Transcript PRECIPITATION PARAMETERIZATION AND MODEL …

SEMI-INTELLEGENT USE OF THE
NCEP MODELS (FALL 1999)
Wes Junker
e-mail: [email protected]
Presented at COMET
for Hydromet 00-1
Friday, 15 October 1999
Why models have forecast problems
• Initialization and quality control smooth data fields, but
some of the lost detail may be important.
• May have poor first guess
• Lack of data over the oceans and Mexico.
• Atmospheric processes are non-linear; small changes in
initial conditions can lead to large forecast variations (this
is the basis for ensemble forecasting).
• Model physics are approximations (radiation, cloud
physics, convection, boundary layer, etc.)
– for lower resolution models (the current operational
models), convection is parameterized
– for higher resolution models (models with a resolution
below 5 km) the micro-physical processes are
parameterized
INTELLIGENT USE OF THE MODEL REQUIRES
THAT THE FORECASTER
• COMPARE THE INITIAL 00HR FORECAST WITH
DATA
– MAY BECOME HARDER TO DO AS NEW DATA STREAMS
ARE USED
• BE FAMILIAR WITH CHARACTERTIC MODEL
ERRORS AND BIASES.
– THESE MAY VARY BY SEASON AND REGIME
• HAVE A ROUGH UNDERSTANDING OF HOW
APPROXIMATIONS OF THE PHYSICS MAY
NEGATIVELY IMPACT A FORECAST.
Understanding how the physics may impact a
forecast is tough because the atmosphere is
complicated and acts in a non-linear fashion. For
example, whenever the parameterization for
convection kicks in, it
•
•
•
•
•
Changes the vertical stability
redistributes and generates heat
redistributes and removes moisture
redistributes momentum
makes clouds
Adapted from notes of Bernard Meisner
Eta Model Physics
• Eta model calculates grid-scale precipitation
using a simplified explicit cloud water
scheme
– includes super-cooled water, simplified snow
processes and the advection of cloud water and
cloud ice
– but does not include horizontal advection of
snow and rain.
• In fast flow snow can advect 50 to 100 km
downwind of its source region (Rauber, 1992))
EXPLICIT CLOUD PREDICTION
SCHEME (large scale)
• Cloud condensation is allowed to occur when the RH
reaches a critical value
• Cloud evaporation is allowed to take place only when the
RH falls below the critical value
– 70% over land, 80% over water
– the difference in the critical value between land and water can
produce discontinuities along the coast
– this may be one of the reasons the Eta over predicts cold season
precipitation along the Gulf and Atlantic Coasts.
The BMJ Convective Scheme
• 1st looks for deep convection
– step 1 is to look for most unstable layer within
the lowest 130 mb
– Next calculates LCL to get cloud base
– then lifts parcel to Equilibrium Level to get
cloud top
– then looks to see if the cloud layer is at least
290 mb in depth
• If the cloud is not 290 mb then it searches for
shallow convection
• does not need low level convergence to develop
convection
The BMJ scheme
• Was developed for tropical systems
– does not handle elevated convection well
• the convection may not extend through a deep enough layer
• does not develop realistic
downdrafts/outflow boundaries
– therefore, during summer it sometimes predicts the convective
development too far north
• the saturation pressure deficits that allow
condensation in the scheme are different
over land and water
THE ETA OFTEN FORECASTS TOO MUCH RAINFALL
NEAR THE GULF AND SOUTHEAST COASTS
BECAUSE OF THE PROBLEMS WITH THE WAY THE
ETA HANDLES THE LAND-SEA INTERFACE
24 HR PRECIPITATION ANALYSIS
V. T. 12Z 1 APR
12-36 HR PRECIPITATION
FORECAST V. T. 12Z 1 APR
A forecaster needs to know how the model terrain
compares to the actual terrain
THE MODEL’S TERRAIN IS AVERAGED OVER
THE GRID BOX SO THE SLOPE OF THE
TERRAIN IS USUALLY NOT STEEP ENOUGH
THIS CAUSES THE VERTICAL MOTION FIELD TO
BE SHIFTED AWAY FROM THE MOUNTAINS
THINGS TO REMEMBER ABOUT MODEL QPFS
IN COMPLEX TERRAIN DURING WINTER
BECAUSE OF THE SIMPLIFIED
MICROPYSICS AND
INADEQUATE RESOLUTION OF
MOUNTAINS.
