First let me make a disclaimer saying that both forecasts you compare are neither mine and actually they are both done by the same US government agency, the National Weather Service. In fact, on line, they are separated by one click of the mouse on the following NWS forecast website for Deerlodge Park;"Hydrograph" vs "Probability Information".
As stated, the data plotted on my WESTERN RIVERS hydrograph array is a product of the National Weather Service. I have gone through considerable pain to process the data they allow for download. May I repeat, considerable pain. I kinda have to call it poaching. If they wanted to allow me to do it easily they'd at least made it f'in easier.
I only offer accurate forecasts, my own, when I'm invited on a Yampa trip launching that week
or any week(just kidding).
The two "for-comparison" data sets(charts
) are not out of agreement when examined closely. The data I plot from the NWS is a refinement of the most current conditions(today!) interpolated forward to generate a 10 day forecast in one hour increments. The bar chart hydrograph that you extract the 9K flow from is a "WEEKLY" statistical probability projected months forward. You've got a lot of averaging going on when considering the average flow for an entire week. However, when you draw the lines between the forecasts there is a perfect agreement.
Unfortunately, when smoothing to a weekly average, as presented on the bar histogram, you might miss a peak, such as a 20k spike for a day smoothed to perhaps 14K for the week on a histogram. It is just the way it is.
Let me know where my short explanation can be improved. I welcome any opportunity to discuss Western River flows. A further disclaimer, I'm a geologist. Although, I have intimate knowledge of data processing in most earth sciences. I will provide or seek an answer if I can.