Predictability of weather regimes
Data and methods:
Reanalysis: ERA-Interim (Z500 for clustering, UV250, UV850, T850, T2m, and PMSL for composites)
Observation: E-OBSv10 (precipitation for composites over the Euro-Atlantic sector)
Forecasts: TIGGE (ECMWF, JMA, NCEP, and UKMO)
[details]
and
GEFS reforecasts
Clustering method: Jung et al.'s K-means clustering (the original K-means method was proposed by Michelangeli et al. (1995, JAS))
Jung, T., T. N. Palmer, and G. J. Shutts (2005), Influence of a stochastic parameterization on the frequency of occurrence of North Pacific weather regimes in the ECMWF model, Geophys. Res. Lett., 32, L23811, doi:10.1029/2005GL024248.
Michelangeli, P-A, R. Vautard, B. Legras, 1995: Weather Regimes: Recurrence and Quasi Stationarity. J. Atmos. Sci., 52, 1237-1256.
Signigicant test: Straus et al.'s method
Straus, D.M., S. Corti, and F. Molteni, 2007: Circulation Regimes: Chaotic Variability versus SST-Forced Predictability. J. Climate, 20, 2251-2272.
Analyses of ERA-Interim and control analyses
EOFs
regimes on phase spaces
regimes on 3D phase spaces (Euro-Atlantic summer regimes, clus5)
regimes on 3D phase spaces (East-Asian winter regimes, clus5)
Regime ID comparison:
Europe:
[NDJFM]
[JJA]
[JJAS]
[MJJAS]
composite map (JJA)
Pacific:
[NDJFM]
East Asia:
[NDJFM]
composite map
North Pole:
[JJA]
[JJAS]
[MJJAS]
Reconstructed Z500
Cluster Centroids (Z500)
Straus's significant test
Area
Season
PCs
Expl. Var.
k=2
k=3
k=4
k=5
k=6
Euro-Atlantic region
DJF
1-20
91.8%
71.4%
97.6%
100.0%
100.0%
100.0%
NDJFM
1-20
91.8%
72.4%
96.8%
99.8%
100.0%
100.0%
JJA
1-20
87.5%
32.8%
36.6%
82.2%
71.4%
50.4%
JJA
1-3
33.1%
52.8%
69.6%
97.2%
99.6%
98.8%
JJAS
1-3
33.4%
64.8%
76.8%
97.4%
99.8%
99.8%
MJJAS
1-3
33.5%
37.2%
79.8%
95.0%
98.8%
95.4%
East-Asian region
DJF
1-20
97.0%
67.8%
72.6%
79.0%
88.2%
97.2%
NDJFM
1-20
96.9%
44.6%
61.2%
96.4%
98.4%
100.0%
JJA
1-20
93.6%
3.4%
2.2%
30.8%
16.6%
17.0%
JJAS
1-20
93.9%
2.6%
3.2%
27.6%
0.0%
0.4%
MJJAS
1-20
94.2%
0.6%
0.0%
0.4%
0.0%
0.0%
Pacific region
DJF
1-20
86.8%
31.0%
75.2%
96.0%
100.0%
100.0%
NDJFM
1-20
86.0%
20.2%
67.2%
96.2%
98.6%
100.0%
JJA
1-20
79.4%
0.0%
0.0%
0.0%
0.0%
0.0%
JJAS
1-20
80.0%
0.0%
0.0%
0.0%
0.0%
0.0%
MJJAS
1-20
79.6%
0.0%
0.0%
0.0%
0.0%
0.0%
North Pole
DJF
1-20
95.7%
3.0%
73.2%
85.2%
78.2%
75.2%
NDJFM
1-20
95.5%
0.4%
52.0%
86.4%*
54.6%
67.2%
JJA
1-20
92.6%
45.6%
79.2%
87.0%
94.8%
83.4%
JJAS
1-20
92.7%
46.8%
80.6%
94.8%
98.0%
96.4%
MJJAS
1-20
92.8%
59.6%
83.2%
60.8%
79.4%
56.2%
Australia-New Zealand region
DJF
1-20
90.3%
51.0%
8.6%
32.8%
33.4%
62.4%
NDJFM
1-20
90.0%
53.4%
34.6%
68.0%
86.4%
91.4%
JJA
1-20
90.3%
37.6%
82.6%
90.8%*
92.0%*
94.2%*
JJAS
1-20
90.3%
34.2%
69.0%
77.8%
95.8%
95.0%
MJJAS
1-20
90.2%
31.7%
67.4%
85.0%
98.6%
99.8%
South America
DJF
1-20
93.6%
77.2%
61.6%
87.2%
94.6%
98.2%
NDJFM
1-20
93.5%
87.2%
87.6%
98.8%
100.0%
100.0%
JJA
1-20
94.4%
97.0%
98.0%
99.8%
100.0%
100.0%
JJAS
1-20
94.4%
99.4%
99.6%
100.0%
100.0%
100.0%
MJJAS
1-20
94.3%
99.8%
100.0%
100.0%
100.0%
100.0%
*North Pole, NDJFM: ERA-Interim and GEFS analyses show different patterns for clus4
*Australia-New Zealand: large differences between JJA and JJAS/MJJAS
Composite maps (Z500,Z250,Z10,UV850,UV250,T850,T2m,SF&VP850,SF&VP250,PMSL,and precipitation)
YSNOW(WM) vs SSNOW (WP) around East Asia
Observed frequency of regimes
Observed frequency of regimes during El Nino/La Nina (table)
Observed duration of regime
Observed relations between regimes
New!
