UK Sunshine card
These are the hourly sunshine totals for yesterday across the UK (fig 1). As you can see Shoeburyness was marginally the sunniest place with 13.0 hours, but many place across southern areas were close to 100% of the maximum possible. Places like Shoeburyness with the sun rising over the North Sea in the east, and setting over a flat Essex to the west must have a slight advantage over places like Exeter for example, where the sun is slightly blocked by the Blackdown hills to the east and by Dartmoor to the west, which may shave off 0.2 hours on a cloudless day.
Interestingly in my application I made the 13.0 hours at Shoeburyness only 92.2% of the theoretical maximum. Checking on the timeanddate.com website tells me that the day length at Shoeburyness was 14:07:30 for yesterday, so the 13.0 hours of reported sunshine was indeed only 92.0% of the total possible, so the algorithm I use is quite accurate. But what happened to the missing 1:07:30 seconds of sunshine? The total day length would of course start when the top limb of the sun first appeared over the horizon until it disappeared below the horizon in the west, this process may account for 15 minutes or so, but I can’t imagine haze in the atmosphere making up for remaining 0:52:30 on a cloudless day with good visibility like yesterday (fig 2). Who knows perhaps there was some thicker cirrus that went undetected by the LCBR, or maybe we have the threshold for bright sunshine set too high these days, in order to mimic the totals we used to get from the old Campbell-Stokes recorders?
Adding hourly totals works
Reassuringly, as long as you collect all the hourly SYNOP observations, then adding the hourly sunshine totals does produce a figure that’s very close to the one reported at 06 UTC the next day (fig 3).