Transcript RegionalModelingUpdate10.ppt

Regional Modeling Update
March 31, 2010
Who is using our model output?
Hits
Canadians
Utah
File Transfers
Utah!!!
New Hardware
• We have acquired 8 nodes of the new Intel Newhalem
processors (last week)..64 processors. They have the
memory bandwidth that will allow much more
effective use of resources
• These will be used to make high-resolution local data
assimilation a reality for us (4-km), making use of as
much of our local observations as possible.
• Exhausting enclosure for the last cluster. Hot air all
heading outside.
• Replaced an old (failing) RAID disk storage array.
• System has been very, very stable
Major Changes and Improvements
• Beginning with the 2009123100 run, the WRFGFS has switched to the latest version (3.1.1.)
– Bug fixes in several of parameterizations we use.
– Thompson microphysical scheme is now double
moment in rain as well as ice (number concentration
as well as mixing ratio calculated). This may help with
a major issue…our light rain starts too late.
– Lot more pbl options
– Gravity wave drag parameterization.
– More stable.
But is it better?
High Resolution 1.33 km Nest
• Western WA only
• Once a day (0000 UTC cycle) to 36 h
• Uses the gap period after we finish all the real-time
work on SAGE.
• Attempt to answer questions:
– What is the payoff in getting the land-water boundaries
and smaller scale terrain much better (4-km hardly has
Puget Sound!)
– Does ultra high resolution improve objective verification or
subjective structures?
– Can it provide a better feed to the NWS for use in their GFE
system (which is going to 2.5 km spacing)?
4 km
1.3
6-hr forecast, 10m wind
speed and direction
4 km
1.3 km
Precip Verification
Wind Direction
Future Evaluation
• Improving PBL and surface drag may
preferentially help 1.3 km (more later)
• Using 1.3 km as a testbed (some problems are
more acute at higher resolution)
A Major Issue Has Been Excessive Wind Speeds Over
Land and Excessive Geostrophy at the Surface –either
too much mixing in vertical or not enough drag. Winds
over land and water too similar
• No magic bullet in PBL tests.
• Recently, we tried something that really looks like
it has potential to help…increasing the friction
velocity….ustar.
• Essentially adds drag, without messing other
things up.
• Perhaps it is realistic, mimicking the effects of
hills.
During the past few months we
have continued our testing
program of various PBL schemes,
vertical diffusion options, etc.
• A test case has been one in which the 4 and
1.3 km created unrealistic roll circulations.
http://www.atmos.washington.edu/~ovens/wrf
_1.33km_striations/
1 km visible
Problem
• Instead of getting open cellular convection,
there are these period cloud streets.
• Look like roll circulation, but of too large a
scale (if you look at sat pics you can see hints
of them).
• Sometimes apparent (but less so) in 4-km.
• Occurs only in unstable, post-frontal
conditions.
Through the kitchen sink at it and
consulted heavily with Dave
Stauffer at Penn State
• Tried a range of PBL schemes (YSU, QNSE, ACM2, MYNN,
MYJ, MYJ with Stauffer mods)
• Added 6th order diffusion and played with diffusion
coefficent.
• Fully, interactive nesting
• Upper level diffusion and gravity wave drag
• Monotonic advection
• Varying vertical diffusion, both more and less
Results
• ACM2 (Pleim PBL and LSM) was the only thing
that helped reduce the rolls.
• It created this stratiform cloud mass that
wasn’t very realistic.
• We asked NCAR to help—nothing yet.
• Did any of these tests suggest fixes for some
of our recurring problems?
Example: Cut vertical diffusion 10
1/8th of normal value
Vertical Diffusion cut to 1/4
Standard
Low Diffusion
Ensemble Kalman Filter
• During the fall, we have tested the EnKF data
assimilation system at 4 km resolution with a
three hour update cycle.
• Results are promising, but it became clear we
lacked the computational power to do this
consistently and dependably (loss of a single
node and we ran behind).
• We also lacked the resources to go to a onehour cycle, which people want.
Some Initial Results
• Bottom Line: We can create a better analysis
• But it doesn’t hold long…we relax back to the
current 4-km skill within a few hours.
• We may be able to do better than this…more
later…but there is a real limitation without
having much upstream mesoscale data.
• The coastal radar might alter this situation.
Plans During the Month
• With new hardware, go to a dependable 3-hr
assimilation cycle at 4-km.
• Switch to the NCAR DART system, not our
homegrown EnKF infrastructure.
• DART has many options that we don’t have
that may improve the assimilation (vertical
localization, satellite radiance and radar assets
assimilated)
• Experiment with 1hr updates
EnKF
• The target is to have by June a dependable
operational system.
• High-quality 3D analyses at 4-km will be
available and archived.
• Short-term forecasts available for other uses.