MJO Simulation Diagnostics Application of Diagnostics to Models Part II

1
MJO Simulation Diagnostics Application of
Diagnostics to ModelsPart II
Eric D. Maloney1 and the US CLIVAR MJO Working
Group2 1Oregon State University 2http//www.uscliv
ar.org/Organization/MJO_WG.html
Special thanks Daehyun Kim
2
Application to climate models
Motivation for Diagnostics

• Level 1
• Basic indication of spatial and temporal
  intraseasonal variability that can be easily
  understood and/or calculated by the non-MJO
  expert.
• Ease of use dictated that the analytic procedures
  be as simple as possible.
• Level 2
• Provide a more comprehensive diagnosis of the MJO
• Physical relationships

3
Application to climate models
Goals/Motivation of this Talk

• Demonstrate ability of diagnostics to distinguish
  between good and bad MJO simulations
• No models are perfect, and so I will tend to be
  evenhanded in criticism of models
• The SP-CAM and ECHAM models appear to produce the
  most realistic MJO simulations of the models we
  analyzed

4
Application to climate models
Outline

• Multivariate EOF-based indices (e.g. Wheeler and
  Hendon 2004)
• Composite structure in several variables
• Others, if time
• Conclusions

5
Application to climate models
Combined EOF Indices, NCEP 1 and AVHRR
EOF1
U200
Variance Explained
U850
OLR
EOF2
6
Application to climate models
Combined EOF Indices, Models
SP-CAM
CAM3.5
7
Application to climate models
Variance Explained
SP-CAM
CM2.1
Obs
ECHAM
CFS
8
Application to climate models
Lag Correlation of Leading PCS
SP-CAM
CM2.1
Obs
CFS
ECHAM
9
Application to climate models
Coher2, Phase Between the Leading uPCs
Projections of EOFs Onto Unfiltered Data
CM2.1
SP-CAM
Obs
ECHAM
CFS
10
Application to climate models
Phase Diagram Relating PC1 and PC2 Observations
Selection Criteria
PC2
PC1
11
Application to climate models
November-April U850 Composites NCEP1
Phase 1
Phase 2
Phase 3
Phase 4
Phase 5
Phase 6
Phase 7
Phase 8
12
Application to climate models
November-April U850 Composites Models
SP-CAM
CM2.1
13
Application to climate models
May-October U850 Composites NCEP1
Local E. Pacific Maximum
Northward and Eastward propagation
14
Application to climate models
May-October U850 Composites Models
CAM3.5
SP-CAM
15
Application to climate models
November-April Precip Composites GPCP
16
Application to climate models
November-April Precip Composites Models
ECHAM
GEOS5
17
Application to climate models
May-October Precip Composites GPCP
18
Application to climate models
May-October Precip Composites Models
ECHAM
CFS
19
Application to climate models
Specific Humidity Composite Observations
Kiladis et al. (2005)
1.1 g kg-1
20
Application to climate models
Specific Humidity Composite NCEP1 Models
ECHAM
NCEP1
SP-CAM
21
Application to climate models
Omega Composite NCEP1 Models
CM2
SP-CAM
NCEP1
22
Application to climate models
OLR and LHFLX Composites NCEP1/AVHRR Models
NCEP1
ECHAM
CAM3.5
23
Application to climate models
Conclusions

• Demonstrate ability of diagnostics to distinguish
  between good and bad MJO simulations
• No models are perfect, although a couple of
  models have nice successes at simulating the MJO
  (SP-CAM, ECHAM)
• Diagnostics allow first-glance looks at physical
  relationships underlying good and less-good MJO
  simulations

24
Thanks!
25
Application to climate models
OLR and SST Composites TMI/AVHRR Model
CFS
TMI/AVHRR
26
Application to climate models
Coher2 and Phase, Symmetric OLR vs. U850
SP-CAM
NCEP1 U850/AVHRR
27
Application to climate models
Climate models