MODELS USUALLY:
1) PREDICT PRECIPITATION
TOO FAR WEST AWAY FROM
MOUNTAIN PEAKS
2) DO NOT ALLOW ENOUGH
PRECIPITATION ON THE
IMMEDIATE DOWNWIND SIDE
OF MOUNTAIN RANGES
THIS IS ESPECIALLY TRUE IN WINTER
DESPITE ITS RECENT PROBLEMS, THE ETA IS
STILL USUALLY BETTER THAN THE AVN OR
NGM FORECASTING PRECIPITATION OVER
COMPLEX TERRAIN DURING WINTER IN A
ZONAL PATTERN.
12-36 H ETA V.T. 12Z 3 JAN 97
12-36 H NGM V.T. 12Z 3 JAN 97
ANALYSIS V.T. 12Z 3 JAN 97
Note that the Eta max in California is a little too far west. It also often under
predicts precipitation over the Siskiyou Mountains of northern California.
FOR ANY MODEL, ALWAYS
BEWARE OF THE 1ST GUESS
00 HR ETA SURFACE AND 1000500 MB THICKNESS
V.T.
00Z 3 SEP 1998
00 HR ETA 500 MB HEIGHT
AND VORTICITY
V.T.
00Z 3 SEP 1998
SURFACE ANALYSIS
V.T. 00Z 3 SEP 1998
TROPICAL STORM EARL WAS LOCATED JUST SOUTHWEST
OF THE FLORIDA PENINSULA. THE 1ST GUESS WILL
SOMETIMES OVERRIDE DATA WHEN INTENSE SMALL SCALE
FEATURES ARE PRESENT.
WHAT HAPPENED?
THE MRF 1ST GUESS FIELD TRIED TO DRAW TO THE DATA BUT THE FIRST
GUESS FIELD OVERWHELMED IT. IF NCEP HAS A COMPUTER FAILURE,
THE ETA DATA ASSIMILATION MAY BE CANCELLED AND THE ETA MAY
RUN ON AN EARLIER GLOBAL MODEL 1ST GUESS
A POOR INITIAL ANALYSIS CAN
PRODUCE HUGE FORECAST ERRORS
ETA 48 HR V.T. 00Z SEP 5
EARL
ETA 48 HR V.T. 00Z SEP 5
EARL
EARL
EARL
ETA 00 HR V.T. 00Z SEP 5
ETA 00 HR V.T. 00Z SEP 5
WHEN THE MODEL FIRST GUESS THINKS THE SOIL MOISTURE
IS HIGH,
THEN, THE MODEL FORECASTS SURFACE DEWPOINTS TOO HIGH AND
SURFACE TEMPS TOO LOW. FORECAST CAPES WILL BE TOO HIGH
32
28
24
SURFACE
TEMPERATURE
OBSERVED
20
24
20
16
ETA FORECAST
DEWPOINT
TEMPERATURE
12
1024
1020
1016
18/00
SURFACE
PRESSURE
18/06
18/12
1012
18/18
19/00
19/06
19/12
19/18
THE MODEL
UNDERPREDICTS
THE BOUNDARY
LAYER WINDS.
HOWEVER, MODEL
FORECASTS OF 850
MB WINDS ARE
OFTEN TOO
20/00
STRONG
WHEN THE FIRST GUESS THINKS THE SOIL
MOISTURE IS LOW IN SUMMER IN THE PLAINS, THE
SURFACE DEWPOINT WILL BE LOW AND THE
TEMPERATURE WILL BE TOO HIGH. FORECAST
CAPES WILL BE TOO LOW
OKLAHOMA CITY
36
32
28
24
20
SURFACE
TEMPERATURE
OBSERVED
ETA FORECAST
20
DEWPOINT
16
TEMPERATURE
12
1016
SURFACE
PRESSURE
19/00
19/06
MAY 1998
19/12
1008
19/18
20/00
20/06
20/12
20/180
21/00
ETA SURFACE
WINDS WERE
TOO WESTERLY,
WAS THERE
TOO MUCH
DOWN-SLOPE?
Forecast -Vs- Observed Best Cape
Spring 96
Line x=y
Line x=y
Forecast
precipitation
1 - less than .25”
2 - more than .25”
Note the large spread.
