WORLD METEOROLOGICAL ORGANIZATION

WORLD WEATHER WATCH

TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION

2003

I WMO-No. 957 I

Secretariat of the World Meteorological Organization - Geneva - Switzerland © 2003, World Meteorological Organization ISBN 92-63-10957-5

NOTE

The designations employed and the presentation of material in this publication do not imply the expression of any opinion whatsoever on the part of the Secretariat of the World Meteorological Organization concerning the legal status of any country, territory, city or area, of its authorities, or concerning the delimitation ofits frontiers or boundaries. CONTENTS

Page

FOREWORD...... v FORTY YEARS OF WWW...... 1 EXECUTIVE SUMMARy...... 5

CHAPTER I- INTRODUCTION 7 Purpose and scope of the WWW Programme 7 Components of the WWW system..... 7 Organization ofWWW programmes '" 7 Relationship ofWWW with other programmes 8

CHAPTER 11- THE GLOBAL OBSERVING SYSTEM 9 Requirements for observational data...... 9 Implementation of the surface-based subsystem...... 9 Upper-air observations 10 Climatological observations.... 12 Marine meteorological observations 15 Aircraft meteorological observations '" 15 Other observation stations/systems 16 The space-based subsystem...... 16 Implementation goals for satellite receivers... 17 Quality of observational data 18 Upper-air data quality...... 19 Land-surface data quality 19 Marine surface data quality 19 AMDAR data quality 21

CHAPTER III - THE GLOBAL TELECOMMUNICATION SYSTEM AND WWW DATA MANAGEMENT: INFORMATION SYSTEMS AND SERVICES 23 Global Telecommunication System...... 23 Main Telecommunication Network...... 23 Regional Meteorological Telecommunication Networks...... 23 Multipoint telecommunication services via satellite and radiobroadcasts 31 Data-communication techniques and procedures...... 32 World Weather Watch data management...... 32 Radio frequencies for meteorological activities...... 32 iv CONTENTS CHAPTER IV - THE GLOBAL DATA-PROCESSING SYSTEM 33 RSMCs with geographical specialization 33 RSMCs with activity specialization 33 Medium-range weather forecasting...... 33 Tropical cyclone forecasting ,. 34 Provision of transport model products for environmental emergency responses...... 42 NMCs and centres with similar functions...... 42 Verification of numerical weather prediction.. 42 Technical development at GDPS centres 45

CHAPTER V- STATUS OF THE AVAILABILITY OF OBSERVATIONAL DATA FROM WWW...... 47 Results of the analysis 47 Analysis of the availability of SYNOP reports 51 Analysis of the availability ofParts A ofTEMP reports...... 53 Analysis of the availability ofreports from mobile stations 55

CHAPTER VI - OPERATIONAL INFORMATION SERVICE...... 57 Available information...... 57

ANNEXES Annex I- Acronyms 59 Annex II - References 61 FOREWORD

This year we commemorate the 40th anniversary ofWMO's World Weather Watch 0l'fWW) Programme. In April 1963, the Fourth World Meteorological Congress approved the concept ofthe WWW, and set WMO on the journey that dramatically changed and enhanced the development ofmeteorology and the atmospheric sciences. The challenge laid down 40 years ago has resulted in a unique success story ofinternational cooperation and opportunity. The basic systems - comprising the provision ofbasic meteo­ rological data and products, telecommunication services and the management thereof - are the pillars for the delivery of meteorological services worldwide, and have become in many ways the 'core' operational facility on which almost all programmes ofWMO and other relevant international agencies depend. The challenge continues, as increasing technological advancement will place further requirements on the basic systems and require continuous adjustment. This publication is the 21st in a series of biannual reports on the status of implementation ofWWW. It was designed mainly to inform the senior management of the National Meteorological and Hydrological Services (NMHSs) - but also those from academia and the private sector who may be interested - of the operational status ofWWW. It provides information concerning the structure, status and trends of the implementation, as well as performance of the core components of WWW, notably the Global Observing System (GOS), the Global Telecommunication System (GTS) and the Global Data-processing System (GDPS). A special introductory chapter for this anniversary edition gives a brief look at WWW's achievements during the past 40 years. Some parts of this publication have been compacted from detailed information and reports made available by Members to the Commission for Basic Systems (CBS) and the working groups of the Regional Associations (RAs) dealing with WWW. I wish to take this opportunity to express my sincere appreciation to the Members ofWMO for their continuing efforts to­ wards the consolidated further development of WWW, as well as their collaboration in providing the information on which this report is based.

_-==c=~- - ~------

G. O. P. Obasi (Secretary-General) YEARS OF WORLD WEATHER WATCH 1963-2003

INTRODUCTION had increased to 222 circuits. The first circuit upgrade from the initial low telegraphic speed was in 1970, when WWW was conceived 40 years ago, when the leaders of the Washington-Tokyo circuit was upgraded to 2400 the world meteorological community had a vision of bits/sec. Since then, telecommunications advances have international cooperation on an unprecedented scale, continued at a rapid rate and many GTS circuits are now and the will to implement it. No other activity has done implemented with high-speed communications links, more to advance our understanding of the atmosphere including satellite- and Internet-based services. or the science and practice of meteorology than WWW. Regional Basic Synoptic Networks (RBSNs) were It has been possible because of the work and commit­ established in the early 1960s as the core part of the Global ment of people and governments of all nations. Observing System (GaS). The GaS provides observational National meteorological services (NMSs) and meteoro­ data from all parts of the world for use in both operational logical centers of excellence have shared skills and and research work. In 1963, the RBSN included 3780 expertise, working in a spirit of cooperation which stations, which had increased to 3957 by 2002. WWW has fostered and which it carries into the future. Over the past 40 years, the total number of surface synoptic stations reported to WMO as implemented ACHIEVEMENTS increased from almost 8000 stations in 1963 to 10 952 stations in 2002. Figure 1 shows this growth during the One of the notable early achievements of WWW was the period 1964-2002. development and implementation of the Global During the same period, the number of upper-air Telecommunication System (GTS). The original WWW Plan stations implemented has increased from around 800 in envisioned the GTS serving three World Meteorological 1963 to a number fluctuating between 900 and 1000 Centres (WMCs), 21 Regional Meteorological Centres over the last decade. The number of observations (RMCs) (now Regional Telecommunication Hubs (RTHs» received from the upper-air network, however, has not and 147 National Meteorological Centres (NMCs). Today, remained so stable. It declined during the 1990s and has there are three WMCs, 32 RTHs and 185 NMCs. recovered again since 2000. The number of observations The original GTS plan reqUired 247 circuits. By received stabilized in the last three years at somewhat 1976, 196 circuits had been established. By 2002, this below 1050 reports daily (Figure 2).

14000

12000

2000 l Figure 1 - Surface synoptic o station implementation 1964 1968 1972 1976 1980 1984 1988 1992 1996 2000 2004 during the period Year 1964-2002 2 WORLD WEATHER WATCH - TWENlY-FIRST STATUS REPORT ON IMPLEMENTATION

1400

1200 -- - ..... (/) - "W"" c: 1000 -r .Qro --- 2: <1> 800 (/) .a 0 '0 600 ~ .a E z'" 400

200

Figure 2 - Daily average 0 number of upper-air reports 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 received during the IS-day Year annual monitoring periods

_____T~hUJLajLcJaJt~orologicaldata relay_(A,--,-M~D",-,-A-,-,R-'J) ~T-,-,h~e-.lp~r~ogressive development of the space-based programme, which began in the late 1980s, provides system from the rudimentary polar orbiting satellites at additional upper-air data. AMDAR reports now number the beginning to the current operational configuration more than 100000 observations per day, greatly surpass- of three to four polar orbiting systems and five to six ing aircraft weather report (AIREP) reports, which satellites in geostationary orbit, together with the inclu- number about 4000 per day. sion of Research and Development satellites in 2002, has Surface and upper-air synoptic observations from been perhaps the single greatest accomplishment in our the oceans are proVided by ships of the Voluntary ability to observe the global atmosphere and ocean Observing Ship (VOS) programme, the Automated (Figure 3). The satellite operators have closely coordi- Shipboard Aerological Programme (ASAP) programme nated their plans with WWW, and the provision of and by drifting and moored buoys. The total number of satellite data into the WWW system has been a model of VOS was around 6500 in the 1960s and reached a peak international response to user requirements. of nearly 8000 in the mid-1980s, before decreasing again The Instruments and Methods of Observation to around 6000 at the start by 2001. At the same time, Programme (lMOP) is responsible for considerable work the number of VOS observations available on the GTS in getting instruments and methods of observation has increased to some 3000 per day. This increase has standardized. This has contributed to a significantly resulted largely from improved telecommunications increased quality and homogeneity of meteorological systems. The ASAP programme, introduced in the mid- data that are applied for various operational applica- 1980s, provides some 6000 upper-air soundings per year tions as well as for research, especially in the field of distributed on the GTS. Serious development of drifting climate change. Significant progress has been made in ocean data buoys began in the 1970s, culminating in getting reliable information on the performance char- more than 250 such drifters being operational during acteristics of instruments, sensors and equipment, and the First Global Atmospheric Research Programme for ensuring global compatibility of observations and (GARP) Global Experiment (FGGE). Numbers declined data homogeneity by carrying out instrument inter- sharply after the FGGE, until the establishment of the comparisons. Data Buoy Cooperation Panel in 1985 acted as a stimu- A series of major international field experiments, Ius to renewed, coordinated global deployments. including GARP Atlantic Tropical Experiment (GATE), Atmospheric pressure observations from drifters on the GARP, FGGE and Tropical Ocean and Global Atmosphere GTS increased from 30 000 per month in 1990 to nearly Programme (TOGA), to name but a few, have tremen- 200 000 per month in 2002. Several hundred moored dously advanced meteorological science in general, and buoys are also deployed and reporting meteorological numerical weather prediction (NWP) capability in variables in real time, mainly from coastal waters or particular. These research programmes were possible equatorial zones. because of the worldwide infrastructure provided by the The advent of satellites was one of the triggers that WWW system of observations, communications and led to the development of the WWW programme in data processing. Many of the systems and technologies 1963. The launch by the former Soviet Union of Sputnik developed and deployed during the experiments have [ in October 1957, followed by the launch by the United become integral parts of the operational WWW itself. States of Explorer [ in January 1958, initiated tremen- WWW implementation of the Global Data- dous interest and an intense desire to ensure that this processing System (GDPS) to proVide data and numerical new technology would be used for peaceful purposes. model output fields to the RMCs and then to the The first meteorological satellite, Tiros I, was launched National Meteorological and Hydrological Services by the United States in 1960. (NMHSs) has greatly assisted in mitigating the impact of 40 YEARS OF WORLD WEATHER WATCH 3 climate events through the national services' provision anomaly correlation coefficient for geopotential height of skilful forecasts and warnings. should be larger than 0.6. The anomaly correlation coeffi­ The progress made in the performance of the cient measures the correlation between the forecast GDPS has manifested itself in a continual improvement anomaly from climatology and observed anomaly. of both accuracy and lead-time of weather forecasts. Figure 5 shows the use of anomaly correlation Five-day forecasts today have better accuracy than two­ coefficients to compare the skill levels of northern and day forecasts had in the 1970s. This has been possible southern hemisphere 500 hPa height forecasts, based on due to tremendous developments in remote sensing, in operational three-, five- and seven-day forecasts at the mathematical techniques for modeling turbulent fluids, European Centre for Medium-Range Weather Forecasts in the science of atmospheric physics and atmospheric (ECMWF) from January 1980 to September 2002. The chemistry, in data assimilation techniques and in figure shows sustained improvements in annual-mean computer science and technology. The prediction of forecast skill in both hemispheres over the last 20 years. extreme events - such as the occurrence, intensity and In the northern hemisphere, the gain has been 1.5 to track of tropical cyclones - has also shown significant two days in 20 years The overall rate of forecast improvement during the past decades. improvement has been higher in the southern hemi­ Figure 4 shows that forecast skill for the large-scale sphere than in the northern. In the early 1980s, the skill synoptic flow in the Northern Hemisphere winter has levels of the three- and five-day forecasts for this hemi­ improved by five days over the last 30 years, and by three sphere were only a little better than those of the five­ days over the last 20 years. This is based on the generally and seven-day northern hemisphere forecasts. Today, accepted fact that for a forecast to have useful skill, the however, the skill at days three and five in the southern

