Tropical cyclone forecast verification - northern hemisphere 2009

1. Introduction

A summary of tropical cyclone activity in the Northern Hemisphere for the 2009 season is presented below together with an assessment of the performance of the Met Office global model in predicting the tracks of these tropical cyclones.

Tropical cyclones are experienced in the North Pacific, North Atlantic and North Indian Oceans and nearby tropical seas. For the purpose of tropical cyclone verification the northern hemisphere is divided in to four basins; the North-West Pacific (west of 180° E to the Malay Peninsula), the North-East Pacific (east of 180° E), the North Atlantic and the North Indian Ocean (west of the Malay Peninsula). Mean error statistics for each basin are presented together with a table of statistics for the whole Northern Hemisphere. Verification is performed at 12-hour intervals up to forecast time T+144, although statistics are only presented at 24-hour intervals in this report. The global model resolution in operation during the season was 0.5625° x 0.375° x 50 levels. This is equivalent to a horizontal resolution of 62km x 42km at the equator. The vertical resolution was increased to 70 levels in November 2009.

An explanation of error statistics and their sign conventions.

Advisory positions received in real time from RSMC Tokyo, JTWC Hawaii, NHC Miami and CPHC Honolulu are used as verifying observations of storm location. Best track data from these centres will be obtained once they become available. Past experience shows that use of best track rather than real time data usually only makes minor differences to seasonal error statistics. Some mean error statistics for last season are also included for the purposes of a comparison. Forecast tracks are only verified when a depression reaches tropical storm status.

2. Tropical cyclone activity

Tropical cyclone activity
 NWPNEPNATNITOTAL
Tropical depressions (<34 knots)4(3)3(2)2(1)0(0)9(6)
Tropical storms (>34 knots, <64 knots)10(13)12(10)6(8)4(6)32(37)
Hurricanes/typhoons (>64 knots)14(12)8(7)3(8)1(1)26(28)
Total28(28)23(19)11(17)5(7)67(71)

Categories assigned based on 1-minute averaged winds

Basin name abbreviations:
NWP : North-west Pacific (west of 180° E)
NEP : North-east Pacific (east of 180° E)
NAT : North Atlantic
NI : North Indian (west of Malay Peninsula)

The number in brackets indicates the figure for the 2008 season.

3. Summary of all northern hemisphere storms

3.1 North-west Pacific basin storms

Table of mean error statistics
 T+0T+24T+48T+72T+96T+120T+144
Possibly verified2662191791401057960
Detection rate (%)1001001009910099100
Mean DX1-26-59-121-204-256-371
Mean DY-5-11-33-76-153-257-347
Mean AT-11-39-70-102-201-293-404
Mean CT0-2-27-70-73-82-166
Mean skill (%)*****415354***************
2008 skill (%)*****304440***************
Mean DPE36115205311471664863
* 2008 DPE 69113195321477695593
Intensity skill (%)*****494641333637

* DPE for all north-west Pacific storms in 2008 season

Plot of the observed tracks of all storms in the north-west Pacific basin
Forecast positional errors in the north-west Pacific basin
Forecast skill in the north-west Pacific basin

There were a similar number of tropical cyclones to 2008 in this region, but a greater proportion of these were typhoons. Tracks were longer and started further east - probably due to the El Niño which developed during the year. Track forecast errors were very similar to those seen in 2008, except at T+144 which were much higher. Skill against CLIPER continued to rise - scores at T+24 and T+48 were the highest ever and at T+72, the second highest. There was a slow bias in forecasts overall. The intensity tendency skill was 39% overall.

3.2 North-east Pacific basin storms

Table of mean error statistics
 T+0T+24T+48T+72T+96T+120T+144
Possibly verified145103684631208
Detection rate (%)10010010093816538
Mean DX2-29-77-215-479-877-1687
Mean DY-15-34-74-126-193-216-430
Mean AT-14-25-64-115-168-2451218
Mean CT-8-28-66-122-177-221-517
Mean skill (%)*****253845***************
2008 skill (%)*****-12217***************
Mean DPE3811723137763310221750
* 2008 DPE 51112184264314320371
Intensity skill (%)*****576572673833

* DPE for all north-east Pacific storms in 2008 season

Plot of the observed tracks of all storms in the north-east Pacific basin
Forecast positional errors in the north-east Pacific basin
Forecast skill in the north-east Pacific basin

The 2009 season in this basin was slightly more active than 2008. There were more tracks in the central Pacific due to the developing El Niño. Track forecast errors at lead times longer than T+24 rose sharply compared to 2008. In particular, errors at T+120 and T+144 were huge. This was primarily due to very poor forecasts for two hurricanes - Rick and Neki. Skill scores against CLIPER up to T+72 were actually good and at higher levels than in 2008. There was a slight slow and left-of-track bias in forecasts. The intensity tendency skill was 57% overall.

