Just for fun, let's see if we can work out the relationship between the following four statements:
1. "The coming summer is 'odds on for a barbecue summer', according to long-range forecasts. Summer temperatures across the UK are likely to be warmer than average and rainfall near or below average for the three months of summer." (Met Office press notice 30 April 2009)
2. "You will need a brolly on holiday in the UK in August - the Met Office is issuing a revised forecast for more unsettled weather well into the month. It is a far cry from the "barbecue summer" it predicted back in April. The news will raise questions about the Met Office's ability to make reliable seasonal forecasts." (Roger Harrabin, BBC environment analyst, 29 July 2009)
3. "The BBC has held a high-level seminar with some of the best scientific experts, and has come to the view that the weight of evidence no longer justifies equal space being given to the opponents of the consensus." ("Safeguarding impartiality in the 21st century", BBC Trust Report 2007, justifying the BBC's overtly partial coverage of global warming; and see here).
4. "What of the... corridor of uncertainty? It must be related in some way to the ‘expanding funnel of doubt’ of which we hear so much in our line of work. Perhaps... ‘corridor of uncertainty’ is a generic term, used to describe a ‘funnel of doubt’ whose expansion/contraction properties are less well defined?" (The Actuary, 1 April 2005)
You see, the thing is, if the £180m pa Met Office's three months weather forecast is no better than than the flip of a coin, HTF can they possibly reckon to forecast our climate 50-100 years ahead? And given the rain lashing down on the barbecue, how can anyone possibly believe they are capable of that?
Admittedly, Tyler is no scientist (although I dare say he knows more about number krunching than the BBC's Environment analyst, who has an English degree). But in all the many forecasting exercises in which Tyler has been involved over the years, there is always this thing called the Funnel of Doubt.
The Funnel starts out nice and narrow - anchored to what we understand as today's reality.
But as you extend your forecast further into the future, the funnel expands rapidly. Known unknowns bombard the forecast from all sides, so that by the time you've got out a couple of years - at most - the funnel is so wide you're pretty well back to coin flipping.
A typical example is this chart showing 20 year projections of US government bond yields, ranging from roughly zero right up to 14%:
And that's without even considering the unknown unknowns, which have a nasty habit of cropping up... well... unknowably. You can't even model them.
So what of that sodden barbecue?
"Some now ask if the Met Office risks its reputation by attempting to popularise its work this way. Certainly, at the time of the forecast there was pressure on the Met Office from tourism chiefs in the UK to be positive about holidays at home. Did Met Office staff feel an obligation to put on a sunny face?"