Title: Variability of Tropical Intraseasonal Convective Anomalies
1Variability of Tropical Intraseasonal Convective
Anomalies and their Statistical Forecast
Skill Charles Jones, Leila M. V. Carvalho, Wayne
Higgins, Duane Waliser and Jae-K Schemm
- Objectives
- What is the seasonal frequency of TICA
occurrences? - How does the TICA frequency depend on ENSO phase?
- Are there differences between TICA
characteristics? - What is the seasonal frequency of major TICA
events? - What is the forecast skill of TICA events?
- Are there significant differences in the forecast
skill of different TICA types - Does the forecast skill depend on ENSO phases?
2Equatorial Eastward EE
- Wang and Rui (1990)
- Synoptic climatology of TICA
- OLR 1975-1985 (1978 missing)
- 5x5 Latitude/longitude
- Life span longer 4 pentads minimum OLRA -25 W
m-2 - Tracking done with minimum OLRA
- Large diversity of TICA occurrences
Equatorial Northeast/Southeastward N(S)E
Equatorial Northward EN
3MASCOTTE
- Data
- OLR and Zonal wind (U) at 200 hPa and 850 hPa
- Pentads January 1979 December 2002 5x5
lat/lon - Filtering
- For each gridpoint, fit harmonic analysis and
remove components with T 100 days - One pass of 1-2-1 filter
- OLR is further smoothed with weighted spatial
filter - OLRA Screening
- For each pentad, contiguous regions with OLRA
-10 Wm-2 are identified - The minimum within the region -15 Wm-2
- OLRA regions are tracked in time with MASCOTTE
-10 Wm-2
t
-
-
t 1
Maximum spatial correlation between OLRA regions
in successive images
-
- TICA Events
- Duration T 4 pentads
- Minimum OLRA during life cycle -25 W m-2
- Eastward propagation and end eastward of 50E
Carvalho, L. M. V., and C. Jones, 2001 A
Satellite Method to Identify Structural
Properties of Mesoscale Convective Systems based
on Maximum Spatial Correlation Tracking Technique
(MASCOTTE). Journal of Applied Meteorology, 40,
1683-1701.
4- Properties of TICA events
- Date and duration (T)
- Lat/lon of center of gravity (-10 Wm-2, -15 Wm-2)
and minimum OLRA - Number of clusters OLRA -15 Wm-2
- Minimum OLRA and OLRA variance
- Fraction of the contiguous OLRA region
intercepting 5 key regions (Rk,t k1,5)
For each pentad (t)
Intersection Ratio in region K and time t
t 1
t
Cumulative Intersection Ratio in region K
Significant Interaction in region K ?k 1
5TICA trajectories
Total 110
IND (19)
ISO (53)
MJO (38)
6What is the seasonal Frequency of TICA events?
7What is the seasonal Frequency of TICA events?
How does the TICA occurrence depend on the ENSO
phase?
8Major MJO Zonal Displacement Median Value in
the quarter and Minimum OLRA Median Value in
the quarter
What is the frequency of major MJO events? 34.21
Increased wind speed is the rectification
effect calculated according to Shinoda and Hendon
(2002)
9TICA Statistical Forecast Skill
- Model Development
- Extended Winter (17 Nov 15 May) and Summer (16
May 16 Nov) Seasons - OLRA U200 and U850 Anomalies
- Develop Combined EOF analysis
U850
Retain First Two EOF Modes
Space
A
U200
OLRA
Time
- Fit multiple linear regression
? 1,10
1979
1980
2000
1999
1990
Training
Forecast
22 years of forecasts
Training
Forecast
10Winter Validation
11Winter Validation
All pentads in the winter season
12(No Transcript)
13- Conclusions
- Eastward propagating TICA exhibits a large degree
of variability. - Frequency of one or more 0.9 (JFM, AMJ, OND)
- Neutral and Warm ENSO phases
- TICA statistical forecast skill
- Useful skill extends to about 5 pentads over most
of Indian and western Pacific - MJO has slightly higher forecast skill than ISO