Spring Snowmelt Forecasting
Weekly snowmelt forecasts are provided each spring
to assist water management entities in dealing with the timing and
magnitude of the spring snowmelt peak. Coordination is required
with many cooperators. The CADWR Snow Survey section participates
in the process for watersheds in California. Weekly forecasts are
provided for all major reservoirs and several important control
points.
To generate the weekly forecasts, the CNRFC uses
the flood forecast model described above in an ensemble mode. This
approach, called Ensemble Streamflow Prediction or ESP, provides
a distribution of future streamflow outcomes that can be sampled
in any way desired. The process starts with the current model conditions
(or states), and then develops a set of forecast scenarios based
on the historical observations of temperature and precipitation
(Figure 1). The approach is extremely flexible in that any time
window and any attribute of streamflow can be analyzed. The most
typical use is for weekly, monthly, and seasonal volumes and peaks.
In addition, the system can be used to describe a whole host of
information such as the distribution of low flow stages during the
upcoming summer.
Figure 1 - Ensemble Streamflow Prediction technique
Figure 2 shows the distribution of future outcomes
for one of the CNRFC forecast points. Spring snowmelt forecasts
are available on the CNRFC Home Page.
Figure 2- Trace outcomes of ESP procedure
|