(3-12 DECEMBER 2002)

WMO Report/TD No. XXX

Secretariat of the World Meteorological Organization
Geneva, Switzerland



Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . i

Foreword. . . . . . . . . . . . . . . . . . . . . . . . . . . . iii

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

CHAPTER 1: Introduction. . . . . . . . . . . . . . . . . . . . 1

CHAPTER 2: Summary and Recommendations
by S. Ready (New Zealand), E. Ritchie-Tyo (USA), and
J. Heming (UK) . . . . . . . . . . . . . . . . . 4

CHAPTER 3: Summary of Keynote Session on Current, Imminent, and
Future Satellite Observations for Tropical Cyclone
Forecasting and Research

by C. Velden (USA), J. LeMarshall (Australia), and
J. Kossin (USA) . . . . . . . . . . . . . . . . . 10

CHAPTER 4: Topic 1 - Tropical Cyclone Structure and
Structure Change

by R. Elsberry, USA . . . . . . . . . . . . . . . . 12

CHAPTER 5: Topic 2 – Tropical Cyclone Landfall Processes
by G. Holland, Australia. . . . . . . . . . . . . . 16

CHAPTER 6: Topic 3 – Tropical Cyclone Motion
by J. C.-L. Chan, Hong Kong, China. . . . . . . . . 19

CHAPTER 7: Topic 4 - Tropical Cyclone Formation and
Extratropical Transition

by P. Harr, USA . . . . . . . . . . . . . . . . . . 22

CHAPTER 8: Topic 5 – Tropical Cyclone Impacts
by R. Falls, Australia. . . . . . . . . . . . . . . 25


Appendix A: Daily Schedules for IWTC-V . . . . . . . . . . . . 28
Appendix B: Global Climate Change and Tropical Cyclones
by T. Knutson, USA . . . . . . . . . . . . . . . . 33
Appendix C: A Survey of Tropical Cyclone Forecast Center
Present Uses and Needs of Satellite Data

By R. L. Elsberry (USA) and C. Velden (USA) . . . . 35
Appendix D: List of Participants . . . . . . . . . . . . . . . 41


The Fifth International Workshop on Tropical Cyclones (IWTC-V) continues a series of Workshops that have brought together forecasters and researchers from around the globe to summarize progress, to discuss needs and requirements, and to plan for future advances. Two of these IWTCs resulted in textbooks and a third IWTC produced a global forecast guide. A web-based update of this global forecast guide is being recommended as a product of IWTC-V.

The design and implementation of the workshop has been carried out by an International Committee and supported by an active Local Organizing Committee under the able direction of G. Foley with assistance by K. Wright. The production of an extensive workbook, which is one of the strengths of the Workshop, depended on the considerable efforts by the Topic Chairs and Rapporteurs, who were assisted by Working Groups. This workbook summarized the advances in forecasting and research during the past four years and made suggestions for future activities. This workbook was published as WMO/TD No. 1136 (which is No.67 in the Tropical Meteorology Research Programme report series) and distributed at the beginning of the Workshop. In addition, a compact disk (CD) version was produced for those who preferred that medium. P. Jones of the USA Naval Postgraduate School coordinated the production of the workbook and the CD.

In addition to the contributions of WMO, especially for the attendance of many of the forecasters, financial assistance for the attendance of the researchers was supplied by their home government. This list includes Canada, Australia, USA, China, France, Hong Kong (China), India, Germany, United Kingdom, Japan, and Korea, plus multi-national organizations such as the European Centre for Medium-range Weather Forecasting. In addition, the USA National Weather Service International Affairs Office, the USA Office of Naval Research International Field Office, and the Pacific Science Association served as co-sponsors and provided funding for several participants. The social activities and outings were supported in part by the Bureau of Meteorology, Aerosonde, Vaisala, and Environmental Systems and Services (Australia).

A Keynote session on Present, Imminent, and Future Uses of Satellite Observations for Tropical Cyclone Forecasting and Research was arranged by C. Velden (USA) and J. LeMarshall (Australia). Special Focus sessions in the form of tutorials on the imminent satellite systems were arranged primarily for forecasters. Other Special Focus sessions were arranged primarily for researchers. A Special Focus session on Global Climate Change and Tropical Cyclones was arranged to determine if new research since 1998 has resulted in a need for a new WMO statement on this topic.

Recommendations (see Chapter 2) are directed to the WMO, to the National Meteorological Service forecast offices, and to the international research community.

In summary, I consider the Workshop to have been a resounding success. The sharing of knowledge, the recommendations, and the collaborations developed will have a beneficial effect for forecasters and researchers for years to come.

Russell L. Elsberry
Chair, International Committee



R. L. Elsberry (USA)

1. Organization and historical background

Activities related to tropical cyclone research are one of the major components of the WMO Commission on Atmospheric Science (CAS) Tropical Meteorology Research Program (TMRP). The TMRP develops projects that will provide scientific advice and guidance for the development and implementation of programs in collaboration with the WMO Tropical Cyclone Program that will improve forecasts.

Following the directive of the Ninth Congress (1983), the International Workshop on Tropical Cyclones (IWTC), now referred to as IWTC-I, was organized in Bangkok, Thailand from 25 November to 5 December 1985 under the International Programme Committee (IPC) chaired by W. M. Gray (USA). This workshop was intended to review the current state of knowledge on tropical cyclones and survey the status of forecasting practices around the globe based upon extensive prepared materials. The reports with recommendations were published as the Proceedings of the WMO International Workshop on Tropical Cyclones (TMRP Report Series No. 21). Another outcome of IWTC-I was the preparation of a text book, A Global View of Tropical Cyclones (GVTC), with R. L. Elsberry as editor and chapter contributions from R. L. Elsberry, W. M. Frank, G. Holland, J. D. Jarrell, and R. L. Southern. Though the establishment of a Standing Advisory Committee on Tropical Cyclones was recommended as the highest priority, EC-XL (1986) decided that the IPC for IWTC-II should cover the functions of the proposed committee.

The IWTC-II was organized under a newly established IPC chaired by G. Holland with a main objective of implementations that can be made to improve forecast office procedures and reduce forecast errors. The report on the IWTC-II was published as proceedings (TMRP Report Series No. 37) with recommendations, some of which followed up those of IWTC-I. New aspects of the recommendations included a statement on the potential impact of climate change on tropical cyclones, a request for action to install PC-based work stations in tropical cyclone forecasting offices, and a statement on the International Decade for Natural Disaster Reduction (DNDR). Another important recommendation was to create the International Committee on Tropical Cyclones with terms of references. Through the discussion in the tenth session of CAS (CAS-X) and TCP regional body meetings, the International Committee for IWTC was established with the same terms of reference as mentioned above.

As an outcome of the IWTC-II, the preparation of the Global Guide for Tropical Cyclone Forecasting was initiated by G. Holland as the editor and with chapter contributions by C. Neumann, R. Merrill, G. Holland, C. Jelesnianski, W. Gray, G. Foley, R. Southern, and K. Puri. It was published just before the IWTC-III.

The IC for IWTC-III was chaired by J. Jarrell (USA) with members G. Holland, R. L. Elsberry, L.-S. Chen, D. F. Best, and S. C. Ready. Since IWTC-III was organized with the cooperation of ICSU under the ICSU/WMO demonstration project “Tropical Cyclone Disasters” for the IDNDR, ICSU was also represented by K. Emanuel on the IC.

The major tasks of IWTC-III held in Huatulco, Mexico during 22 November – 1 December 1993 were to update the textbook “A Global View of Tropical Cyclones,” develop a statement concerning global climate change and tropical cyclones, and to review progress on recommendations from IWTC-II and to develop further recommendations. A new textbook “Global Perspectives on Tropical Cyclones” (WMO/TD—No. 693) was edited by R. Elsberry with chapters by G. Foley, H. Willoughby, J. McBride, R. Elsberry, I. Ginis, and L.-S. Chen. A symposium under the chairmanship of J. Lighthill (ICSU) on global climate change and tropical cyclones led to a paper by J. Lighthill, G. Holland, W. Gray, C. Landsea, G. Craig, J. Evans, Y. Kurihara, and C. Guard. A subsequent report was commissioned by CAS to provide a post-IPCC assessment of climate change and tropical cyclones under a committee chaired by Ann Henderson-Sellers. This report was presented to CAS-XII and published in the Bulletin of the American Meteorological Society. A forecaster forum was also organized to highlight issues and request research that addresses specific forecast needs.

The IC for IWTC-IV was chaired by G. Foley (Australia), with co-chair L.-S. Chen (China) and members G. Holland, R. L. Elsberry, Sir James Lighthill, S. Ready, A. V. R. K. Rao, J. Jarrell, and R. R. Vaghee. When Mr. Rao of India retired just prior to the IWTC-IV, he was replaced by S. R. Kalsi. The IWTC-IV was held in Haikou, Hainan Province, China, during 21-30 April 1998.

The major tasks of IWTC-IV were to revise the Global Guide to Tropical Cyclone Forecasting, review progress in tropical cyclone research and operational practices, and to identify opportunities that may prove fruitful to explore or develop. A forecasters meeting produced their perspectives on operational issues and research needs. An innovation in IWTC-IV was informal special focus groups in the afternoons and evenings organized by a facilitator. Nine such groups met and the response of the participants was enthusiastic.

2. International Committee (IC)

The IC for IWTC-V consisted of:

3. Objectives of IWTC-V

As in the previous international workshops, the overall objective was to provide a forum for discussion between researchers and forecasters on the current status of tropical cyclone knowledge and on priorities and opportunities, and in particular:

4. Keynote session

Because of the importance of satellite observations for tropical cyclone analysis and forecasting, and with an imminent, large increase in the amount of satellite platforms, sensors, and data streams, a Keynote session on Present, Imminent, and Future Satellite Observations for Tropical Cyclones was organized principally by Chris Velden (USA) with assistance by John LeMarshall (Australia). As shown in the Daily Schedules for 3 and 4 December in Appendix A, the Keynote session had four parts: (i) Forecast office present uses; (ii) Imminent uses and data assimilation; (iii) Future satellite instruments and opportunities for tropical cyclones; and (iv) Satellite data assimilation plans at selected forecast centres. Representatives of five forecast offices (Brisbane, Australia; RSMC-Fiji; RSMC-La Reunion; RSMC-Miami; and USA Joint Typhoon Warning Centre) described the present uses and future needs. The imminent satellite products presentations covered the Tropical Rainfall Measurement Mission microwave imager and precipitation radar, Quikscat scatterometer, Advanced Microwave Sounding Unit (AMSU), and new geostationary infrared products. The presentation on future satellite instruments covered the whole range of meteorological sensors and indicated via timelines when each was expected to become available. Ten-fold increases of data flow are expected within the next five years. Planning needs to begin now for methods to handle these data streams in forecast offices and numerical weather prediction centers. Researchers need to be planning how to use these new satellite data for improved physical understanding, for assimilation in models, and for developing new forecast guidance products. Thus, one objective of IWTC-V was to inform participants about these opportunities for tropical cyclone analysis and forecasting. Information on the satellite data assimilation plans were presented by seven centres (Japan Meteorological Agency, European Centre for Medium-range Weather Forecasts, National Centers for Environmental Prediction, UK Meteorological Office, China Meteorological Administration, Naval Research Laboratory-Monterey, and Meteo-France).

A brief summary of the Keynote session and recommendations that arose from the discussions at IWTC-V is given in Chapter 3.

5. Workshop topics and organization

As shown in the Daily Schedules in Appendix A, the next five days were devoted to presentation of the Rapporteur and Topic Chair reports in the following topics:

Topic 1 - Tropical Cyclone Structure and Structure Change
(R. Elsberry)
Topic 2 - Tropical Cyclone Landfall Processes
(G. Holland filled in for W. Frank)
Topic 3 - Tropical Cyclone Motion (J. Chan)
Topic 4 - Tropical Cyclone Formation and Extratropical Transition
(P. Harr)
Topic 5 – Tropical Cyclone Impacts (R. Falls)

A change from previous practice was to include an immediate discussion period following each Rapporteur report rather than delaying discussion until all reports were given. The purpose was to ensure all topics were discussed and avoid only a few topics being discussed in plenary session.

The order of these topics reflects the increasing importance of structure and structure change (this includes intensity change as well as outer structure change such as the radii of 34 kt, 50 kt, and 64 kt winds). Similarly, the Topic 2 includes precipitation during and following landfall, which is being recognized as a primary factor in loss of lives and damage. For the first time, hydrologists were invited to IWTC-V (see Topic 2.3) to make the connection between tropical cyclone precipitation and flooding. Another emphasis in Topic 2 was storm surge, which is still a major threat in many countries. About 15 Australian disaster managers joined the Workshop for the Topic 5 impacts discussion, which emphasized the need to address public and other customer requirements in the total warning system.

Small Discussion Groups of about 15 participants were organized for each Topic. These Groups gave opportunities for more detailed discussions. The Topic Chairs provided summaries of potential recommendations from discussions by the Groups.

The Recommendations Committee (S. Ready - New Zealand, E. Ritchie-Tyo - USA, and J. Heming - UK) collected these inputs from the Discussion Groups and from the plenary session. They also reviewed the recommendations from IWTC-IV. Their synthesis of these inputs was presented in plenary (see Daily Schedule for 11 December in Appendix A), modified, and endorsed. These recommendations in three categories (for WMO, for research community, and for forecast community) are included in Chapter 2.




Steve Ready (New Zealand), Elizabeth Ritchie-Tyo (USA), and Julian Heming (UK)


Major General Recommendations

1. The future of the Global Guide to Tropical Cyclone Forecasting was discussed at a special meeting and in plenary. The guide was seen as a valuable forecast reference and it was recommended that it undergo an evolutionary revision and be reissued. Several specific aspects were addressed:

The structure of the Guide should be revised slightly with some name changes, amalgamation of two chapters, and addition of a new chapter. The final content is recommended to be:

The bibliography should be expanded to become a general, tropical bibliographical resource, with submitted references being added if they are in the correct format.

It was further recommended that a forecaster's Web site be established as part of the Guide Web site on which operational studies, forecasting rules, and other relevant material could be posted.

2. The amount of remotely sensed data from satellite and ground-based observations (e.g., microwave sounders and imagers, Doppler radars) is expected to increase markedly in the next few years, along with high-resolution models that can assist with the analysis and forecasting of tropical cyclone structure and position. The meeting recommends that WMO actively work to ensure that:

  1. These data be made available as widely as possible to operational forecasting centres. Network communications need to be improved to guarantee accessibility of data and RSMC products for all forecast centres during high traffic periods.
  2. Tools and techniques be developed to combine the most useful
    observational and model data to aid the forecaster in real-time analysis and forecasting of tropical cyclone structure and intensity.
iii) Training be provided to operational forecasters in the best use of
these new observations and tools that become available.

Other WMO Recommendations

1. Both in-situ and high-resolution radar measurements of tropical cyclones at landfall and over land are considered a high priority by the meeting. The WMO World Weather Watch should support development and improvement of observing networks in countries affected by tropical cyclone landfall through such programmes as the VCP. A particular concern is to improve the capacity of instruments to survive severe weather events. In addition, WMO should encourage all countries to make freely available the highest resolution observations of tropical cyclone landfall, including rainfall data, for research and model development and verification.

2. Storm surge and wave height forecasting is still a significant problem in many tropical cyclone-affected countries. WMO should endorse and encourage the establishment of storm surge techniques and models, including river flooding and wave action (especially for small islands), for regions that do not have this capability. In addition, vulnerability assessments should be conducted for all countries threatened by storm surge.

3. Although scatterometer data were not specifically developed for tropical cyclone applications, IWTC-V recognises the valuable contribution these data have provided to both the operational and research communities. The meeting recommends that WMO encourage the development of future plans for deployment of scatterometer sensors, and other satellite surface wind vector retrievals within the tropical cyclone.

4. IWTC-V recognises the importance of widespread communication of official tropical cyclone warnings.

i) WMO should ensure that tropical cyclone warnings are disseminated using communication methods that are most easily accessed by the local community at risk. It is important that alternate communication systems are available as a backup.

  1. WMO should continue to emphasise the use of the official RSMC tropical cyclone forecasts and actively promote the use of the Severe Weather Information Centre (SWIC) web site when it becomes fully operational.

5. Tropical cyclone intensity should not be defined solely by a single parameter such as central pressure or maximum wind. A more detailed structural analysis is required. The WMO should encourage forecast centres to report within the current WMO format structural information such as quadrant gale radii, eye size, and radius of outer closed isobar in international exchanges of both real-time and best track data. In addition, the time of occurrence and the value of the minimum pressure, and an indication of the occurrence of an eye passage should be added to the synoptic code. This information is needed to determine the maximum wind-minimum central pressure relationship associated with tropical cyclones.

6. IWTC-V recognises there is a gap between research findings and the ability to transfer this information into operational techniques. WMO should facilitate the transition of research findings on all aspects of tropical cyclone analysis and forecasting into operations. In addition, training of management, forecasters, researchers and users should be provided using current WMO training programmes. The meeting endorses the USWRP testbeds and recommend similar demonstration programmes for other countries. The meeting also endorses the approach used in the implementation of the “systematic approach” as an example of how to transition research to operations.

7. IWTC-V recognises that a consistent definition of extratropical transition (ET) of a tropical cyclone does not exist and recommends that WMO support the development of an operationally and physically consistent definition for ET for use by the operational and research communities, and that this ET definition be presented at IWTC-VI. This should include conceptual models of wind, precipitation and ocean surface wave distributions.

8. The meeting considers small focus workshops to be a useful means to organise research-operational-hydrological interaction on topics of particular interest to the operational community. The meeting endorses the increasing use and application of ensemble prediction systems (EPS) in forecasting tropical cyclones. WMO should organise a series of thematic workshops on (e.g.,)

  1. flood forecasting models
  2. forecasting of regional and local flooding due to TCs
  3. storm surge forecasting
  4. extratropical transition of tropical cyclones
  5. ensemble models for QPF of tropical cyclone landfall

9. WMO should encourage socio-economic policies and programs to increase societal resilience to tropical cyclone impacts such as storm surge and inland flooding. To document the value of the Tropical Cyclone Warning Centres to the community, there should be a careful assessment of the economic value of the tropical cyclone warning. In addition, it is imperative that a strong effort in public education be implemented as a permanent, continuous program. WMO should foster the following initiatives:
i) Education of groups involved in disaster preparedness and mitigation (e.g., media and emergency managers) in the tropical cyclone warning process including the difference between official and unofficial forecasts and warnings.
ii) Co-ordinate development and sharing of products (e.g., pamphlets,
videos, interactive CDs, Web pages) to promote public awareness
and education.

  1. Workshops to facilitate the education process.

10. It was recommended that the organisational structure for IWTC-VI follow IWTC-V procedure in which each topic has forecaster and researcher rapporteurs as appropriate. The continuation of special focus sessions is strongly recommended.

11. Following the practice with the previous IWTCs, it was recommended that the next IWTC review the recommendations of the previous IWTC in the pre-workshop documents.


1. The meeting recommends continued development of statistical-dynamical, ensemble strategies/techniques, dynamical and conceptual models for use in research and real-time operational forecasting of tropical cyclone structure and intensity change over the ocean, during and following landfall.

i) International co-operation is encouraged to address the major
modelling issues such as environmental effects, convective
processes, air-sea/land interface, and atmospheric and oceanic
boundary layers.
ii) Special attention should be afforded to the issue of vortex
initialisation and of advanced data assimilation techniques in
models, making best use of in situ and radar observations, aircraft
reconnaissance data and current and future satellite data.
iii) Ensure that attention is given to accurate prediction of both
tropical cyclone inner and outer structure, intensity and track
in the development of models.

iv) Multidisciplinary co-operation is encouraged to address
prediction of rainfall associated with tropical cyclones over
land at high spatial and temporal resolution and the associated
and subsequent effects of flooding, mudslides, and debris flows.

2. The meeting considers the validation and utility of both models and new satellite techniques and data sources to be an important issue:

i) Model representation of tropical cyclone structure should be validated against all available data, including aircraft reconnaissance and radar. Particular attention should be paid to validation of convective asymmetries.
ii) Validation of tropical cyclone formation and intensity changes in
regional and global models should be encouraged including development of appropriate validation techniques and turbulent kinetic energy formulations for ocean mixing parameterization schemes.
iii) New sources of satellite data that can be used to represent tropical cyclone structure should be validated against aircraft reconnaissance and radar data, and through in-situ observation-based verification field campaigns in all basins lacking reconnaissance programs.

3. Future research on tropical cyclone structure, intensity change, motion, and predictability during formation, maturation, landfall, decay, and extratropical transition should focus on, but not be restricted to, the effects of the following:

i) The environmental flow
ii) Wind shear
iii) Vortex Rossby waves
iv) Mid-latitude circulation
v) Tropical upper tropospheric troughs
vi) Air-sea-land interactions
vii) Upper-ocean and atmospheric boundary layers
viii) Convective asymmetries
ix) Vertical tilt
x) Thermal anomalies
xi) Potential vorticity tendency

The use of observations is considered important as part of this research.

4. The meeting recognises the value of data obtained from process-related studies and field campaigns (e.g., CBLAST, NOAA/HRD, CAMEX) and Global Positioning System dropsonde data and encourages their utilisation to improve the understanding of the atmospheric and oceanic boundary layers, air-sea-land interaction, tropical cyclone motion, and the effects of landfall on tropical cyclone structure and intensity.

5. The meeting recognises the need for a comprehensive rainfall climatology of the tropics including temporal and spatial characteristics. Thus, this meeting recommends that a precipitation climatology should be developed and disseminated to all countries.

6. While considerable progress has been made in storm surge forecasting in the last several years, the meeting recognises that this capability needs to be improved and extended to real-time forecasts of lowland inundation. Thus the meeting recommends that current storm surge and wave setup models should be improved via validation using measurements, through intercomparison studies, and through improvement of parametric wind models.

7. Parametric wind models form a basis for a range of forecast and diagnostic applications. Yet many such models are kept confidential or have not been adequately tested. The meeting recommends that a public domain parametric wind field model, fully tested and verified by peer review, be developed to provide the standard for comparison purposes.

8. There is a need for Numerical Weather Prediction (NWP) centres to evaluate the skill and utility of operational NWP and mesoscale model systems in predicting tropical cyclone formation.

9. Research should continue to identify intraseasonal, synoptic, and mesoscale factors associated with basin and seasonal factors influencing tropical cyclone formation.

10. It is recommended that seasonal tropical cyclone prediction be continued through the development of dynamical and dynamical-statistical models, and ensemble prediction techniques, and the value of seasonal forecasts be increased by including information on tracks, timing and landfall characteristics on seasonal timescales.

11. To advance the understanding of the process of extratropical transition, it is recommended that a comprehensive research program be developed including the use of existing data sets, and field experiments in co-operation with existing programs such as THORPEX.

12. An international database should be developed that would track the social and economic impacts of tropical cyclones as well as the costs associated with tropical cyclone forecasting, information services, and mitigation methods. This information could then be used to justify additional resources for tropical cyclone forecasting and disaster mitigation.

13. The meeting recommends that the research community explore the use of ensemble forecasting techniques for tropical cyclone forecasting including track, intensity, Quantitative Precipitation Forecasts (QPF), storm surge, wind waves and flood forecasting.

14. IWTC-V
recommends that a suitable format be developed to enable summary rainfall information for tropical cyclones to be archived, along with automated techniques for deriving such an analysis.
The format should allow the data to be useable for objective verification of rainfall forecasts for climatological, hydrological and risk studies, and for post-season summaries.

15. A parametric model of precipitation associated with a landfalling storm should be developed combining:

16. The lack of observational data has caused difficulties with the calibration of the Dvorak intensity analysis technique. It is recommended that calibration or re-calibration of the Dvorak technique, and all pressure-wind relationships, be undertaken in all basins.


1. Although significant improvements have been made in the accuracy of tropical cyclone track forecasts by operational Numerical Weather Prediction (NWP) models, large forecast errors do occasionally occur and systematic errors still exist. All NWP and major forecast centres are urged to isolate the causes of these large errors, and evaluate the sources and impacts of the systematic errors. These investigations should include examination with respect to initial conditions, vortex specification techniques, data assimilation, track type, and model resolution. It is recommended that the research community conduct model intercomparison exercises with cases of tropical cyclones that have caused major forecasting errors.

2. Consensus forecasting techniques have been demonstrated to improve track forecasts provided an adequate number of skillful forecasts is available. It is recommended that all major NWP centres make available track and intensity forecasts, and radius of gale/storm force winds out to 120 hours or beyond with access via GTS and controlled access on a suitable website (e.g., WMO). The WMO should request the European Centre for Medium-range Weather Forecasts to make available forecasts of tropical cyclones in their deterministic model and their tropical ensemble when it is operational.

3. Consensus forecasts should be closely evaluated to identify

  1. The minimum number and optimal combination of forecast members
that adds value to the forecast process
ii) Scenarios that yield small errors
  1. Strategies to deal with situations when the consensus forecast does not work (e.g., bifurcation scenario when members are in two different track types)

4. It is recognised that ensemble forecasting techniques may provide an important opportunity to improve tropical cyclone track predictions. To be able to utilize these techniques in an optimum way, it is recommended that

i) Ensemble-based products be made available to all RSMCs and Tropical Cyclone Warning Centres, who should undertake verification of these techniques.
ii)Training be provided on how to apply these techniques in an
operating setting.

5. All centres should create best-track (post-storm) data sets using data that might not have been available in real time. All efforts should be made to access non-operational data from local tropical cyclone regions that may increase the fidelity of best-track data sets. Best-track data sets should be in the WMO format (subject to revision). In addition, eye diameter should be recorded. Centres should provide an explanation of the originating point of each best track. The meeting commends the work done on re-analysis, including United Nations Environment Program (UNEP) for its global database of tropical cyclones. This work should be continued, and the integrity of historical best-track data is likely to be increased by careful re-analysis using current understanding and new techniques.

6. Close co-operation is encouraged among operational forecast centres, hydrological agencies and disaster management officials in planning the mitigation of tropical cyclone impacts, development of response strategies, and in the exchange of information during tropical cyclone events.

7. There is a need for a standard conversion chart that enables users to convert between different wind-averaging periods and gust factors. The meeting endorses the RSMC recommendation for updating the conversion chart in
the Global Guide on Tropical Cyclone Forecasting and requests that the updated values be distributed to TCWCs as soon as they become available. In addition, the chart should be updated on a regular basis as more research into these conversion factors comes to hand.

8. Tropical cyclone forecast offices are encouraged to evaluate the skill and utility of operational NWP and mesoscale model systems in predicting tropical cyclone formation.

9. One of the main impacts of tropical cyclones is the inundation of coastal areas. To address this problem, it is recommended that techniques for forecasting inundation areas be considered and applied, including the use of inundation maps with the combined effects of river floods and storm surge.



J. Kossin and C. Velden

The recurring theme of the Keynote session on forecaster use of satellite observations addressed the need for objective methods for dealing with (expertly synthesising) and presenting the vast amount of data that is now available to forecasters. Analyses must be pared down to the germane features, while further research and communication between researchers and forecasters is needed to establish what those germane features are.

Lixion Avila (U.S. National Hurricane Center) discussed the time constraint that the forecasters are under, and noted that “point and click” images such as the presently available Cooperative Institute for Meteorological Satellite Studies (CIMSS) and Naval Research Laboratory-Monterey products are very useful. He suggested that the dilemma of a lack of conventional wind data in oceanic regions needs to be further addressed using satellite data. In particular, a need for better estimation of the gale-force wind radii (R34) was pointed out. Philippe Caroff (La Reunion, France) pointed out the need for more objective guidance, particularly with Tropical Rain Measurement Mission microwave imager (TMI) and Special Sensor Microwave Imager (SSMI) imagery.

The need for more objective guidance was particularly emphasized by Rich Jeffries (Joint Typhoon Warning Center, USA), who suggested that a “data overload” was occurring and would only get worse for forecasters with the advent of more satellite sensors and derived products. He suggested that visualization is a key issue, and noted that the CIMSS vertical wind shear products are a good example of a useful product. Similar to Lixion Avila, Rich Jeffries pointed out that efficient use of forecaster’s time is critical, and suggested that subjective interpretation methods are too hard to train on, particularly with transient staff. He noted that an expert system that gives objective guidance is sorely needed, particularly with the TMI and the SSMI. For example, the Dvorak technique for estimating tropical cyclone intensity would greatly benefit from information from microwave data, particularly during eyewall replacement events. Other specific recommendations made by Rich Jeffries were to automate the centre fix required by the Dvorak technique, and to perform post-season verifications of products used for operational forecasting.

Alipate Waqaicelua (Fiji Meteorological Service) also recommended post-season validation of satellite-based techniques and products, and he pointed out the need for better objective guidance for TC intensity, track, precipitation, size (e.g., R34), and associated sea height interactions in the region of small islands. He expressed a need for formal training for forecasters on current and future satellite-derived products.

During the discussion period, Lixion Avila expressed a desire to have all products used by forecasters available at a unified website. Jim Kossin posed the question of what sort of forecaster training will be needed as objective expert systems achieve widespread use. For example, a completely automated method would technically not require training to execute. A subsequent point was made that the training may shift emphasis from understanding subjective products to understanding the methods of the expert system.

2. Imminent uses

Satellite-based products that are in an advanced state of development or in “beta-test” at a forecast centre were presented. The first presentation was on the Japan-USA Tropical Rainfall Measurement Mission (TRMM) Precipitation Radar (PR) data that are now more widely available. Two presentations were on the Quikscat surface wind vectors, one on the Advanced Microwave Sounding Unit (AMSU), and one on an extension of the Objective Dvorak Technique for intensity estimation. Tutorials on each of these new products (except the TRMM PR) were offered to give more detail and to allow more questions by forecasters. These special focus sessions were well attended, which indicates the wide interest in these new satellite-based products.

3. Data assimilation session - John LeMarshall and Chris Velden

It is apparent that satellite data assimilation has been afforded high priority at Numerical Weather Prediction (NWP) centres, as all of the centre’s plans include the use of advanced methodologies to exploit current and future satellite observations. Several examples were provided of the benefit of properly assimilated satellite data to NWP in general, and tropical cyclone (TC) forecasting in particular. Of note was a recent study by the United Kingdom Met. Office that demonstrated the importance and considerable impact of satellite data on global NWP forecasts. It was also noted that full exploitation of satellite data will require continued and expanded international collaboration to provide the necessary science and software tools to NWP centres.

In relation to use of the increasing amounts of satellite data becoming available to NWP, it is felt guidance and information needs to be more widely available in terms of the optimal use of the data, in particular for TC modelling and prediction. This guidance should include data selection, quality control, and assimilation methodologies. Development efforts involving major NWP centres should be coordinated to achieve cost-effectiveness in light of the expected increase in the magnitude of the satellite observations coming soon. For example, U. S. Joint Centre for Satellite Data Assimilation (JCSDA) is a pro-active collaborative effort that is beginning to address assimilation issues in anticipation of the new satellite data sources expected over the next decade.

It is recommended that other operational NWP centres establish collaborations to exploit the JCSDA advancements. It is also recommended that increasing research and development activity be directed to the optimal assimilation of current and future satellite data for NWP application. This will assist in the introduction of these data in a timely and optimal fashion at operational NWP centres.



R. L. Elsberry (USA)

  1. Introduction

The purpose of this Topic 1 Chair report is to summarise additional issues raised at the IWTC-V during plenary and Discussion Group sessions that reviewed the Rapporteur and Topic Chair reports. It is not the intent here to repeat the material contained in the Topic 1.0 report. (The reader is referred to WMO/TD No. 1136 for the Rapporteur and Topic Chair reports). Rather, new items that were raised during the IWTC-V sessions will be the primary focus of this report.

Another source of information for consideration by the IWTC-V participants was recommendations from the 4th Technical Coordination Meeting of the TC Regional Specialised Meteorological Centres (RSMC) in Nadi, Fiji. Of the ten recommendations, six were related to intensity and intensity change (and thus part of this topic), and this included three of the top five recommendations. Another three of the RSMC recommendations were related to precipitation prediction. This set of recommendations reflects the changing requirements from the operational forecast community. During earlier IWTCs, the number one focus had been TC motion and track prediction. The new importance of structure change and precipitation is reflected in the organisation of IWTC-V by making these Topics 1 and 2. Whereas TC motion is shifted to Topic 3, this does not mean that track prediction does not continue to be an operational requirement; rather, significant advances in track prediction have been made and it now makes the accurate structure (or precipitation) prediction more viable if the track is better known. As one of the Discussion Groups pointed out, an accurate track prediction is required to correctly predict the effects on the TC structure of an approaching mid-latitude trough or region of high vertical wind shear.

As a final introductory note, this topic is closely related to Topic 2 TC Landfall Processes. It is unlikely that dynamical model guidance for TC-related precipitation will be successful unless the model correctly predicts the wind structure. Clearly, the surface flux of moisture from the ocean and the inward advection of the moisture to where it rises in clouds and precipitates depends on the correct representation of the atmospheric boundary layer of the vortex. Thus, some overlap between the two topic reports should be expected.

  1. Structure Analysis and Forecasting

The Topic 1.6 on structure analysis and forecasting techniques was presented first at IWTC-V to provide the motivation for this topic, which is almost exclusively devoted to research issues. That is, the operational intensity forecasts at forecast centres around the globe have little skill relative to a combined climatology and persistence technique. As summarised in the Topic 1.6 report, many forecast centres are using (or developing) statistical-dynamical techniques for intensity prediction. Since no dynamic model has shown significant skill in intensity prediction, the statistical-dynamical models need to be improved. Several of the imminent and future satellite products presented in the Keynote session are expected to provide better estimates of the initial intensity, and this will be a first step in improved intensity change prediction.

One of the shortcomings in the historical records (and continuing in many forecast centres) that might be used to develop improved statistical-dynamic techniques is that only minimum sea-level pressure (MSLP) or maximum surface wind speed have been used to characterise the structure. This record is insufficient since these two measures are not uniquely related. For the same MSLP, a different radius of maximum winds (where the horizontal pressure gradient is steepest) will give a different maximum wind speed, and vice versa. Thus, a radius of maximum winds (RMW) should be a part of the storm description, and the radii to other threshold surface wind speeds (34 kt, 50 kt, and 64 kt) should be provided when applicable. This requirement presents a problem for operational centres when the TC is not within radar range or no microwave imagery revealing the eyewall is available to at least provide an estimate of the RMW. Evidence presented in the keynote session suggests that the Quikscat surface winds may be used to define the outer wind structure inward to at least the 34 kt radius (and perhaps even 50 kt). What is needed are techniques to describe the intermediate wind structure between the 34 kt radius and the RMW (derived from the microwave satellite imagery) with the maximum wind speed (from infrared satellite imagery). The wind structure in this intermediate region is important for the storm motion, the supply of heat and moisture from the oceans, the ocean surface wave generation, etc.

One hypothesis to be explored is that the high-resolution, regional models have sufficiently accurate physical process representations that the outer wind structure should be predictable given an accurate initial structure specification. That is, the prediction of outer wind structure change is believed to be considerably more tractable than prediction of the eyewall region, which may include eyewall mixing events and involve highly complex, overshooting inflows at the base of the eyewall. As indicated above, a combination of Quikscat surface winds in the outer region, plus a means of specifying the intermediate zone to the radius of maximum winds, may be adequate for vortex initial conditions at the surface. A consistent free-atmosphere vortex would have to be specified. If this hypothesis is correct, the next few years should see predictions of the radius of gale-force winds, which would be useful to the emergency managers in setting the timing of evacuations or other disaster prevention measures.

As described in the Topic 1.5 report, a new era of research models with extremely high resolution (1-2 km in the horizontal) and full physics has arrived. These models need to be validated with observations on this horizontal scale. Fortunately, the aircraft radar data have the horizontal and vertical resolution to validate several aspects of these new models. For example, the dynamical model prediction should have similar probability density functions as the observed vertical motion and rain rates, and similar convective versus stratiform cloud distributions. A limited set of microphysics distributions in tropical cyclones are available from the NASA Convection and Moisture Experiments (CAMEX) and the NOAA Hurricane Research Division (HRD) field experiments. Since comparisons of the model microphysics and these experimental data sets have revealed some significant discrepancies, more data sets and model comparisons are required for a variety of storm intensities and vertical and horizontal positions within the storm circulation. Clearly, the relative fractions and locations of convective rain and stratiform rain with radius from the centre will also be a requirement for representing the latent heat release, and also for accurate quantitative precipitation forecasts (Topic 2). These microphysical distributions are also an important factor in satellite retrieval model validations for tropical cyclones.

  1. Environmental Effects

Forecasters in the IWTC-V Discussions Groups and the RSMC recommendations emphasised the need for better understanding and forecast tools for rapid intensification, and for rapid weakening. This high priority item requires prediction of the onset, duration, and magnitude of these intensifying and weakening events. While the Topic 1.2 report on environmental effects indicates research progress in understanding the vertical wind shear effects on intensification and weakening, no operational model provides accurate guidance. Vertical wind shear products derived from satellite-based winds described in the Keynote session may provide some forecaster guidance. The extratropical transition of a TC to an extratropical cyclone (discussed in Topic 4.6) is another source of rapid intensity decreases.

As described in the Topic 1.2 report, the concept of “good troughs” (i.e., contributing to intensification) and “bad troughs” have been introduced into the folklore. However, the dynamical models have a difficult time accurately predicting the effects of adjacent trough interactions. Thus, this topic (specifically interaction with mid-latitude baroclinic systems) is item five of the 10 RSMC recommendations.

One of the IWTC Discussion Groups emphasised the need to provide good intensity forecast guidance over the range of environmental conditions. The specific recommendation was to study weak storms as well as mature storms.

  1. Convective Scale Effects

The asymmetric structure of convection in tropical cyclones was described in Topics 1.1, 1.2, and 1.3. In some cases such as storms experiencing strong vertical wind shear, the asymmetry in convection is evident in visible and infrared imagery. In other cases, the asymmetry may be obscured in those imageries because of thick cirrus clouds. Thus, microwave imagery (now available on web sites such as the Naval Research Lab – Monterey, CA) is required to detect partial eyewall structures and asymmetric rainband convection where the storm is not within the range of coastal radars (or aircraft reconnaissance in the Atlantic). However, it is not clear how a snapshot of asymmetric convection at a single time may be utilised by the operational forecaster to predict the future convective structure and its effect on the vortex structure. Clearly, the asymmetry in convection must also be assimilated as part of the initial conditions of the dynamical model if an accurate structure forecast is to be produced. However, it has not been demonstrated that dynamical models can predict such structure evolutions.

The amount and distribution of small cumulus clouds along the inflow trajectory appears to be an important factor in the accumulation of heat and moisture. One of the IWTC-V Discussion Groups expressed a concern that the inward heat and moisture flux may not be accurately predicted if these small cumulus clouds are not well represented on outer grids (which have coarser resolution) of a nested model. In addition, the possibility exists that the nested grid interface boundary condition will impede the inward flux of heat and moisture. Thus, alternate numerical approaches such as an adaptive grid or a much larger domain for the high resolution, inner grid may be necessary.

The possible role of eyewall mixing events in changing the vortex structure is receiving considerable research attention. In many cases, the inner wind structure and MSLP are considerably altered by such mixing events. Various aspects such as triggering mechanism(s), magnitude of the mixing, resulting effects on the vortex structure, time to “reload” for another event, etc. will need to be understood and predicted for guidance to the forecaster. Another research topic is predictability of such eyewall mixing events, or will they be included as a stochastic function in the model forcing?

  1. Air-Ocean Boundary Layer Influences

The issues, related to the Topic 1.4 report, are summarised in my Topic Chair 1.0 report. The main addition at IWTC-V was a Special Focus session on the U.S. Office of Naval Research Coupled Boundary Layer Air-Sea Transfer (CBLAST). Ten scientists presented short synopses of their research and how it will contribute to determining the air-sea fluxes and leading to a better parameterisation in the high winds of a hurricane. This program of field experiments, analyses, and modelling is expected to significantly advance understanding of interface processes and the changes in both the atmospheric and oceanic boundary layers.

  1. Numerical Model Guidance

Many aspects of the research modelling have been included in the above sections. It was also emphasised that these models must be validated by observations (e.g., HRD aircraft radar probability density functions). Many of the modelling efforts in the U.S. are tied to the CBLAST and CAMEX field experiments that will provide validation data sets, including the microphysics distributions that are being parameterised in the dynamical models.

The challenges of producing an operational forecast model capable of predicting TC structure changes are many. In addition to having to address scales of phenomena ranging from sea spray droplets to interactions of the TC with mid-latitude systems, the operational modelling centres must deliver the guidance product on a strict time schedule. Both explicitly resolved convection on the inner grid and parameterised convection on the outer grid must be addressed. These and other aspects are included in the Topic 1.5 report and Topic Chair 1.0 report. It is only emphasised here that the data assimilation problems on the vortex scale and in the environment are very difficult. In addition, the land-surface model, ocean surface wave model, and the ocean model must be initialised. If these aspects can be addressed, we may expect improved structure change guidance products for the forecasters.

Many national operational modelling centres have programs to develop improved TC structure models in the next few years. This IWTC can assist from the sharing of ideas, and hopefully collaborations in the future.




G. J. Holland (Australia)

1. Introduction

This report summarises the issues raised at IWTC-V during the presentations and related discussions. Please refer to WMO/TD No. 1136 for the full Rapporteur and Topic Chair (prepared by Bill Frank, who could not attend) reports.

Here we define landfall to include the approach of the cyclone to the coast, the coastal crossing, and the subsequent decay and transition over land. Within this envelope, the major damage (both wind and flooding) and economic losses tend to be in highly developed coastal zones, which continue to increase in value and complexity. However, the major loss of life in recent decades has been from inland rainfall and its associated flooding, land slip and mud slides.

The landfall of tropical cyclones has gained substantial prominence and attention across all fields of activity in recent years. The available observing systems and techniques have reached the level of sophistication to enable detailed analysis and research into the related processes and effects on the cyclone structure and associated impacts. We have seen an enormous increase in relevant data from the USA field programs and increasing operational use of remote sensing systems, both satellite and land-based, which provide a wealth of information for research activities. Numerical models with higher resolution allow direct calculation of mesoscale (and even microscale) processes, and lead to improved parameterizations, and thus provide a capacity for both improved research and future operational forecasts. Innovative basic research approaches are improving our understanding and capacity to analyze important features of the boundary layer wind field. In addition, the impacts/response and meteorological communities have moved closer together in improving the warning message and its communication.

Nevertheless, all is not rosy. We have very little real skill in forecasting the intensity and structure of tropical cyclones as they approach land, especially when rapid intensity changes occur (up or down), and no skill in predicting any details of the rapid structural changes that occur at landfall. No real skill exists in forecasting tropical cyclone rainfall, and in many cases such forecasts are little more than a general indication of rainfall conditions. This deficiency is the major limitation to hydrological modeling of the run-off and flooding that ensues. Whereas the storm surge modeling is relatively sophisticated, the high dependence on the tropical cyclone track and the wind structure and the poorly modeled interactions with river flooding, are critical limitations.

Overlaying all of these aspects is the great dependence on track forecasts, where even relatively small track errors can result in substantially different rainfall, storm surge, precipitation, and localized wind-field changes. Thus we can provide no more than a general warning of the damage that will occur at a specific location. Even so, the prospects for improvements are excellent.

2. Rainfall

As mentioned in the introduction, cyclone-induced rainfall is a topic of major community concern and a growing research area. Numerical models are considered to now have sufficient capability in terms of resolution and explicit cloud physics to address many issues of this rainfall production. Promising advances are being made in digital analysis of the available data sources (radar, rain gauge, and satellite) for operational and post-event research in several countries. Satellites such as the Tropical Rainfall Measurement Mission (TRMM) also have given us a major boost in capability to observe the rainfall structure over the oceans. However, the data sources remain sub-optimal for most communities. For example, radars are often not well calibrated, and few have dual polarization, which is critical for accurate rainfall estimates.

For these reasons, discussion on this topic tended to focus on the synoptic and storm-scale structure of rainfall. Discussion on high rainfall concentrations tended to be more of a descriptive nature (e.g., associated with topography). The potential for radar and in situ data was discussed and recommended for continued expansion and development.

The growing capability and role of mesoscale models in providing forecasts of rainfall during and after landfall were emphasized. Even if the track forecast errors could be eliminated, there remain other issues. Such limits include the proper representation of rainfall associated with the surface wind/land/ topographical interactions, the interactions with adjacent synoptic-scale systems, and the transformations during extratropical transitions.

Some statistical/empirical techniques have shown promise in the prediction of general rainfall amounts and trends. More work in this area would provide a useful comparative approach to the more complex mesoscale modeling approaches.

3. Hydrological Models of Precipitation

There is little doubt that flooding causes the major loss of life in tropical cyclones and in recent decades the largest contributor has been inland flooding. The manifestations vary from widespread flood outbreaks from major rivers that extend over several days or weeks to local flash floods that can develop in minutes.

There are two phases to flood prediction: predicting the location and timing of rainfall and then modeling the hydrology of the resultant water runoff and percolation through the substrate. A comprehensive review of models and modeling approaches indicated that while the science was well advanced, the predictions were hampered by rainfall prediction errors, lack of adequate basin data and model calibration capacity, and real-time changes associated with the evolving landfall event. Tropical cyclone rainfall is intense and is often very localized, and these factors tend to make the hydrological forecasts difficult. Little real-time information is usually available to assess how well the rainfall forecasts are performing. Inadequacy of model calibration is particularly a problem in smaller basins with complex terrain and strong local gradients of rainfall. This calibration also tends to be lost and additional factors arise when the cyclone landfall is underway. These factors include problems in receiving adequate and timely data on changes in the river characteristics, soil wetness conditions in the catchment, and errors in predictions and handling of coastal ocean/river flooding.

As a result of these factors, much uncertainty remains in forecasting flooding from tropical cyclones.

4. Coastal and Inland Wind Structure Changes

Dramatic changes occur to the low-level wind fields as the high wind region of the cyclone moves over land. There is a relatively slow change as the cyclone loses its source of energy and decays. However, superposed on this trend are rapid changes that occur as the surface generally becomes rougher and more heterogeneous and the surface fluxes of sensible heat and moisture are dramatically reduced. Several processes then occur: the boundary layer cools and stabilizes; a variety of mesoscale circulation features are often created; and orographic effects may greatly change the local wind structure via blocking or gap flows. We are aware of several potential mesoscale processes that may be contributing locally, including relocation of the maximum wind zone, development of roll vortices, development of mesoscale vortices, and breakdown of the eyewall wind structure. However, our capacity to forecast even the general aspects of such features is virtually non-existent. Some skill is apparent for well-defined orographic effects provided the track forecast is sufficiently accurate.

Major limitations to both our knowledge and forecasting of these changes lie in the available boundary layer models and observing techniques. New remote sensing techniques, including mobile Doppler radar and deployable instrument packs have helped with observing under quasi-research conditions. However, the sparse routine observing systems often fail or provide incorrect readings when subject to high wind conditions. Such failings can arise from poor instrument siting and inadequate maintenance, but storm-related effects, such as flying debris, can also cause failures.

The group also discussed a range of issues concerned with observations of landfalling tropical cyclones: a review of the newer remote sensing techniques for monitoring the storms; problems with in-situ wind observations during extremely high winds over land; and the problems that beset the performance of boundary layer models when applied under these extreme conditions. It was acknowledged that major advances have been achieved in our knowledge of boundary layer processes under high wind conditions over the ocean, but much more needs to be done for the landfall process.

5. Storm Surge

Storm surge is the net elevation of sea level that occurs along the coast as a tropical cyclone makes landfall or moves along a track that is close to the shore. Although the manner in which a tropical cyclone causes death and destruction varies greatly from one storm to the next, the storm surge (sometimes combined with fresh water flooding) has been responsible for the greatest losses of life. Storm surge theory is relatively well advanced compared to many aspects of tropical cyclone forecasting. When the storm track and wind distributions are well forecast and the ocean bathymetry is well known, surge models can produce impressively accurate forecasts of the pattern of flooding. However, a great many problems still remain related to the forecasting of storm surge and the measures that need to be taken to minimize its effects.

Recent developments in storm surge modeling were presented together with methods of using them more effectively. Because of the conditions and locations where storm surges occur, it is often very difficult to obtain accurate indications of the actual surge height and extent. Verification methods were extensively reviewed especially those used in conjunction with remote sensing techniques.

A significant public knowledge gap exists on the effects of storm surges and the need to evacuate critical areas. This gap could be alleviated by improved public education programs for both the general public and civil defense officials.

6. Where to Next?

The problems of landfalling tropical cyclones require comprehensive international programs to alleviate them. Much can be learned by the cross-fertilisation of ideas and technologies from one region or country to another. Thus, strong support for the World Weather Research Program Tropical Cyclone Landfall Program and the U. S. Weather Research Program Hurricane Landfall program was indicated. Implementation of regional tropical cyclone coastal impact programs was recommended to provide a means of transferring the international information to the local community level.



J. C.-L. Chan (Hong Kong, China)

1. Introduction

This report summarises the discussions and recommendations related to Topic 3 – Tropical Cyclone Motion. While most of these items are centred around the Rapporteur reports, some additional comments have also been made.

As a prelude, it is worth noting that in all of the Discussion Groups, the main focus was on track forecasting, which differed from the previous four IWTCs in which theory of motion and developing dynamical models were the foci. This track forecasting focus is a clear reflection of the significant advances the research community has made over the last 15 years in understanding the physics of tropical cyclone motion. As pointed out in the Topic 3.0 summary and the Rapporteur report of Topic 3.4 theoretical advancements, relatively few researchers are now engaged in such studies. Thus, this report discusses almost exclusively the forecasting aspects and only brief comments are made on future theoretical research.

An additional point not mentioned in any of the Rapporteur reports is the question of the limit of track predictability. Even if we were able to make further improvements in the models and understanding in the physics of tropical cyclone motion, what is the limit to which errors in track forecasts can be reduced? This is both a scientific and an economic issue because such a limit determines the amount of resources that should be invested in the improvement of tropical cyclone track prediction or understanding. The issue of predictability should therefore be a topic for future research.

2. Numerical and Statistical Model Guidance and Improvement

Significant improvements have been made during the last few years in the accuracy of track forecasts from numerical weather prediction (NWP) models. Mean 72-h track forecast errors today are comparable to mean 48-h errors a decade ago, mainly due to improved model physics and initialisation (including the insertion of synthetic or bogus observations). However, major model failures occasionally plague forecasters, especially for erratic track cases in which the 48-h forecast errors may exceed 500 km. Therefore, one workshop recommendation was a systematic evaluation of such failures to identify model problems. These failures could arise from a poor representation of the vortex (i.e., the insertion of synthetic observations), from possible systematic biases or false alarms of individual models, and certain synoptic situations in which such large errors tend to occur. One question was whether the model performance might be linked to the occurrence of El Niño events. Information on how the synthetic observations are created should be shared among the different numerical prediction centres. Research should also be carried out to understand the causes of erratic tracks and oscillatory motion, as well as the interaction between the tropical cyclone and its neighbouring mesoscale and subsynoptic-scale systems.

Although tremendous improvements have been made in numerical track forecasts, systematic errors are found in individual models. The workshop recommended detailed evaluation of model predictions to identify such systematic biases and to correct them either through modification of the model or post-processing of the forecasts. Because an important aspect of track forecasting is the determination of the landfall location, it was also suggested that special attention should be made for the evaluation and verification of landfall forecasts.

Since the performance of different models may vary under different situations, the workshop recommended that the predictions from the various operational centres be distributed and shared as widely as possible. Individual forecast centres could then make use of more information in their decision-making. In addition, evaluations of the model performance can be done more extensively and intercomparisons between models can also be made.

A strong recommendation is for WMO to request the major NWP centres to transmit track predictions up to 120 hours. However, such predictions need to be evaluated in detail and might be more appropriately labelled as “outlooks” rather than “forecasts”.

Much of the discussion on the improvements in track prediction through NWP models focused on global models. One group noted the lack of discussion on the use of regional models in track prediction and questioned whether it is still necessary to employ such models given the problems of boundary conditions in regional models and the high resolution already present in global models.

3. Consensus Forecasts

The workshop found the consensus approach (i.e., combining a number of numerical forecasts)to be useful in improving track forecasts. The main questions are the minimum number of such forecasts for the consensus and the number of consensus members beyond which the improvement in track forecasts becomes minimal. The desire to answer these questions is one motivation for the workshop recommendation that the major operational centres provide their forecasts for evaluation. At the same time, those centres that produce consensus forecasts should make such forecasts available to other warning centres. A suggested approach was to use statistical weights in the combination of the forecasts, with these weights being based on the previous track record of individual models. This approach is similar to the “superensemble” application to tropical cyclone track forecasts, which was discussed in Topic 3.3.

One group expressed a concern that strict use of non-selective consensus risks rendering the forecaster “useless.” The forecaster’s value should be in his/her experience in interpreting the consensus as well as the individual forecasts. Training in the application of the consensus approach was therefore considered crucial.

4. Ensemble Forecasts

Ensemble forecasting of tropical cyclones is still in its infancy, as evidenced by the small number of researchers working in this area. A number of NWP centres are generating an ensemble of tropical cyclone track predictions as a byproduct of their global ensemble prediction systems. A number of issues remain unresolved: whether ensemble perturbations that were designed for use in global models or forecasting midlatitude systems are applicable to tropical cyclone prediction, and under what conditions the ensemble technique produces a better forecast than a single, higher resolution deterministic forecast. For the ensemble technique to become a viable alternative, such issues have to be dealt with through more research. In other words, more evaluation and validation are necessary before such a technique can be declared to be a useful operational alternative. Given such uncertainties, it is no surprise that warning centres have not made extensive use of ensemble forecasts of tropical cyclone tracks even when they are available. Techniques must be developed to enable the forecasters to utilize such forecast products as well as for them to evaluate the usefulness of ensemble forecasts. For forecasters to be able to use such information in operational forecasting, extensive training will be required, which is a strong recommendation from the workshop. Ensemble forecast products should also be made widely available for both forecasters and researchers to evaluate their utility.

Some discussion groups addressed the utility of the superensemble approach mentioned above. Despite the apparent improvement in the numerical track forecasts, a general question has been how the training sample for the superensemble is to be developed. This question is difficult because all models undergo continuous development. Given the potential usefulness, the workshop suggested that the superensemble approach be explored further.

5. Theoretical Advancements

As mentioned in the introduction and the Topic 3.4 Rapporteur report, very little research had been done in this area since IWTC-IV. The main focus has been on the application of the concept of potential vorticity tendency. Both theoretical and observational studies have demonstrated the utility of this tendency to indicate tropical cyclone movement. The question is whether such studies can be considered as conclusive. A general recommendation is that more research is necessary in this area, especially through in situ observations. However, the possibility of a field experiment was questioned because of the general lack of funding.




P. Harr (USA)

1. Introduction

During the IWTC-V plenary session on Tropical Cyclone Formation and Extratropical Transition, Rapporteurs presented summaries of progress on several aspects of this topic since IWTC-IV. The Rapporteur and Topic Chair reports are contained in WMO/TD No. 1136. In this Topic Chair Summary, additional issues that were identified during the plenary session and during individual Discussion Groups are summarized.

A common theme associated with progress on environmental influences, numerical prediction, and seasonal forecasts is the interactions among a variety of space and time scales that range from global-scale interannual and intraseasonal circulations, synoptic-scale features over each basin, and mesoscale factors related to individual cloud clusters. Also, a strong need for a consistent definition of tropical cyclone formation was identified such that operational priorities may be satisfied, and high-quality data sets will be defined based on physically relevant formation characteristics. Finally, several tropical cyclone warning centres have intentions to extend operations to five-day forecasts. Therefore, it is important to recognize that a tropical cyclone may form, move a long distance, and potentially undergo extratropical transition in a five-day period. Therefore, this topic encompasses a significant portion of the tropical cyclone life cycle during which there are many limits to understanding due to sparse data coverage and complex physical interactions that present difficult forecast challenges.

2. Formation Definitions

Overall, the need for a physically realistic and consistent definition of tropical cyclone formation was recognized. Because of a variety of operational constraints, and the need to provide appropriate warning for protection of life and property, there are several regional differences in the definition of tropical cyclone formation. These regional differences may cause some deviation from standardized physically-based definitions of tropical cyclone formation, because they have evolved to provide optimum utility to the respective tropical cyclone warning centre. A region-independent definition would be desirable as it would provide a consistency in several tropical cyclone-related data bases that are utilized for both policy and research. However, it would be extremely difficult to get all tropical cyclone warning centres to agree upon such a definition that would preserve the practices of each region that have evolved to provide an optimum warning of a tropical cyclone threat. However, it is the responsibility of each region to convey their unique specifications to the general tropical cyclone community such that a systematic assessment of several tropical cyclone formation-related factors may be documented. It is the responsibility of the tropical cyclone community to use the regional information to define the most consistent tropical cyclone data base possible such that accurate and consistent tropical cyclone characteristics are recorded. The post-process by which assessment of tropical cyclone formation is made should be one that is basin-independent.

The creation and maintenance of such a data set for specification of tropical cyclone formation characteristics would greatly benefit the research community for studies that range from seasonal prediction to synoptic-scale and mesoscale environmental influences on tropical cyclone formation. Furthermore, these data sets could be utilized to test theories of tropical cyclone formation in observation and modeling frameworks.

3. Observations in Addition to Satellites

Both forecasters and researchers recognize the critical need for observations of tropical cyclone formation from a variety of platforms over a wide range of space and time scales. However, it is generally recognized that global conventional observing systems will not expand and that several factors are contributing to a reduction in standard atmospheric and oceanographic observations. Therefore, systematic studies are needed of the impact to the prediction (both numerical and non-numerical) of tropical cyclone formation due to reduced observations worldwide. Furthermore, a strong need exists to insure that current observational capabilities are utilized fully. This includes investigation of the utility of various observation types and locations to numerical weather prediction model predictions of the large-scale tropical environment that influences tropical cyclone formation. Factors important to the process of tropical cyclone formation should be examined carefully such that appropriate observation requirements are identified to resolve the important characteristics such that improved understanding of the formation process is achieved.

4. Large-scale Control and Mesoscale Influences; Intraseasonal

It has become clear that the large-scale influence on tropical cyclone formation is dependent on interactions among a variety of space and time scales. Researchers stressed the need to define factors that would provide diagnostic evaluation of potential tropical cyclone formation, which should include the systematic examination of tropical cyclone formation in numerical prediction models. Furthermore, it is imperative that the examination address a measure of uncertainty in the ability for tropical cyclone formation to be related to large-scale factors in numerical model simulation. This may also be examined with ensemble techniques to assess relative roles of important environmental factors that may be associated with tropical cyclone formation. Furthermore, the increased use of re-analysis data to identify large-scale influences on tropical cyclone formation was advocated. However, caution is advised because of potential problems associated with data assimilation aspects of re-analyzed data sets.

While recognizing the importance of the above research, forecasters stressed that transition of research findings on large-scale factors in tropical cyclone formation to operational forecast centres has been slow. Significant training is required to move new research findings to the operational forecast process.

5. Prediction of Tropical Cyclone Formation with Numerical Models

In recent years, significant effort has been expended to examine factors associated with tropical cyclone formation through numerical simulation. Both researchers and forecasters recognized that collection of high-resolution (spatial and temporal) observations is critical to evaluate numerical results. However, questions were raised as to whether there has been adequate use of data sets from operational analyses and previous field programs. These sources should be exploited before additional specialized programs are initiated.

The ability of operational numerical models to forecast tropical cyclone formation must be systematically identified. Furthermore, the ability of a numerical model to forecast tropical cyclone formation must be represented in factors that have operational relevance, and should include the rates of false alarms, hits, and misses in terms of the forecasts of tropical cyclone formation at various time intervals.

The use of ensemble techniques associated with operational global numerical prediction models should be increased to identify areas where tropical cyclone formation is likely. Regional models should then be utilized to examine the timing of imminent tropical cyclone formation.

6. Seasonal Prediction

While seasonal prediction of tropical cyclone formation presents an interesting and challenging scientific problem, forecasters stressed that prediction of just the seasonal number of tropical cyclones in a particular basin may not have operational significance to the forecaster. However, these seasonal forecasts may have a high degree of significance for certain industries and the general public, because release of the seasonal forecasts increases public awareness of the tropical cyclone threat. To be more effective, education of the general public is needed on what the seasonal forecast means to them.

In the research community, debate continues as to what methodology provides the most promise of accurate seasonal prediction. Currently, statistical methods, dynamical models, and combinations of statistical methods and dynamical models are being used to produce seasonal forecasts. Statistical models require identification of specific factors that may be related in a predictive manner to seasonal formation characteristics. Researchers generally agreed that if there is hope for dynamical model-based seasonal predictions, the models must include a coupling of the atmosphere and ocean.

7. Extratropical Transition

Because of the severe impacts associated with expanded precipitation, wind, and wave fields during the transition of a tropical cyclone to an extratropical cyclone, researchers and forecasters both agreed that a need exists for a focused research program that examines the important physical characteristics associated with extratropical transition. This research program should promote analysis of existing observations, analyses, and forecast data sets. In particular, uncertainties associated with predictions of extratropical transition should be examined to identify specific physical characteristics that impact the numerical prediction of extratropical transition. Furthermore, the overall research program should be coordinated with existing programs (e.g., operational field programs, planned specialized field programs) to obtain detailed observations of the evolutionary structure of extratropical transition. Studies such as these would identify optimal observation types, locations, and densities to map the extratropical transition process.

Finally, it was recognized that the role of the ocean in extratropical transition may be highly variable and is not understood. A severe lack of observations of atmospheric and oceanic characteristics in the environment exists during extratropical transition events. More observations are needed to specify the sensitivity of the transition process to changes in the overall environment.




R. Falls (Australia)

1. Introduction

The purpose of this Topic 5 Chair report is to summarise additional issues raised at IWTC-V during the plenary and Discussion Group sessions that reviewed the Rapporteur and Topic Chair reports, and is primarily focused on the discussion without repeating material included in those reports, which are included in WMO/TD No. 1136. Expanded bibliographies provided with the Rapporteur reports should prove very useful. About 15 Australian disaster managers and researchers joined the Workshop for this topic.

Topic 5 plenary session also included two special presentations:

i) Dr. Pascal Peduzzi described a United Nations Environment Program (UNEP) project to define global risks, which has included the construction of a global tropical cyclone best-track data base and its use for tropical cyclone impact assessments. This data base provides an opportunity for a possible collaboration between WMO and UNEP to achieve global data bases recommended by IWTC-IV and IWTC-V.

ii) Dr. S. R. Kalsi of the RSMC New Delhi provided an interesting overview of the India Meteorological Department program to develop an early warning system.

The three sub-topics within Topic 5 are inter-laced, within the context of the "Total Warning System." As defined in Topic 5.1 by Linda Anderson-Berry, a Total Warning System is comprised of accurate and timely tropical cyclone forecasts and warnings, efficient communication mechanisms to transmit warning messages, and public education encouraging appropriate disaster preparedness and defensive actions. Such a Total Warning System is essential if citizens are to be empowered to reduce their vulnerability and build community resilience. That is, accurate information as to the threat must be effectively communicated, processed by all citizens, and acted upon in an appropriate manner to decrease the adverse human and economic impacts of tropical cyclones.

The IWTC-V discussions highlighted that meteorological services are increasingly concerning themselves with the end-to-end approach that characterises the Total Warning System. These services are finding that the socio-economic issues related to the response to warnings are a huge challenge. It was suggested that large gains in the effectiveness of warning systems are to be found in this area. Moreover, unresolved longer-term mitigation issues such as unsafe land use and poor (or no) building standards have placed large burdens on the effectiveness of the Total Warning System in attempting to reduce human and economic exposure. Unless mitigative approaches to human expansion or re-development in coastal areas are implemented, this aspect will continue to be the major shortcoming in the Total Warning System.

The Workshop re-endorsed with some modification the recommendations from IWTC-IV concerning improved best tracks and socio-economic data base. While a re-analysis of the best-track data has been accomplished for Atlantic hurricanes, and is being discussed for the western North Pacific, a comprehensive costs and benefits assessment has not been done for any basin. A number of new recommendations were also formulated (see Chapter 2).

2. Societal Impacts

Societal impacts of tropical cyclones tend to be far greater in the developing countries where the capacity for responding to the warnings and the infrastructure recovery capability is more limited. The great societal vulnerability existing in small island countries where warning communications can be tenuous was dramatically illustrated in Linda Anderson-Berry's presentation. She also highlighted the limited penetration of public education initiatives in countries such as Australia. On Australia's east coast, a cyclical downturn in severe tropical cyclone impacts over the past 25 years is a particular concern for public awareness because of the large socio-economic expansion and lack of cyclone experience of people moving to the coastal area.

Discussions highlighted a requirement for collaborations of social scientists and economists with the cyclone warning centres to accomplish the extension components of the Total Warning System and to address the multi-disciplinary approaches to cyclone risk management that have emerged over time. This requirement was specifically linked to the need for a comprehensive, on-going effort in public information and education activities, and activities with the media and disaster prevention agencies as outlined by Max Mayfield (USA). In some countries, legal issues may arise in the future if damage occurs because the public information program has been relaxed, or public officials have failed in their responsibilities during tropical cyclone events.

3. Economic Impacts

Whereas insurance data documented that direct economic losses from tropical cyclones are far greater in the developed countries, the human toll is much larger in the developing countries. With the growth of economic losses has emerged requirements for improved modelling of economic impacts and assessments of the overall societal vulnerability. These needs lead to a requirement for suitable data bases. Considerable discussion occurred on parameters for inclusion in best-tracks both here and under Topic 1. The parameters previously recommended in the Global Guide for Tropical Cyclone Forecasting are considered suitable with the addition of eye diameter deduced from satellite imagery that may be considered a proxy for the radius of maximum wind, which is normally only available from aircraft reconnaissance. Although a format recommended by WMO has been in existence for some time, blanks are often left in the structure parameters provided by operational centres. Even crude estimates are much better than nothing, provided the quality of the estimate is also indicated. Remotely sensed data such as scatterometer winds and microwave imagery provide a means to improve best-track data now and in the future, even though the coverage of these data may not yet be optimal.

Some weather services noted a tension between the economic costs of over-warning and the warning lead-times needed to ensure human safety. For instance, an economic shut-down in Hong Kong with estimated costs of US $0.5B per day follows the issuance of a cyclone warning, and the city is able (perhaps uniquely) to cope with a short warning lead-time of less than 24 h. Rex Falls provided a discussion paper and schematic that characterised costs and losses from a tropical cyclone scenario as a function of warning lead-times. He suggested that the lead-time could be optimised to minimise the economic impact. Careful attention needs to be given to include time for preparations necessary to minimise the human toll.

4. Warning Systems

Discussions that led to the recommendations in Chapter 2 covered the need to advance both the technical quality of warnings and the socio-economic aspects of responses to warnings. On the science side, improvements are still needed in track predictions, particularly approaching landfall, and intensity prediction. Even relatively small deviations in the track forecast can have huge socio-economic ramifications near sensitive areas such as large cities, populous fertile river deltas, or islands.

Forecaster issues included the need for communication and media training. Support from the WMO Public Weather Services Program was noted with appreciation, and should be continued. Importance was placed on developing partnerships with the media and disaster prevention agencies. What happens in the period between cyclone seasons and during the pre-season period can be as important as what happens during the season.

Improved means of presentation of the warnings via the Internet were demonstrated. One challenge is the depiction of uncertainty at all warning time periods, and especially in medium-range forecasts out to 72 hours that have become possible in some basins with recent improvements in track prediction. With the growth in ensemble forecasting techniques, the appropriate interpretation of bifurcation track cases and the ensemble "spread" is one of the challenges.

The critical problem of time management for the forecaster to answer external enquiries versus spending the time in warning preparation still exists in many centres with limited human resources. These competing time demands also highlight a continuing need for streamlined access to information and tools for warning preparation through systems such as the Automated Tropical Cyclone Forecast workstation.

There was healthy discussion on the Internet as a means of dissemination. Although the Internet provides opportunities for new warning products, the public access to forecasts from a variety of sources on the Internet could possibly lead to confusion. The view was expressed that attempts to suppress such forecasts may suppress potentially useful forecast products from the research community. Discussions suggested the best solution lay in education of the public and emergency managers in the difference between official warnings and unofficial forecasts (which may involve interpretation of one particular model), and by exploring all available means of disseminating official warnings. It was also suggested that unofficial sources could include a link to official warnings such as the experimental WMO National Warnings page.

Warning centres in most countries have become increasingly reliant on the Internet to receive data, particularly the expanding processed data sets from satellites. However, the Internet can become clogged in some countries during tropical cyclone events. There was a caution against becoming too reliant on the Internet, and the need to continue to strengthen other means of communications was emphasised.

5. Concluding Remarks

Warning systems, and their rightfully perceived socio-economic value, are one of the central drivers for tropical cyclone science. The "Impacts" topic brings welcome inter-disciplinary perspectives to IWTC that could be further developed in future meetings. For example, building standards and land use have emerged as very big issues for the Total Warning System. Improving understanding of the mechanisms through which warnings are more or less successful in prompting the appropriate response and reducing impacts, and better quantification of the socio-economic value of warnings, remain fertile areas for study.

Special thanks go to the Topic 5 Rapporteurs Jim Weyman and Linda Anderson-Berry, Rick Murnane, and T.C. Lee for making the Session so relevant and successful.


December 2-12, 2002

0800 Registration
0830 Opening Ceremony
0930 Morning Tea
Forecast office present uses -- Jeff Hawkins, Session Chair
1000 Jeff Callaghan, Australia Bureau of Meteorology
Forecast office present uses
1020 Alipate Waqaicelua, Fiji Meteorological Service
Forecast office present uses
1040 Philippe Caroff, La Reunion, France
Forecast Office Use at La Reunion
1100 Lixion Avila, U. S. National Hurricane Centre
Satellite applications for TC analysis and forecasting at TPC/NHC
1120 Rich Jeffries, Joint Typhoon Warning Centre, USA
Satellite applications at the Joint Typhoon Warning Centre
1140 Discussion
1215 Lunch

0.2 Imminent uses and data assimilation -- Chris Velden, Session
1315 Tetsuo Nakazawa, Japan Meteorological Research Institute
1335 Roger Edson, University of Guam
Use of scatterometer data in tropical cyclone forecasting and
1355 Kurt Brueske, U. S. Air Force Academy
AMSU applications
1415 Jim Kossin, University of Wisconsin
Geostationary IR
1435 John LeMarshall, Australia Bureau of Meteorology Research Centre
Tropical cyclone forecasts using high-resolution satellite data and
1455 Discussion
1515 Afternoon Tea

0.3 Future satellite instruments and opportunities
Chris Velden,
Session Chair
1530 Jeff Hawkins, Naval Research Laboratory, Monterey and
Chris Velden, University of Wisconsin, CIMSS
1630 Discussion
1730 End of session
1800 Ice breaker

[NOTE: Session 0.4 is Wednesday night]

Plenary Session -- Masashi Nagata, Session Chair
0815 Edwin S. T. Lai, Hong Kong Observatory
1.6 TC structure analysis and forecasting techniques
0845 Frank Marks, Hurricane Research Division
1.1 Observations
0915 Elizabeth Ritchie-Tyo, University of New Mexico
1.2 Environmental effects
0945 Roger Smith, University of Munich
1.3 Role of convective processes
1015 Morning Tea
1030 Lynn Shay, University of Miami
1.4 Air-ocean interface and boundary processes
1100 Shuyi Chen, University of Miami
Naomi Surgi, National Centers for Environmental Prediction
1.5 Numerical model guidance
1130 Russell Elsberry, Naval Postgraduate School
1.0 Tropical cyclone structure and structure change
1150 Discussion
1215 Lunch
1315 Discussion Groups – Topic 1
1515 Afternoon Tea
1530 Special Focus 1a: Coupled Boundary Layer Air-Sea Transfer (CBLAST)
Simon Chang, Office of Naval Research
Three minute presentations by the following presentors:
Greg Holland
Nick Shay
Isaac Ginis
Kevin Tory
Mike Banner
Jeff Kepert
Yuqing Wang
Eric Uhlhorn
Ed Walsh
Shuyi Chen
1530 Special Focus 1b: Tutorial on QUIKSCAT
Roger Edson, University of Guam
1730 End of session


Plenary Session 2a – Q-u-Z Chaudhry, Session Chair
0815 Chen Lianshou, China Academy of Meteorological Sciences
2.1 Observing and forecasting rainfall
0845 Woo-Jin Lee, Korean Meteorological Administration
2.2 Data assimilation and numerical prediction of precipitation
0915 Sun-Fatt Chong, Malaysia Hydrology Division
2.3 Hydrological models of precipitation
0945 Jeff Kepert, Australian Bureau of Meteorology Research Centre
2.4 Coastal and inland wind structure changes
1015 Morning Tea
Plenary Session 2b - S. R. Kalsi, Session Chair
1030 Shishir Dube, Indian Institute of Technology
2.5 Storm surge
1100 Bill Frank, Pennsylvania State University
(Greg Holland,
Aerosonde Corporation)
2.0 Tropical cyclone landfall processes
1130 Discussion
1215 Lunch
1315 Discussion Groups – Topic 2
1515 Afternoon Tea
1530 Special Focus 2a: Tutorial on Advanced Microwave Sounding Unit (AMSU)
Kurt Brueske, U. S. Air Force Academy
Chris Velden, University of Wisconsin CIMSS
1630: Special Focus 2b: Tutorial on Advanced Objective Dvorak Technique
Jim Kossin, University of Wisconsin, CIMSS
Chris Velden, University of Wisconsin, CIMSS
1730 End of session
Special Focus 0.4: Satellite Data Assimilation Plans at Selected Forecast Centers
John LeMarshall, Session Chair
1930 Hideo Tada, Japan Meteorological Agency
0.4a Data assimilation at JMA
1945 David Richardson, European Centre for Medium-range Forecasts
0.4b Data assimilation at ECMWF
2000 Naomi Surgi, National Centers for Environmental Prediction
0.4c Data assimilation at NCEP
2015 Julian Heming, United Kingdom Meteorological Office
0.4d Data assimilation at UKMO
2030 Xue Jishan, China Meteorological Agency
0.4e Data assimilation at CMA
2045 Jim Goerss, Naval Research Laboratory, Monterey
0.4f Data assimilation at FNMOC
  1. Antoine Lasserre-Biggory, Meteo-France
0.4g Data Assimilation at Meteo-France
2115 Discussion
2130 End of session



Plenary session – Hung Kwan Lam, Session Chair

0815 Noel Davidson, Australia Bureau of Meteorology Research Centre
3.1 Numerical and statistical model guidance and improvements
0845 Rich Jeffries, Joint Typhoon Warning Center
3.2 Consensus approach to track forecasting
0915 Lance Leslie, University of Oklahoma (David Richardson, ECMWF)
3.3 Ensemble prediction system
0945 Johnny Chan, City University of Hong Kong
3.4 Theoretical advancements
1015 Morning Tea
1030 Johnny Chan, City University of Hong Kong
3.0 Tropical Cyclone Motion
1100 Discussion
1215 Lunch
1315 Discussion Groups – Topic 3
1515 Afternoon Tea
1530 Special Focus 3a: U. S. Weather Research Project Hurricane Landfall
(2004) Collaborations
Russell Elsberry, Naval Postgraduate School
Frank Marks, Hurricane Research Division
1530 Special Focus 3b: Tutorial on Track Consensus Forecasting and
Systematic Approach Forecasting Aid (SAFA)

Rich Jeffries, Joint Typhoon Warning Center
1730 End of session


Plenary Session – Greg Holland, Session Chair
0815 Mark Lander, University of Guam
4.1 Formation definitions
0845 Jeff Callaghan, Australia Bureau of Meteorology
4.2 Observations in addition to satellites
0915 Jenni Evans, Pennsylvania State University
4.3 Large-scale control and mesoscale influences; Intraseasonal
0945 Mike Montgomery, Colorado State University
4.4 Predictions of formation with numerical models
1015 Morning Tea
1030 Bin Wang, University of Hawaii
4.5 Seasonal formation predictions
1100 Jim Abraham, Meteorological Service of Canada
4.6 Extratropical transition
1130 Patrick Harr, Naval Postgraduate School
4.0 Tropical cyclone formation and extratropical transition
1150 Discussion
1215 Lunch
1315 Discussion Groups – Topic 4
1515 Afternoon Tea
1530 Special Focus Topic 4: Global Climate Change and Tropical Cyclones
Basis for an Updated Statement
1530 Bill Gray, Colorado State University
Observational studies of tropical cyclones and climate variability
1545 Tom Knutson, Geophysical Fluid Dynamics Laboratory
Regional model simulations of impact of climate change on tropical cyclone intensity and precipitation
1557 Greg Holland, Aerosonde Corporation
Maximum Potential Intensity theory and its application to the climate change and tropical cyclone problem
1609 Jun Yoshimura, Frontier Research System for Global Change
GCM simulations of tropical storm frequency – is it realistic? – and response to CO2 doubling
1621 Rick Murnane, Risk Prediction Institute
Progress in paleotempestology; Tropical cyclone contributions to meridional heat transport
1633 John Persing, Colorado State University
Hurricane superintensity
1645 Discussion – Is an updated statement appropriate at this time? If so, how to proceed?
1710 Discussion—Recommendations for future research – What are the highest priorities?
1730 End of session



Plenary session – Max Mayfield, Session Chair

0815 Linda Anderson-Berry, James Cook University
5.1 Societal impacts
0855 Richard Murnane, Risk Prediction Initiative (Bermuda)
5.2 Economic analysis
0935 T. C. Lee, Hong Kong Observatory
5.3 Effective warnings
1015 Morning Tea

  1. Pascal Peduzzi, United Nations Environmental Programme
    Computation of global risk, and identification of
vulnerability factors
  1. S. R. Kalsi, RSMC New Delhi Warning Centre
Early warning considerations at India Meteorological Department
1100 Rex Falls, Australia Bureau of Meteorology (retired)
5.0 Tropical cyclone impacts
1120 Discussion
1215 Lunch
1315 Discussion Groups – Topic 5
1515 Afternoon Tea
1530 Special Focus 5a: Australian Tropical Cyclone Coastal Impacts Program

Jim Davidson, Australia Bureau of Meteorology
1530 Special Focus 5b: Pennsylvania State Cyclone Characterization Diagram
Jenni Evans, Pennsylvania State University
  1. Special Focus 5c: Tropical Cyclone Formation in Relation to Tropical
Wave Dynamics
John McBride, Australia Bureau of Meteorology
Research Centre
1730 End of session

Evening: Cairns Zoo Night Dinner


0900 Plenary session
Presentation of recommendations
1030 Morning Tea
1045 Discussion of recommendations
1215 Lunch
1315 Discussion of recommendations
1515 Afternoon Tea
1530 Special Writing Group begins to prepare final report
1530 Special Focus 6a: Microwave imagery interpretation for tropical
cyclone analysis

Jeff Hawkins, Naval Research Laboratory-Monterey
1630 Special Focus 6b: Synoptic ocean-atmosphere measurements during
Isadore and Lili

Nick Shay, University of Miami


0900 Plenary session
Discussion of final report
1015 Morning Tea
1030 Discussion of final report
1145 Lunch
1330 Special Writing Group prepares final report



T. Knutson (USA)

A special focus session was held at IWTC-V on Global Climate Change and Tropical Cyclones. Since the assessment of this topic by Henderson-Sellers et al. (1998), a number of studies have provided relevant new information. This research was reviewed in the session, and the question of whether a new assessment should be undertaken was discussed. During the discussion time, some recommendations for future research efforts in this topic area were advocated.

1. Summary of Presentations

During the presentation part of the session, six talks were given on the topic of Global Climate Change and Tropical Cyclones. The first talk by Bill Gray of Colorado State University (USA) presented a skeptical view of the ability of climate models to address the issue of climate change in general. Gray argued that the climate system is too complicated to be adequately modeled by climate models at this time, and that the response of the climate system to increased carbon-dioxide (CO2) would be small due to the negative water vapor feedback mechanism proposed by Richard Lindzen.

Tom Knutson (USA) of the Geophysical Fluid Dynamics Laboratory (GFDL) presented results of regional nested model experiments using the GFDL hurricane model along with CO2-induced changes in environmental boundary conditions taken from a climate change experiment with the GFDL global climate model. According to those experiments, the simulated hurricanes were about 5-10% more intense and had 16-28% more rainfall than storms for present-day conditions.

Greg Holland (Australia) reviewed some climate change results using his Maximum Potential Intensity (MPI) theory to predict the potential change in tropical cyclone intensity due to changes in the thermodynamic state under 2xCO2 conditions. He also showed some new results from work by Kendall McGuffie applying the MPI framework to interannual and seasonal climate variations from a number of global climate model experiments –- with specified sea-surface temperatures (SSTs)-- from a model intercomparison project.

Jun Yoshimura of Japan’s Frontier System for Global Change Research presented an overview of tropical storm simulation capabilities of global climate models. Results from several models indicated that experiments in which a grid spacing as fine as T106 (about 1 deg. lat.) is used simulated fewer tropical storms for high CO2 conditions versus present-day conditions. Using lower resolution models gave much more mixed results of either increases or decreases in frequency.

Rick Murnane of the Risk Prediction Initiative (Bermuda) gave an overview of recent work in paleotempestology, in which attempts are made to reconstruct the record of pre-historical hurricane strikes based on geological proxy data such as overwash sediments in near-coastal lakes. Other related work has included examination of ancient historical records of typhoon activity from China. Murnane also briefly reviewed recent work by Emanuel that argues that tropical cyclones are responsible for a substantial part of the meridional ocean heat transport through the vigorous upper-ocean mixing they induce.

The final speaker was John Persing of Colorado State University who showed some new results on hurricane “superintensity” in which intensities substantially exceeding those predicted by Emanuel’s MPI theory are simulated using a high resolution axisymmetric model.

2. Summary of Discussions

During the discussion period, a number of questions were raised or opinions given concerning the research results presented. One question was whether the paleotempestology records were reliable recorders of hurricane intensity due to effects of tidal variations, etc. A participant questioned whether there was any evidence in observed records for trends in tropical moist instability measures and referred to the Sun and Held study that showed more vertical coherence of water vapor interannual variations in global models than in observations. It was noted that the Sun and Held results had been questioned due to their dependence on a lack of radiosonde coverage over much of the tropical oceans. Another participant questioned whether the GFDL hurricane model intensity inferences from the climate change experiments were credible since the model had essentially no skill in operational intensity prediction. Another participant noted a strong dependence of simulated tropical storm frequency in global models on the physical parameterizations used. Another participant advocated that since the GFDL climate model and hurricane model were both “highly tuned,” many experiments should be conducted with different parameter settings to test the robustness of the results. In response, Tom Knutson noted that some new climate change-related experiments with the GFDL hurricane model were underway using different convective parameterizations. Another participant expressed reservations about drawing conclusions about future tropical cyclone intensity or precipitation changes since the results at this time were too reliant on a single model -- the GFDL hurricane model. A participant argued that Emanuel’s estimates of tropical cyclone contributions to meridional ocean heat transport were greatly overestimated. One participant noted that El Nino had a tremendous impact on Pacific island nations and that there was evidence that El Nino activity would become more intense with global warming. Jenni Evans presented some results from a climate model experiment with the NCAR CSM1 model that supported the Henderson-Sellers et al. conclusion that as CO2-induced warming occurs the regions of deep tropical convection do not expand along with the expansion of the 26C isotherm. In response to a comment concerning the IPCC 2001 report’s treatment of the tropical cyclone/climate change issue, a brief summary of the IPCC’s summary conclusions on the topic was shown.

3. Discussion and Recommendations for Next Steps

The question put to the attendees was whether it was now appropriate for the tropical cyclone community to develop and publish a new statement on tropical cyclones and climate change. The responses and opinions expressed were mixed. Some members of the audience voiced agreement with Bill Gray that the Henderson-Sellers et al. (1998) assessment was not conservative enough with regard to likely future tropical cyclone intensity changes and that no estimate of future changes could be given. Greg Holland stated that the tropical cyclone community should make periodic public statements on the issue, or the “void” would be filled by less informed statements from outside the tropical cyclone community. He questioned whether evidence exists for a decrease in tropical cyclone intensities for high CO2 conditions. John McBride stated that insufficient time was available to address these issues and concerns at the present workshop, and suggested that a group of experts be formed to examine the issues and report back to the community. That group could recommend when it was appropriate to issue a new statement and could work to formulate any new statement. Greg Holland proposed that McBride could lead such a group as part of his duties in the Tropical Meteorology Research Program Working Group. The session concluded with the recommendation that McBride form such a group as the next step.



Russell L. Elsberry Chris Velden
Naval Postgraduate School Cooperative Institute for
Meteorological Satellite Studies


Taking advantage of the presence of representatives from 36 tropical cyclone forecast centres at the fifth International Workshop on Tropical Cyclones (IWTC-V), a survey was constructed to document present uses of satellite data in the forecast process. Where the centre did not have access, information was requested on what would be necessary for that centre to gain access and utilize the data. Because of the great increase in utilizing the Internet websites for accessing satellite data, the degree of access and any concerns or issues with Internet access were requested. Based on discussions at IWTC-V, training (or lack thereof) is a major issue at almost all centres. Thus, the training requirements and desirable attributes of the training in the use of satellite data were included as one section of the survey.

1. Introduction

The Keynote session at the fifth International Workshop on Tropical Cyclones (IWTC-V) was on the present uses, imminent products, and future opportunities for using satellite data for tropical cyclone forecasting and research. The IWTC-V is the fifth in a series of World Meteorological Organization (WMO)-sponsored workshops that bring together tropical cyclone forecasters and researchers each four years. The primary objectives are to assess the progress made during the past four years and to plan for the future. Because of the importance of satellite data for tropical cyclone forecasting, and an impending huge increase in the satellite data streams, it was natural to have satellites as the theme for the Keynote session. The session was primarily organized by Chris Velden of the Cooperative Institute for Meteorological Satellite Studies (CIMSS) with assistance from John LeMarshall of the Australia Bureau of Meteorology Research Centre.

The first segment of the Keynote session was five invited talks from tropical cyclone forecasters to provide background on the present uses of satellite data. These talks were from centres that might be classified as being relatively well-equipped to being very well-equipped in terms of satellite observations and products. In the latter case, an issue was that they had more satellite data than their processing and display systems could present in a timely and forecaster-friendly manner to extract optimum benefits!

It was known from the beginning that the centres represented by the five invited talks were not representative of the global uses and needs for satellite data. However, the approximately 45 forecaster participants at IWTC-V represented 36 warning centres. Most of these forecasters were sponsored by the WMO and were selected to represent the five WMO regions. Thus, it was decided to prepare a survey to explore how these 36 warning centres used satellite data. Having heard the five invited talks, these forecasters were also asked to identify deficiencies in their satellite data acquisition and what would be required for them to gain access and use the present satellite data.

A major change in the past four years has been the availability on Internet websites of satellite imagery and digital data geo-referenced to the tropical cyclone position. Two of the best-known USA websites are at the CIMSS and the Naval Research Laboratory-Monterey (NRL). It has become evident that these websites are being accessed by many tropical cyclone warning centres. Thus, a second segment of the survey explored the degree of website satellite data acquisition among the centres and what was most useful. Another question explored possible issues or concerns of the centres about their dependency on such websites over which they have no control, and which may be unreliable or even withdrawn.

Another key issue for most of the forecast centres represented at IWTC-V is the need for training, and this especially applies for new satellite products. The third segment of the survey explores the training needs of the warning centres. Various options (local, national, or international) are available. Some training-related issues include the cost (in time and money) for the trainer to come to multiple locations or the trainees to come to a central training site. In a small forecast office, the absence of one forecaster for training will cause an overload on other forecasters.

2. Survey responses on satellite data availability

The warning centres who completed the survey are listed in Table A-1. A total of 31 centres returned the survey out of 36 distributed. Australia, China, and the USA have multiple tropical cyclone warning centres. The representative from Barbados indicated that her survey responses are probably characteristic for other Caribbean forecast offices: Antigua, Jamaica, St. Lucia, Grenada, Trinidad/Tobago, and Guyana. Input from the Taiwan office was received via a Pacific Science Association (PSA) participant.

a. Basic geostationary imagery

The first question explored the availability of visible, infrared, and water vapor imagery from an appropriate geostationary meteorological satellite. Such imagery is essential for tropical cyclone positioning, structure analysis, cyclogenesis, and for synoptic assessments in the environment of the tropical cyclone. One pleasing result is that all centres reported access to the geostationary imagery, although some small centres indicated the imagery is lower resolution than desirable. While these positive responses are encouraging, universal availability at tropical cyclone warning centres may not be true if the IWTC-V participants responding to the survey are not representative.

The survey also explored the acquisition and use of the Objective Dvorak Technique (ODT) for estimating tropical cyclone intensity. This technique, which was developed by CIMMS and the Cooperative Institute for Research in the Atmosphere (CIRA), works best for cyclones with well-defined eyes. Two centres in China and Cuba indicated they have a national system similar to the ODT. Although the ODT has been available since 1998, only nine centres indicated that the ODT or national version has been utilized. Three warning centres indicated that they have the technique and would be more interested in the ODT if it was more accurate for the weaker tropical cyclones. An Advanced ODT (AODT) and a multivariate linear regression technique that are being developed by CIMMS were presented in a Special Focus session at the IWTC-V. The AODT is being tested at the USA National Hurricane Center and Joint Typhoon Warning Center, and appears to perform better than the ODT for the Tropical Storm and Tropical Depression stages. If the AODT and regression technique do well in pre-operational testing, they would complement the manual Dvorak-type intensity estimates that nearly all warning centres use. Even small warning centres with PC-type computers could apply these two techniques.

b. Polar-orbiting satellite imagery

Nearly all (29 of 31) warning centres have access to the polar-orbiting visible, infrared, and water vapor data. As might be expected, the centres in developed countries have receivers for direct receipt of the satellite data. It might not have been expected that several smaller countries (Bangladesh, Fiji, Madagascar, and Sri Lanka) also have their own receivers for direct acquisition. The remainder of the warning centres receive the polar-orbiter satellite data via Internet website access, and the two negative responses were from centres that did not know of the websites. Six of the centres indicated they had the data but did not use it, mainly because of the availability of higher temporal resolution geostationary imagery. Lack of training was given as one reason this imagery was not used. Other reasons cited include that the imagery was not timely (communication problem ?) and it was irregular. Thus, the polar-orbiter data appear to be used, although in many cases as a back-up to the geostationary imagery.

c. Microwave imagery and soundings

The microwave imagery and microwave soundings (e.g., Advanced Microwave Sounding Unit, AMSU) availability was explored in two questions but are combined here as the answers were generally common. Whereas 21 of 31 centres indicate they have access, eight centres indicated that the microwave information was not used. Lack of training was cited as a primary reason it was not used. However, many centres are just becoming aware of the advantages of microwave imagery for detecting the tropical cyclone centre during early stages when the cirrus shield obscures the centre. The interest of IWTC-V participants in the microwave observations and their interpretation was indicated by their attendance at a voluntary Special Focus session on this topic. The availability of the microwave imagery geo-referenced to the tropical cyclone centre on the NRL-Monterey website appears to be one of the popular uses of the Internet. Whereas this experimental website was sometimes not available due to weekend maintenance problems, a fully maintained (seven days, 24 hours) public website at the Fleet Numerical Meteorology and Oceanography Centre in Monterey, California will eliminate these outage problems. The microwave imagery is also useful for precipitation estimation, although only a few warning centres have taken advantage of this aspect. One new product from the USA Hurricane Research Division that may increase usage is a tropical cyclone intensity-dependent rain-rate climatology and persistence (R-CLIPER) aid. This R-CLIPER will also serve as a skill measure for other tropical cyclone precipitation prediction techniques.

Use of microwave sounding data to detect the tropical cyclone warm core and infer the corresponding minimum central pressure (intensity) has been attempted for two decades. The new Advanced Microwave Sounding Unit (AMSU) has more promise of achieving this objective because of its better horizontal resolution (~ 25km at nadir). Both a Keynote presentation and a Special Focus session at IWTC-V raised awareness of the capabilities and processing procedures of the AMSU. At least three algorithms that relate AMSU data to tropical cyclone intensity are being evaluated. Many forecasters indicated that this was their first exposure to AMSU and other imminent satellite products. Consequently, the survey revealed many warning centres would like to get access to the AMSU-based intensity estimates.

d. TRMM Precipitation Radar

The Tropical Rainfall Measurement Mission (TRMM) carries an active Precipitation Radar-PR (in addition to the passive microwave instrument-TMI). This instrument was only intended for monthly precipitation totals, so provision of the PR data in near-real time has required some effort and time. The PR gives rain-rates at multiple levels with very high horizontal resolution. The disadvantage is that the PR has a narrow swath and is on a polar-orbiting satellite so that the tropical cyclone is missed on many orbits. Only eight warning centres indicated real-time access to the PR data and/or use in the forecast process. Where available, the primary use was in estimating the precipitation distribution relative to the tropical cyclone centre as it approaches land. Eleven centres indicated that they had access, but did not use the product. Although this may be due to the lack of timeliness and limited tropical cyclone coverage, several centres indicated a need for training. Twelve centres indicated that they did not have access.

e. Scatterometer data

Although scatterometer-based estimates of surface wind vectors have been available from the short-lived SEASAT and NSCAT, and from the European Research Satellites (ERS-1 and ERS-2), the applicability near tropical cyclones has been limited. The broader swath of the Quikscat has led to many more interceptions of tropical cyclones. Both the two Keynote presentations and a Special Focus session at IWTC-V on the Quikscat data and its interpretation generated great interest among forecasters. In many cases of tropical cyclone formations over remote ocean areas, the Quikscat gives the first evidence of the surface wind circulation centre, and this sometimes results in major relocations of the centre. The Quikscat surface wind distributions have also improved the analysis of the outer vortex structure (e.g., 35 kt wind radius), which is useful for estimating ocean surface wave generation, and wind structure changes during extratropical transition of a tropical cyclone. A variety of opinions are expressed as to the maximum wind speeds in a tropical cyclone that can be reliably measured by the Quikscat. Questions also arise as to how to interpret the rain-flagged wind estimates and the various wind direction ambiguity solutions. These Quikscat data are available from the NESDIS website.

Twenty-five of the 31 warning centres responding to the survey indicated that Quikscat data are acquired and are an important data source for their tropical cyclone wind analysis (and general marine wind analysis). Almost all of the other six centres indicated a desire to gain access to Quikscat data and a need for training in its interpretation. Because of concerns that future satellites may not carry scatterometers, an IWTC-V recommendation is that all options be explored to continue and expand the scatterometer data because of their applicability for tropical cyclone position and structure analysis.

f. Radar altimeter

The IWTC-V presentation on the air-sea influences on tropical cyclone structure change was the first introduction for many of the participants to the use of radar altimeter data to infer the ocean heat content, which is a better measure than the sea-surface temperature alone. In addition, the radar altimeter provides an estimate of the ocean surface wave heights, albeit along the satellite track nadir.

Only three of the tropical cyclone warning centres indicated in the survey that they have used the radar altimeter data. Five centres indicated they had access, but it was not used for ocean heat content analysis. Twenty-three centres indicated they did not have access. Some of the responses indicated they were not convinced that ocean heat content was a concern for the SST decrease associated with tropical cyclones because of the deep ocean mixed layers in their region. Limited interest was expressed in the ocean surface wave estimates from the radar altimeter because of the rare cases that the satellite will pass directly over the tropical cyclone.

g. Infrared or microwave rainfall estimates

With growing recognition of tropical cyclone-related precipitation damages, various over-ocean rain estimates have been developed based on geostationary infrared or polar-orbiting microwave data, or a combination of both. These techniques are generally calibrated with radar estimates or gage measurements.

The survey responses indicate that eighteen centres have access to these satellite-based rainfall estimates. Some of those centres that have access and use the estimates indicate that further evaluation is needed that will demonstrate their accuracy and robustness. The seven “not used” responses were generally that the accuracy was not sufficient. Other centres indicated some interest in the techniques if they prove to be accurate, and if training was available. Two centres who forecast for mountainous islands suggested that such open-ocean precipitation estimates are not expected to be that useful for them.

3. Survey responses on Internet uses

As indicated in the survey responses in Sections 2b-2f above, many warning centres do not have direct access to polar-orbiting satellite data, and especially the specialized instruments such as AMSU, TRMM, Quikscat, and radar altimeter. Consequently, their access has only been via the Internet websites. Thus, these websites have been widely used, which is a major change in the operational forecast process over the past four years.

In response to a survey question on Internet access of satellite imagery by the warning centres, 29 out of 31 centres answered positively. By far the most common satellite data acquisitions mentioned were the Quikscat (22) and visible, infrared, and microwave imagery (19). Other data sets mentioned were TRMM (6) and water vapor winds (6). This short list should not be considered exclusive as several positive responses did not list specific systems.

Other survey questions requested information on Internet access of satellite products, and specifically those CIMSS products that are most useful. Of the 30 responses, 22 centres indicated that they did access the CIMSS or other websites for satellite products. Nearly all of the eight negative responses indicated a desire for gaining access. In several cases, the warning centre had not been aware of the Internet availability and would make inquiries for future use. Several of these centres indicated they would need training in the use of these satellite-based products, and this will be discussed further in Section 4.

As these experimental websites providing satellite-based products were being developed, some operational forecast centres have been hesitant to use these websites due to lack of control over the source, and especially possible interruptions or terminations of the source. Being an experimental product, concerns were also expressed about lack of timeliness and poor reliability of communications. Consequently, one of the survey questions explored these issues and concerns. Of the 22 warning centres that indicated use of the websites for acquiring satellite-based products, only four stated no concerns. Twelve centres listed one or more of the above concerns. Four centres indicated their communications with the websites became unreliable during tropical cyclone events, and one of these centres suggested an international “mirror website” to ensure access when the local website is saturated. Lack of timeliness or availability, especially due to lack of maintenance on weekends, was cited as a concern by four centres. Fear of termination was cited by two centres, and one centre indicated for this reason such website products would have to be only an auxiliary tool. Finally, four centres admitted concerns, but that the products were such an important part of their analysis that they were dependent on them, and the benefits outweigh the risks.

4. Survey responses related to training

Discussions at the IWTC-V revealed widespread concerns about the training of forecasters to effectively use the satellite observations and products. Consequently, the third segment of the survey addressed these concerns.

Satellite training is often done locally under the direction of a mentor. In some countries, a national training program is organized for basic training needs. International training in tropical cyclone analysis with some satellite training is available in WMO-sponsored programs for Northern Hemisphere (generally at USA National Hurricane Center) and Southern Hemisphere (generally in Australia).

Only the Guam office of the USA National Weather Service and the New Zealand Meteorological Service indicated no need for additional training. All of the other centres that responded to the survey indicated need for additional training, with the needs ranging from instruction in the basic Dvorak-type satellite interpretation to training in the new products on the Internet websites (see Section 3 above). The most common training requirement mentioned in the survey was for using the microwave imagery and soundings.

Perhaps the most surprising response is the training concerns are not just in warning centres in small countries – major centres such as in Australia indicate a need. The RSMC-La Reunion indicates this is a “permanent issue” due to the arrival of forecasters who are not trained in tropical cyclone analysis. The Joint Typhoon Warning Centre in Hawaii has a continual training requirement because of the three-year rotation cycle of the satellite analysts and forecasters. Other centres that do not have full-time tropical cyclone forecasters cite a need for “refresher” training, especially when the cyclones are relatively rare or are highly seasonal. In a small centre, travel to a training session by a forecaster may create a gap that has to be filled by other forecasters.

The most frequently mentioned obstacle for training is the cost to travel to the training site. Some centres expressed the opinion that it is more cost-effective for the satellite expert to travel, than for the forecasters to come to the expert’s site. However, the time the expert has to devote to training activity may be limited and travel to a number of warning centres can require a lot of time. For example, researchers developing new satellite-based products are generally not being paid to do training, and special arrangements must be made. One suggestion was to record the training in the form of a tutorial and make it available at a website.

Table A-1
Tropical cyclone warning centres that responded to the survey

Australia Brisbane Darwin Perth France La Reunion Philippines
Bangladesh Hong Kong Solomon Islands
Barbados (represents eastern Caribbean) Japan Sri Lanka
Canada Korea Taiwan*
China Beijing Guangzhou Macao Thailand
Cook Islands Madagascar United States Guam JTWC Miami
Costa Rica Mexico Vanuatu
Cuba New Zealand Vietnam
Fiji Pakistan

*Information received via Pacific Science Association (PSA).



December 2-12, 2001

Mr. Jim AbrahamMeteorological Service of Canada2121 TransCanada Highway Dorval, Quebec Canada H9P 1J3 4.6 R Telephone: 514.421.4751Fax: 541.421.4758e-mail:
Dr. Joseph Adejokun7 bis, avenue de la PaixCase Postale 2300Geneve 2, Switzerland WMO Telephone: 41 22 730-8213Fax: 41 22 733-2326e-mail:
Dr. Linda J. Andersen-BerryJames Cook University, Centre for Disaster StudiesP. O. Box 6811Cairns, Queensland 4870 Australia 5.1 R Telephone: 61 7 4042 1215Fax: 61 7 4042 1214e-mail: au
Dr. Lixion AvilaTPC/NHC11691 S. W. 17th StreetMiami, FL 33165-2149 USA IC 1. P 3.2 WG Telephone: 305.229-4410Fax: 303-553-1901e-mail:
Mr. Peter Baddiley Bureau of Meteorology GPO Box 413 Brisbane, 4001 Australia 2.3 WG Telephone: 617.3239.8768 Fax: 617.3239.8687 e-mail:
Dr. Mike Banner University of New South Wales Sydney, NSW, 2052 Australia 1.4 WG Telephone: Fax: e-mail:
Dr. Peter BowyerCanadian Hurricane Center 45 Alderney Dr., Dartmouth, Nova Scotia, B2Y 2N6 Canada 4.2 WG Telephone: 902.426.9181Fax: 902.426.4873e-mail:
Mrs. Barbara BrumitDepartment of Atmospheric ScienceColorado State UniversityFt. Collins, CO 80523 Admin Telephone: 970.491.8681Fax:e-mail:
Dr. Kurt BrueskeDepartment of Physics2354 Fairchild Drive, Suite 2A31USAF Academy, CO 80840 2. P 1.6 WG Telephone: 719.333.3412Fax: 719.333.3182e-mail:
Ms. Kathy-Ann CaesarCaribbean Institute of Meteorology & Hydrology P. O. Box 130 Bridgetown, Barbados, W. I. 1.6 WG Telephone: 246.425.1362 /3/5Fax 246.424.4733e-mail:
Mr. Jeff CallaghanBureau of Meteorology295 Ann St. GPO Box 413Brisbane, 4001 Queensland Australia 1. P 4.2 R Telephone: 617.3239.8624Fax: 617.3221.4895e-mail:

Mr. Philippe CaroffDivision CYCDirection Interrgionale de La ReunionB. P. 4 97491 Sainte ClotildeLa Reunion, France 1. P 2.1 WG Telephone:
Mr. Vernon CarrBureau of Meteorology 150 Lonsdale StreetMelbourne, 3000 Victoria Australia Admin Telephone: 613.9669.4634 Fax: 613.9669.4670 e-mail:
Dr. Qamar-uz-Zaman ChaudhryPakistan Meteorological Department Headquarters OfficeP. O. Box 1214 Sector H-8/2, Islamabad, Pakistan IC 2.1 WG Telephone: (+92-51) 9257314Fax: (+92-51) 4432588e-mail:; or
Professor Johnny ChanDept. of Physics & Material Sci.City University of Hong Kong83 Tat Chee Ave, Kowloon, Hong Kong, China 3.0 C 3.4 R Telephone: 852.2788.7820Fax: 852.2788.7830e-mail:
Dr. Simon ChangOffice of Naval Research800 N. Quincy St.Arlington, VA 22217 USA SF1 L Telephone: 703.588.2553Fax: 703.696.3390e-mail:
Professor Lianshou ChenChina Academy of Meteorological Sciences46 Zhongguancun South StreetHaidian District, Beijing 100081 China IC 2.1 R Telephone: 86 10 68407056 Fax: 86 10 62175931 e-mail:
Dr. Shuyi S. ChenUniversity of Miami/RSMAS/MPO4600 Rickenbacker CausewayMiami, FL 33149-1098 USA 1.5 R 2.2 WG 4.4 WG Telephone: 305.361.4048Fax: 305.361.4696e-mail:
Mr. Sun Fatt ChongHydrology DivisionDepartment of Irrigation and DrainageKM-7, Jalar Ampang68000 Ampang, Kuala LumpurMalaysia 2.3 R Telephone: 603.425.65828Fax: 603.425.63735e-mail: or
Mrs. Thi Hong Nga DangHydroMeteorological Service of Viet Nam No. 4 Dang Thai Than StreetHanoi, Viet Nam Telephone: 844.826.4087Fax: 844.826.0779e-mail:
Dr. Richard DareBMRCP. O. Box 1289KMelbourne, 3001 Australia 2.2 WG Telephone: 613.9669.4443Fax: 613.9669.4660e-mail:
Mr. Jim DavidsonBureau of Meteorology GPO Box 413Brisbane, Queensland 4001 Australia 5.3 WG Telephone: +617.3239.8739Fax: +617.3221.4895Mobile: 0409 641 216e-mail:

Dr. Noel DavidsonBMRC AustraliaP. O. Box 1289KMelbourne 3001, Australia 3.1 R 4.1 WG Telephone: 613.9669.4416Fax:
Ms. Li DuanHainan Meteorological Bureau No. 60 Haifu Rd. Haikou, 570203 China 1.6 WG Telephone: 86 898 65347582Fax:e-mail:
Dr. Yihong DuanShanghai Typhoon Institute#166 Puxi Rd. Shanghai, 200030 China 1.4 WG Telephone: 86.21.6438.6700 x6480Fax: 86.21.64391966e-mail: or
Professor Shishir Dube Director, Indian Institute of Technology Kharagpur, India 721-303 2.5 R Telephone: 91 3222 82002Fax: 91 3222 82000e-mail:
Mr. Roger EdsonPMB 101-226415 Chalan San Antonio Rd.Tamuning, Guam 96913 2. P 1.6 WG 4.1 WG Telephone: 671.735.2695Fax: 671.734.8890e-mail:;
Dr. Russell L. ElsberryDepartment of Meteorology, MR/EsNaval Postgraduate School589 Dyer Rd. Room 254Monterey, CA 93943-5114 USA IC 1. C SF 3 L 3.2 WG Telephone: 831.656.2373Fax: 831.656.3061e-mail:
Dr. Jenni EvansPennsylvania State UniversityDepartment of Meteorology503 Walker Bldg.University Park, PA 16802 USA 1.2 WG 1.3. R 4.6 WG Telephone: 814.865.____Fax: 814.865.3663e-mail:
Rex Falls11 Manly StreetBirkdale, Queensland 4159 Australia 5.0 C Telephone: 617.3207 2014Fax:e-mail:
Dr. Mike FiorinoLLNL, Bldg 1700 L-103700 East Ave. Livermore, CA 94550-9234 3.1 WG Telephone: 925.423.8505Fax: 925.422.7675e-mail:
Mr. Gary FoleyBureau of MeteorologyP. O. Box 1370West Perth, Australia 6872 IC SF2 L Telephone: 619.401.7317Fax: 619.263.2297e-mail:
Dr. Michael FoleyBureau of MeteorologyP. O. Box 40050Casuarina NT 0811 Australia 4.2 WG Telephone: 618.8920.3820Fax: 618.8920.3840e-mail:
Professor Klaus FraedrichMeteorologisches InstitutUniversitaet HamburgBundesstr. 55Hamburg D-20146 Germany 4.3 WG Telephone: 49 40 42838 5064Fax: 49 40 42838 5066e-mail:
Ms. Reggina GarzaNWS Southeast River Forecast Center4 Falcon DrivePeachtree City, GA 30269 2.1 WG2.3 WG 2.5 WG Telephone 770.486.0028 ext. 404Fax: 770.486.0930e-mail:

Dr. Isaac GinisGraduate School of Oceanography University of Rhode Island Narragansett, RI 02882 USA 1.5 WG 2.5 WG Telephone: 401.874.6484Fax: 401.874.6728e-mail:
Dr. James GoerssNaval Research Laboratory-Monterey7 Grace Hopper Ave.Monterey, CA 93943-5502 USA 3.1 WG 3.2 WG Telephone: 831.656.4739Fax: 831.656.4769e-mail:
Dr. William GrayDepartment of Atmospheric SciencesColorado State UniversityFt. Collins, CO 80523-1371 USA 4.5 WG Telephone: 970.491.8681Fax: 970.491.8449e-mail:
Mr. Bruce Gunn National Manager, Special Services Unit Bureau of Meteorology150 Lonsdale St.Melbourne, Victoria 3000Australia Telephone: 61 3 9669 4990 Fax: 61 3 9669 4670 e-mail:
Dr. Akhilesh Gupta Director, Scientist-F NCMWFDepartment of Science & TechnologyMausam Bhawan Complex, Lodi RoadNew Delhi 100 003 India 3.1 WG Telephone: 91 11 469 4639, 461 9815Mobile: 91 98 104 22249 Fax: 91 11 469 0108e-mail: or
Barry HanstrumBureau of MeteorologyP. O. Box 1370West Perth, 6872 Australia 1.6 WG 4.6 WG Telephone:Fax:e-mail:
Dr. Bruce HarperSystems Engineering Australia LTD7 Mercury CourtBridgeman DownsQueensland 4035, Australia 2.4 WG Telephone: 61 7.3353.0288Fax: (same as above)e-mail:
Dr. Patrick A. HarrDepartment of Meteorology, MR/HrNaval Postgraduate School589 Dyer Rd. Room 254Monterey, CA 93943-5114 USA 4.0 C Telephone: 831.656.3787Fax: 831.656.3061e-mail:
Mr. Jeff HawkinsNaval Research Laboratory-Monterey7 Grace Hopper Ave.Monterey, CA 93943-5502 USA 1.3. P 1.1. WG 1.6 WG Telephone: 831.656-4833Fax:831.656-4769e-mail:
Mr. Julian HemingUK Met Office/Forecasting SystemsLondon Road, Bracknell RG12 2SZ United Kingdom 3.1 WG Telephone: 44 1344 854494Fax: 44 1344 856446e-mail:
Dr. Greg HollandAerosonde Ltd.41-43 Normanby Rd.Notting Hill Victoria 3168 Australia IC SF2 L SF4 WG Telephone: 61 3 9544 0866Fax: 61 3 9544 0966e-mail:

Mr. Rich JeffriesNaval Pacific Meteorology and Oceanography Center425 Luapele Road, Bldg. 398Pearl Harbor, HI 96860-3103 3.2 R Telephone: 808.471.2157Fax: 808.471-4581e-mail:
Mrs. Penny JonesGSEAS Editorial Assistant Naval Postgraduate School589 Dyer Rd., Room 254Monterey, CA 93943-5114 USA Admin. Telephone: 831.656.2925Fax: 831.656-3061e-mail:
Mr. S. R. KalsiDirector, RSMC New DelhiIndia Meteorological DepartmentMausam Bhawan Lodi Road New Delhi, 110 003 India 5.3 WG Telephone: 91 11 4611068Fax: 91 11 464 3128e-mail: or
Mr. Samarendra KarmakarBangladesh Meteorological DepartmentDeputy Director, Storm Warning CenterAbbawa Bhaban, Agargaon, Dhaka-1207 Bangladesh Telephone: 88 02 914 388Fax: 88 02 811 8230e-mail:
Mr. Takahiro KayaharaNational Research Institute for Earth Science & Disaster Prevention3-1 Tennodai, TsukubaIbaraki 305-0006 Japan 5.1 WG Telephone: 81 298 51 1611Fax: 81 298 51 1610e-mail:
Mr. Jeffrey KepertBureau of MeteorologyGPO Box 1289KMelbourne, Victoria 3000, Australia 1.4. WG 2.4 R Telephone: 631.9669.4492Fax: 613.9669.4660e-mail:
Mr. Kriengkrai KhovadhanaDirector, Weather Forecast DivisionThai Meteorological Department4353 Sukhumvit Road, BangnaBangkok 10260 Thailand 5.3 WG Telephone: 662 398 9801Fax: 662 398 9836e-mail:
Ms. Naoko KitabatakeTyphoon Research DepartmentMeteorological Research InstituteJapan Meteorological AgencyNagamine 1-1Tsukuba 305-0052, Japan 4.6 WG Telephone: 81 298 52 9167Fax: 81 298 53 8735e-mail:
Dr. John KnaffCIRA Colorado State UniversityWest Laporte Ave, Foothill CampusFt. Collins, CO 80523-1375 USA 1.2 WG4.1 WG4.3 WG Telephone: 970.491.8881Fax: 970.491.8241e-mail:

Mr. Tom KnutsonGFDL/NOAAP. O. Box 308Forrestal CampusPrinceton, NJ 08542 USA SF 4 L Telephone: 609.452.6509Fax: 609.987.5063e-mail:
Dr. Jim KossinUniversity of Wisconsin-CIMSS1225 West Dayton St., Room 219Madison, WI 53706 USA 0.2 P1.1 WG Telephone: 608.265.5356Fax: 608.262.5974e-mail:
Edwin ST LaiHong Kong Observatory134A Nathan Road, Kowloon, Hong Kong China 1.6 R Telephone: 852.2926.8461Fax: 852.2375.2645e-mail: or
Dr. Mark LanderWater and Energy Research InstituteUniversity of GuamUniversity of Guam StationMangilao, Guam 96923 USA 4.1 WG Telephone: 671.735.2695Fax: 671.734.8890e-mail:
Dr. Hung Kwan LamDirector, Hong Kong Observatory134A Nathan RoadKowloon, Hong Kong, China 5.3 WG Telephone: 852.2926.8221Fax: 852.2721.6557e-mail:
Mr. Antoine Lasserre-BigorryCellule Recherche Cyclones (CRC)Direction Regionale de La ReunionB. P. 4 97491 Sainte-Clotilde La Reunion, France 4.4 WG Telephone: 262 2 62 92 11 86Fax: 262 2 62 92 11 47e-mail:
Dr. Zhaochong Lei7 bis, avenue de la PaixCase Postale 2300Geneve 2, Switzerland WMO Telephone: 41 22 730-8213Fax: 41 22 733-2326e-mail:
T. C. LeeHong Kong Observatory134A Nathan Road Kowloon, Hong Kong, China 5.3 R Telephone: 852.2926.8442Fax: 852.2377.3472e-mail:
Dr. Woo-Jin LeeKorean Meteorological Administration460-18 Shindaebang-dongDongjak-gu, Seoul 156,720 Republic of Korea 2.2 R Telephone: 822.831.0365Fax: 822.836.3157 e-mail:
Mr. Leong Weng Kun Rampa do ObservatoriaCaixa Postal No. 93 Macao, China 1.6 WG Telephone: 853.898.6275Fax: 853.850.557e-mail:
Dr. Chi-Sann Liou Naval Research Laboratory-Monterey7 Grace Hopper Ave.Monterey, CA 93943-5502 USA 1.5 WG Telephone: 831.656.4735Fax: 831.656.4769e-mail:
Dr. John LeMarshallBureau of Meteorology Research CentreGPO Box 1289K Melbourne, Australia 1. C 0.2 P Telephone: 61.3.9669.4420Fax: 61.3.9669.4660e-mail:

Mrs. Nanette Lomarda7 bis, avenue de la PaixCase Postale 2300 CH1211 Geneve 2,Switzerland WMO Telephone: 41 22 730 8384Fax: 41 22 730 8021e-mail:
Professor Zhexian LuoNanjing Institute of MeteorologyDepartment of Atmospheric SciencesNanjing 210044, China 1.2 WG Telephone: 86.25.340.2556Fax: 86 25.873.1191e-mail:
Dr. John L. McBrideBureau of Meteorology Research CentreGPO Box 1289KMelbourne 3001, Australia 4.4 WG Telephone: 61-3-9669-4421Fax: 61-3-9669-4660e-mail:
Dr. Frank MarksNOAA/HRD4301 Rickenbacker CausewayMiami, FL 33149 USA 1.1 R Telephone: 305.361.4321Fax: 305.361.4402e-mail:
Dr. Max MayfieldTPC/NHC11691 S. W. 17th StreetMiami, FL 33165-2149 USA 2.1 WG 5.3 WG Telephone: 305.229-4402Fax: 305.553-1901e-mail:
Dr. Meng ZhiyongChinese Academy Meteorological Sciences46 Zhongguancun South StreetHaidan District, Beijing 100081 China 2.2 WG Telephone: 86 10 6840 6586Fax: 86 10 6217 5931e-mail:
Craig MitchellBureau of Meteorology GPO Box 413Brisbane, Queensland 4001 Australia Telephone:Fax:e-mail:
Dr. Dominique MoellerUniversity of MunichTheresienstr. 37 80333 Munich, Germany 1.3 WG4.4 WG Telephone: 4989 2180 4327Fax: 4989 2180 4182e-mail: or nique@(above)
Dr. Michael MontgomeryDepartment of Atmospheric SciencesColorado State UniversityFoothills CampusCampus Mail Delivery-1371Ft. Collins, CO 80523-1371 USA 4.4 R Telephone: 970.491.8355Fax: 970.491.8449e-mail:
Russel MorisonCentre for Environmental Modelingand Prediction, Applied MathematicsUniversity of New South WalesSydney, NSW, 2052 Australia 1.5 WG Telephone: 61 2 9 385 7080 Fax: 61 2 9385 7123 e-mail:

Richard J. MurnaneRisk Prediction Initiative Bermuda Biological Station for Research, Inc.P. O. Box 405 (11312 Rokeby Ave.)Garrett Park, MD 20896 USA 5.2 R Telephone: 301.962.5053Fax: 301.942.1886e-mail:
Dr. Masashi NagataHead, National Typhoon Center (RSMC Tokyo)Forecast Division, Forecast DepartmentJapan Meteorological AgencyOte-machi 1-3-4, Chiyoda-kuTokyo 100-8122 Japan IC 1.6 WG 3.1 WG Telephone: 81 3 3212 8341 (x 3148) Fax: 81 3 3211 4923e-mail:
Dr. Tetsuo NakazawaMeteorological Research Institute/JMA1-1 NagamineTsukuba, 305-0052, Japan 0.2 P 1.3. WG 4.3 WG Telephone: 81. 298.53.8671Fax:
Dr. Prisco NiloPAGASA1424 Quezon Ave.Quezon City, 1100 Philippines 2.1 WG Telephone: 632.929.4570Fax: 632.929.4865 e-mail: or
Mr. Pascal PedduzziHead, Early Warning UnitUNEP/DEWA/GRID-Geneva11, ch. Des Anemones,1219 Chatelaine, Geneva, Switzerland Admin Telephone: 4122 917 82 37Fax: 4122 917 80 29e-mail:
Dr. Tao-Yong PengGuangdong Meteorological Observatory6 Fujin Road Guangzhou 510080 China 1.6 WG Telephone: 86 20 8777 4919 (O)Fax: 86 20 8777 4919e-mail:
Dr. Ramon Perez -SuarezCentro del ClimaInstituto de Meteorologia/AMA/CITMA17032, Habana 17 C.P. 11700 Cuba 4.5 WG Telephone: 537.867.0718Fax: 537.338.010 e-mail:
Dr. John PersingDepartment of Atmospheric SciencesColorado State UniversityFoothills CampusCampus Mail Delivery-1371Ft. Collins, CO 80523-1371 USA Telephone: 970.491.7714Fax: 970.491.8449e-mail:
Rajendra PrasadDirector, Fiji Meteorological ServiceRSMC Nadi Fiji 5.3 WG Telephone: 679-6724888 (x5001)Fax: 679-6720430e-mail:

Dr. Vladimir PudovInstitute of Experimental MeteorologySPA “Typhoon”82 Lenin AvenueObninsk, Kaluga Region, 249038 Russia 1.4 WG Telephone: 07.0843971751Fax: 07.0843940910e-mail: or
Mr. Qian ChuanhaiNational Meteorological CenterBaishiqiaolu No. 46Western Suburb, Beijing, China 3.2 WG Telephone: 86 10 6840 6614Fax: 86 10 6217 2908e-mail:
Gary QuinnAustralia Environmental Systems and Services 20 Council St.Hawthorn Vic 3128 Australia Telephone:613.8420.8923Fax: 613.8420.8900e-mail:
Mr. Alan RadfordGroup Head, IT & Service DeliveryMet. OfficeLondon Road, Bracknell RG12 2SZUnited Kingdom 5.3 WG Telephone: 44.1344.856289Fax: 44.1344.854462e-mail:
Mr. Rishi RajDirector, Public WorksP. O. Box 3740Samabula, Fiji Telephone: 679 315244Fax: 679 303023e-mail:
Alain Solo RazafimahazoDirector Generale de la MeteorologieBP 1254 Antananarivo 101Madagascar Telephone:
Mr. Stephen C. Ready, Director, Tropical Cyclone Warning Center 30 Salamanca Rd.P. O. Box 722Wellington, New Zealand 5.3 WG 4.6 WG Telephone:Fax: e-mail:
Dr. Paul ReasorDepartment of Atmospheric SciencesColorado State UniversityFoothills CampusCampus Mail Delivery-1371Ft. Collins, CO 80523-1371 USA Telephone: 970.491.8613Fax: 970.491.8849e-mail:
David RichardsonECMWFShinfield ParkReading, Berkshire RG2 9AXUnited Kingdom 0.4 P3.1 WG3.3 WG4.5 WG Telephone: 44.118.949.9750Fax: 44.118.986.9450e-mail:
Dr. Elizabeth Ritchie-TyoElectrical/Computer Engineering Dept.Room 125, EECE BuildingUniversity of New MexicoAlbuquerque, NM 87131-1356 USA 1.2 R 4.4 WG Telephone: 505.277.8325Fax:
Dr. Michel RosengausServicio Meteorologico Nacional Av. Observatorio 192 Col. Observatorio Deleg. Miguel Hidalgo D.F. 11860, Mexico 2.1 WG 5.3 WG Telephone: 525.55 626 8600 x 3404 Fax: 525 55 626 8695 e-mail:

Dr. Jose RubieraDirector, National Forecasting CenterInstituto de MeteorologiaLoma de Casa Blanca, Regla,Ciudad de la Habana, Cuba C.P. 11700 1.3. WG Telephone: 53 7 867 07 08Fax: 53 7 33 80 10e-mail:
Matt SaundersonBureau of Meteorology GPO Box 413Brisbane, Queensland 4001 Australia Telephone:Fax:e-mail:
Dr. Lloyd ShapiroUniversity of MunichTheresienstr. 3780333 Munich Germany 1.3. WG 1.2 WG Telephone: 4989 2180 4327Fax: 4989 2180 4182e-mail:
Dr. Lynn ShayUniversity of Miami, RSMASDivision of Met & Phys Oceanogr4600 Rickenbacker CausewayMiami, FL 33149-1098 USA 1.4 R2.5 WG Telephone: 305.361.4075Fax: 305.361.4696e-mail:
Ethige Sunil SilvaDepartment of MeteorologyBauddhaloka MawathaColombo 7, Sri Lanka Telephone: 941 694 104Fax: 941 252 319e-mail:
Dr. Keon-Tae SohnDepartment of StatisticsPusan National University 30 Jangjeon-dong Geumjung-gu Pusan 609-735 Korea 1.4. WG 3.1 WG Telephone: 82.51.510.2215Fax: 82.51.581.1459e-mail:
Professor Roger SmithUniversity of MunichTheresienstr. 37 80333 Munich Germany 1.3 WG Telephone: 49 89 2180 4383Fax: 49 89 2180 4182e-mail: roger@
Mr. Brian StrahlFleet Numerical Meteorology & Oceanography Center7 Grace Hopper Ave., Stop 1Monterey, CA 93943-5505 Telephone: 831.656.4484Fax: 831.656.4363e-mail:
Mr. Helder SueiaInstitueo Nacional de Meteorologia (INAM)Rua de Mukumbura 164C. Postal 256Maputo, Mozambique Telephone: 618 8177 0259Fax: 618 8201 3261 (Attn: Elaine Kane)e-mail:
Dr. Naomi SurgiNCEP/EMC5200 Auth Rd.Camp Springs, MD 20746 1.5 R2.2 WG 3.1 WG 1.5. WG Telephone: 301.763.8000 (x 7285) Fax: 301.763.8545 e-mail:
Mr. Hideo TadaNumerical Prediction DivisionJapan Meteorological AgencyOte-machi 1-3-4 Chiyoda-kuTokyo 100-8122 Japan 1.1. P 2.2 WG Telephone: 813.3212.8341 (x 3316)Fax: 813.3211.8407e-mail:
Dr. Kevin ToryBMRCP. O. Box 1289KMelbourne 3001, Australia 4.4 WG Telephone: 613.9669.4578Fax: 613.9669.4660 e-mail:

Dr. Steve TractonNCEP5200 Auth Rd.Camp Springs, MD 20746 3.3 WG4.6 WG Telephone:Fax:e-mail:
Mr. Mitsuru UenoTyphoon Research DepartmentMeteorological Research InstituteJapan Meteorological AgencyNagamine 1-1Tsukuba 305-0052, Japan 1.5 WG Telephone: 81 298 53 8665Fax: 81 298 53 8735e-mail:
Dr. Eric UhlhornNOAA-HRD4301 Rickenbacker CausewayMiami, FL 33149 USA SF1 P Telephone: 305.361.4320Fax: 305.361.4402e-mail:
Lic. Norman VegaNational Meteorological InstituteCalle 17, Avenida 9 bisSan Jose, Costa Rica Telephone: 506.222.5616Fax: 506.223.1837e-mail:
Mr. Chris VeldenUniversity of Wisconsin--CIMSS1225 West Dayton St.Madison, WI 53706 USA 1. C 0.3 P 1.2 WG 1.6. WG Telephone: 608.262.9168Fax:
Dr. Edward WalshNOAA/ETL R/ET1325 BroadwayBoulder, CO 80305-3328 USA 1.4 WG2.5 WG Telephone: 303.497.6357Fax: 303.497.3577e-mail:
Dr. Kevin WalshSchool of Earth SciencesUniversity of MelbourneMelbourne, Victoria 3010 Australia 1.4. WG 5.1 WG Telephone: 613 8344.6523Fax: 613 8344 7761e-mail:
Mr. Richard WalshVaisalaP. O. Box 619Hawthorn 3122, VictoriaAustralia Telephone: 613.9818.4200Fax: 613.9818.4522e-mail:
Professor Bin WangIPRC, University of Hawaii1680 East-West Rd., Post Bldg. 401Honolulu, HI 96822 USA 3.4 WG4.5 R Telephone: 808.956.2563Fax: 808.956.9425e-mail:
Ms. Wang JianjieCMA National Meteorological CenterNo. 46, South Zhong Guan Cun St.Hadian District, Beijing 100081China 3.1 WG Telephone: 86.10.6840.8584Fax: 86.10.6840.8584e-mail:
Dr. Wang ShourongZhejiang Meteorological Bureau 5.3 WG Telephone: 86 571 8706 7973Fax: 86 571 8707 1884 e-mail:
Dr. Yuqing WangIPRCUniversity of Hawaii – SOEST2525 Correa RoadHonolulu, HI 96822 1.5 WG Telephone: 808.956.5609Fax: 808.956.9425e-mail:

Alipate WaqaiceluaFiji Meteorological ServicePrivate Mail Bag NAP0351 Nadi Airport Republic of Fiji 0.1 P 1.6 WG Telephone: 679.672.4888Fax: 679.672.0430e-mail:
Dr. Harry WeberUniversity of MunichTheresienstr. 3780333 Munich, Germany 3.2 WG Telephone: 49 89 2180 4577Fax: 49 89 2180 4182e-mail:
Mr. Frank H. WellsP. O. Box 13105 Santa Rita, Guam 96915-3105 USA 1.1 WG Telephone: 671. 564.2571Fax: 671.472-0980 or 671.564.2571e-mail:
Mr. Jean-Marie WillemetMeteo France, BP497491 Ste. ClotildeLa Reunion, France 4.3 WG Telephone: 00 262 262 921185Fax: 00 262 262 821147e-mail:
Mrs. Kate WrightAerosonde Limited41-43 Normandy Rd. Notting HillVictoria 3168, Australia Admin Telephone: 61 3 9544 0866 Fax: 61 3 9544 0966e-mail:
Dr. Chun-Chieh WuPSA Meteorology Committee c/o Prof. C.-P. Chang Dept of Meteorology, Code MR/Cp Naval Postgraduate School Monterey, CA 93943 1.1 WG2.2 WG Telephone: 886 2 2363 2303 Fax: 886 2 2363 2303 e-mail:
Professor Xu XiangdeChinese Academy of Meteorological SciencesNo. 46, Zhongguancun South StreetHadian District, Beijing, 100081 China 4.6 WG Telephone: 86 10 68407445Fax: 86 10 6840 8656e-mail:
Professor Xu YimingShanghai Typhoon Institute166 Puxi Rd.Shanghai 2000 30 China 3.1 WG Telephone: x6446 Fax: e-mail:
Professor Xue JishanChina Meteorological AgencyNo. 46, Zhongguancun South StreetHadian District, Beijing, 100081 China 0.4 P Telephone: 86.10.6840.8706fax: 86.10.6217.5931e-mail:
Dr. Tien-Chiang YehPSA Meteorology Committee c/o Prof. C.-P. Chang Dept of Meteorology, Code MR/Cp Naval Postgraduate School Monterey, CA 93943 2.1 WG 3.1 WG Telephone: 886 2234 91200Fax: 886 2234 91207e-mail:
Mr. Jun YoshimuraFrontier Research System for Global Change3173-25 Showa-machi, Kanazawa-kuYokohama-city 236-0001Japan SF 4 Telephone: 81 45 778 5700Fax: 81 45 778 5707e-mail:

Mr. Yong YuGuangdong Meteorological Bureau6 Fujin Road Guangzhou 510080 China 1.1 WG Telephone: 86 20 8777 6918 ext. 311 (O)Fax: 86 20 8777 3952e-mail:
Dr. Michiaki YumotoNational Research Institute for Earth Science & Disaster Prevention3-1 Tennodai, TuskubaIbaraki 305-0006 Japan 5.2 WG Telephone: 81 298 51 1611Fax: 81 298 51 1610e-mail:

revised: 12/12/02 (pj)