGRI of using “destructive wave of actions to block change.”"
Feb 17, 2020 at 8:45 AM Mark Hodgson

This is sad, and proves that the Green Blob and EU have combined to ensure self destruction.

The original purpose of the Common Market and its post War predecessor was to ensure member states did not suffer food shortages leading to civil strife and ultimately war.

If well salaried EU officials think that the EU Electorates will be happy with food shortages and price inflation, they should be fed cattle cake for a year.

Mark I think you maybe correct that weather forecasting hour by hour for days ahead in regions of high relief may be extremely difficult and could be impossible on a "micro" scale even with future computer technology. I hate to tell you this,but in my flat part of the world we have it for most of the time. I can look at a detailed forecast for several days hence and be reasonably assured that if it's predicted to rain at a certain time, that has a reasonable chance of being accurate.
I am reminded of the accuracy of wristwatches. When I was young, watches I could afford were terrible timekeepers losing minutes every few hours. With the appearance of digital watches you knew the time accurately almost to the second and life changed.
I'm sorry the new super-duper computer is not for the likes of you, it's for us flat-earthers and for the climatically stressed.