3.3 North Atlantic basin storms

Table of mean error statistics
 T+0T+24T+48T+72T+96T+120T+144
Possibly verified574027191384
Detection rate (%)100100100100100100100
Mean DX-8-33-57-107-111-143-381
Mean DY-6-41-130-245-371-587-1068
Mean AT-7-32-82-163-281-492-997
Mean CT-9-36-64-128-183-245-424
Mean skill (%)*****516262***************
2008 skill (%)*****516060***************
Mean DPE401081772834106201142
* 2008 DPE 5295186279376513619
Intensity skill (%)*****35115800

* DPE for all North Atlantic storms in 2008 season

Plot of the observed tracks of all storms in the North Atlantic basin
Forecast positional errors in the North Atlantic basin
Forecast skill in the North Atlantic basin

2009 saw the lowest activity in this region since 1997 by some measures. Track forecast errors were near to last season's values at short lead times, but much higher at longer lead times. However, there were very few long lead time forecasts to verify. Skill scores against CLIPER were the highest or second highest ever recorded. There was a slow and left-of-track bias in forecast tracks. The intensity tendency skill score was 16% overall.

3.4 North Indian Basin Storms

Table of mean error statistics
 T+0T+24T+48T+72T+96T+120T+144
Possibly verified201130000
Detection rate (%)10010067********************
Mean DX-7-108-409********************
Mean DY-9-91-234********************
Mean AT6-112-20********************
Mean CT14175********************
Mean skill (%)*****-23*************************
2008 skill (%)*****345266***************
Mean DPE68197481********************
* 2008 DPE 74117226338418718342
Intensity skill (%)*****45100********************

* DPE for all North Indian storms in 2008 season

Plot of the observed tracks of all storms in the North Indian basin
Forecast positional errors in the North Indian basin
Forecast skill in the North Indian basin

Cyclone activity was much lower in 2009 than the previous few years. Storms were short lived, so no forecasts beyond T+48 were verified. Track forecast errors were, on average, higher than the last few years for the few forecasts that were verified. The intensity tendency skill was 60% overall.

3.5 Combined statistics for whole northern hemisphere

Table of mean error statistics
 T+0T+24T+48T+72T+96T+120T+144
Possibly verified48837327720514910772
Detection rate (%)10010010098969393
Mean DX-8-23-54-102-244-329-430
Mean DY-8-23-54-102-180-278-393
Mean AT-11-36-69-111-202-303-366
Mean CT-3-13-39-86-101-113-197
Mean skill (%)*****385053***************
2008 skill (%)*****344948***************
Mean DPE38117210322494708919
* 2008 DPE59106190293401556596
Intensity skill (%)*****504844363334

* DPE for all northern hemisphere storms in 2008 season

Forecast positional errors for the whole northern hemisphere
Forecast skill for the whole northern hemisphere
Along-track errors for the whole northern hemisphere
Cross-track errors for the whole northern hemisphere
Intensity skill for the whole northern hemisphere.

Overall, activity in the northern hemisphere in 2009 was lower than 2008, although there were regional variations. Track forecast errors showed a slight increase on previous seasons at short lead times, but a large increase at longer lead times. The latter was primarily due to poor forecasts for a few long lived storms. Skill scores against CLIPER were similar to the high values achieved in 2008. The slow and left-of-track biases seen in 2008 were larger in 2009. Detection percentages were fairly high with a small percentage of forecasts being missed from T+72 onwards. The intensity tendency skill was 41% overall, which is the highest value recorded since intensity verification started in 2001.

Further tropical cyclone information

The Met Office web pages contain information on tropical cyclone forecasting at the Met Office. Monthly updates of tropical cyclone activity and forecasts are made, together with observed and forecast track information of recent storms, track prediction error statistics, lists of names and real time tropical cyclone forecast guidance.

Tropical Cyclone information can be found on the home page under the "Weather/World" heading.

Seasonal summaries of tropical cyclone activity and forecasts have been issued since the 1994-5 Southern Hemisphere season. To obtain these or any further information on tropical cyclone forecasting email the Met Office.

________________________

Acknowledgements

Mr.S.Lord, NMC, Washington, USA and Mr.C.Mauck, FNOC, Monterey, USA supplied CLIPER models for various basins.
Mr.S.Lord and Dr.M.Fiorino supplied GrADS software used to produce track plotting charts.

Last updated: 17 December 2013