Milestone 1960 1965 1970 1975 1980 1985 1990 1995 2000

First meteorological satell~e (MetSat) ~ TIROS·1 (USA)

First operational MetSat ~ ESSA-l (USA) I First geostationary MetSat W ATS-1 (USA) I I First Soviet MetSat ~ Meteor-1m (USSR) I I Introduction 01 Automatic Picture ~ I Transmislon (APn ITOS-1 (USA) , l-d~ First atmospheric sounder ~NOAA-2(USA) l 1 '"'' .,." Coordination Group for Meteorological First meeting of too 6GMS '~ ~~ Satell~es (CGMS) ;t,/ r First operational geostationary MetSat I ~ SMS-1 (!J.SA) .....1 First Japanese geostationary MetSat 1" ~ GMS-1 (Japan) ~ , n~-- First European geostationary MetSat I -.)._,,8 e"" M~eOsat-1 (ESA) r .~ First water vapour imager I ~Meteosat-1 (ESA) r First operational atmospheric sounder J ~. ~ TiROS-N (USA)">.- I Global coverage achieved, for the First Satell~es from the USA, USSR.)~~ -~h GlobalGARP Experiment (FGGE) (. ~Illhary sa:~~~~ , New generation of geostationary satellites 'I ,,~ .tW ~GOES-8(USA) First Russian geostationary MetSat I --- GOMS-1/Elektro (Russia) ~ New generation of atmospheric sounders "~ NOAA-K (USA) ~ New multi-functional satellite concept I I I I MTSAT (Japan) ~ Figure 3 - Milestones in Space-based component of GOS I the evolution of , R&D satel~es included in GOS ~ satell~es ~ expanded to include R&D I meteorological and Next generation of European MetSat I I I MSG-1 (EUMETSAn~1!Y environmental satellites

0.9 -.

...... 0.8 ·..· , . ·.·... 0.7 c o ~ 0.6 -+----T----T-~iC-;..---;..---'l~-T----T-----;.---'lU~-----j ~ o ~ 0.5 ro Winter 2001/02 Figure 4 - Anomaly E correlation scores for 500 g 0.4

0.3 .1 - ••• - -. - ~ _. --- -. - ~ ••••• -- .'. -. from Miyakoda et al (1972), labelled 1972; from the 0.2 .{ .....{ :- .; ; . 1979/80 daily ECMWF operational forecasts for Dec. 1979-Feb. -~ .~.------~------~ 1980 labelled 1979/80; and 0.1 --- -$-:- #. --·-f· .----. - .------f·- -.- -·--f----- .. -.- -- 1972 ...... from the daily ECMWF o -+---,.c--~.r__--r----r-.---,-'----;-.-----r.----,.-----i operational forecasts for 2 3 4 5 6 7 8 9 10 Dec. 2001-Feb. 2002 Days labelled Winter 2001/02. 4 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION

hemisphere is only a little lower than the skill at the CONCLUSION same range in the northern hemisphere, and skill levels have recently become very similar at day seven. This Weather does not respect political boundaries and hence corresponds to a gain of some three days in skill for the no single country is self-reliant in regard to meteorolog­ southern hemisphere over the last two decades. ical data. Every meteorological service in the world Extreme climate events such as the 1997-98 El-Niiio­ shares meteorological data. WWW has created impor­ related floods in eastern Africa and the 1998-2000 tant pillars for the provision of worldwide La Niiia-related drought, also in eastern Africa, confirm meteorological services, which address the creation of that many countries are very vulnerable to extreme the data (the GOS); data and product distribution (the climate events. Such events often translate to disasters and GTS); and data processing, analysis and provision of have shown a marked increase in recent years (Figure 6). forecast products (the GDPS). Originally conceived to One quality of the WWW programme which was support NMHSs' operational weather services, over the not originally anticipated, but may prove to be its most years WWW has become integral to many additional valuable asset and strongest reason for its continuance, is activities, not only of NMHSs, but also other interna­ the scope and quality of the data set that WWW is tional organizations. Such activities include generating. These data now form the basis of the meteorological research, seasonal and inter-annual atmospheric component of the Global Climate Observing climate monitoring and prediction, the support to System (GCOS). It is precisely the spatial and temporal private sector meteorological operations and the provi­ ___--'-c.QYerage_oLthes_e_data and the qualiry: control and quality sion of other services not originally envisaged. These assurance processes that have been used for some 40 years will continue to multiply and be of importance in a by WWW in their collection, that are needed to measure world where weather, climate and the research needed climate variability and detect climate change. to predict them become ever more vital.

---Northern hemisphere ---Southern hemisphere Figure 5 - Anomaly correlation coefficients of three-, five- and seven-day ECMWF SOOhPa height forecasts for the Day 5 extratropical northern and 80 southern hemispheres, plotted in the form of annual running means of archived monthly-mean Day7 scores for the period January 1980-September 2002. Values plotted for a 50 particular month are averages over that month and the 11 preceding 40 months. The shading shows the differences in scores 3O+--.--.---r--.--.---r--.--.---r--.--i----r--.-----r--.--.---r--.--.---r--r-- between the two 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 hemispheres at the forecast ranges indicated. (Simmons & Hollingsworth (2002)) 800,------,

700 .hydromet .total

600 -f------

500 -l------

300 Figure 6 - Disasters in Africa for the period 1992­ 200 2001. Hydromet includes droughts, floods and similar weather-related disasters. 100 The total includes addi­ tional disasters like 0 earthquakes, landslides and N (") ,.,. ....

STATUS OF THE WORLD WEATHER WATCH

INTRODUCTION observations over the oceans. The total number of ships recruited by WMO Members for the VOS programme has 1. WWW is an international cooperative programme decreased slightly in recent years, while the number of that oversees the gathering and distribution of meteoro­ ship reports received at MTN centres has remained the logical data and products to WMO Members. This 21st same during the past six years. In May 2002, the total Status Report on Implementation documents the number of active drifting buoys reporting data on the progress and changes made in 2001 and 2002 within its GTS had increased to 750, compared with 612 in 1997. various components. 8. There were about 140 000 AMDAR observations per day in mid-2002. Because about 50 per cent of the GLOBAL OBSERVING SYSTEM (GOS) aircraft proViding AMDAR data fly between Europe and North America and within those continents, there are 2. GOS consists of facilities for making observations still relatively few AMDAR observations over Regions I on land and at sea, and from aircraft and satellites. and 11. 3. Each WMO Regional Association (RA) draws up an 9. During 2002 the space-based component of the RBSN, a regional network of observing stations, to meet GOS was expanded to include appropriate Research and the collective needs of its Members. The level of imple­ Development (R&D) satellite missions. The addition of mentation of RBSN surface stations in 2002 varied from these R&D satellites has resulted in a major expansion of 40 per cent in Region III to 95 per cent in Region VI, satellite systems and associated data and products. The with a global average of 80 per cent. The number of R&D constellation joins the existing geostationary and SYNOP reports actually received at Main polar-orbiting constellations. Telecommunications Network (MTN) centres on the GTS varied regionally, from 50 per cent of those reqUired GLOBAL TELECOMMUNICATION SYSTEM (GTS) in the RBSN of Region I, to 93 per cent in Region VI, AND DATA MANAGEMENT (DM) with a global average of 75 per cent. While surpassing the 2000 level, the overall results showed that almost 10. The GTS is comprised of the MTn Regional one quarter of expected reports were still missing in the Meteorological Telecommunication Networks (RMTNs) international exchange. The main reason for this data and National Meteorological Telecommunication gap continues to be either the absence of observations Networks (NMTNs). The MTN is the backbone of the or telecommunication problems. GTS, connecting regions via major RTHs. The RMTNs 4. In addition to stations in the RBSNs, a large number connect WWW centres within a Region, and the NMTNs of supplementary stations have been established to meet connect meteorological stations or centres of a particu­ regional and national needs. Most of these stations are lar country. automated and record observations hourly. There were a 11. All but one of the 23 MTN circuits are in operation total of 2391 automated stations as of October 2002. with X.25, Transmission Control Protocol/Internet 5. The percentage of upper-air reports actually Protocol (TCP/IP) or a combination of both. Fourteen of received at centres on the GTS varied from 27 per cent the 23 circuits use digital technology at speeds ranging of those reqUired in Region I to 87 per cent in Region IV, from 64 to 256 kbits/s. The migration to pure TCP/IP proViding a global average of 64 per cent. As in preVious operation is continuing. Implementation of the years, the number of southern hemisphere stations Improved MTN project (IMTN) has commenced, with 10 making two observations per day is considerably less WMCs and RTHs interconnected through advanced than those in the northern hemisphere. Gaps in upper­ data-communications network services. air data coverage persisted over Africa, Asia, South 12. Significant progress has been made in the imple­ America, and the oceans, mainly due to obsolete equip­ mentation of RMTNs and connecting these to the MTN, ment and lack of consumables. but serious shortcomings still exist at the regional and 6. New Regional Basic Climatological Networks national levels in some Regions. In Region I, several GTS (RBCNs) have been established for each Region. As of circuits were upgraded via leased lines, satellite-based July 2002, all Regions, including the Antarctic, comprise systems or public data networks, including the Internet. a total of 3086 stations. 'Of these, 2575 are listed as There are planned upgrades to the satellite-based CLlMAT stations and 511 as CLlMAT TEMP stations. Meteorological Data Distribution (MDD) and RETIM 7. WMO relies on ships, moored and drifting buoys, 2000 data distribution systems. Most of Region 11 GTS and stationary platforms for synoptic and upper-air circuits are operating at medium or high speed, but a 6 WORLD WEATHER WATCH - TWENTY FIRST STATUS REPORT ON IMPLEMENTATION number of low-speed connections remain. There is a industry. The agenda ofthe upcoming WRC-2003 includes plan for an improved RMlN based on a cluster of several items of importance to meteorology, including a networks using cost-effective network services such as threat to parts of the 1675-1690 MHz band used for mete­ Frame Relay network services. Several satellite-based orological satellites and radiosonde operation. data-distribution systems are operating in Region Il, 16. The table driven code forms (TDCF) are still not including MDD and TV-Inform-Meteo. In Region Ill, widely used and many NMCs can therefore not benefit significant progress has been made with the 64 kbit/s from the full range of data and products available within digital connections of RTHs Brasilia and Buenos Aires. WWW. Therefore, a comprehensive plan for a WMO-wide The Regional Meteorological Data Communication migration to TDCFs has been developed. TDCFs offer Network (RMDCN) is entering its implementation many advantages over traditional alphanumeric codes phase. All 13 NMCs in the Region receive World Area (TAC), including increases in data quality and quantity. Forecast System (WAFS) and operational meteorological The migration is planned to be very flexible, with imple­ (OPMET) information via the International Satellite mentation dates spanning the period 2005-2015. Communications System (ISCS). In Region V, progress was made with the implementation of Frame Relay GLOBAL DATA·PROCESSING SYSTEM (GDPS) services. Several satellite-based telecommunications systems were either expanded or upgraded. Use of the 17. The GDPS generates nearly all the NWP products Internet also increased, particularly for the linking of reqUired by Members. It is made up of WMCs, Regional small Pacific nations. In Region VI, the RMDCN inter­ Specialized Meteorological Centres (RSMCs) and NMCs. connects 33 RTHs and NMCs using a shared, 18. Most RSMC operations have shown sustained commercially-provided managed network service to improvement by enhancing their forecast systems and which NMCs and RTHs connect via a national access computer facilities, although some RSMCs in Region I point. The RMDCN has proven to be highly reliable and still need to upgrade their data-processing eqUipment. very secure. The other RA VI Members are expected to 19. Global models are now running at 16 GDPS join the RMDCN as soon as possible. centres; 67 centres run NWP models operationally; 34 13. Satellite-based multipoint telecommunication run Limited Area Models (LAMs) (with resolution coarser systems are an essential component of the GTS, and than 35 km); and 44 run mesoscale models (resolution each WMO Region is completely covered by at least one 35 km and finer). Several centres are running Ensemble satellite distribution system. These systems effectively Prediction Systems (EPSs) for short-, medium- and complement pOint-to-point circuits, particularly in the extended-range forecasts, and an increasing number of delivery of processed meteorological information to centres are using EPS for long-range forecasting. NMCs. 20. Steady progress has been made in RSMC forecasts of 14. High frequency (HF) radio broadcasts have high tropical cyclone positions. Most RSMCs apply statistical recurrent operational costs and limited efficiency; conse­ models to supplement deterministic numerical models. quently, several Members have discontinued their 21. Eight RSMCs share the responsibility for dissemi­ operation. In some instances, these have been replaced nating atmospheric transport model products in the by satellite distribution systems. framework of the international coordinated response 15. The radio frequency bands allocated to meteorolog­ plans for nuclear emergencies, thus achieving a global ical aids (radiosondes, satellites, radars and wind profilers) coverage. were successfully protected at the World 22. The monthly exchange of forecast verification Radiocommunication Conference 2000 (WRC-2000), scores has continued among nine GDPS centres. Of the where adequate frequency bands were provided for mete­ six centres whose root-mean-square (RMS) errors are orological requirements. However, the radio frequency reported herein, most showed a small but steady allocations to radiosondes and meteorological satellites improvement in forecasting 72- and 120-hour 500 hPa face a continuing threat by the telecommunications heights over Asia, Europe and Australia/New Zealand. CHAPTER I INTRODUCTION

