Re: assimilating data correctly
Completely unrelated, but it should show how important proper initialization is. Before June 20th (Enderlin EF5 day), most models were very consistent in showing zero prefrontal development in the potent warm sector due to very warm 700mb temperatures. However, the night before, the HRRR was the only one consistently initializing temps. Other models were running 1-2 °C too warm, which obviously led to incorrectly choking out prefrontal development. Because the HRRR was correctly initializing cooler, it was consistently showing robust prefrontal development ~18 hours pre-event, while it took the other models till mere hours before to start showing prefrontal development. If you follow svwx you know what happened later that day...