University of Washington Modeling Infrastructure Available for Olympex

1
University of Washington Modeling Infrastructure
Available for Olympex

• Cliff Mass
• University of Washington

2
GoalsThe Local Modeling Effort at the UW

• Provide high-resolution forecasts for mission
  planning.
• Provide high-resolution simulations to drive
  hydrological modeling
• Assimilate a wide-range of mesoscale and synoptic
  observational assets to produce the best possible
  description of the mesoscale structures over the
  region.

3
GoalsThe Local Modeling Effort at the UW

• Evaluate model fidelity, particularly for cloud
  and precipitation fields. Work with partners to
  improve microphysics and other model
  deficiencies.
• Demonstrate the value of combining the model with
  observations to produce skillful snowpack and
  water-related fields.

4
Important Points

• The Olympex area offers substantial precipitation
  and terrain ideal for a GPM testbed.
• Terrain offers the potential to place assets in
  crucial locations, with certainty that you will
  catch the cloud/precipitation structures you
  want.
• Models are very good in the dynamics of
  orographic flows, so you can get the winds right
  fairly easily.
• Then you can tear the microphysics/PBL physics
  apart to find the flaws and fix them.
• Rivers offer a wonderful integration of moist
  processes, both on a short-term and long-term
  basis.

5
NW Modeling Resources

• High-resolution WRF ARW forecasts at 36, 12, 4,
  and 1.3 km grid spacing completed twice a day.
• High-resolution (4-km) WRF-DART Ensemble Kalman
  Filter (EnKF) data assimilation system run on a
  three-hour cycle, with intermittent 24-h
  forecasts.
• 12-km mesoscale ensemble system based on the
  initializations and forecasts of major modeling
  centers.
• Collection of all real-time data assets over the
  region.

6
(No Transcript)
7
(No Transcript)
8
(No Transcript)
9
(No Transcript)
10
(No Transcript)
11
(No Transcript)
12
Optimized Physics for the Region

• Based on testing hundreds of physics
  combinations, domains, and numerical options.
• Best performance plus reliability
• Physics
• SAS Convection on 4, 12, and 36 km
• YSU PBL
• Thompson Microphysics
• RRTM IR, RRTMG solar radiation
• NOAH LSM
• MODIS land use

13
(No Transcript)
14
MODIS Land Use
15
Regional Data Assimilation and Forecasting
16
EnKF System

• Based on a large (64 member) ensemble of
  forecasts at 36 and 4 km grid spacing. WRF model
  and DART Ensemble Kalman Filter (EnKF) System
• Every three hours assimilate a wide range of
  observations to create 64 different analyses.
• Then we forecast forward for 3 hours and then
  assimilate new observations.
• Thus, we have a continuous cycle of probabilistic
  analyses.

17
EnKF Ensemble Forecasting System

• We can run ensemble of forecasts forward to give
  us probabilistic forecasts for any period we
  want. Now doing 24h ahead, four times a day.

18
Improvement in short-term forecast using our
local assimilation system
19
WRF 4 km at same time
20
NWS NAM
21
(No Transcript)
22
(No Transcript)
23
There is Room for Improvement

• Using the data from the IMPROVE-2 experiment, UW,
  NCAR, Stony Brook and others put a lot of effort
  in improving moist physics.
• In general, we do an excellent job on the
  windward side of barriers but often overpredict
  in the lee.
• Probably a microphysical explanation, but PBL
  problems could also be involved.
• OLYMPEX will provide a comprehensive data set for
  the next round of improvements of model moist
  physics.

24
Small-Scale Spatial Gradients in Climatological
Precipitation on the Olympic Peninsula Alison M.
Anders, Gerard H. Roe, Dale R. Durran, and
Justin R. Minder Journal of Hydrometeorology
Volume 8, Issue 5 (October 2007) pp. 10681081
25
Annual Climatologies of MM5 4-km domain
26
2011-2012
27
(No Transcript)
28
2012-2013
29
(No Transcript)
30
Verification of Small-Scale Orographic Effects
31
Dungen
Buck
32
(No Transcript)
33
Cascade Cumulative Precipitation
West LIne
East LIne
34
Work for the Next Year

• Provide model data sets to Jessica Lundquist and
  colleagues to test ability to determine snowpack
  from model output.
• Improve model precipitation/cloud physics
• Intensive model verification year. Use gauges,
  snow measurement, and hydrological verification
• Enhance local data assimilation to include all
  regional radars and additional observational
  assets (e.g., TAMBAR aircraft).

35
The End