I wasn't going to blog today...but had to mention the convection--heavy showers and thunder--that rolled through on both sides of the Cascades. The radar showed it nicely---here are a few samples.
And all of this during one of the climatologically driest weeks of the year. A number of locations along the eastern slopes of Cascades, around the Tri-Cities, and over the south Sound got .15 to .20 inches in the downpours.
And I hate to admit it, but the forecasts were not good regarding these thunderstorms--both the human predictions and computer models. The NWS forecast yesterday was correct about a major cool down today as the upper trough pushed in, but pretty much missed the heavy showers. The models also did not do well. At least we have plenty to work on!
Now this was a classic convective pattern, with a sharp trough moving in:
But clearly some ingredient was missing in the models--either how the models were initialized or their descriptions of the physical processes.
Perhaps a hint of the origin of the failure can be found from the infrared satellite sequence today. Here are images from last night through this afternoon. My eye sees two features: one associated with the trough moving in from offshore and another feature initially over Oregon. It is when they join that the convection blew up in a way that Friday's computer runs missed. My own examination of the model output suggests that the eastern convection was associated with a weak upper trough over land that the models underplayed....but I only looked at the situation briefly. It takes some detective work to conclusively discover what is behind such forecast failures.