The model stability
forecasts are worst
when precipitation is
forecast
The performance characteristics of
the eta have changed dramatically
during the past year.
• QPF forecasts during the past winter deteriorated
when compared to the AVN or NGM.
• ETA surface and 500 mb forecasts were also
worse compared to the other models.
– April 1999 ETA 500 h and 250 mb forecasts usually
verified worse than the AVN.
• June-August 1999, The eta model QPF usually
verified better than the AVN or NGM.
ETA AND STORM TRACKS
DURING 1999 TENDED TO BE TOO FAR SOUTH
WITH LOWS AS THEY REFORMED EAST OF
ROCKIES. THIS ERROR USUALLY CONTINUED
UNTIL THE LOW MOVED EAST OF THE
MISSISSIPPI RIVER.
BY CONTRAST, THE AVN IS SOMETIMES TOO
FAR NORTH AND TENDS TO SOMETIMES
TRACK LOWS TOO FAR NORTH AND WEST
WITH LOWSALONG THE EAST COAST.
ESPECIALLY DURING MAJOR
CYCLOGENESIS WHEN A COASTAL
TROUGH IS PRESENT
LOWS TO THE LEE OF THE
ROCKIES
• THE AVN AND NGM USUALLY PREDICT
THEM TO FORM TOO FAR NORTH
• THE ETA IS SOMETIMES A LITTLE TOO FAR
SOUTH
• USE THE 300 MB UPPER LEVEL JET. THE
SURFACE LOW IS USUALLY FOUND IN THE
LEFT EXIT REGION OF THE JET, USUALLY
JUST TO THE NORTH
28 ETA model runs were evaluated during the
period from 00Z March 30-12 Z April 13.
During the entire period the mean 500h
pattern was similar to the one shown below.
A RIDGE AND POSITIVE ANOMALY
NEAR 160W, BELOW NORMAL
HEIGHTS OVER ALASKA AND A
TROUGH NEAR OF JUST INLAND
FROM THE WEST COAST WITH
BELOW NORMAL HEIGHTS
EXTENDING EASTWARD INTO THE
SOUTHWESTERN U.S.
THE ETA SHOWED A CONSISENT
CHARACTERISTIC ERROR DURING
THE PERIOD. THE NEXT FEW
SLIDES WILL DESCRIBE THE
ERROR
As the upper trough digs into the west the
ETA did not dig the shortwaves strongly
enough once the trough reached the ca coast.
Note how much lower the heights are across
NV and CA.
48 hr ETA valid 00Z 1 April
00 hr eta valid 00Z 1 April
The eta underplays the second shortwave
diving into the mean trough and
overplays the first one.
48 hr ETA 500 h and vorticity v.t. 12Z 4 April
00 hr ETA 500 h and vorticity v.t. 12Z 4 April
THE ETA PREDICTED THE UPPER LOW ASSOCIATED WITH THE FIRST SHORTWAVE
TOO FAR SOUTH AND EAST IN THE PLAINS. INSTEAD THE INITIAL SHORTWAVE
LIFTED MORE TO THE NORTH BEFORE BEING FORCED EASTWARD. THIS HAPPENED
SEVERAL TIMES DURING THE STUDY.
The eta was generally too fast and far southeast with
the 500h low over the Plains with 120 meter errors
over MO and IA. This can have a very serious
impact on frontal speed and on the position of the
low level convergence and resulting convection.
546
48 hr Eta v.t. 12Z 10 Apr
558
00 hr Eta v.t. 12Z 10 Apr
The Eta surface low and associated fronts can also be
affected. The slower eastward movement of the
ridge axis may allowed for the flow along the east to
be more northwesterly which allowed the surface
boundary to sink farther to the south
48 hr Eta v.t. 12Z 10 Apr
00 hr Eta v.t. 12Z 10 Apr
COMMON ETA ERROR ALONG EAST COAST
WHEN A CLOSED UPPER LOW APPROACHES THE COAST THE ETA
SOMETIMES HAS PROBLEMS FORECASTING THE LOCATION OF THE
SURFACE LOW. NOTE WHERE THE UPPER LOW IS CENTERED AND
WHERE THE STRONGEST UPPER-LEVEL DIVERGENCE IS IMPLIED.
48 H ETA 500H V.T. 12Z 23 APR 98
48 H NGM 500H V.T. 12Z 23 APR 98
Based on the 500 h and vorticity pattern, where would you predict the surface low?
NOTE THAT THE ETA SURFACE LOW IS A LITTLE
WEST OF ITS 500 MB CENTER. THE NGM HAS A
MUCH BETTER FIT TO THE 500 MB PATTERN.
THE STRONG EASTERLY COMPONENT TO THE WINDS NORTH OF THE
ETA MODEL LOW ALLOWS IT TO WRAP MOISTURE AND PRECIPITATION
TOO FAR WEST
48 H ETA SURFACE V.T. 12Z 23 APR 98
48 H NGM SURFACE V.T. 12Z 23 APR 98
THE LOW VERIFIES A LITTLE NORTH AND
EAST OF THE NGM. REMEMBER, THE NGM IS
TYPICALLY TOO SLOW WITH LOWS ALONG
THE COAST.
VERIFYING SURFACE V.T. 12Z 23 APR 98
L
ETA SURFACE LOW
VERIFYING 500H V.T. 12Z 23 APR 98
When the NGM and AVN sheared 500 troughs approaching
the east coast in 1999, the eta often amplified the trough and
overdeepened the surface low. An example:
48 HR ETA 500
48 HR ETA SFC
48 HR NGM 500
48 HR NGM SFC
The Eta predicted a major east
coast snowstorm. The NGM and
AVN predicted light snow at best
36-48 hr ETA precipitation
36-48 hr NGM precipitation
HOW THE MODEL VERIFIED. NO
MAJOR SNOWSTORM DEVELOPED.
48 HR ETA 500
VERIFYING 500 MB
48 HR ETA SFC
VERIFYING SFC
MORE ON ETA PERFORMANCE
• TOO WET IN FLORIDA
• SOMETIMES OVERDEVELOPS LOW-LEVEL JET
• DURING WINTER HAS BEEN TOO FAST BRINGING
SHORTWAVES THROUGH THE ROCKIES INTO THE
PLAINS.
• HAS BEEN TOO FAR SOUTH WITH CLOSED LOWS
COMING EASTWARD INTO THE PLAINS
• OVERFORECASTS THE STRENGTH OF
ANTICYCLONES
• HAS PROBLEMS INITIALIZING TROPICAL
SYSTEMS
Models have problems with arctic airmasses.
The reasons why are listed below
• Terrain is averaged
• Initialization process sometimes robs shallow
airmass of its coldness
• Models have problems handling strong inversions
• Models have problems handling cold air damming
• The sigma coordinate system, the Eta coordinate
system does better
• The leading edge of the ETA LI gradient is often
the best indicator of the frontal position
THE NGM AND AVN/MRF HAVE SERIOUS PROBLEMS WITH
ARCTIC AIRMASSES.
L
36 HR NGM V.T. 00Z APR 09,
1995
36 HR AVN V.T. 00Z APR 09,
1995
AVN ANALYSIS
V.T. 00Z APR 09, 1995
TEMPERATURES ACROSS KANSAS WERE IN THE LOW TO MID
50s WITH STRONG NORTH WINDS. SOUTH OF THE FRONT
TEMPERATURES WERE IN THE UPPER 70s TO LOW 90s.
WHEN THE ETA 500 H FORECAST IS COMPARABLE
TO THE OTHER MODELS IT WILL DO A BETTER JOB
IN HANDLING THE COLD AIR SURGE
VERIFYING PRECIPITATION
• BIAS=FORECAST/OBSERVED
• EQUITABLE THREAT=(H-E)/(F+O-H-E)
• THREAT SCORE=H/(F+O-H)
– N=NUMBER OF HITS, F=NUMBER OF GRID POINTS FORECAST,
O=GRID POINTS OBSERVED, E=(F*O)/N
MODEL BIAS AND THREAT
SCORE
• IS DEPENDENT ON RESOLUTION OF
MODEL
• HOW THE MODEL IS DISPLAYED.
THE FAX VERSION OF ETA IS NOT
DISPLAYED WITH FULL MODEL
RESOLUTION!
• HOW THE MODEL IS VERIFIED
– WHETHER VERIFIED AT A POINT, OR
AVERAGED OVER A GRID BOX
ETA THREAT SCORES WERE LOWER THAN THOSE FROM
THE SUBJECTIVE AND AVN GUIDANCE DURING WINTER,
AVN LAGS ETA IN SUMMER
Regional ETA verification using
model grid (80 km)
WARM SEASON 1.00” OR MORE VERIFICATION
VERIFIED TO AN 80 KM GRID
.64
.15
.59
.19
.35
.09
.98
.15
.65
.14
.47
.08
.93
.17
.83
.12
BIAS TOP NUMBER,
EQUITABLE THREAT BOTTOM
.97
.18
From 1998 data
Regional ETA verification using
model grid (80 km)
COLD SEASON 1.00” OR MORE VERIFICATION
VERIFIED TO AN 80 KM GRID
.69
.17
.71
.27
.58
.10
.94
.18
.74
.09
.71
.15
BIAS
ETS
1.07 1.36
.23 .22
1.04
.19
AGAIN NOTE HIGH BIAS ALONG EAST COAST
AND LOW BIAS OVER WEST
Regional ETA verification using
model grid (80 km)
.01” OR GREATER AMOUNTS DURING COLD SEASON
VERIFIED TO AN 80 KM GRID
1.43
.25
.81
.37
.95
.26
1.05
.35
1.23
.23
.79
.32
1.07
.35
1.11
.34
1.07
.35
HIGHEST THREATS ALONG WEST COAST.
HIGH BIAS OVER UPSLOPE AREAS EAST
OF ROCKIES AND OVER PLAINS
Regional ETA verification using
model grid (80 km)
.01” OR GREATER AMOUNTS DURING WARM SEASON
VERIFIED TO AN 80 KM GRID
1.11
.28
1.00
.37
.82
.23
.92
.37
1.21
.19
1.01
.32
.81
.34
.96
.39
.99
.38
BIG DIFFERENCES WITH POINT VERIFICATION. USING A
POINT VERIFICATION, YOU SEE THE HUGE BIASES OVER
THE SOUTH
ETA .50” OR MORE PERFORMANCE
DURING WARM SEASON
VERIFIED TO AN 80 KM GRID
.77
.21
.82
.28
.88
.12
1.09
.25
.82
.17
1.07
.24
1.10
.23
.62
.14
BIAS
ETS
.86
.20
DURING SUMMER ETA UNDERPREDICTS
.50” OR GREATER AMOUNTS IN PLAINS.
ETA PERFORMANCE FOR .50 OR
GREATER AMOUNTS APR 96-NOV 97
VERIFIED TO AN 80 KM GRID
.89
.23
1.00
.83 .15
.35
1.10
.23
.97
.13
.90
BIAS
THREAT
1.13
.31
1.10
.26
.23
ETA OVERPREDICTS .50 OR
GREATER ACROSS SOUTH AND
ALONG EAST COAST. MESO-ETA
HAS SAME BIAS
1.32
.31
AVN/MRF APPROXIMATED PHYSICS
• THE AVN/MRF USE A MODIFIED ARAKAWASHUBERT SCHEME
• THIS USES THE CHANGE IN STABILITY TO
DETERMINTE WHEN TO RELEASE ENERGY AS
CONVECTION.
• NO DIRECT MIXING BETWEEN THE CLOUDY AIR
AND ENVIRONMENTAL AIR.
(except at the cloud top and bottom)
• NO CLOUD WATER EXISTS, THEREFORE ALL
WATER IS CONVERTED TO RAIN.
A NUMBER OF AVN/MRF
PERFORMANCE CHARACTERISTICS
HAVE CHANGED IN THE PAST YEAR.
• SINCE JUNE THE AVN/MRF AGAIN UNDERPREDICTS
HEAVIER PRECIPITATION THRESHOLDS.
• THE AVN/MRF NO LONGER “OFTEN” UNDERPREDICTS
SURFACE LOWS OVER OCEANS EXCEPT IN THE
SUBTROPICS AT LONGER TIME RANGES.
• ISOLATED PRECIPITATION “BULLSEYES” ARE STILL
SOMETIMES A PROBLEM, ESPECIALLY DURING THE WARM
SEASON.
• THE PROBLEM IS MORE LIKELY WHEN THERE IS SLOW
SYSTEM MOVEMENT
• THE AVN CONTINUES TO HAVE A WARM BIAS AT MID
LEVELS OVER THE PLAINS.
LATEST AVN/MRF CHANGES
• June 15, 1998: INCREASED
HORIZONTAL RESOLUTION TO 170
AND LAYERS TO 42
• THIS LED TO A WARM BIAS AND THE DEVELOPMENT OF
SPURIOUS PRECIPIATION BULLSEYES/TROPICAL SYSTEMS
• July 21, 1998: EMERGENCY MODEL
IMPLEMENTATION TO REDUCE
ERRORS IN THE JUNE 15TH CHANGE
The MRF has spin-up precipitation bombs and
develops tropical systems erroneously,
especially at longer time ranges.
24-h MRFX v.t. 12Z 27 May 1998
36-h MRFX v.t. 00Z 28 May 1998
10”+ bullseye
24-36-h MRFX v.t. 00Z 28 May 1998
SFC ANALYSIS v.t. 00Z 28 May 1998
MRF PRECIPITATION
Convective - dashed
Gridscale - solid green
(inches -Vs- time)
BEFORE 7/21
AFTER 7/21
During Oct-Feb 1999, the
AVN/MRF had a bias of
• About 1.00 for 1.00” or greater amounts
• .47 for 2.00” or greater amounts
• .27 for 3.00” or greater amounts
The Aviation significantly
underpredicted heavier rainfall
amounts
During Late Spring and Summer
of 1999
• The AVN has often predicted convection too
far north in the Plains
• has produced spurious precipitation bombs
• has had a warm bias at mid levels. This
bias is much stronger in the summer than
during winter and is strongest over the high
plains and Rocky Mountains. This has
caused it problems in handling MCCs
How do I tell if the AVN/MRF
precipitation bomb is helping to
produce an erroneous vort max.
• The best way is to look at the convective
part of the precipitation and compare it with
the total precipitation
• if the convective portion of the precipitation
is only a small percentage of the total
precipitation and the model is predicting 2
or 3 inches of rain. (BEWARE!!!!!!!!)
• the AVN is probably having problems.
During OCT-JAN 1999 4 INCH OR
GREATER ISOHYETS
• WERE ANALYZED 22 TIMES
• WERE PREDICTED 2 TIMES
• THE AVIATION FORECAST 3” OR
GREATER AMOUNTS ONLY 6 TIMES.
One of the typical errors of the AVN is to
predict convective systems too far north
12-36 HR QPF V.T. 1200Z 29
JAN 1999
ANALYSIS V.T. 1200Z 29 JAN
1999
12-36 HR QPF V.T. 1200Z 30
JAN 1999
ANALYSIS V.T. 1200Z 30 JAN
1999
The AVN/MRF may be right on the synoptic scale
features but cannot handle outflow boundaries, etc.
36 HR QPF V.T. 1200Z 29 JAN 1999
ANALYSIS V.T. 1200Z 29 JAN 1999
36 HR QPF V.T. 1200Z 29 JAN 1999
A SLOW MOVING 500 CLOSE LOW AND
QUASISTATIONARY FRONT WERE WELL
FORECAST BY THE MODEL. HOWEVER,
THE EFFECTIVE BOUNDARY REMAINED
SOUTH OF THE MODEL FORECAST
IN CONCLUSION
• THE ETA MODEL HAS HAD SERIOUS PROBLEMS SINCE
3DVAR WAS IMPLIMENTED.
• PRIOR TO 3D-VAR ETA QUANTITATIVE PRECIPITATION
FORECASTS WERE BETTER THAN THOSE OF THE AVN.
– SINCE 3DVAR….AVN PRECIPITATION FORECASTS HAVE BEEN
SUPERIOR DURING THE COLD SEASON. ETA PRECIPITATION
FORECASTS HAVE BEEN BETTER DURING THE WARM SEASON.
• BETTER VERIFICATION IS NEEDED OF OPERATIONAL
MODELS. THE VERIFICATION NEEDS TO BE SHARED WITH
FORECASTERS.
– EMC AND HPC ARE NOW MAINTAINING A VERIFICATION
SECTION ON THEIR HOMEPAGE.
• EACH TIME A MODEL IS CHANGED, IT MAY AND PROBABLY
WILL CHANGE THE PERFORMANCE CHARACTERISTICS.