There are some very hot members in the GEFS 06z (likely including some 40C+ maxes), so although we've had a couple of MEHish runs from several models in the last 24 hours the overall pattern remains.
In 2022 I got into the habit of calculating the "average max" 850hpa across all the ensemble perturbations. Because they all peak at different times the average 850 line is always lower than where most runs will end up.
On this 00z the maxes for London are, from P1-P30 plus CON and OP: 17, 20, 20, 18, 23, 21, 15, 20, 18, 14, 23, 16, 23, 15, 20, 16, 14, 17, 18, 15, 23, 20, 17, 18, 17, 20, 19, 16, 20, 16, 22, 16
The average (mean) max is 18.34C. At this time of year and assuming sunshine that equates to a surface max of around 33-34C. So the average model perturbation is peaking at some point at 33-34C. There are 13 runs with a max of 20C+, i.e. at least 35-36C, and 4 runs that hit 23C: getting close to 40C. The coolest peak in the set is 14, which still gives us at least 30C.
For GEFS 06C we have an average max of 17.5C and a few runs only getting to 13C. So the peaks are a bit lower although there is one member at 24C.
Originally Posted by: TimS