Prediction of Tropical Cyclone Activity over the Western North Pacific in 2000
Issued in April 2000
1. Background
Chan et al. (1998, hereafter CSL) developed a statistical regression scheme to predict tropical cyclone (TC) activity over the western North Pacific (WNP) and the South China Sea (SCS). In their scheme, three sets of predictors were chosen:
sea-surface temperature anomalies over
the central and eastern Pacific (as a proxy for the El Ni
o/Southern Oscillation (ENSO) signal)
indices that represent the
characteristics of the circulation over Asia and the western Pacific from summer of the
previous year to March of the current year, and
trend of the interannual variations in TC activity (climatology and persistence).
Predictors in each set that pass the statistical tests are then used to make predictions and the final prediction is a linear, weighted combination of these forecasts. The weights are determined from the error characteristics of each predictor estimated through the jackknife technique. The predictands include TC activity over the entire WNP and that over the South China Sea (SCS) for the entire year as well as the active season (May to December). Three types of activity have been considered: all activity, only tropical storms and typhoons and only typhoons.
Real-time predictions of the 1997 and 1998 TC seasons were made using the CSL scheme. The results (Table 1) show that in 1997, the scheme predicted the total activity quite well although it over-predicted TC activity over the SCS. Almost the reverse is true in 1998 (Table 2). Since 1997 (1998) was a strong warm (cold) ENSO year, these results suggest that perhaps the scheme could not capture the precursor signal of either a strong cold or warm event. It is also possible that such signals do not exist prior to April of the current year. Nevertheless, these results suggest that the CSL scheme can provide some indication of the level of TC activity.
2. Predictions for 2000
To increase the sample size, the coefficients in the individual regression equations are re-derived by including the data in 1995 to 1999. The predictions (Table 3) suggest that while the overall number of TCs in the WNP is close to normal, most of these will reach at least tropical storm intensity and the number of typhoons is likely to be above normal. Tropical cyclone activity over the SCS is predicted to be near normal.
If the La Ni
a conditions that have been present since May 1998 subside, as
predicted by a number of researchers, this year could be considered as the year after La
Ni
a. Then, according to Chan (2000), TC
activity over the SCS is likely to be near or above normal in the earlier part of the
season (May to July). In the past six occasions (year after a La Ni
a event), three of them (1965, 1971, 1989) had above-normal number of TCs over the
SCS between May and July, two of them (1974, 1976)
had a normal number and one (1996) below. Most of the TCs developed
west of 145oE and moved across the Philippines into the SCS, generally making
landfall around Hainan Island (Fig. 1).
References
Chan, J. C. L., J. E. Shi and C. M. Lam,
1998: Seasonal forecasting of tropical cyclone activity over the western North Pacific and
the South China Sea. Wea. Forecasting, 13, 997-1004.
Abstract
Chan, J. C. L., 2000: Tropical cyclone
activity over the western North Pacific associated with El Ni
o and La Ni
a
events. J. Climate, 13, 2960-2972.
Abstract
Table
1. Verification of the 1997 TC activity forecasts made by the CSL scheme.
The boundaries are defined as follows: WNP: 100-180oE, 0-40oN;
and SCS: 100-120oE, 0-25oN.
The observed numbers are from the Joint Typhoon Warning Center and the
"normal" is the average for the years 1959-94.
1997 |
Forecast |
Observed |
Normal |
Annual for the WNP: |
|||
No. of TCs |
33 |
33 |
31 |
No. of TCs with at least tropical storm intensity |
30 |
31 |
27 |
No. of typhoons |
19 |
21 |
17 |
May to December for the WNP: |
|||
No. of TCs |
30 |
30 |
28 |
No. of TCs with at least tropical storm intensity |
27 |
30 |
25 |
No. of typhoons |
18 |
20 |
16 |
Annual for the SCS: |
|||
No. of TCs |
12 |
7 |
13 |
No. of TCs with at least tropical storm intensity |
9 |
7 |
10 |
Table 2. As in Table 1 except for 1998. The verifications are all for the annual activity since no TC formed before May 1998.
1998 |
Forecast |
Observed |
Normal |
WNP: |
|||
No. of TCs |
32 |
27 |
31 |
No. of TCs with at least tropical storm intensity |
28 |
17 |
27 |
No. of typhoons |
20 |
9 |
17 |
SCS: |
|||
No. of TCs |
13 |
14 |
13 |
No. of TCs with at least tropical storm intensity |
11 |
7 |
10 |
Table 3. Predictions of the number of tropical cyclones over the western North Pacific and the South China Sea in 2000.
2000 |
Forecast |
Normal |
WNP: |
||
No. of TCs |
30 |
31 |
No. of TCs with at least tropical storm intensity |
29 |
27 |
No. of typhoons |
22 |
17 |
SCS: |
||
No. of TCs |
13 |
13 |
No. of TCs with at least tropical storm intensity |
10 |
10 |
Fig.1
All Tracks |
![]() |
Track of TCs |
![]() |
Track of TCs |
![]() |
Track of TCs |
![]() |
Track of TCs |
![]() |
Track of TCs |
|
Track of TCs |
![]() |