investigating height assignment type errors in the ncep...
TRANSCRIPT
1
Investigating Height Assignment Investigating Height Assignment Investigating Height Assignment Investigating Height Assignment
Type Errors in the NCEP Global Type Errors in the NCEP Global Type Errors in the NCEP Global Type Errors in the NCEP Global
Forecast SystemForecast SystemForecast SystemForecast System
James JungCooperative Institute for Meteorological Satellite Studies
John Le MarshallCentre for Australian Weather and Climate Research, Australia
Jaime DanielsNational Oceanic and Atmospheric Administration
NESDIS, Center for Satellite Applications and Research
Lars Peter RiishojgaardJoint Center for Satellite Data Assimilation
2
Outline
• Height Assignment types
• Definitions of Statistics used
− Speed
− Direction
• Height Assignment Statistics
− Water Vapor Images
− Infrared Images
• NCEP Future Work
• Comments and Conclusions
3
Height Assignment Types
• CO2 Slicing − GOES-12 and beyond
− Menzel et al 1983
• Water Vapor Intercept− Szejwach 1982
• Histogram− Nieman et al 1993
• IR Window− Nieman et al 1993
• Cloud Base− Le Marshall et al 1997
4
Wind Comparison Statistics
• Vector Difference
• Mean Vector Difference
• Standard Deviation
( ) ( )∑=
+= −−
N
i
mimi VVUUN
MVD1
221
( ) ( )VVUUVD mimi −+−=22
( ) ( )[ ]∑ −=
=
N
ii MVDVD
NSD
1
21
* Nieman et al 1997 i = observation, m = model
5
• Root Mean Square Error
• Normalized Root Mean Square Error
• Speed Bias
Wind Comparison Statistics
( ) ( )SDMVDRMSE22
+=
SpeedRMSENRMSE =
∑=
= +−+
N
i
mmii VUVUBIASN
1
22221
* Nieman et al 1997 i = observation, m = model
6
New Directional Statistics
• Directional RMSE
• Directional Bias
B
A
C
(Xob,Yob)
(Xmod,Ymod)
∆Θ
∆X
∆Y
( )∆Θ−+= cos2222
ABBAC
∑=
=
N
i
iCBIASN
1
1
∑=
−=
N
i
iCDRMSEN
1
2
1
1
7
Height Assignment Statistics
• Ocean only vectors used
• Monitored all height assignment types available for each observation.
• Separated by Satellite and Image type− GOES-11, GOES-12
− Water Vapor winds, Cloud-Drift IR winds
• Combined two seasons− Days 1-20 in July and December 2009
• Stratified by Latitude− NH (20N – 60N)
− Tropics (20N – 20S)
− SH (20S – 60S)
8
Speed Bias
GOES-12
Water Vapor Winds
100
200
300
400
500
600
700
800
900
1000
-10 -5 0 5 10 15
Bias [m/s]
Pre
ssu
re [
hP
a]
CO2 H2O HIST WIN
Normalized MVD RMSE
GOES-12
Water Vapor Winds
100
200
300
400
500
600
700
800
900
1000
0 0.5 1 1.5 2 2.5
RMSE [m/s]
Pre
ssu
re [
hP
a]
CO2 H20 HIST WIN
Normalized MVD RMSE
GOES-11
Water Vapor Winds
100
200
300
400
500
600
700
800
900
1000
0 0.5 1 1.5 2 2.5
RMSE [m/s]
Pre
ssu
re [
hP
a]
H20 HIST WIN
Water Vapor Winds
Speed NRMSE and Bias
Speed Bias
GOES-11
Water Vapor Winds
100
200
300
400
500
600
700
800
900
1000
-10 -5 0 5 10 15
Bias [m/s]
Pre
ssure
[hP
a]
H2O HIST WIN
9
Directional Bias
GOES-12
Water Vapor Winds
100
300
500
700
900
-10 -5 0 5 10 15
Bias [degrees]
Pre
ss
ure
[h
Pa
]
CO2 H2O HIST WIN
Directional RMSE
GOES-12
Water Vapor Winds
100
200
300
400
500
600
700
800
900
1000
0 20 40 60 80 100
RMSE [degrees]
Pre
ssu
re [
hP
a]
CO2 H2O HIST WIN
Directional Bias
GOES-11
Water Vapor Winds
100
200
300
400
500
600
700
800
900
1000
-10 -5 0 5 10 15
Bias [degrees]
Pre
ssu
re [
hP
a]
H2O HIST WIN
Directional RMSE
GOES-11
Water Vapor Winds
100
200
300
400
500
600
700
800
900
1000
0 20 40 60 80 100
RMSE [degrees]
Pre
ssu
re [
hP
a]
H2O HIST WIN
Water Vapor Winds
Directional RMSE and Bias
10
Speed Bias
GOES-12
Cloud-Drift IR Winds