PURPOSE AND SCOPE OF THE WWW PROGRAMME rapid, reliable collection and distribution of obser­ vational data and processed information; 1. Meteorological services are required for the safety of (c) GDPS, consisting of World, Regional/Specialized life and property, the protection of the environment, and and National Meteorological Centres to provide for the efficiency and economy of a wide range of processed data, analyses and forecast products. weather-sensitive activities. Central to the provision of 4. Implementation and integration of the three core these services is the receipt by NMCs of observational elements are supported through the: data, analyses and forecasts. WWW is the international (a) WWW Data Management (WDM), which cooperative programme that arranges for the gathering addresses standards and practices for the efficient and distribution in real-time, on a worldwide scale, of handling and flow of data and products within meteorological information required by individual the WWW system; Members, by other WMO programmes and relevant (b) WWW System Support Activities (SSA), which programmes of other international organizations. The full provide guidance, assistance and training to those description of the WWW Programme, including the involved related to the planning, development implementation goals, are contained in Chapter 6.1 of the and operation of WWW. Fifth WMO Long-term Plan (2000-2009) (WMO-No. 908). 2. The overall objectives of WWW are: ORGANIZATION OF WWW PROGRAMMES (a) To maintain an effective worldwide integrated system for the collection, processing and rapid 5. The WWW core elements are managed through exchange of meteorological and related environ­ quasi-independent programmes, each supported by a mental data, analyses and forecasts; corresponding organizational unit in the WMO (b) To make available, in real-time and non-real-time, as Secretariat. The SSA are carried out as integral parts of appropriate, observational data, analyses, forecasts the individual core programmes. The approved opera­ and other products to meet the needs of all Members, tional procedures and practices of WWW are given in of other WMO programmes and of relevant the regularly updated Manual on the Global Data-process­ programmes of other international organizations; ing System (WMO-No. 485), Manual on the Global (c) To arrange for the introduction of standard meth­ Observing System (WMO-No. 544), Manual on the Global ods and technology which enable Members to Telecommunication System (WMO-No. 386), Manual on make best use of the WWW system and ensure an Codes (WMO-No. 306), and the associated Guides adequate level of services and also the compatibil­ (Annex 11). ity of systems for cooperation with agencies 6. In addition to the core element and support func­ outside WMO; tion programmes, WWW incorporates five other (d) To provide the basic infrastructure for GCOS and programmes that complement and enhance the WWW other WMO and international programmes for core elements in important specific areas: climate monitoring and studying of climate (a) IMOP, aimed mainly at improving both the accu­ issues. racy and homogeneity of observational data. IMOP is managed under the technical responsibil­ COMPONENTS OF THE WWW SYSTEM ity of the Commission for Instruments and Methods of Observation (CIMO); 3. WWW operates at global, regional and national (b) WMO Satellite Activities, which define environ­ levels. It involves the design, implementation and mental satellite data requirements and services, further development of three closely linked and increas­ and strengthen Members' capabilities to receive ingly integrated core elements: and effectively use satellite data; (a) GOS, consisting of facilities and arrangements for (c) Tropical Cyclone Programme (TCP), designed to making observations at stations on land and at assist NMSs in minimizing loss of life and prop­ sea, and from aircraft, environmental observation erty damage caused by tropical cyclones and satellites and other platforms, is designed to associated phenomena; provide observational data for use in both opera­ (d) Emergency Response Activities (ERA), which assist tional and research work; NMSs to respond effectively to man-made (b) GTS, composed of an increasingly automated environmental emergencies, including nuclear network of telecommunication facilities for the accidents; 8 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION

(e) WMO Antarctic Activities, aimed at promoting range of WMO programmes and related activities of and coordinating the implementation and opera­ international organizations. Within WMO, these tion of the core elements of WWW in the include TCP, the Marine Meteorology and Antarctic. Oceanography Programme (MMOP), the Public Weather 7. The WMO bodies primarily concerned with WWW Services Programme (PWSP), Aeronautical Meteorology, are the Commission for Basic Systems (CBS), CIMO and as well as the World Climate Programme (WCP) and the the RAs. CBS is responsible for technical matters relating Integrated Global Ocean Services System (IGOSS). The to worldwide cooperation in the planning, implementa­ WWW implements and operates the basic infrastructure tion, operation and further development of the WWW for the atmospheric component of GCOS. Of particular system. At the regional level, working groups on the importance is the coordination of plans and activities regional aspects of the WWW assist the RAs in coordi­ between the WWW and international organizations, nating the implementation of WWw. CIMO is notably the Comprehensive Nuclear Test-Ban Treaty responsible for the scientific and technical standards rele­ Organization (CTBTO), Bureau International des Poids vant to instruments and observing systems deployed et Mesures (BIPM), the Food and Agriculture within WMO programmes, primarily WWW. Organization of the United Nations (FAO), International Atomic Energy Agency (lAEA), International Civil RELATIONSHIP OF WWW WITH OTHER Aviation Organization (lCAO), Intergovernmental PROGRAMMES Oceanographic Commission (IOq, International Telecommunication Union (lTU), United Nations 8. WWW proVides both the common infrastructure Environment Programme (UNEP) and World Health and the meteorological database to support a broad Organization (WHO). CHAPTERII

THE GLOBAL OBSERVING SYSTEM

REQUIREMENTS FOR OBSERVATIONAL DATA global, regional and national. The first two data types are mainly used to define the initial conditions of global 1. Although GOS requirements are dictated to a large weather prediction models. Individual countries use degree by the needs of numerical techniques, GOS also national data as supplementary observations for has the responsibility of meeting the needs of WMO nowcasting, severe weather warnings and other special­ programmes dealing with climatology and climate ized services. Each WMO Regional Association draws up change, aviation and agricultural forecasts and environ­ an RBSN to meet the collective needs of its Members. mental quality monitoring. The formulation of data Together, these regional networks form the global reqUirements is an evolving process based on experience network. Generally, surface synoptic stations are with observing systems and improvements in data expected to report every six hours for global exchange assimilation techniques. The process balances user and every three hours for regional exchange, while demand with the technical feasibility of data resolution. upper-air stations are reqUired to report at least twice per The reqUirements for observational data for GDPS day. The details of all stations operated by Members are centres are given in the Manual on the Global Data­ given in Weather Reporting (WMO-No. 9) Volume A, processing System (WMO-No 485). The complete set of which is available on the Internet via the WMO observational requirements for other WMO-supported Operational Information Service (OIS) home page at programmes such as GCOS can be found in the http://www.wmo.ch/web/www/ois/ois-home.htm. Committee on Earth Observation Satellites (CEOS)j WMO online database maintained at http://alto-stra­ IMPLEMENTATION OF THE SURFACE·BASED tus.wmo.ch/sat/stations/SatSystem.html. SUBSYSTEM 2. The frequency and density of observations needed depend on the scale of the meteorological phenomena Surface synoptic observations being analyzed and forecast, whether small scale, mesoscale, large scale, or planetary scale. Short-range 4. The status of implementation of RBSN surface weather forecasts require more frequent observations stations as of 1 October 2002, according to information from a denser network over a limited area in order to prOVided by Members, is presented in Table II-l. The detect small-scale phenomena and their development. level of implementation of the RBSN surface stations in As the length of the forecast period increases, so does 2002 that make the eight observations per day varied the area over which observations are reqUired. from 40 per cent in Region III to 95 per cent in Region 3. The WMO requirements for synoptic observa­ VI, with a global average of 80 per cent. All Regional tional data are generally divided into three categories - Associations have reviewed and modified their RBSNs

Table 11-1 Surface synoptic stations in the RBSN as of1 October 2002 compared to those in 2000. The numbers of stations expected to report every three hours, every six hours, and less frequently, as committed to by Members in Weather Reporting (WMO-No. 9) Volume A, are shown.

WMORegion Required Making the complete Making observations Making some Stations not yet in theRBSN observing programme, at the main hours observations daily established or otherwise at least 8 observations (0000, 0600, 1200 non-operational per day (0000, 0300, and 1800 UTC) 0600,0900,1200, 1500,1800 and 2100 UTC) Number Number % Number % Number % Number % 2000 2002 2000 2002 2002 2000 2002 2002 2000 2002 2002 2000 2002 2002 Region I 593 588 364 399 68% 51 60 10% 165 125 21% 13 4 1% Region II 1195 1234 1101 1126 91% 40 43 3% 43 51 4% 11 14 1% Region III 467 435 166 174 40% 5 2 0% 267 238 55% 29 21 5% Region IV 511 512 382 394 77% 25 32 6% 95 70 14% 9 16 3% Region V 396 394 303 303 77% 53 41 10% 34 39 10% 6 11 3% Region VI 714 769 681 732 95% 8 10 1% 18 20 3% 8 7 1% Antarctica 81 72 71 58 81% 8 9 13% 0 3 4% 2 2 3% GLOBAL 3957 4004 3068 3186 80% 190 197 5% 622 546 14% 78 75 2% Global 1988 4040 3042 75% 238 6% 612 15% 148 4% 10 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION during the past four years and major changes were made 6. In addition to stations in the RBSNs, a large to some networks. In Region I, there has been a slight number of supplementary stations have been estab­ decrease to 588 stations, while in Region VI there has lished to meet mostly regional and national needs. been an increase from 714 stations in 2000 to 769 Table 11-2 shows the increase in the number of stations stations in 2002. Region 11 also increased the number of established since 1988. The number of automatic synoptic stations from 1195 to 1234. Overall, there has stations reached 2391 in October 2002, compared to been an increase of almost 2 per cent in the number of 1948 stations in October 2000, an increase of 22 per cent stations in the RBSN, which reverses the trend for the in two years. All Regions show an increase in the period 1998-2000, when there was a decrease of 4 per number of automatic stations (Figure 11-3), with an cent in RBSN surface stations. increase of more than 100 per cent in Region IV and a 5. The 2002 Annual Global Monitoring of the smaller but significant increase in Region VI. Of these Implementation of the WWW indicates the number of 2391 automatic stations, 577 ofthem are included in the SYNOP reports actually received at centres on the GTS. RBSN. Thus, 15 per cent of surface synoptic stations in Highlights of that report are in Chapter V. Figure V-3 the RBSN are automatic stations. shows that the percentage ofreports received compared to those required from the RBSNs in 2002 ranged from UPPER-AIR OBSERVATIONS 50 per cent in Region I to 93 per cent in Region VI, with a global average of 77 per cent. This is a 2 per cent 7. Some 90 per cent of all established upper-air improvement over the 2000 level. Nevertheless, the 2002 stations are included in the RBSNs. Table 11-3 shows the results show that 23 per cent of reqUired reports are miss­ status of implementation of all upper-air stations in the ing in the international exchange. The main reason for 2002 RBSNs with comparable figures for 2000. It should this data loss is that almost 10 per cent (361 stations) be noted in this table that all upper-air stations are included in the RBSNs did not report during the 2002 radiowind stations (919 stations globally), but that not monitoring period. However the majority of these so­ all upper-air stations are radiosonde stations called 'silent stations' are implemented according to (839 stations globally). The difference between these Weather Reporting (WMO-No. 9) Volume A. Reports are two, namely 80 stations, is the upper-air stations that also not received due to telecommunication problems, make only wind observations. Table 11-3 shows that mainly at the national level, where NMCs are not able to during the two-year period 2000-2002 the number of gather the reports from their stations. Figure 11-1 shows fully operational stations (making two observation per the total number of SYNOP observations received from day) again decreased, continuing the trend from the RBSN stations during October monitoring periods from previous two-year period. Radiowind stations dropped 1992 to 2002. The 2002 result is the highest-ever number from 639 to 625, and radiosonde stations dropped from of SYNOP reports received since the start of monitoring, 595 to 588. However, the implementation as a percent­ and the fifth consecutive year that the number ofreceived age of the RBSN did not change significantly, remaining reports has increased. Figure 11-2 shows the percentage of just over 70 per cent. The geographical distribution SYNOP observations received from each Region, together remains a concern, with the southern hemisphere with the relative surface land area of each Region. It is haVing consistently lower percentages of stations apparent that Region I and Region 11 have a sparcity of proposed for making two observations a day than the SYNOP reports compared to their land areas. Region I has northern hemisphere. There is also a persistence of data­ 23 per cent of the global land but only 6 per cent of the sparse areas over some parts of Africa and South observations, whereas Region VI, with 8 per cent of the America. global land area and 40 per cent of the observations, has 8. The number of upper-air reports actually received a very dense coverage of SYNOP reports. during the 2002 monitoring period at MTN centres

Table II-2 All Surface Synoptic Stations counted for the period 1988 to 2002. The percentage change for the last decade is indicated as well as the numbers of stations making hourly and CUMAT observations.

