and evolution of the transition of precipitation type (e.g., rain to snow, snow to
rain, etc.).
g) The addition of a blowing snow potential product allowed users to get a better feel
for the duration of an event.
There were also some issues that arose during the field season that were problematic.
These issues and lessons learned also provide useful information for future researchers
and developers of MDSS like capabilities. Some of the outstanding issues are listed
below.
a) Forecast Period: The decision to shorten the forecast period of the mesoscale
models from a 24 to 15 hour period in favor of hourly runs was flawed because
storm staffing decisions were made about 15 hours beforehand, typically 12 to 3
PM the day before the event. Many of the decisions were based on forecast data
from 15-24 hours, which was the period not covered by the mesoscale models.
There were cases where the mesocale models were correctly predicting rain up to
15 hours and then the MDSS precipitation type switched to snow at hour 16
because of the data source changed to the older runs of the NWS models (Eta or
GFS).
b) Road Frost: The lack of guidance on road frost remains a hole in the prototype
MDSS. Users have frequently asked for improved road frost guidance. There are
several road frost research efforts that could be leveraged toward the development
of a more mature product. The inclusion of a road frost product (based in part on
work performed by Iowa State University) in the 2004-2005 MDSS prototype will
be beneficial to the user of the Colorado MDSS Test Bed.
c) Winter Precipitation Measurements: The lack of quality winter precipitation
measurements impacts the ability to identify snowy regions, determine where
blowing snow may be occurring, and impacts the skill of predicting quantitative
precipitation amount.
d) Weather Prediction Models: Weather prediction models are evolving over time
and improving. Care must be taken to ensure that the data used is what is
expected. The MDSS team discovered several issues related to short wave
radiation fields from GFS, Eta, and WRF that were not resolved or addressed until
mid way through the demonstration. NCEP corrected some of the Eta problems in
late March, too late for the field demonstration. Those using data from both the
research and operational models are cautioned to monitor the output and track
updates to ensure the data are actually what was expected.
Based on the analyses presented herein, we can conclude that the second demonstration
of the MDSS was successful as the performance of the prototype was significantly
improved, particularly the RCTM. Several outstanding issues were resolved between the
first and second year. Several key capabilities (e.g., methods, techniques, code,
165
Previous Page