100
200
300
400
500
600
700
800
900
1000
-10 -5 0 5 10
Bias [m/s]
Pre
ssu
re [
hP
a]
CO2 H20 HIST WIN
Normalized MVD RMSE
GOES-12
Cloud-Drift IR Winds
100
200
300
400
500
600
700
800
900
1000
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
RMSE [m/s]
Pre
ssu
re [
hP
a]
CO2 H20 HIST WIN
Speed Bias
GOES-11, IR Window
100
200
300
400
500
600
700
800
900
1000
-10 -5 0 5 10
Bias [m/s]
Pre
ssu
re [
hP
a]
H20 HIST WIN
Normalized MVD RMSE
GOES-11
Cloud-Drift IR Winds
100
200
300
400
500
600
700
800
900
1000
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
RMSE [m/s]
Pre
ssu
re [
hP
a]
H20 HIST WIN
Cloud-Drift IR Winds
Speed NRMSE and Bias
11
Directional Bias
GOES-12
Cloud-Drift IR Winds
100
200
300
400
500
600
700
800
900
1000
-10 -5 0 5 10 15
Bias [degrees]
Pre
ssu
re [
hP
a]
CO2 H2O HIST WIN
Directional RMSE
GOES-12
Cloud-Drift IR Winds
100
200
300
400
500
600
700
800
900
1000
0 10 20 30 40 50 60 70 80
RMSE [degrees]
Pre
ssu
re [
hP
a]
CO2 H2O HIST WIN
Directional Bias
GOES-11
Cloud-Drift IR Winds
100
200
300
400
500
600
700
800
900
1000
-10 -5 0 5 10 15
Bias [degrees]
Pre
ssu
re [
hP
a]
H2O HIST WIN
Directional RMSE
GOES-11
Cloud-Drift IR Winds
100
200
300
400
500
600
700
800
900
1000
0 10 20 30 40 50 60 70 80
RMSE [degrees]
Pre
ssu
re [
hP
a]
H2O HIST WIN
Cloud-Drift IR Winds
Directional RMSE and Bias
12
Speed Bias
GOES-12, H2O Intercept
Water Vapor Winds
100
200
300
400
500
600
700
-6 -4 -2 0 2 4 6 8 10
Bias [m/s]
Pre
ssu
re [
hP
a]
SH Trop NH
Normalized MVD RMSE
GOES-12, H2O Intercept
Water Vapor Winds
100
200
300
400
500
600
700
0 0.5 1 1.5 2
RMSE [m/s]
Pre
ssu
re [
hP
a]
SH Trop NH
Speed Bias
GOES-11, H2O Intercept
Water Vapor Winds
100
200
300
400
500
600
700
-6 -4 -2 0 2 4 6 8 10
Bias [m/s]
Pre
ssu
re [
hP
a]
SH Trop NH
Normalized MVD RMSE
GOES-11, H2O Intercept
Water Vapor Winds
100
200
300
400
500
600
700
0 0.5 1 1.5 2
RMSE [m/s]
Pre
ssu
re [
hP
a]
SH Trop NH
Latitude error dependence
Water Vapor Winds
13
Speed Bias
GOES-12, H2O Intercept
Cloud-Drift IR Winds
100
200
300
400
500
600
700
-12 -10 -8 -6 -4 -2 0 2 4
Bias [m/s]
Pre
ssu
re [
hP
a]
SH Trop NH
Normalized MVD RMSE
GOES-12, H2O Intercept
Cloud-Drift IR Winds
100
200
300
400
500
600
700
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
RMSE [m/s]
Pre
ssu
re [
hP
a]
SH Trop NH
Speed Bias
GOES-11, H2O Intercept
Cloud-Drift IR Winds
100
200
300
400
500
600
700
-12 -10 -8 -6 -4 -2 0 2 4
Bias [m/s]
Pre
ssu
re [
hP
a]
SH Trop NH
Normalized MVD RMSE
GOES-11, H2O Intercept
Cloud-Drift IR Winds
100
200
300
400
500
600
700
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
RMSE [m/s]
Pre
ssu
re [
hP
a]
SH Trop NH
Latitude Error Dependence
Cloud-Drift IR Winds
14
NCEP Continuation Work
• Modify analysis code to process all height
assignment types
• Generate O-B/O-A statistics for each height
assignment type
• Derive assimilation techniques specific to
height assignment type
− Investigate QC techniques using multiple height
assignments
15
Comments and Conclusions
• The cloud-drift IR winds have a slow speed bias not found in the water vapor winds.
− Why, how can this be resolved?
• Are improvements possible to the CO2 and H2O intercept methods?
− What are the upper and lower limits of the techniques?
• O-B errors in the tropics are greater than at mid-latitudes
− This is a model problem. How can data providers help?
• Can other data providers include height assignment type in the BUFR file?
16
Questions ?