This tutorial will focus on a series of storms that occurred over the Russian River basin in California in Early-Mid January 2023. These storms were atmospheric river events that are common to the region. The data in this example is Multi-Radar Multi-Sensor (MRMS) precipitation data and is therefore not a QPF. While the data is not actually a QPF, the methods described within the tutorial are translatable to QPFs in a real-time forecasting scenario. The MRMS data was gathered from Cumulus for the entire extents of the California Nevada River Forecast Center (NCRFC). The extents of the CNRFC are much larger than the Russian River watershed which will allow for the spatial translation demonstrated in Part C of the tutorial.

Throughout this tutorial, the Ukiah subbasin will be used as an example. In this subbasin, the average hyetograph during the simulation is shown in the figure below that demonstrates that a series of large storms hit the area.

Running this precipitation data through HEC-HMS yielded the flow estimates show below. In this figure, the peak flow was near the end of the lookback, with a flow of ~7600cfs. 

Simulation results for lower in the Russian River basin (Johnson's Beach Gage) are shown in the figure below, where fewer peak flows are observed than higher in the basin. In this hydrograph, the peak flow is just before the Forecast Time, but a forecasted peak on January 14th is also very high. This may lead a modeler to consider "what-if" scenarios, where the forecasted peak flow may be altered if the QPF is missing something. Given the lead time of nearly 5 days to the next peak, there is considerable uncertainty in the flows.