METRo immediately diverges from obs
I've been looking into a road-T fcst run-to-run consistency issue and it comes down to the fact that during the day (especially mid-morning), METRo quickly diverges from the road-T observations provided in the obs-history. This causes the road-T forecast time-series to get increasingly warmer and warmer from run to run. In the example I have provided here:
For the 18z run: the obs-history value valid at 18z is 41.07 the first road-T fcst values valid at 18:20z is 45.77 the road-T fcst value at 19z is 48.86
I does not seem realistic that the road-T jumps 4.7 degC in 20 minutes.
Any insight would be greatly appreciated.
I have attached the example input and output file(s).