Year Total number of Making observations at Climate stations 0000 UTC 0600 UTC 1200 UTC 1800 UTC Hourly reporting 1988 9525 6958 7390 7904 7255 3849 1830 1990 9649 7016 7483 7999 7323 3965 2247 1992 9762 7168 7597 8065 7420 4162 2264 1994 9950 7314 7786 8313 7634 4449 2310 1996 10106 7469 7898 8460 7776 4585 2528 1998 10 214 7697 8017 8595 7926 4712 2499 2000 10602 8072 8376 8974 8333 5027 2729 2002 10932 8365 8661 9247 8617 5422 2628

Increase since 1988 15% 20% 17% 17% 19% 41% 44% THE GLOBAL OBSERVING SYSTEM 11

12500 -,------,

12000 +------.------

10000

Figure H-l - SYNOP reports 9500 received on the GTS from RBSN 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 stations during the 15 day 2002 Year AGM period

45% ,------;======;;------l I m relative percentage of Region's surface area 40% -I-'------1 m% SYNOP received i 35% +- ----1---'0='-·~%'__T~E::'M"'P---"'re~c~ei~ve':"d:'.-. - __..J_---~

30% -j------~ Ql 1§' 25% -j------c

~ 20% I [L 15%

10% Figure H-2 - The percentage of 5% SYNOP and TEMP observations for each Region received dUring 0% the 2002 AGM, together with Region I Region 11 Region III Region IV Region V Region VI the relative surface land area of WMO Region each Region

Table II-3 Status of implementation of RBSN upper-air stations showing the number of stations requested and the number and percentages implemented for 2002 compared to 2000. 0N = Radiowind, R = Radiosonde)

WMO Requested in Making observations at Making one observation Stations not yet established or Region theRBSN 0000 and 1200 UTC per day otherwise non-operational Number Number % Number % Number % 2000 2002 2000 2002 2002 2000 2002 2002 2000 2002 2002 w R WRWRWR WR WRW R WRW RW R W R

Region I 113 96 106 91 31 25 24 24 23% 26% 51 47 53 48 50% 53% 31 24 29 19 27% 21%

Region II 327 296 327 294 256 252 250 241 76% 82% 31 35 37 42 11% 14% 40 9 40 11 12% 4%

Region III 56 56 58 58 16 16 15 15 26% 26% 25 25 34 34 59% 59% 15 15 9 9 16% 16%

Region IV 143 142 143 142 124 124 126 126 88% 89% 11 11 11 11 8% 8% 8 7 6 5 4% 4%

Region V 124 93 118 87 77 44 81 54 69% 62% 33 43 34 38 29% 44% 14 6 3 5 3% 6%

Region VI 143 143 136 136 127 126 123 122 90% 90% 10 11 9 10 7% 7% 6 6 4 4 3% 3%

Antarctic 13 13 13 12 8 8 6 6 46% 50% 5 5 6 6 46% 50% 0 0 1 0 8% 0%

GLOBAL 919 839 901 820 639 595 625 588 69% 72% 166 177 184 189 20% 23% 114 67 92 53 10% 6%

GLOBAL 988 855 720 642 73% 72% 147 166 15% 19% 121 87 12% 10% 1988 12 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION

1000 --.----..-- -.---..- -- --.------.-- -- - ..-.- --.-..-.-·..·__ ··..·_·_..·_..·__···········.._··.. ·1

800

(J) c: o "" 600 +------10;'•.• ~ '0 :;; ~ 400 -t------::J Z

200 Figure II-3 - Number of Automatic Weather Stations (AWS) operational in 2000 and 0 2002 according to Weather RAI RAil RA III RA IV RAV RA VI Antarctic Reporting (WMO-No. 9) WMORegion Volume A

1200

1150

(J) 1100 c: 0 ""Cl! ~ 1050 (J) .0 0 ] 1000 .0 E z::J 950

900 Figure II-4 - TEMP reports 850 received on the GTS from RBSN 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 stations during the IS-day 2002 Year AGM period varied from 27 per cent of those required from the apparent that there is a definite lack of coverage of TEMP RBSNs in Region I to 87 per cent in Region IV; with a observations in Region I. global average of 64 per cent (Figure V-8). For the third year in a row, the total number of TEMP reports received CLIMATOLOGICAL OBSERVATIONS has decreased, albeit only slightly. However, rationaliza­ tion of the regional upper-air stations included in the 10. New RBCNs have been established for each RBSNs from 839 radiosonde stations in 2000 to 820 Region. This network includes the GCOS Surface radiosonde stations in 2002, caused the global average of Network (GSN) and the GCOS Upper-Air Network received observations to increase slightly. In addition, (GUAN) stations, supplemented by other CLlMAT and 19 per cent of the radiosonde network (156 stations) CLlMAT TEMP reporting stations needed to meet national included in the RBSNs did not report during the 2002 and regional requirements. These RBCN stations now monitoring period, even though the majority of these serve as the target list for WWW monitoring. As of July 'silent stations' are implemented according to Weather 2002, these networks comprised a total of 3086 stations, Reporting (WMO-No. 9) Volume A. of which 2575 are CLlMAT stations and 511 are CLlMAT 9. Figure II-4 shows the total number of RBSN TEMP TEMP stations. The regional breakdown is shown in observations received during October monitoring periods Table II-4. from 1992 to 2002. From this figure it is apparent that 11. The number of CLlMAT and CLlMAT TEMP reports the number of TEMP reports received is about 10 per cent available during the last decade was far from an ideal lower than during the early 1990s. Big gaps in upper-air network and reached only about 40 per cent of the total reports still persist in certain regions of Africa, South number of RBSN stations. In addition, about 20 per cent America, and Central and Northern Asia, mainly due to of stations initially selected for the GSN were not in the lack of consumables and obsolete equipment. Figure II-2 RBSNs and consequently were not monitored during shows the percentage of TEMP observations received WWW monitoring. The establishment of the new from each Region, together with the relative surface land RBCNs, which are now included in the WWW monitor­ area of each Region. As for SYNOP observations, it is ing, is expected to increase the availability of CLlMAT THE GLOBAL OBSERVING SYSTEM 13

100

90 - ---- 0 -0. il Iy·v 0 0 to 0 0 0 -0- -0. 0'''0 .f"\. .1'"\. A 80 N - - 1/ 1,,1 l.eJ. I... .J,. lo- O d ~ _I'- --1-. ~~ ~ re- I':- -re 70 .. .. I- 0 18:~ .. lP- lP- o re' lA 1,,- ~ l,e- ~ ,..I ~ :'~r .. ~ In Ij, ~ ~ I~ ,Q; .../ 60 - -. 1\-11 lAt - F& ,., F8 ,. ~ V .. -. .. ~ '-' £r I~ .. ~ I?r RI.. ~ I1 I'*' J).. 10' .... I~ I'-. ..().~ (f L-. f). JY, ~ IrJ i\.. 50 J\ l-d M - --' f...I '-' ...... ~ ~ -. ~Iv ~ re I~~ ITV" .(). 0, -- ~ .. ll:'-' 0 0, I/ot-o 1\ -. J ~~ KJ R Ip- 40 ~ ~ - 11:;1" ~'k .. l/'" V" -v -u lD, V"" """'!I\ ~ • -0, ~. ) ('o- f'I [r ~- 'rei 1.--0, ~ ra, ~ ·v ~ ~ f)'" I\t ft V h 30 DJ 'c5 u.. ~ v '""",, rx V\ I'~ '0' ~ }( ~ 20 d

10 I

o I I o 0 0 0 0 0 000 o 0 0 000000000000 g 8 8 g g g g g g 888 0 0 0 0 0 0 0 0 0 0 0 0 NNNNNNNNN (\J (\J (\J (\J (\J (\J (\J (\J (\J (\J (\J (\J (\J (\J (\J I- > () Z al a:: a:: ...J ~ a. I- () ~ z ~ > ~ ffi ~ ~ ~ 5 5 ~«~ fu () 0 w « w « a. ~ ~ w () 0 w , ~ ~ « ~ " w o z 0 , ~ ~ « ~ , , « w 0 Z 0

- RAI -+-RAIII -+- RA V -0- Antarctica - RAil -RAIV -+- RA VI Total

Figure Il-S - Percentage of CLlMAT reports received from GSN-stations, January 2000 - June 2002

250000--.------,

200 000 ------. ------. --., ------. ----....------..- --..------. --. ------. -.-. -- -- -...

150 000 ---- .._..-..------. --. ------.. -. --- -- ... ------. ------... ------. -. -- ---. ------..--.----

100 000 ------. ------. ------. --. -. ------

50 000 ------""

o 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

Figure 11-6 - Number of monthly air pressure reports from drifting buoys during the period 1992 to 2002 (from ECMWF) 14 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION

l< '" ",'" " '" '" '" I'l.'" " '" lie "'""~:lE "k '" ~ " "",./ft·· ..c... ' " .. '" " ..

