Progress and Problems with Forecasting Orographic Precipitation over the Pacific Northwest and Southwest Canada Clifford F. Mass, University of Washington, Seattle, WA

1
Progress and Problems with Forecasting Orographic
Precipitation over the Pacific Northwest and
Southwest CanadaClifford F. Mass, University
of Washington, Seattle, WA
AMS Mountain Meteorology Conference, August 2008
2
Orographic Precipitation is an essential part of
the regional meteorology
3
Few Areas of North America Experience Such Large
Amounts and Gradients of Precipitation
4
(No Transcript)
5
Northwest Orographic Precipitation Has Major
Societal Impacts
Flood Control on Dozens of Dams (Wynochee Dam
shown)
6
Billion-Dollar Storms Are All Associated with
Orographic Precipitation
7
Mount Rainier National Park18 inches in 36 hr
(Nov 8, 2006)
8
Dec. 3, 200720 inches in two days over coastal
terrain of SW Washington
The results massive landslides and river
flooding
9
And, of course, the 2010 Olympics will depend on
our understanding and predictive capabilities for
orographic precipitation
10
Northwest U.S. and S.W. Canada an excellent
testbed for studying orographic precipitation

• Relatively simple terrain of various
  configurations
• Olympicsan orographic island
• Vancouver Island and portions of Cascades (linear
• Undisturbed flow approaching the barriers
• Accessible with a large number of surface
  observing stations
• Major high resolution real-time simulation
  efforts at the UW and University of British
  Columbia.
• Lack of deep convection.

11
There have been major progress in understanding
and predicting orographic precipitation over this
region during the past several decades

• A number of regional field experiments have led
  to substantial advances in understanding.

12
Major Regional Orographic Precipitation Field
Experiments

• CYCLES (1970s)
• COAST (Dec. 1993, Dec. 1995)
• IMPROVE 1 (Jan.-Feb. 2001)
• IMPROVE 2 (Nov.-Dec. 2001)
• COASTAL OLYMPICS (2003-2004)
• Proposed OLYMPEX 2010

13
Progress

• Long-term real-time NWP and case-specific
  numerical experiments have examined the strengths
  and weaknesses of orographic NWP in the region.
• Prior to roughly 1995-2000 operational center
  models lacked the resolution and physics to even
  begin to handle the regional precipitation.
• NWP is now resolving major orographic
  precipitation features of the region.

14
NGM, 80 km,1995
15
NGM, 1995
16
2001 Eta Model, 22 km
17
2007-2008
12-km UW MM5 Real-time 12-km WRF-ARW and
WRF-NMM are similar December 3, 2007 0000 UTC
Initial 12-h forecast 3-hr precip.
18
2007-2008
4-km MM5 Real-time
19
NWS WRF-NMM 12-km
20
NWS WRF-NMM (12-km)
21
UW Real-Time Prediction System

• Running the MM5 and WRF-ARW at 36-12-4 km since
  1996
• Thompson Microphysics
• NOAH LSM
• Run twice a day to 72h
• Verified with thousands of stations from over 70
  networks. Long record of model biases and issues
  over terrain.

22
Domains
23
A Few Major Lessons

• There are several key horizontal scales that
  influence orographic precipitation. The first is
  the scale of the major mesoscale barriers (e.g.,
  west slopes of Cascades, mountains of Vancouver
  Island).
• In order to resolve the influence of the these
  features, one needs grid spacing of 12-15 km.

24
100 km
25
36-km
12-km
26
Major Lessons

• Then there are smaller scale features, produced
  by the corrugations in the terrain associated
  with the river valleys, and smaller-scale
  features forced by terrain such as the Puget
  Sound convergence zone.
• Such features require 4-km or better grid spacing
  to get a reasonable handle on the precipitation
  distributions.

27
10-km
28
12-km
4-km
29
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
30
Annual Climatologies of MM5 4-km domain
31
Verification of Small-Scale Orographic Effects
32
But not so perfect for individual events (issues
of resolution, model physics, and initialization,
among others)
33
Perhaps the most detailed look at this scale
separation of orographic flows was presented by
Garvert, Smull and Mass, 2007 (IMPROVE-2 paper)
34
Garvert et al.

