The Pacific Northwest Mesoscale Forecasting System: An Update

Download Report

Transcript The Pacific Northwest Mesoscale Forecasting System: An Update

Overview of the Pacific Northwest
Environmental Prediction System
Supported by the Northwest Modeling
Consortium…the regional modeling effort centered at
the UW is
• Running the MM5 at 36, 12, and 4 km resolution
• Running the new WRF model at 36, 12 km and 4 km
resolution
• Running TWO high resolution regional ensemble systems
to provide probabilistic forecasts and data assimilation
• Gathering all local weather observations from dozens of
networks. Plus quality control.
• Running a wide range of weather applications dealing with
air quality, hydrology, transportation weather and fire
weather.
http://www.atmos.washington.edu/mm5rt/
36 km
12 km
4 km
NWNet: Regional Real-Time Collection of Over 60
Networks Over the Pacific Northwest
The UW Quality Control System
• A major task continues to be the gathering
of all real-time observations of the region
into one place
• Right now we acquire over 60 networks in
real time for displaying on our web site,
verification, and many other uses
• Quality Control is essential for such a
heterogeneous network of networks.
The UW Quality Control and
Warning System
• We have developed an advanced QC system
suitable for an area of complex terrain Have
also created an automated QC display
system that one can check on the web and
which can automatically tell the manager of
a network when their data is suspect
The effort has
roughly three
clusters of Linux
machines and 120
TB of storage
The “Audience” for NW MM5 Products
Continues to Increase
The UW Ensemble System
• The UW ensemble system was borne out of experience
from the high-resolution local MM5 effort (36-12-4 km
resolution)
• Specifically, although high resolution in general
produced better (sharper, high amplitude) structures, the
forecasts verified only marginally better than lower
resolution forecasts using traditional measures.
• UW research on forecast verification and evaluation
revealed large differences, and thus uncertainty, in the
initializations and forecasts of major operational
forecasting systems.
• Also apparent that there is considerable uncertainty in
the model physical parameterizations.
UW Ensemble System
• Previous results showed that approximately 12-km
resolution was needed to get the major regional
mesoscale features “right.”
• Thus, it was natural to create a 12-km mesoscale
ensemble system for the Northwest.
UW Mesoscale Ensemble System
• Single limited-area mesoscale modeling system
(MM5)
• 2-day (48-hr) forecasts at 0000 UTC in real-time
since January 2000. New 12 UTC cycle
• 36 and 12-km domains.
a)
b)
36-km
12-km
Configurations of the MM5 short-range ensemble grid domains. (a) Outer 151127 domain with 36-km
horizontal grid spacing. (b) Inner 103100 domain with 12-km horizontal grid spacing.
UW Ensemble System
• UW system is based on the use of analyses and
forecasts of major operational modeling centers.
• The idea is that differences in initial conditions of
various operational centers is a measure of IC
uncertainty.
• These IC differences reflect different data
inventories, assimilation schemes, and model
physics/numerics and can be quite large, often much
greater than observation errors.
• In this approach each ensemble member uses
different boundary conditions--thus finessing the
problem of the BC restraining ensemble spread.
“Native” Models/Analyses Available
Resolution (~ @ 45 N )
Abbreviation/Model/Source
Type
avn, Global Forecast System (GFS),
Spectral T254 / L64
~55 km
National Centers for Environmental Prediction
cmcg, Global Environmental Multi-scale (GEM),
Computational
Distributed
1.0 / L14
~80 km
Objective
Analysis
SSI
3D Var
Finite
Diff
0.90.9/L28 1.25 / L11
~70 km
~100 km
3D Var
Finite
Diff.
32 km / L45
90 km / L37
SSI
3D Var
Spectral T239 / L29
~60 km
1.0 / L11
~80 km
3D Var
Spectral T106 / L21
~135 km
1.25 / L13
~100 km
OI
Spectral T239 / L30
Fleet Numerical Meteorological & Oceanographic Cntr.
~60 km
1.0 / L14
~80 km
OI
tcwb, Global Forecast System,
1.0 / L11
~80 km
OI
Canadian Meteorological Centre
eta, limited-area mesoscale model,
National Centers for Environmental Prediction
gasp, Global AnalysiS and Prediction model,
Australian Bureau of Meteorology
jma, Global Spectral Model (GSM),
Japan Meteorological Agency
ngps, Navy Operational Global Atmos. Pred. System,
Taiwan Central Weather Bureau
ukmo, Unified Model,
United Kingdom Meteorological Office
Spectral T79 / L18
~180 km
Finite
Diff.
5/65/9/L30 same / L12
~60 km
3D Var
Relating Forecast Skill and Model Spread
Mean Absolute Error of Wind Direction is Far Less When
Spread is EXTREME (Low or High)
Ensemble-Based Probabilistic Products
Local Data Assimilation using an
EnKF System
• The system produces 90 different analyses that can be
combined to produce the best guess at what is there and tell
us the uncertainty in the analyses.
• These analyses can be integrated forward in time to give us
probabilistic predictions of the future
• We now have it running at 36 and 12 km resolution…
A Vision of an Integrated
Regional Prediction System
Output from the UW MM5 is now being fed into
a number of modeling and diagnostic systems:
• Distributed Hydrological Model for Western
Washington
• Calgrid Air Quality Model
• Land Surface Model for Surface Temperature
Prediction
• Smoke, Ventilation, and Fire Guidance
• Transportation Information System
The UW
Coupled
MM5DHSVM
Hydrological
Prediction
System
DHSVM: Distributed Hydrology Soil Vegetation Model
• Terrain - 150 meter
aggregated from 30
meter resolution DEM
• Land Cover - 19
classes aggregated from
over 200 GAP classes
• Soils - 3 layers
aggregated from 13
layers (31 different
classes); variable soil
depth from 1-3 meters
• Stream Network based on 0.25 km2 source
area
DHSVM
DHSVM Distributed Hydrological Prediction System
11/25
12/01
12/07
12/13
December 11-12, 2001 Santium River
12/19
The UW/Washington State
University Coupled MM5-Air
Quality
Prediction System
AIRPACT Regional Air
Quality Modeling System
IC/BC
landuse
terrain
MM5
u, v formatted
for each layer
of CALMET
landuse
terrain
IC/BC
emissions
chem mech
dry dep
p
CALMET
CALGRID
3D met field:
u, v, w, T,
BL variables
3D species field:
O3, VOC, NOx,
primary PM
Calgrid Air Quality Prediction System
AIRPACT Current
Developments
• Expand domain
• Add air toxics
• Improve PM emissions
inventory
– woodstoves & other
primary PM sources
• Improve web graphics and
GIS content
• Long term: convert to
CMAQ
AIRPACT Output Products
PNW 2001 O3 Enumclaw Forecast
Aug 10 - 14, 2001
0.080
0.070
O3 (ppm)
0.060
0.050
Obs
Pred
0.040
0.030
0.020
0.010
0.000
221
222
223
224
225
Day of Year
226
227
228
Road Weather Information System
• This effort is a partnership between the UW and
the Washington State Department in
Transportation, with funding from the US
Department of Transportation.
• An attempt to combine weather data, modeling,
road information, and other data sources into
applications that can serve the public and the
Washington State DOT.
• Rick Steed will provide a detailed briefing.
Washington State DOT Traveler Information System
U.S. Forest Service Smoke and Fire Management System
Ventilation
Index
U.S. Forest Service
• MM5 grids are sent to the field for running
Eulerian and Lagrangian smoke
plume/dispersion models.
• MM5 output used for fire fighting
operations.
BlueSky
Simulating Wildland Fire
in Real-Time
(www.fs.fed.us/bluesky)
Susan O’Neill, Sue Ferguson
USDA Forest Service
Rob Wilson
US EPA
BlueSky
Smoke Modeling Framework
What is it?
• Real-time Smoke Concentration
Predictions: Prescribed, Wild,
Agricultural Fires
• Daily Emission Tracking from MultiAgency Burn Reporting Systems
• Quantitative Verification
• Automated, centralized processing
– Forecasts for 5 domains daily
• Web-access output products
FIRE Characteristics
Area Burned
Fuel Moisture
Fuel Loadings
Fire Location
Fire Ignition Time
BlueSky
Smoke
Modeling
Framework
Emissions
Calculate fuel consumption
and
EPM/COMSUME
v1.02
variable
BURNUP rate emissions of: Heat
Released, PM2.5, PM10, PM,
CO, CO2, and CH4
Meteorology
3-d Wind/Temp/Moisture
UW MM5 Forecast System
12 km Domain
72 Hour Forecast
Smoke Dispersion
CALPUFF
Visibility
Chemistry
HYSPLIT
PM Concentrations
(CMAQ)
Plume Rise
Web Display of Output Products (RAINS)
Animations, Zoom In/Out, Concentration Fields,
Trajectories, Meteorological data,
Overlay GIS Data
BlueSkyRAINS Output Products
BlueSkyRAINS Output Products
BlueSkyRAINS Output Products
BlueSkyRAINS Output Products
BlueSkyRAINS Output
Military Applications
• The NW MM5 is now the main source of
regional forecasts for Navy and Air Force
operations at Whidbey NAS and McChord
Air Force Base, as well as the Everett
Carrier homeport.
The End