Figure I1-7 - Automated Shipboard Aerological Programme (ASAP) coverage during 2001 (from Meteo-France)

Figure I1-8 - Satellite system coordinated within the Coordination Group for Meteorological Satellites (CGMS), including geostationary, polar-orbiting and research and development satellites THE GLOBAL OBSERVING SYSTEM 15

Table 11-4 ships adequately cover the major shipping lanes. Moored Status of Regional Basic CIimatoIogicaI Network (RBCN) and drifting buoys play a very important role in provid­ (as ofJuly 2002) ing observations from the large data-void ocean areas. 14. The total number of ships recruited by Members WMO Region CLlMAT CLIMATTEMP for the WMO VOS programme and actively reporting Region I 616 19 observations has decreased in recent years. At present, Region II 593 194 52 Members operate VOS, with the total number of Region III 344 49 active ships at the end of 2001 being 6967. The number Region IV 242 72 of Ship reports received at MTN centres has not changed Region V 188 74 significantly during the past six years, and over this Region VI 520 91 period has averaged about 5500 reports for all hours per Antarctica 72 12 day. This can also be seen in Figure V-16, where SMM Global 2575 511 monitoring at 0000, 0600, 1200 and 1800 UTC during the period October 1996 to October 2002 showed a fairly and CLlMAT TEMP reports, both overall and for GSN and constant reception of about 2700 reports per day. During GUAN requirements. Monitoring conducted in 2002 of this same monitoring period, the daily receipt of SHIP the number of reports received compared to the number TEMP reports has remained essentially constant at 15 to of reports expected from the RBCNs showed that 62 per 20 reports per day. cent of the CLlMAT reports and 71 per cent of the 15. There has been a continuing increase in the CLlMAT TEMP were received. Comparison with previous deployment of other types of sea stations. By May 2002, years is difficult, but there does appear to be a consider­ the total number of active drifting buoys deployed able improvement in the number of reports received in globally and reporting data to the GTS was around 750, 2002 compared to previous years. compared with 612 in 1997. Coverage of buoy reports 12. The GUAN, composed of 150 stations, and the during the October 2002 monitoring period can be seen GSN, composed of 989 stations, were established in 1996 in Figure V-B. During the past 12 years the number of and 1999, respectively, to provide a global backbone air pressure reports from drifting buoys increased network of homogeneous, long-term and high-quality fivefold, from approximately 30 000 to around 200 000 atmospheric observations. For both networks, monitoring per month in mid-2002 (Figure II-6). The collection of centres have been established. The GUAN performance is buoy data via satellite is carried out mostly through the monitored by the UK Meteorological Office (UKMO) Argos System, a cooperative undertaking between the Hadley Centre with respect to monthly CLlMAT TEMP Centre nationale d'etudes spatiales (CNES) of France, and reports and by the National Climate Data Center (NCDC) the National Oceanic and Atmospheric Administration in Asheville, United States with respect to TEMP reports. (NOAA) of the United States. The GSN is monitored jointly by theJapan Meteorological 16. Automated marine stations on moored buoys and Agency OMA) with an emphasis on temperature, and the fixed platforms supplement the GOS to an increasing Deutscher Wetterdienst (DWD) with an emphasis on extent. In addition to meteorological parameters, these precipitation. With regard to GUAN, an improvement in stations provide oceanographic and other environmen­ the availability of CLlMAT TEMP reports was reported, with tal data, including wave height and direction, sea 77 per cent received at the Hadley Centre in June 2002. temperatures, water and air pollution data, and surface During the period July-December 2002, NCDC received and underwater currents. A total of 224 fully automated 93 per cent of all TEMP reports from GUAN stations, from systems were operating under the ASAP during 2001. which they can create monthly statistics if necessary. These systems, together with related sounding units, With regard to the availability of CLlMAT reports from were operated by nine countries, primarily in the North GSN stations, since the beginning of monitoring in Atlantic, although there were some observations from January 2000, an improvement in the performance of the all ocean basins. The ASAP Panel has implemented a GSN had been noted, reaching 60 per cent in June 2002. round-the-world ASAP line, which proVides much­ Figure II-5 shows the availability of CLlMAT messages on a needed upper-air data from southern hemisphere waters. regional basis for the period 2000 to 2002, where big ASAP coverage in 2001 is shown in Figure II-7. differences in the performance between the regions are evident. Reasons for this include effects from changes in AIRCRAFT METEOROLOGICAL OBSERVATIONS the network design, improvements of the monitoring software, and actions taken by NMHSs responsible for 17. AIREP has been declining (Figure V-16) as disseminating CLlMAT reports from GSN stations as a reac­ automated reporting of meteorological data from aircraft tion to monitoring results. has been increasing. Monitoring in 2002 showed that less than 4000 AIREP reports were exchanged daily on the MARINE METEOROLOGICAL OBSERVATIONS GTS. The volume of Aircraft to Satellite Data AcqUisition and Relay (ASDAR) and Aircraft Communications 13. WMO relies on ships, moored and drifting buoys, Addressing and Reporting System (ACARS) data, and stationary platforms for synoptic surface and upper­ collectively known as AMDAR, increased from about air observations over ocean areas. Observations from 78 000 observations in 2000 to about 140000 in 2002. 16 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION

Over 50 per cent of the aircraft prOViding AMDAR data fly OTHER OBSERVATION STATIONS/SYSTEMS between Europe and North America and within these continents. While only Germany and the United States 19. Observations from weather radar stations, along exchange data in BUFR format and seven other countries with geostationary satellite data, constitute one of the in the text-based FM42 AMDAR format, BUFR data forms best means of studying small-scale and mesoscale cloud more than 80 per cent of data exchanged on the GTS. precipitation systems. They are essential for the effective BUFR coded reports are not yet counted dUring monitoring and reliable detection, tracking and forecasting/warning periods, but conventional character coded reports are of dangerous weather phenomena such as tropical monitored. There is a smaller, but significant, number of cyclones and tornadoes. In some countries, systems AMDAR reports over Australasia, Asia and South Africa. combining the output of radar networks and the infor­ ASDAR data, from a small number of aircraft, continues to mation received from geostationary meteorological be provided from data-sparse areas of the world such as satellites are in operational use, or are at an advanced South America, the central Atlantic, Africa, and parts of stage of development. Combined systems such as these Asia. ASDAR aircraft also contribute data to parts of the now number in the low thousands. In addition, some north Atlantic not covered by AMDAR aircraft. Automated Members have indicated that they are operating air reporting from the North Atlantic and Pacific regions Spherics Detection Systems for the detection and loca­ carried out in association with the ICAO Automated tion of lightning flashes. The increasing skill of Dependent Surveillance (ADS) transmitted about 6000 forecasters, along with the availability of radar, satellite, meteorological reports per month to the World Area and now Spherics data, both in situ and through the Forecast Centres (WAFCs) in London and Washington. GTS and the Internet, have enhanced the quality of The Meteorological Data Collection and Reporting System severe weather prediction and warning where these data (MDCRS) in the United States received more than 70 000 are made available. Wind-profiling and Doppler radars automated air reports per day. Over half of these reports are proving to be extremely valuable in proViding data were obtained at cruising level, but an increasing number of high resolution in both space and time, especially in of reports were being obtained during the ascent and the lower layers of the atmosphere. Wind-profilers are descent phases of flights. These data were distributed via especially useful in making observations at times the GTS in WMO standard BUFR format. The Canadian­ between balloon-borne soundings, and have great AMDAR Data Acquisition System (C-ADAS) became potential as a part of integrated networks. Doppler operational inJuly 2002. Twenty aircraft servicing Eastern radars are used extensively as part of national and, Canada are reporting AMDAR observations and these will increasingly, of regional networks, mainly for short­ shortly be distributed over the GTS. In Region V in 2001, range forecasting of severe weather phenomena. the number of AMDAR-equipped aircraft operating in Particularly useful is the Doppler radar capability of Australia was reduced by 26 as the result of airline making wind measurements and estimates of rainfall restructuring. These aircraft had flown predominantly amounts. domestic routes. This resulted in about a 55 per cent reduction in the number of AMDAR observations over THE SPACE-BASED SUBSYSTEM Australia. New Zealand's programme of four AMDAR­ eqUipped aircraft contributes about 27 000 observations 20. During 2002, the most significant and important per month. In Region VI, the European Meteorological change since the early 1990s occurred. The space-based Services Network (EUMETNET) AMDAR (E-AMDAR) component of GOS expanded to include appropriate R&D programme, involVing 18 European countries, continued satellite missions. In 2001, WMO's Executive Council to increase observations and improve data quality. The E­ approved the expanSion. In 2002 WMO was informed by AMDAR programme generated about 25 000 reports per several R&D space agencies of their commitment to day from more than 100 airports. participate in the space-based component of the GOS. In 18. Several countries are either actively planning or particular NASA's Aqua, Terra, NPp, QuikSCAT and GPM implementing AMDAR programmes including the missions; ESA's ENVISAT mission; NASDA's ADEOS Il and Russian Federation, Saudi Arabia, China, Hong Kong GCOM series; and Rosavikosmos's research instruments China, Japan, the Republic of Korea, Chile and on board ROSHYDROMET's operational METEOR 3M N1 Argentina. South Africa is planning to extend its national satellite as well as on its future Ocean series are all now programme into a regional programme with the partici­ part of the R&D constellation. This major expanSion of pation of other countries from the Southern Africa satellite systems and associated data and products can be region. Saudi Arabia is also leading the development of a seen in Figure Il-8 regional programme expected to involve other countries 21. The meteorological satellites comprising the pres­ in the Middle East region. An agreement has been ent space-based component of GOS, both polar-orbiting completed between the ASECNA group of countries in and geostationary, have continued to prove invaluable Central and West Africa and Madagascar and the E­ to WMO NMHSs through the provision of a multitude of AMDAR programme for the implementation of a services including imagery, soundings, data collection programme of targeted AMDAR observations to be and data distribution. During 2002, the space-based proVided by European aircraft. This programme is constellation, in addition to the R&D satellites, was expected to commence in mid 2003. comprised of the follOWing geostationary and THE GLOBAL OBSERVING SYSTEM 17 polar-orbiting satellites: GOES-8, GOES-lO, NOM-IS, IMPLEMENTATION GOALS FOR SATELLITE NOAA-16 and NOAA-17 operated by the United States; RECEIVERS GMS-S operated by Japan; GOMS N-l, METEOR 2-20, METEOR 2-21, METEOR 3-S and METEOR 3M NI oper­ 23. Both polar-orbiting and geostationary satellites ated by the Russian Federation; Meteosat-S, Meteosat-6 proVide visible and infrared cloud images. Geostationary and Meteosat-7 operated by the European Organization satellites also proVide water vapour images and indica­ for the Exploitation of Meteorological Satellites tions of wind structure. The latter are of particular value (EUMETSAT); and FY-2B, FY-lC and FY-lD operated by in the tropiCS. Polar-orbiting satellites are eqUipped with China. NOAA-lS, NOAA-16 and NOM-17 are the first of vertical profile radiometers for temperature and humidity the TIROS N series to fly the Advanced TIROS soundings. Both types of satellite operate data collection Operational Vertical Sounder (ATOVS) including the and dissemination services. It should be noted that the Advance Microwave Sounding Unit (AMSU). There were goals for the percentage for WMO Members are 100 per several satellite launches in 2002; China launched FY-1D cent for polar-orbiting satellite data receivers, either in May, NOAAjNational Environmental Satellite, Data Automatic Picture Transmission (APT) or High-Resolution and Information Service (NESDIS) (United States) Picture Transmission (HRPT); and 100 per cent for geosta­ launched NOAA-17 in June and EUMETSAT launched tionary satellite receivers, either Weather Facsimile the first of the Meteosat Second Generation (MSG) satel­ (WEFAX) or High Resolution (HR). This means that each lites, MSG-l, in August. WMO Member should be equipped with at least one 22. Current information related to satellite status, polar-orbiting satellite data receiver and one geostation­ broadcast schedules and future plans can be found on ary satellite data receiver. In general, the percentage of the WMO home page (www.wmo.ch) under 'Satellite implementation has continued to increase in most WMO Activities'. The current and future status of polar-orbit­ Regions during 2001 and 2002. In this regard, WMO ing and geostationary satellites is shown in Table II-S Members have achieved an overall implementation of 88 and Table II c 6 respectively. per cent, 94 per cent, 100 per cent, 76 per cent, 86 per

Table II-S Satellites in orbit as of November 2002 coordinated by the Coordination Group for Meteorological Satellites (CGMS)

Current Polar-Orbiting Satellites Coordinated within CGMS Orbit type (equatorial Satellites in orbit Operator Crossing time Launch Status crossing times) (+ operation mode) (+ altitude) date Sun-synchronous NOM-17 (Op) USA/NOM 1002 (D) 06/02 Functional 'Morning' .. 812 km (0600-1200) NOM-IS (B) USA/NOM 0704 (D) 05/98 Functional (problems with AVHRR + HIRS) (1800-2400) 810 km DMSP-F15 (Op) USA/NOM 2131 (A) 12/99 Defense satellite. Data available to 850 km civilian users through NOM DMSP-FI4 (B) USA/NOM 2014 (A) 04/97 Defense satellite. Data available to 852 km civilian users through NOM DMSP-F12 (L) USA/NOM 1856 (A) 08/94 Defense satellite. Non-operational 850 km (no onboard recorders). RESURS-01-4 (P) Russian 0930 (A) 07/98 Temporarily out of operation Federation 835 km METEOR-3M-Nl (Op) Russian 0915 (A) 10/01 Functional. In commissioning phase Federation 860 km till end of 2002 Sun-synchronous NOM-16 (Op) USA/NOM 1355 (A) 09/00 Funtional, no APT 'Afternoon' 851 km (1200-1600) NOAA-14 (B) USA/NOM 1807 (A) 12/94 Functional, one OBP is unusable (0000-0400) 847 km NOAA-ll (B) USA/NOM 2242 (A) 09/88 SBUV instrumental data limited 843 km Sun-synchronous DMSP-F13 (Op) USA/NOM 1818 (A) 03/95 Defense satellite. Data partly available to 'Early morning' 850 km civilian users through NOAA (0400-0600) (1600-1800) Sun-synchronous FY-1C (B) China 0750 (A) 05/99 Replaced by FY-lD 'Morning' 862 km Non sun-synchronous METEOR 3-N5 (Op) Russian 1200 km 08/91 Functional (APT transmissions of visible or unspecified orbits Federation images)

