How Much

1
How Much Skill was there in Forecasting the
1998-2001 La Nina Event and the 2002 El Nino
Onset?

• Chris Landsea
• NOAA/Hurricane Research Division
• Miami, Florida, USA
• 21 October, 2002
• Climate Diagnostics and Prediction Workshop

Co-Author John Knaff (NOAA/CIRA)
2
1997-98 El Nino Forecasts
3
Skill Assessment of the 1997-98 El Nino
4
ENSO Predictions for Summer 2002 from December
2001 (2 Season Lead 6-8 Months) Statistical
Models Analog -0.1C NEUTRAL CCA
0.6C WEAK EL NINO CLIPER 0.3C NEUTRAL LIM -
0.2C NEUTRAL Markov 0.4C WEAK EL
NINO Neural 0.3C NEUTRAL N.lin.-CCA 0.1C NEUT
RAL SSA-MEM 0.1C NEUTRAL Numerical Models
COLA -1.5C STRONG LA NINA COLA-Coup. 0.6C WEA
K EL NINO HCM 0.6C WEAK EL NINO LDEO -0.2C NE
UTRAL NCEP 0.8C MODERATE EL NINO Neural/Hybrid
-0.8C MODERATE LA NINA SCR/MPI 0.2C NEUTRAL BMR
C 0.5C WEAK EL NINO NASA
1.2C STRONG EL NINO
5
(No Transcript)
6
(No Transcript)
7
(No Transcript)
8
(No Transcript)
9
(No Transcript)
10
(No Transcript)
11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
(No Transcript)
21
(No Transcript)
22
Late 1996-Late 2001
23
Late 1996-Late 2001
24
Summary ENSO-CLIPER can be utilized as a
no-skill benchmark prediction For 1998-2002,
the most skillful models were (0-1 seasons)
CCA, Markov (2-3 seasons) Analog, LIM,
Markov For 1998-2002, numerical models are still
having difficulty outperforming CLIPER and thus
have little to no skill Significant biases Too
warm BMRC, CCA, NCEP Too cold COLA-Anom,
LIM, Neural, SCR/MPI Improvements in ENSO
predictability can achieve through
optimal combination of existing models