• Used aircraft radar and in situ data from the
  IMPROVE-2 field experiment, as well as high
  resolution (1.3 km grid spacing) MM5 output.
• Documented and simulated small scale mountain
  waves and their microphysical/precipitation
  implications.

35
(No Transcript)
36
(No Transcript)
37
Proposed Olympex 2010-2011will hopefully
continue this work
38
During the 1990s it became clear that there were
problems with the simulated precipitation and
microphysical distributions over Northwest terrain

• Apparent in the daily UW real-time MM5 forecasts
  at 12 and 4-km
• Also obvious in research simulations of major
  storm events.

39
Early Work-1995-2000 (mainly MM5, but results are
more general)
Colle and Mass, 1999Colle, Mass and Westrick
,2000

• Relatively simple microphysics water, ice/snow,
  no supercooled water, no graupel. (explicit
  moisture scheme of Hsie et al. 1984, with
  ice-phase microphysics below 0C Dudhia 1989) was
  applied in for 36, 12, and 4-km domains.
• Tendency for overprediction on the windward
  slopes, even after considering undercatchment.
  Only for heaviest observed amounts was there no
  overprediction.
• Tendency for underprediction to the lee of the
  barrier and in major gaps.

40
MM5 PrecipBias for24-h90 and 160 lines are
contoured with dashed and solid lines
For entire Winter season
41
Problems Were Obvious in the Lee of the Olympics

• Lack of clouds and precipitation in model on the
  lee side in light to moderate events.
• Too much precipitation moving over mountains
  under strong winds.

42
Testing more sophisticated schemes and higher
resolution 2000

• Testing of ultra-high resolution (1 km) and
  better microphysics schemes (e.g., with
  supercooled water and graupel), showed some
  improvements but fundamental problems remained
  e.g., lee dry bias, overprediction for light to
  moderate events, but not the heaviest.
• Example simulations of the 5-9 February 1996
  flood of Colle and Mass 2000.

43
5-9 February 1996
44
Colle and Mass, 2000Little Windward Bias, Too
Dry in Lee
Windward slope
Lee
Bias 100-no bias
45
Higher Resolution changes lee precipitation,
but lee bulls eyes of heavy precip develop
mountain waves too strong?
46
Varying Microphysics

• Modest changes, with graupel causing high
  intensity areas in the immediate lee.

Most sophisticated microphysics did not
necessarily produce the best verification
47
Flying Blind
48
IMPROVE

• Clearly, progress in improving the simulation of
  orographic precipitation demanded better
  observations
• High quality insitu observations aloft of cloud
  and precipitation species.
• Comprehensive radar coverage above the barrier
• High quality basic state information (e.g., wind,
  humidity, temperature)
• The IMPROVE field experiment (2001) was designed
  and to a significant degree achieved this.

49
British Columbia
Legend
Washington
UW Convair-580
Airborne Doppler Radar
Cascade Mts.
Two IMPROVE observational campaigns I.
Offshore Frontal Study (Wash. Coast,
Jan-Feb 2001) II. Orographic Study
(Oregon Cascades, Nov-Dec 2001)
S-Pol Radar
Offshore Frontal Study Area
BINET Antenna
Olympic Mts.
Olympic Mts.
Paine Field
Univ. of Washington
NEXRAD Radar
Area of Multi-Doppler Coverage
Wind Profiler
Rawinsonde
Westport
Cascade Mts.
WSRP Dropsondes
Special Raingauges
Columbia R.
PNNL Remote Sensing Site
90 nm (168 km)
Washington
Ground Observer
S-Pol Radar Range
S-Pol Radar Range
Portland
0
100 km
Oregon
Terrain Heights
Coastal Mts.
Salem
lt 100 m
Orographic Study Area
100-500 m
500-1000 m
1000-1500 m
Newport
1500-2000 m
2000-3000 m
gt 3000 m
Rain Gauge Sites in OSA Vicinity
Santiam Pass
OSA ridge crest
Santiam Pass
Orographic Study Area
S-Pol Radar Range
Cascade Mts.
Coastal Mts.
Oregon
SNOTEL sites CO-OP rain gauge sites
Medford
California
50 km
50
The NOAA P3 Research Aircraft Dual Doppler Tail
Radar Surveillance Radar Cloud Physics and
Standard Met. Sensors
Convair 580 Cloud Physics and Standard Met.
Sensors
51
(No Transcript)
52
Convair-580 Flight Strategy
9000
8000
7000
Slope matches that of an ice crystal falling at
0.5 m/s in a mean cross-barrier flow of 10 m/s,
which takes 3 h.
6000
5000
Terrain ht. (m)
4000
3000
Total flight time 3.4 h
2000
1000
0
PARSL Site
S-POL Radar
Santiam Junction
Santiam Pass
Camp Sherman
0
50
100
-50
-100
Distance (km)
53
The S-Pol Doppler Radar
54
(No Transcript)
55

• Pacific Northwest National Lab (PNNL)
• Atmospheric Remote Sensing Laboratory (PARSL)
• 94 GHz Cloud Radar
• 35 GHz Scanning Cloud Radar
• Micropulse LIDAR
• Microwave Radiometer
• Broadband radiometers 
• Multi-Filter Rotating Shadowband Radiometer
  (MFRSR)
• Infrared Thermometer (IRT)
• Ceilometer
• Surface MET
• Total Sky Imager

S-Band Vertically Pointing Radar
56
An IMPROVE-2 Sample Dec. 13-14, 2001

• Strong, extremely well sampled event on the
  Oregon Cascades
• Varied biases on the windward slopes, and now
  overprediction over the lee.
• Overprediction at valley stations on windward
  side
• Little bias on windward crest stations

1.3 km
Garvert et al., 2005a
4 -km
57
But now, we had the microphysical data aloft to
determine what was happening
Model
Observations
58
The Diagnosis

• Too much snow being produced aloft
• Too much snow blowing over the mountains,
  providing overprediction in the lee
• Too much cloud liquid water on the lower windward
  slopes
• Too little cloud liquid water near crest level.
• Problems with the snow size distribution (too few
  small particles)
• Several others!

59
In Comparison The Weaker Dec. 4-5, 2001 Event

• Based on WRF Model
• Overprediction over windward slopes
• Too much precip in the immediate lee of the crest
• Underprediction to the east of the Cascades
• Excessive generation of snow aloft

Yanluan and Colle 2008
60
Lots of activity in improving microphysical
parameterizations

• New Thompson Scheme for WRF that includes a
  number of significant improvements.
• Higher moment schemes are being tested. (e.g.,
  new Morrison two-moment scheme)
• Microphysical schemes are being modified to
  consider the different density and fall speed
  characteristics of varying ice habits and degrees
  of riming (work of Woods, Hafen, and Stoelinga,
  UW)

61
Another Major Question

• What is the importance of unresolved small scale
  orographic features and sub-grid scale motions on
  mesoscale orographic precipitation?
• Do these features enhance precipitation? Do they
  need to be parameterized for coarser simulations?
  Or do we need ultra high resolution to get the
  orographic precipitation right?

62
The Influence of Small Scale Ridges (Colle 2008)
Small net windward enhancement by small scale
features
63
The Influence of Shear-Induced Turbulence on
Microphysics
Houze and Medina, JAS, 2005
64
The problems with the simulation of orographic
precipitation are not limited to microphysics and
resolution

• The MM5 and WRF V1-2.1 lacked positive definite
  advection schemes for moisture variables.
• The result of such numerics is a lack of
  conservation of moisture, producing essentially
  an unphysical source of water. Thus, lack of PD
  advection explains part of the overprediction
  problem in MM5/WRF
• COAMPS and CSU RAMS have PD schemes.

65
Recent Work of Robert Hahn, UW, for Dec. 13-14,
2001 IMPROVE 2 eventPD-NOPD
Domain 36km 12km 4km 1.33km
Coast Water -4.0 -2.5 -6.5 -6.6
Coast Mountains -4.1 -4.4 -7.9 -9.8
Willamette Valley -3.5 -3.9 -13.0 -15.6
Cascade Windward -4.1 -5.0 -13.5 -17.2
Cascade Leeward -4.3 -8.0 -10.2 -11.4
DOMAIN TOTAL -3.9 -4.4 -10.9 -13.4
66
Benefits Appear to Be Apparent In UW Real-Time
Prediction
MM5 and WRF have similar bias
WRF has lesser bias
Positive Definite Advection Initiated
67
Problems and deficiencies of boundary layer and
diffusion schemes can significantly affect
precipitation and microphysics

• Boundary layer parameterizations are generally
  considered one of the major weaknesses of
  mesoscale models (as noted at recent WRF users
  group meeting in Boulder).
• Deficiencies in the PBL structures were noted
  during IMPROVE.
• Errors in boundary layer structure can
  substantially alter mountain waves and resultant
  precipitation.

68
Garvert, Mass, and Smull, 2007Improve-2Dec13-14,
2001Changes in PBL Schemes substantially
change PBL structures, with none being correct.
69
Impacts of Boundary Layer Parameterization on
Microphysics
Snow-diff
CLW-diff
Graupel-diff
Microphysics Differences ETA - MRF
70
The Next Major Challenge Probabilistic
Orographic Precipitation Prediction

• The atmosphere is not deterministic and there are
  substantial uncertainties in initial conditions
  and physics parameterizations, and continued
  approximations in the numerics.
• Over the next several years, we need to perfect
  approaches for probabilistic prediction of
  orographic precip that produce sharp and reliable
  probability density functions.

71
Special Challenges and Advantages of
Probabilistic Prediction Over Terrain

• Less observations that over flatland, making
  calibration more difficult. (disadvantage)
• More frequent precipitation (an advantage).
• Less of a phase space, since orography does
  constrain possible atmospheric states.
  Orographic flow often controlled by interaction
  of synoptic scale flow with mesoscale terrain.
  (advantage).

72
Probabilistic NWP over NW terrain is already well
along

• Current Operational Systems
• University of Washington UWME system (36-12 km)
• University of Washington EnKF System (36-12km)
• NWS Multi-Model SREF System (32 km)

73
UWME

• Core 8 members, 00 and 12Z
• Each uses different synoptic scale initial and
  boundary conditions
• All use same physics
• Physics 8 members, 00Z only
• Each uses different synoptic scale initial and
  boundary conditions
• Each uses different physics
• Each uses different SST perturbations
• Each uses different land surface characteristic
  perturbations
• Centroid, 00 and 12Z
• Average of 8 core members used for initial and
  boundary conditions

74
Current International Multi-Analysis Collection
Resolution ( _at_ 45 ?N )
Objective Abbreviation/Model/Source
Type Computational Distributed Analysis
gfs, Global Forecast System, Spectral T254 /
L64 1.0? / L14 SSI National Centers for
Environmental Prediction 55km 80km 3D
Var   cmcg, Global Environmental Multi-scale
(GEM), Finite 0.9? / L28 1.25? / L11 3D
Var Canadian Meteorological Centre Diff. 70km
100km   eta, Eta limited-area mesoscale model,
Finite 12km / L60 90km / L37 SSI National
Centers for Environmental Prediction Diff. 3D
Var   gasp, Global AnalysiS and Prediction
model, Spectral T239 / L29 1.0? / L11 3D
Var Australian Bureau of Meteorology 60km 80km
jma, Global Spectral Model (GSM), Spectral T106
/ L21 1.25? / L13 OI Japan Meteorological
Agency 135km 100km   ngps, Navy Operational
Global Atmos. Pred. System, Spectral T239 /
L30 1.0? / L14 OI Fleet Numerical Meteorological
Oceanographic Cntr. 60km 80km tcwb, Global
Forecast System, Spectral T79 / L18 1.0? / L11
OI Taiwan Central Weather Bureau 180km 80km  
ukmo, Unified Model, Finite 5/6??5/9?/L30 same
/ L12 3D Var United Kingdom Meteorological
Office Diff. 60km
75
Ensemble domain
76
Post-Processing of Ensembles

• Uses Bayesian Model Averaging to optimally
  combine the various ensemble members to produce
  reliable and sharp probabilistic forecasts.
• The output provides spatially varying PDFs of
  precipitation and other parameters.

77
(No Transcript)
78
Probability Density Function at one point
Ensemble-Based Probabilistic Products
79
Work Cut Out for Us

• Large amount of work yet to be done to perfect
  ensemble-based probabilistic prediction of
  orographic precipitation.
• Quantification of uncertainty in
  parameterizations
• Higher resolution
• Many others.

80
The End
81
High (4-km or higher) resolution also need for
some small scale orographically forced
precipitation features
82
(No Transcript)
83
(No Transcript)
84
(No Transcript)
85
(No Transcript)
86
(No Transcript)
87
(No Transcript)