P = pre-operational B = backup A= North Op = operational L = limited availability D = South 18 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION

Table ll-S Continued Current Geostationary Satellites Coordinated within CGMS Sector Satellites currently Operator Location Launch Status in orbit (+ type) date East-Pacific GOES-10 (Op) USA/NOM 135°W 04/97 Inverted, solar array anomaly, DCP (1800 W-108°W) interrogator on backup West-Atlantic GOES-8 (Op) USA/NOM 75°W 04/94 Replaced by GOES-12 March 2003 (108°W-36°W) Minor sounder anomalies, loss of redun- dancies on some subsystems GOES-ll (B) USA/NOM lOsoW 05/00 In orbit backup, 48 hours availability GOES-9 (L) USA/NOM 106°W 05/95 In standby, attitude control problems GOES-12 (B) USA/NOM lOsoW 07/01 Placed in storage mode in Dec. 2001 East-Atlantic METEOSAT-6 (B) EUMETSAT 9SW 11/93 Rapid Scanning Service, minor gain (36°W-36°E) anomaly on infrared imager METEOSAT-7 (Op) EUMETSAT 0° 02/97 Functional METEOSAT-80p) EUMETSAT 0° 08/02 Functional Indian Ocean METEOSAT-S (Op) EUMETSAT 63°E 03/91 Indian Ocean data coverage (IODC), (36°E-108°E) functional but high inclination mode GOMS-N1 (B) Russian 76°E 11/94 Since 09/98 in standby Federation FY-2B (Op) China 105°E 06/00 Image transmission interrupted in eclipse periods FY-2A (B, L) China 86SE 06/97 - INSAT Il-B (B) India 111SE 07/93 Back-up satellite from an inclined orbit mode of operation. IR channel not available INSAT Il-E (Op) India 83°E 04/99 Imagery data from 3 channel CCD payload (lkm res.) available for opera- tional use. 3 channel VHRR not available for use INSAT H-C India 74°E 01/02 No met payload used for dissemination of processed met data in broadcast mode. No WEFAX broadcast capability MTSAT (Op) Japan 74°E 10/02 Dedicated meteorological satellite

West-Pacific GMS-5 (Op) Japan 1400 E 03/95 The back-up of GMS-5 with GOES-9 is (108°£-1800 E) expected to be ready starting in the boreal spring (April) of 2003

P = pre-operational B = backup A= North Op = operational L = limited availability D= South cent and 87 per cent for polar-orbiting and geostationary Revised procedures and formats for the exchange of satellite receivers in WMO Regions I, 1I, Ill, IV, V and VI, monitoring results have been prepared and published in respectively. Over the 10 years starting in 2003, Low Rate Attachment 1I-8 of the Manual on the Global Data-process­ Picture Transmission (LRPT)/ Low Rate Information ing System (WMO-No 485). These include revised and Transmission (LRIT) will replace the present APT/WEFAX updated selection criteria for the determination of receivers. The transition will have a direct and potentially 'suspect' stations. These standardized criteria will assist large impact on existing and planned ground receiving in the intercomparison of the various lead centre reports equipment belonging to WMO Members. in the future. The lead centres produce six-monthly reports on the results of data quality monitoring. These QUALITY OF OBSERVATIONAL DATA reports are distributed to Members so that they can take remedial action as required. AWMO Web site to support 24. The quality of observational data is regularly data quality monitoring is available at http://www.wmo. monitored and reported on under the WWW Data ch/web/www/DPS/Monitoring-home/mon-index.htm. Quality Monitoring Plan as given in the Manual on the Global Observing System (WMO-No. 544). That plan is UPPER·AIR DATA QUALITY based on Recommendation 8 (CBS-IX) approved by the fiftieth session of the Executive Council. The monitor­ 25. As well as regular monthly monitoring of the qual­ ing is carried out by designated lead centres (Table 1I-?). ity of upper-air data conducted by several monitoring THE GLOBAL OBSERVING SYSTEM 19 centres, detailed six-monthly reports are prepared by examples of the types of errors that have been docu­ ECMWF, the designated lead centre for monitoring mented are use of a wrong station elevation, errors in upper-air data. Each report includes details of stations the reduction to mean sea level, use of an incorrect reporting suspect geopotential heights and suspect wind station pressure, and problems with consistency of speed and wind direction, and suspect temperature. In barometer readings. Standardization and quality control addition to monitoring radiosonde data, wind-profiler for Automatic Weather Stations (AWSs) are becoming data is now included in the monitoring. more important with the introduction of new more sophisticated sensors and processing algorithms. Basic LAND·SURFACE DATA QUALITY quality control procedures are applied at the AWS. More extended quality control is applied at national data­ 26. The quality of land-surface reporting continues to processing centres. improve. For some Regions, the number of 'suspect' stations remained effectively constant, but for others MARINE SURFACE DATA QUALITY this number decreased. To assist with consistency, and to allow easy comparison of results from all monitoring 27. RSMC Bracknell, the lead centre for quality moni­ centres, a revised set of monitoring rejection criteria toring of surface marine data, collects statistics on all have been adopted to overcome the discrepancies caused ship call signs and buoy identifiers, from which a list of by centres using disparate selection criteria. Of signifi­ 'suspect' platforms is produced. Currently the six­ cance is the setting of the criteria for suspect minimum monthly lead centre report includes MSLP, wind speed, sea level pressure (MSLP) to 4 hPa. This change will lead wind direction and sea surface temperature (SST). There to closer scrutiny of data quality in the future. Some are plans to expand this list to include air temperature

Table II-6 Future satellites planned for coordination by the Coordination Group for Meteorological Satellites (CGMS)

Future Polar-Orbiting Satellites Coordinated within CGMS

Orbit type Future additional Operator Planned launch date Other information (equatorial crossing times) satellites

Sun-synchronous METOP-1 EUMETSAT 12/2005 (827 km) (0930) AHRPT 'Morning' METOP-2 EUMETSAT 06/2010 (827 km) (0930) AHRPT (0600-1200) METOP-3 EUMETSAT 12/2014 (827 km) (0930) AHRPT (1800-2400)

FY-3A China 2004 (0930) FY-3B China 2006 (0930)

METEOR3M-2 Russian Federation 08/2003 (1030) or (1630) AHRPT

DMSP-S16 USA/NOM OS/2003 (19:54 A) (SSMI/S) DMSP-S18 USA/NOM 10/2006 (SSMI/S)

NPOESS-1 USA/NOM 04/2009 (833 km) (930 D) LRD(AHRPT) NPOESS-4 USA/NOM 11/2015 (833 km) (930 D) LRD(AHRPT)

Sun-synchronous NOM-N USA/NOM 6/2004 (1400) 'Afternoon' NOM-N' USA/NOM 03/2008 (1400) (1200-1600) NPOESS-2 USA/NOM 06/2011 (833 km) (13:30 A) LRD(AHRPT) (0000-0400) NPOESS-5 USA/NOM 01/2018 (833 km) (13:30 A) LRD(AHRPT)

Sun-synchronous DMSP-S17 USA/NOM 10/2004 (SSMI/S) 'Early morning' DMSP-S19 USA/NOM 10/2008 (SSMI/S) (0400-0600) DMSP-S20 USA/NOM 10/2010 (SSMI/S) (1600-1800) NPOESS-3 USA/NOM 04/2013 (833 km) (5:30 D) LRD(AHRPT) NPOESS-6 USA/NOM -2019 (833 km) (5:30 D) LRD(AHRPT) 20 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION

Table 11-6 Continued

Future Geostationary Satellites Coordinated within CGMS

Sector Future additional Operator Planned launch date Other information satellites

East-Pacific GOES-N USA/NOAA 4/2004 135° W or 75° W (1800W-108°W) GOES-O USA/NOAA 04/2005 75°W and West-Atlantic GOES-P USA/NOAA 2008 135°Wor 75°W (1800W-36°W) GOES-R USA/NOAA 2012 135°Wor ?5°W

(East-Atlantic MSG-l EUMETSAT 08/2002 0° (36°W-36°E) MSG-2 EUMETSAT 2004 0° MSG-3 EUMETSAT 2008 0°

Indian Ocean GOMS-N2 Russian Federation 2005 76°E (36°W-108°E) INSAT 3-A India 2/2003 Mixed mission. (Similar to INSAT-2E) INSAT 3-D India 2005 Dedicated meteorological mission

FY-2C China 2004 Improved FY-2 series 5 channel VISSR, LRIT FY-2D China 2006 Improved FY-2 series 5 channel VISSR, LRIT FY-2E China 2009 Improved FY-2 series 5 channel VISSR, LRIT

West-Pacific MTSAT-1R Japan summer 2003 Multi-functional/Transport (108°E-1800E) Satellite l400E MTSAT-2 Japan 2004 (FY) Multi-functional Transport Satellite l400E. Back-up role while MTSAT-1R is Op until 2008. After MTSAT-2 opera- tions, MTSAT-1R will be used as stand-by

Table II-? deviations above a certain level compared with first Lead Centres for monitoring observation data quality guess field) pressure observations over a period of time. Centre Data type Area of The Six-monthly monitoring reports are distributed to responsibility Port Meteorological Officers in those NMSs that have recruited the identified ships, with a request to under­ WMC Washington aircraft and satellite global take remedial action. The effectiveness of both the RSMC/ECMWF upper-air data global monitoring and follow-up actions was confirmed by the RSMC Bracknell surface marine data global fact that the number of ships having consistently bad RSMC Nairobi land surface observations RAI pressure observations fell from 50 to 60 ships per month RSMC Tokyo land surface observations RAIl several years ago to a reasonably constant current level RSMC Buenos Aires land surface observations RA III of 20 to 25 ships per month. The numbers would RSMC Montreal land surface observations RA IV undoubtedly rise again if the monitoring and follow-up WMC Melbourne land surface observations RAV were discontinued. RSMC Offenbach land surface observations RA VI 28. A number of major meteorological centres undertake real-time monitoring of the quality (compared with the first-guess model fields) of variables reported by and relative humidity (which are reqUired for the VOS drifting buoys and received via the GTS. The monitoring Climate Project (VOSClim). An example of the ship results are relayed to buoy operators through an Internet monitoring conducted by RSMC Bracknell is the moni­ mailing list, in accordance with the Quality Control toring of the quality of surface pressure from ships. Guidelines recommended by the Data Buoy Cooperation Monitoring allows the identification of those ships that Panel (DBCP). An example of the positive results of the report consistently low-quality (root mean square (RMS) operation of these gUidelines is seen in the RMS THE GLOBAL OBSERVING SYSTEM 21

Figure H-9 - RMS differ­ 2.5.,------, ence between buoy air pressure observations and the first guess model field for the period 1992 to 2002 (from ECMWF)