Observed relations between regime frequency and MJO phases
New!
MJO lagged composites
New!
Frequency of regime transition:
(plot)
(table)
all regions and seasons:
only ERA-Interim
all regions and seasons:
ERA-Interim and each analyses
all regions and seasons during El Nino/La Nina:
ERA-Interim
Europe, NDJFM:
[clus4]
[clus4, each month]
Europe, JJA:
[clus4]
[clus5]
[clus5, each month]
Europe, summer:
[clus4, each month]
[clus5, each month]
East Asia, NDJFM:
[clus5]
[clus5, each month]
East Asia, DJF:
[clus5]
Pacific, NDJFM:
[clus4]
Pacific, DJF:
[clus4]
Winter daily map (Z500,T2m,SLP, and precipitation with Z500 reconstracted from PCs)
Summer daily map (Z500,T2m,SLP, and precipitation with Z500 reconstracted from PCs, only for the Euro-Atlantic sector)
Analyses of TIGGE forecast and GEFS reforecast
(PDFs of PCs)
Comparicon of observed and predicted PDFs of PCs)
(Example of ensemble forecats on phase spaces)
Ensemble forecasts on phase spaces (only for winter Euro-Atlantic and East-Asian regions)
(Frequencies of regime and regime transition)
Predicted frequency of regimes
Predicted frequency of regimes (each center)
Predicted frequency of regimes transition (observed frequency inclusive)
text file for observed frequency
Expected frequency of regime (Markov chain)
(Verification of probabilistic regime foreacasts)
forecat probability for NDJFM regimes (Euro-Atlantic, Pacific, and East-Asian sectors)
forecat probability for JJAS regimes (only for Euro-Atlantic sector)
Reliability diagram for probabilistic regime forecast
Reliability diagram for probabilistic regime forecast (only regression lines)
Brier Skill Score (BSS) for probabilistic regime forecast (for all cases and initial regime dependency)
Probabilistic skil dependency on Initial and forecast regime ("stamp" map of BS and BSS)
(Verification of Z500 forecasts)
Deterministic and probabilistic scores for Z500 forecast (for all cases and dependency on initial regime)
Skill dependency on Initial and/or forecast regime (Z500 forecasts)
Skill dependency on Initial and/or forecast regime (Z500 forecasts, each lead time)
Skill dependency on Initial and forecast regime (Z500 forecast, "stamp" map)
Skill dependency on Initial and forecast regime (Z500 forecast, "stamp" map, each lead time)
Timeseries of score and regimes for Z500 forecast
(Forecast skill and MJO)
MJO teleconnection
New!
Dependency of deterministic skills on MJO
BSS for probabilistic regime forecast (dependency upon initial MJO phases)
New!
Probabilistic skil dependency on Initial and forecast regime("stamp" map of BS and BSS, dependency upon initial MJO phases))
References
Dawson, A., T. N. Palmer, and S. Corti (2012), Simulating regime structures in weather and climate prediction models, Geophys. Res. Lett., 39, L21805, doi:10.1029/2012GL053284.
Dawson, A., and T. N. Palmer (2015), Simulating weather regimes: impact of model resolution and stochastic parameterization, Clim. Dyn., 44, 2177-2193. doi:10.1007/s00382-014-2238-x.
Ferranti, L., Corti, S. and Janousek, M. (2015), Flow-dependent verification of the ECMWF ensemble over the Euro-Atlantic sector. Q.J.R. Meteorol. Soc., 141, 916-924. doi: 10.1002/qj.2411.
Frame, T. H. A., J. Methven, S. L. Gray and M. H. P. Ambaum (2013), Flow-dependent predictability of the North Atlantic jet, Geophys. Res. Lett., 40, 2411-2416, doi:10.1002/grl.50454.
Frame, T. H. A., Ambaum, M. H. P., Gray, S. L. and Methven, J. (2011), Ensemble prediction of transitions of the North Atlantic eddy-driven jet. Q.J.R. Meteorol. Soc., 137, 1288-1297. doi: 10.1002/qj.829.