openWRF production runs are in the process of being converted to a strategy of dropping the first 6 hours of forecast as this is the spin-up time of the simulation and it tends to have noise and artefacts which disappear after a few hours of simulation. This results in forecasts starting at xx with reference times of xx minus 6hrs.
In the run above there was a setup problem and the 00Z cycle was used instead of the 06Z cycle. Regarding the GRADS charts: the code producing these is common to all openWRF runs and it will be updated to reflect the 6 hours between reference time and forecast start time once all openWRF runs have been converted to the new strategy.
Patience, please.
David