0.5 ------.---.------.---.------.------.------..------•• ---.------••

o-I+tti+ltl-tttrH+ttH+ttttt+l+H-H+ttt-H+ttttttti+ltl+I+HItttliH-H+H+tti+ltl+H-H-tttrH+ttH+ttttt+ltt+tt+H-HH+ttH+ttH+ttttl 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 mailing list, in accordance with the Quality Control major centres such as National Centers for Guidelines recommended by the Data Buoy Cooperation Environmental Protection (NCEP), ECMWF, UKMO, Panel (DBCP). An example of the positive results of the Royal Netherlands Meteorological Institute (KNMI), operation of these gUidelines is seen in the RMS Canadian Meteorological Center (CMC), Australia's difference between buoy air pressure observations and National Meteorological Operations Centre (NMOC) and the first-guess model field, which has decreased from the United States Navy's Fleet Numerical Meteorology around 2.0 hPa in 1992 to 1.2 hPa in 2002 (Figure Il-9). and Oceanography Center (FNMOC). These monitoring The mean difference now probably is close to the centres provide statistics and routine reports in non-real­ uncertainty in the model. The improvement in quality time. Errors in wind speed and direction and of these observations is significant and due primarily to temperature observations are identified by the improvements in the quality of the pressure sensors. monitoring centres and traced for corrective action. Inter-aircraft comparisons for ascent and descent data AMDAR DATA QUALITY are performed in some areas, proViding a powerful means for the identification of aircraft that produce 29. AMDAR data quality remains very high with erroneous data. A Web-based reporting system is being frequent exchange of monitoring reports. Operational developed by the responsible agency for monitoring experience continues to show that this constant vigi­ aircraft reports (NCEP). This Web site will proVide lance on a daily basis is essential to ensure erroneous routine public access to monitoring information. A set data are detected and their distribution inhibited. of uniform monitoring rejection criteria has been 30. The monitoring of the quality of AMDAR data on adopted to overcome the wide range of criteria currently a real- or near-real-time basis is carried out by several in use by the various participating centres. CHAPTERIII THE GLOBAL TELECOMMUNICATION SYSTEM AND WWW DATA MANAGEMENT: INFORMATION SYSTEMS AND SERVICES

GLOBAL TELECOMMUNICATION SYSTEM communication systems. The migration to pure TCP/lP, (GTS) with the use of FTP or sockets, is steadily continuing. 3. A project is underway to replace the MTN with data­ 1. The GTS is organized into three levels - the MTN communications network services. The lMTN, as (global), the RMTNs (regional) and the NMTNs depicted in Figure m-2, aims at the full implementation (national). The structure of theGTSis depicted in Figure of the MTN through these advanced network services. Ill-I. The MTN is the backbone of the GTS, intercon­ Implementation hasstarted andin early 2003 10 WMCs necting the six Regions; the RMTNs interconnect the and RTHswere interconnected with a mix of network various regional and national WWW centres within a services and leased point-to-point circuits. Benefits are Region; and the NMTNs connect the meteorological expected for the whole GTS and include improved cost­ stations or centres to the NMCs ofeach country. effeetiveness,connectivity and flexibility.

MAIN TELECOMMUNICATION NETWORK (MTN) REGIONAL METEOROLOGICAL TELECOMMUNICATION NETWORKS (RMTNS) 2. The MTN interconnects the three WMCs and 18 major RTHs. All but one of the 23 MTN circuits are in 4. Significant progress was made in the implementa­ operation with X.25, TCP/IP or a combination of both. tion of RMTNs,although serious shortcomings still exist Most of them were using digital links at speeds from 64 in some Regions at the regional and national levels. to ZS6 kbit/s (14 circuits). An MTN centres are eqUipped Figures m-3 to lH-S depict detailsofthe status of imple­ with computer-based message sWitching and data- mentation of the RMTNs of the six Regions.

Regional National networks networks

Figure lU-I-Structure ofthe Global Telecommunication System 24 WORW WEATHER WATCH - TI"IEN"'TlETH STATUS REPORT ON IMPLEMENTATION The Improved MTN

Figure HI·2 - The improved MTN phase I as ofjanuary 2003

Region I Region II

5. There are eight RTHs in Region I - Cairo, Nairobi, 6. Region n has nine RTHs - Tehran, Tashkent, Lusaka, Pretoria, Dakar, Algiers, BrazzaviUe and Niamey. Novosibirsk, Khabarovsk, To1..)'o, Bangkok, New Delhi, Despite serious economic difficulties, continuous effort Beijing and]eddah. Most Region II GTS circuits are oper­ has enabled some improvement of GTS circuits via ating at medium or high speed} but there are still nine leased lines, satellite-based telecommunications, in centres with low-speed connections, and three that are particular SATCOM, or public data networks, induding not connected. Twelve centres are connected by circuits the Internet. The RMTN plan comprises 88 circuits, of operating at 64 Kbit/s with the TCP/IP protocoL A plan which 69 are in operation. Twenty GTS circuits (jnterre­ for an improved RMTN in Region U has been developed. gional and regional) are now operating at 19.2 to 128 The RMTN, particularly in its eastern and southern kbit/s and 18 are operating at 2.4 to 9.6 kbit/s. Satellite­ parts, has been improved by the continuing implemen­ based data-distribution systems (MDD, RETIM and tation of advanced data communication systems, Satellite distribution system for information relating to including frame-relay services. This has been comple­ air navigation (SADlS) as part of the ICAO aeronautical mented by satellite-based distribution systems such as fixed service (AFS» and the data-collection system METEOSAT, MDD, TV-Inform-Meteo and the satellite EUMETSAT Series of Meteorological Geostationary systems operated by China and India, and by the use of SateUites (METEOSAT)/ Data Collection System (DCS) the Internet. The exchange of data and products on continued to play a crucial role. There are still some point-to-point circuits is complemented by satellite serious shortcomings, in particular at the national level. distribution systems operating in 16 centres. Eight RTHs To address these shortcomings and to enable sustain­ still operate radio broadcast systems. able development, particularly in meteorological data communications, the RA I Strategic Plan for Enhancing Region In WWW Basic Systems has been developed. The Preparation for the Use of MSG in Africa (PUMA) proj­ 7. There are three RTHs In South America -Brasilia, ect for the implementation of MSG receiving stations, Buenos Aires and Maracay. The 64 kbit/s digital funded by the European Commission} and the RETlM connections of Brasilia and Buenos Aires are operational, Africa system, have entered the implementation stage but shortcomings persist at several NMCs with the lack in 2003. of automation and low speed of telecommunication RTH

D NMC

[[:::::::]1 RTHilDtherregion r·······! NMCinothe:tregion ------:r19.2~~------_l:,:;;;;;;;:::!.:===_ _ L...... J MI'N:ircuit ---

Regiand.rcuit ...... Int:en:Egiaeib=it ~96-... ~temet INO,__ ' ...... ReglO~,I (Africa) RMTN 't/.) pain' t--to pointFigurecircuits',,III-3Implementation- (transmission\' speed in kbl s [¥]

~64 ~

[1~~~:N~

~ __ I

r. 1 ,7.2,

!lilB lib~ :0 -:••• ':

E Mmila ~ i : o RTH D NMC

RTHirotherregion ~~ '. 1 .uyNI I~~I r..·····..: -- PhnomPenh NMCinotherregion :' .. r················..·············..··': L...... J i S:ir.gap::lre ~ E KualaLumpur ~ MrN:ircuit .- --I 'lJ:ans:ni.ssicn : : : : •.J@.·£I speaiM:Ji.ts/s 1--1 ._N~, Notinplemented '--I Adliticmdrcuit ,!::I <:? I Nalperational

Figure III-4 - Region II (Asia) RMTN paint-ta-paint circuits implementation (transmission speed in kbit/s) ...... ::::::.": IL~~~J!

o RTH r--, D NMC I_N! J Notinplernented

rI __NO'J NlIperational RTHirotherregion

WMCinothe:rregion ! Regionailultipoinl:i.rcui.t r...... ····] ~ viaatellitl$VSAT) MTN:ircuit === r - -... T:ransroissicn I .:! ~~ I s.r,:e;dltitsls

. Figure III-5 - Region III (South America) RMTN pomt-to-point drcuits implementation (transmission speed in kbit/s) [~~~~~" V '--'-::2~~:,--- _ IMexico r=--,------1_ .:. _:-1------// 9 6 I,*'-i ~:;

o RTH D NMC

_ .. Bogota ,: th"""egion CIl1lectial = M'.lN:ircuit r ~ionailultipointi.rcuit __ REgionadir' t -' VJ.Batellit~VSAT) CUJ.t '-.:.. 38.4kb/sa. 2kb ...... lilteJ::rEgicnailrcui.t . I grans r___19:-2"i 'I:ta1EIIissionspeaiJ

Figure III-6 - Re . point-ta-point' glOn IV (North and Centr~I ~merica) circuits implementation (transmlSSlOn speed inRMTNkbit/s) :~::::::::::::::::::::::::::::: ~1 ~~ NewDelhi fr=~~~~::i~ ;;m\mm=mm"" g:?::::::::::::':I~::::'':

r: J ... \ l:':l·1 "~ : 1---"--- : 1 2.4 X.25 I 1 --r-

o RTH Noumea I D NMC Peal£peecthrough RTHirotherregion JbI:eimt: cemnitted NMCinotherregion 1---\ I_C~~ I InfomatioFate

MI'N:ircuit 1- -I I t! QI NOlilperational Regionafucuit Interregicmilrcuit Additiona:lrcuit

Figure III-7 - Region V (South-West Pacific) RMTN point-to-point circuits implementation (transmission speed in kbit/s) w ••u • .. o lr"~~;~ihi···jl :: :

I FR128 (CIR32) I

:::::::::::::::::::::::::::::::?'I'I•• ;~·· H Mell:ourne II H::::::::::::::::::::::::::::::

o RTH

D NMC

IL::J.1 RTHlmthe:o:egion '---I NMCinotherregion ,.3~.~ 1

M'IN:ircuit fBI/: ,--I I_N'!'I Not::illplanented '--I ,~Q, NtDperational

Figure HI-8 - RMTN Region VI (Europe) point-ta-paint circuits implementation (transmission speed in kbit/s) THE GLOBAL TELECOMMUNICATION SYSTEM AND WWWDATAMANAGEMENT: INORMATION SYSTEMS AND SERVICES 31 circuits. The RA III RMDCN project aims at radical are essential components of the GTS and comprise both modernization of the RMTN, and has entered its commercial telecommunications and environmental implementation phase, with the finalized technical satellites. As shown in Table Ill-1, each WMO Region is specifications and the international tender in the completely covered by at least one satellite distribution beginning of 2003. The 13 NMCs are also equipped with system. These systems effectively complement point-to­ systems for receiving WAFS and OPMET information via point circuits, particularly in the delivery of processed the ISCS operated by the United States. meteorological information to NMCs. Upgrades of several data-distribution systems were implemented or Region IV firmly planned, such as RETIM-2000 and FAX-E. These upgrades will benefit from advanced digital data trans­ 8. Washington is the only RTH in the Region. The missions, such as satellite-based digital video two-way satellite-based network that is integrated into broadcasting (DVB) services, that are a highly cost-effec­ the ISCS in Region IV remained fully operational. The tive solution in terms of recurrent and investment costs transition to a more modern successor ISCS started in for large capacity (multiple 10 mbit/s) meteorological 2003. This includes the migration to TCP/IP procedures data-distribution. as from August 2003 and the replacement of NMCs' 12. Several Members discontinued their HF broadcasts workstations. because of high recurrent operational costs and limited operational effectiveness. In some centres, for example Region V RTH New Delhi, HF broadcasts were replaced with satel­ lite distribution systems through digital audio 9. There are two RTHs in Region V- Melbourne and broadcasting (DAB) techniques. The WorldSpace Radio Wellington. Significant progress was made in the Region and Internet (RANET) experiment over Africa is also V RMTN with the inclusion of several additional GTS based on DAB services. HF broadcasts may still be circuits, particularly in the Pacific. Also important was reqUired in some areas, mainly at the regional level. In the implementation of frame relay services and the the maritime community, more effective distribution expansion and planned upgrades of satellite-based systems, such as International Maritime Satellite System communications, including the ISCS, the DCS of the (INMARSAT), within the framework of the Global Geostationary Meteorological Satellite (GMS) and Maritime Distress and Safety System (GMDSS), are in Geostationary Operational Environmental Satellite rapid development. (GOES) satellites and the GOES Emergency Managers Weather Information Network (EMWIN). There was also an increasing use of the Internet, in particular for the Table III-I collection of observational reports and for linking small Satellite-based telecommunication systems by WMO Region nations in the Pacific. Region Complete or near-complete Partial Region VI coverage by: coverage by:

I ISCS (Atlantic) 10. There are nine RTHs in Region VI - Bracknell, MDD* Norrk6ping, Toulouse, Offenbach, Moscow, Rome, UKSF/WWW Prague, Vienna and Sofia. The RA VI RMDCN, based on a RANET Experiment shared managed network service managed by the RETIM-Africa (planned 2003) RETIM-2000 ECMWF, interconnects 33 RTHs and NMCs at speeds FAX-E ranging from 8 to 256 kbit/s. The RMDCN met both the II UKSF/WWW** ISCS (Pacific) RA VI GTS requirements and the data exchange require­ TV-Inform-Meteo ments between ECMWF and its Member and III EMWIN-E Cooperating States. The RMDCN has proven to be an ISCS (Atlantic) MDD MTSAT excellent example of a cost-effective implementation of ISCS (Pacific) the GTS with very high reliability, full security, guaran­ IV EMWIN-E teed quality of service, easy scalability of capacity and ISCS (Atlantic)* MDD reduced installation and maintenance costs. Some RA VI ISCS (Pacific) Members still operate leased point-to-point GTS circuits. V ISCS (Pacific)* These centres are expected to join the RMDCN soon. MTSAT EMWIN-W Satellite-based distribution systems (FAX-E, RETIM and VI MDD MDD) continue to play an important role in the Region. FAX-E* ISCS (Atlantic) MULTIPOINT TELECOMMUNICATION SERVICES RETIM-2000* VIA SATELLITE AND RADIOBROADCASTS UKSF/WWW * Component of the RMTN 11. There has been extensive implementation of satel­ ** Pilot project lite-based multipoint telecommunication systems. These 32 WORLD WEATHER WATCH - TWENTIETH STATUS REPORT ON IMPLEMENTATION

DATA-COMMUNICATION TECHNIQUES AND 18. TDCFs, with their self-description, flexibility and PROCEDURES expandability, are the solution to satisfy the demands of rapidly evolving science and technology. The TDCFs FM 13. The GTS adopts international standard data­ 94 BUFR and FM 95 CREX offer great advantages in communication services, eqUipment, techniques, comparison with the TACs, such as FM-12 SYNOP and protocols and applications to the largest extent possible, FM-35 TEMP. The reliability of binary data transmission with a view to improving the cost-effectiveness of provides for an increase in data quality and quantity facilities. This results in reduced costs for equipment received at meteorological centres, which will lead to the purchase and maintenance, as well as in reduced human generation of better products. resources needed for development. TCPlIP and the File Transfer Protocol (FTP) have in many instances already WWW operation monitoring procedures replaced the former X.25 and are being introduced at a fast pace. The rapid expansion and evolution of 19. A large and increasing number of WWW centres dedicated telecommunications services, such as frame­ participate in the annual global monitoring (AGM) of relay networks, has enabled very cost-effective upgrades WWW operations. The special MTN monitoring (SMM) to several parts of the GTS. This trend is expected to proVides complementary results enabling more detailed continue. analysis. CBS developed and is refining an integrated 14. The Future WMO Information System (FWIS) is monitoring plan to further improve monitoring proce­ under development as an overarching approach to bring dures. This plan calls for the monitoring of all data together the diverse and divergent information systems types, including binary data forms, and products. that have been developed to meet WMO requirements. The FWIS will be used for the collection and sharing of RADIO FREQUENCIES FOR METEOROLOGICAL information for all WMO and related international ACTIVITIES programmes. 15. There has been an increasing use of the Internet 20. The current radio frequency allocations and regula­ for meteorological and related data exchange, particu­ tory provisions of the ITU Radio Regulations are addressing larly in developing areas. Sparsely populated areas, such requirements for meteorological and related environmen­ as in the Pacific and Indian Oceans, also benefit from tal activities, through specific radiocommunication the Internet in linking the small NMHSs. The Internet, services: Meteorological Aids (radiosondes)j Meteorological despite its shortcomings in availability, delays and secu­ Satellitej Earth Exploration-Satellite, including passive rity, is often the only affordable telecommunication remote sensing; and Radiolocation for weather and wind means for small NMHSs. profiler radars. However, the threat on the full range of radio frequency bands allocated for meteorological and WORLD WEATHER WATCH DATA MANAGEMENT related environmental systems is continuing with the increasing development and expanSion of new commercial Metadata standards radiocommunication systems. WMO, including CBS, NMHSs, meteorological satellite agencies and the WMO 16. Technical data exchange is the key to the success Secretariat, has continued its active participation in lTU­ of WMO Programmes. As the variety and volume of data Radiocommunication and related fora to ensure that increased, the need for describing those data in an meteorological frequency issues are recognized and unambiguous way became essential, and CBS developed supported. A joint ITU-WMO Handbook on Use of Radio a WMO metadata standard, based on the International Spectrum for Meteorology has recently been published. Organization for Standardization (ISO) standard for The agenda for WRC-03 includes items of importance for geographiC metadata. meteorology. In particular, the threat to portions of the band 1675-1690 MHz, which might seriously hamper the Data representation and codes development of meteorological satellites and the operation of radiosondes in Regions 1I, Ill, IV and V, will be discussed. 17. There has been a continuous development of the Utmost importance should also be attached to ensuring WMO codes and code tables, in particular TDCF FM 92 absolute protection of the special bands allocated to space­ GRIB Edition 2, FM 94 BUFR and FM 95 CREX, in response borne passive sensing, which are a unique natural resource to new and evolving reqUirements, including ensemble for atmospheric measurements and have an increasing and long-range forecasts, satellite imagery, radar data importance in meteorology, such as for observations, NWP and transport model products. and climatology. CHAPTER IV

THE GLOBAL DATA-PROCESSING SYSTEM

1. The GDPS consists of WMCs, RSMCs and NMCs. A 4. In Region I, for example, sustained progress was count of each type from 1990 to 2001 is shown in Table made in NWP at RSMCs Algiers, Cairo, Casablanca and IV-I. WMCs Melbourne, Moscow and Washington Tunis, which all run Limited Area Mesoscale Models. continued to develop their Global Forecasting Systems. RSMC Pretoria currently runs a Global and LAM, an EPS WMCs Melbourne and Washington now run an EPS for for medium range, and a General Circulation Model up medium range and long range. Details of the daily to eight months. Other RSMCs in Region I, however, output of products (analyses and forecasts) of WMCs are still need to upgrade their data-processing eqUipment to given in the annual WWW Technical Progress Reports more fully perform the functions reqUired of them. In on the GDPS available from the GDPS Web site at Region 11, the RSMCs with geographical specialization ­ http://www.wmo.ch/web/www/DPS/gdps.html. Beijing, ]eddah, Khabarosk, Novosibirsk, New Delhi and Tokyo - are all running regional models. Two RSMCs, RSMCs WITH GEOGRAPIDCAL SPECIALIZATION Beijing and Tokyo, and the Indian National Centre for Medium Range Weather Forecast, are running global 2. There are 29 GDPS centres implemented as models. Beijing and Tokyo are running EPS for medium­ RSMCs, of which 26 have geographical specialization, range and for extended-range forecasting. RSMCs in namely Algiers, Beijing, Bracknell, Brasilia, Buenos Region III have upgraded their computing capabilities to Aires, Cairo, Dakar, Darwin, ]eddah, Khabarovsk, run better regional models. RSMC Brasilia runs now a Melbourne, Miami, Montreal, Moscow, Nairobi, New mesoscale model. RSMC Bracknell in Region VI is Delhi, Novosibirsk, Offenbach, Pretoria, Rome, disseminating WAFS Products for aviation worldwide Tashkent, Tokyo, Tunis/Casablanca, Washington and through the SADIS satellite dissemination system, Wellington. These centres proVide regional products to which covers Region I and the western part of Region 11. assist NMCs in the forecasting of small-scale, mesoscale and large-scale meteorological systems. They also RSMCs WITH ACTIVITY SPECIALIZATION provide, upon request, meteorological assistance to United Nations humanitarian missions as defined in 5. RSMCs with activity specialization proVide prod­ the Manual on the Global Data-processing System ucts that meet specific requirements, such as (WMO-No. 485), Appendix 1.5. medium-range forecasting products, tropical cyclone 3. The locations of GDPS centres active in NWP, forecasting products and transport model products for including RSMCs, are indicated in Figure IV-I, and the environmental emergency responses. types of models run at those centres are listed in Table IV-2. The processing equipment available at the GDPS MEDIUM-RANGE WEATHER FORECASTING centres, as reported by them, is listed in Table IV-3. Most RSMC operations in all Regions have shown sustained 6. ECMWF was one of the first among the leading improvement, with enhanced forecasting systems and GDPS Centres to implement a four-dimensional varia­ computer facilities, thereby imprOVing the accuracy of tional (4-D VAR) Data Assimilation System, making use their products. of the massive parallel-processor technology. Currently

Table IV-l Number of GDPSs grouped by type (WMCs, RSMCs and NMCs) and by forecast models for the period 1990 to 2001. The categories are exclusive: i.e. the WMCs are not recounted among the RSMCs or the NMCs; RSMCs are not re-counted among the NMCs

Type ofcentre WMC RSMC NMC

Year 90 92 94 96 98 00 01 90 92 94 96 98 00 01 90 92 94 96 98 00 01 Number of centres 3 3 3 3 3 3 3 25 25 25 26 28 28 29 133 141 155 155 156 156 156 Using computers for NWP 3 3 3 3 3 3 3 15 15 16 20 25 26 26 15 19 22 24 29 33 35 Run global model 2 2 2 2 3 3 3 6 7 8 10 10 10 10 0 1 1 2 2 3 3 Run hemispheric model 2 2 2 1 1 1 1 2 1 1 1 0 0 0 0 1 1 0 0 1 1 Run limited area model 3 3 3 3 3 3 3 14 14 14 20 16 14 13 15 19 20 24 18 16 18 Run mesoscale model 0 0 0 1 3 3 3 1 3 4 4 7 10 12 1 4 7 7 14 23 29 34 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION

Figure IV-l - Location of GDPS centres running numerical models and the types of models run at those centres

WMO/ESCAP Panel ESCAP/WMO on Tropic Cyclones Typhoon Committee 400 E 600E 800E 1000E 120'E 1400E 160'E 180' 1600W """"fijil'"""'\'iii!ii""'''F'''''''''''''''-'''''''+====''''''''''''''''''' P''''''''''~='''''''''1I=--= 600 N 600 N

40'N 40'N

200 N 200 N

EQ EQ

200S 20'$

40'S 40'$

RA ITropical Cyclone RA VTropical Cyclone Committee for the South-West Committee for the South Pacific Indian Ocean and South-East Indian Oceans

Figure IV-2 - Areas of responsibility for RSMCs specializing in tropical cyclone forecasting

ECMWF runs a T511L60 global model. In 2002, ECMWF Honolulu is the latest addition to these RSMCs. All increased its products disseminated on the GTS in GRID centres have specifically assigned roles and international and GRIB codes, which are also available on the centre's and regional responsibilities under WWW and the Web site. Table IV-4 prOVides a summary of these prod­ Tropical Cyclone Programme (TCP). Their respective ucts. A 50"member (T255L40 model) EPS for areas of responsibility are shown in Figure IV-2. medium-range is run daily and selected results are made 8. Position forecast verification statistics for warn­ available on the GTS and on the Internet. ings issued by the centres are shown in Figures IV-3 to IV-7, with comparisons with the Climate and Persistence TROPICAL CYCLONE FORECASTING Models (CLIPER) where available. The forecast error statistics for RSMC Nadi are shown in Figure IV-3. 7. RSMCs Nadi, New Delhi and La Reunion are trop­ Position error statistics for New Delhi are shown in ical cyclone centres; RSMCs Honolulu and Miami are Figure IV-4. Figure IV-5 shows the average track errors of hurricane centres, and RSMC Tokyo is a typhoon centre. RSMC Miami in 2000 for the North Atlantic and Eastern THE GLOBAL DATA-PROCESSING SYSTEM 3S

Table IV-2 Types of numerical models running at RSMCs and NMCs (table based on the latest information available from 2000 to 2002) BT Breeding technique MSM = Mesoscale Model (resolution RSMC Regional Specialized Ens Ensemble Prediction System 35 kilometres and finer) Meteorological Centre with: GM Global Model NMC = National Meteorological Centre MRWF Medium-range weather forecasting HM Hemispheric Model RSM Regional Spectral Model Geo = Geographic specialization LAM = Limited Area Model (resolution SV Singular Vector (peturbation TM Transport model responsibilities coarser than 35 kilometres) technique for ensemble) TC = