first results of the cosmo priority project ’tackle deficiencies in … · 2016-12-20 · pp qpf...
TRANSCRIPT
Federal Department of Home Affairs FDHAFederal Office of Meteorology and Climatology MeteoSwiss
First results of the COSMO Priority Project ’Tackle deficiencies in
quantitative precipitation forecasts’
LM User Seminar, 6 March 2007, Langen
S. Dierer1, J. Achimowicz9, T. Andreadis8, M. Arpagaus1, E. Avgoustoglou8, M. Baldauf2, C. Cacciamani3, U. Damrath2, R.
Dumitrache10, V. Fragkouli8, A. Fuccello4, F. Grazzini3, T. La Rocca4, P. Louka8, P. Mercogliano7, P. Mezzasalma3, M. Milelli5, E. Oberto5, A. Parodi6, I. V. Pescaru10, U. Pflüger2, P. Prohl2, I. Pytharoulis8, A.
Sanna5, Ch. Schraff2, F. Schubiger1, A. Seifert2, K. Starosta9, J. Steppeler2, M. Tesini3, E. Zala1
1MeteoSwiss (CH), 2DWD (D), 3ARPA-ER (IT), 4UGM (IT), 5ARPA-P (IT), 6Uni Genova (IT), 7CIRA (IT), 8HNMS (GR), 9IMGW (PO), 10NMA (RO)
2PP QPF
S. Dierer et al.
Outline
• Aim of COSMO priority project ’Tackle deficiencies in quantitative precipitation forecasts’ (PP QPF)
• Overview of PP QPF• Selection of test cases• List of sensitivity studies • First results of the sensitivity studies• Conclusions and Outlook
3PP QPF
S. Dierer et al.
Aim of PP QPF
The aim of PP QPF is improved knowledge about • most suitable namelist settings or • parts of the model that need to be reformulated
to obtain a better QPF at 7 km horizontal grid size
Good quantitative precipitation forecast is a challenging task – not only but also in the COSMO model:
The project has a focus on model deficiencies – not on errors from e.g. large scale conditions
4PP QPF
S. Dierer et al.
Overview of PP QPF
• Task 1: Selection of test cases representative for„typical“ QPF deficiencies of COSMO model
• Task 2: Definition of sensitivity studies
• Task 3: Run sensitivity studies and draw conclusions
• Task 4: Idealized runs for moist benchmark cases
5PP QPF
S. Dierer et al.
Selection of test cases
6PP QPF
S. Dierer et al.
Selection of cases of poor QPF
• Selection of single cases reflecting general problems of QPF in every country
• Large scale situation should be well predicted in order to concentrate on poor performance of COSMO model
• Concentration on “newer” cases (2005 or later) in order to avoid deficiencies due to an old model version
• To ensure this, cases are rerun with reference version• Every country defines of final list of 3 test cases
7PP QPF
S. Dierer et al.
Example: 02.02.2005 - test case selected by MeteoSwiss
24h precipitation of all situations in 2005 withnortherly large scale flow: average precipitation patternfrom radar and forecast
24h precipitation on 02.02.2005: radar and forecast(northerly large scale flow)
8PP QPF
S. Dierer et al.
List of test cases from all countries
DATE INSITUTION Overestimation(+)/underestimation (-)
Stratiform(strat)/convective(con)
06.12.2004 LM-DWD + strat warm sector
18.03.2005 LM-DWD + strat orography
02.07.2005 LM-NMA +/- strat Cold front
03.05.2005 LM-DWD + Strat+conv Warm front
21.06.2005 LM-DWD - conv Cold front
02.02.2005 aLMo + strat Cold front
22.03.2005 aLMo + strat Warm front
12.07.2005 aLMo + conv -
17.12.2005 aLMo + Mainly strat orography
24.09.2004 LAMI - conv Cold front
10.04.2005 LAMI + conv Occluded front
17.08.2006 LAMI +/- conv front
09.09.2005 LAMI + conv -
27.03.2005 LAMI - strat/conv front
01.12.2005 EuroLM - mainly strat ?
03.12.2005 EuroLM +/- Strat/conv ?
17.12.2005 EuroLM - strat/conv ?
15.09.2005 LM-HNMS + conv orography
23.11.2005 LM-HNMS + Strat+conv Cold front
26.11.2005 LM-HNMS +/- Strat+conv Warm+cold front
03.05.2005 LM-IMGW + strat Warm front
04.05.2005 LM-IMGW +/- conv cold front
09.06.2005 LM-IMGW - conv occluded
09.08.2005 LM-IMGW +/- ? cold front
23.06.2005 LM-NMA + strat+conv cold front
12.07.2005 LM-NMA + conv -
9PP QPF
S. Dierer et al.
Frequency of types with poor QPF‘s
underestimation Mixed(under-/overestimation)
overestimation
Preliminary list 11 (20%) 13 (24%) 30 (56%)Final list 6 (30%) 5 (25%) 9 (45%)
by Ulrich Damrath, DWD
10PP QPF
S. Dierer et al.
Weather situations associated withpoor QPF
Preliminary list Final listWarmfront 8 (15%) 3 (15%)
Mixed (WF/CF) 3 (6%) 1 (5%)Coldfront 22 (41%) 9 (45%)Occlusion 11 (20%) 3 (15%)
Trough 5 (9%) 2 (10%)No front 5 (9%) 2 (10%)
by Ulrich Damrath, DWD
11PP QPF
S. Dierer et al.
List of sensitivity runs
12PP QPF
S. Dierer et al.
Different fields of sensitivity studies
• 1. Changes of initial conditions• 2. Changes of numerical methods• 3.1 Changes of microphysics• 3.2 Changes of convection schemes• 3.3 Changes of PBL schemes
13PP QPF
S. Dierer et al.
List of sensitivity studies
by Axel Seifert, DWD
14PP QPF
S. Dierer et al.
List of sensitivity studies
by Axel Seifert, DWD
15PP QPF
S. Dierer et al.
List of sensitivity studies
by Axel Seifert, DWD
16PP QPF
S. Dierer et al.
List of sensitivity studies
by Axel Seifert, DWD
17PP QPF
S. Dierer et al.
List of sensitivity studies
by Axel Seifert, DWD
18PP QPF
S. Dierer et al.
Runs of sensitivity runs
19PP QPF
S. Dierer et al.
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
06.12.2004
23.06.2005
02.07.2005
18.03.2005
03.05.2005
21.06.2005
02.02.2005
22.03.2005
12.07.2005
17.12.2005
24.09.2004
10.04.2005
17.08.2006
09.09.2005
27.03.2005
01.12.2005
03.12.2005
17.12.2005
15.09.2005
23.11.2005
26.11.2005
03.05.2005
04.05.2005
10.06.2005
09.08.2005
12.07.2005
22
Runs of sensitivity studies (finished end 02/2007)
20PP QPF
S. Dierer et al.
First results from sensitivity studies for German, Swiss and Romanian cases
• All together 10 cases:• 5 cases of overestimated winter precipitation• 4 cases of overestimated summer precipitation• 1 case of underestimated summer precipitation
• Comparison of 24h area average precipitation of sensitivity study, control run and observations
21PP QPF
S. Dierer et al.
Winter cases with overestimated precipitation - change of 24h area average precipitation sum in the sensitivity studies
06.12.2004 18.03.2005
22PP QPF
S. Dierer et al.
Winter cases with overestimated precipitation - change of 24h area average precipitation sum in the sensitivity studies
02.02.2005 22.03.2005
5 dt206 lfsl37 Rksl38 rkbott9 rktp10 oro
11 micro112 micro213 micro314 conmod15 conkfb16 conoff
17 rlam0118 rlam5019 sto05020 sto25021 sea0122 sea40
O ctrl1 wso802 Wso1203 Qv0904 Qv110
23PP QPF
S. Dierer et al.
Winter cases with overestimated precipitation - change of 24h area average precipitation sum in the sensitivity studies
17.12.2005
O ctrl1 wso802 Wso1203 Qv0904 Qv110
5 dt206 lfsl37 Rksl38 rkbott9 rktp10 oro
11 micro112 micro213 micro314 conmod15 conkfb16 conoff
17 rlam0118 rlam5019 sto05020 sto25021 sea0122 sea40
24PP QPF
S. Dierer et al.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
06.12.200418.03.200502.02.200522.03.200517.12.2005
22
Winter cases with overestimated precipitation - change of 24h area average precipitation sum in the sensitivity studies
O ctrl1 wso802 Wso1203 Qv0904 Qv110
5 dt206 lfsl37 Rksl38 rkbott9 rktp10 oro
11 micro112 micro213 micro314 conmod15 conkfb16 conoff
17 rlam0118 rlam5019 sto05020 sto25021 sea0122 sea40
25PP QPF
S. Dierer et al.
Summer cases with overestimated precipitation - change of 24h area average precipitation sum in the sensitivity studies
03.05.200512.07.2005
O ctrl1 wso802 Wso1203 Qv0904 Qv110
5 dt206 lfsl37 Rksl38 rkbott9 rktp10 oro
11 micro112 micro213 micro314 conmod15 conkfb16 conoff
17 rlam0118 rlam5019 sto05020 sto25021 sea0122 sea40
26PP QPF
S. Dierer et al.
Summer cases with overestimated precipitation - change of 24h area average precipitation sum in the sensitivity studies
AVERAGE PRECIPITATION23.06.2005
Obse
rvat
ion
Cont
rol
WSO
080
WSO
120
QV 0
90
QV 1
10DT
20
LFS
13RK
S 13
RK b
ott
RK tp
Mic
ro 1
Mic
ro 2 M
icro
3CO
N m
od CON
off
RLAM
01
RLAM
50
STO
050
STO
250
0
2
4
6
8
10
12
14
(blue: grid-scale rain, red: convective precipitation)
mea
n ac
c.pr
ecip
in m
m
AVERAGE PRECIPITATION 02.07.2005
STO
250
STO
050
RLAM
50
RLAM
01
CON
off
CON
mod
Mic
ro 3
Mic
ro 2
Micr
o 1
RK tp
RK b
ott
RKS
13
LFS
13DT 20
QV 1
10
QV 09
0
WSO
120
WSO
080
cont
rol
obse
rvat
ion
0
5
10
15
20
25
(blue: grid-scale rain, red: convective precipitation)
mea
n ac
c. p
reci
p in
mm
27PP QPF
S. Dierer et al.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
03.05.200523.06.2005 ?02.07.2005 ?12.07.2005
22
Summer cases with overestimated precipitation - change of 24h area average precipitation sum in the sensitivity studies
O ctrl1 wso802 Wso1203 Qv0904 Qv110
5 dt206 lfsl37 Rksl38 rkbott9 rktp10 oro
11 micro112 micro213 micro314 conmod15 conkfb16 conoff
17 rlam0118 rlam5019 sto05020 sto25021 sea0122 sea40
28PP QPF
S. Dierer et al.
Summer cases with underestimated precipitation - change of 24h area average precipitation sum in the sensitivity studies
21.06.2005
29PP QPF
S. Dierer et al.
Conclusions and Outlook• None of the studies completely solves a QPF problem, but some
give a significant improvement for single cases like • changes of snow microphysics for winter precipitation
overestimation• Kain-Fritsch/Bechtold for summer precipitation underestimation
• Initial humidity is of great importance • Runge-Kutta reduces mean precipitation in most of the cases• Decreased vertical exchange of heat and moisture (RLAM) tends to
decrease mean precipitation• Next steps:
• Complete statistics for all runs• Detailed look at runs with changes in order to understand
changes, e.g. why does Runge-Kutta tend to decrease average precipitation? Why decreased vertical exchange?
• Possibly further studies, e.g. using COSMO model version 3.22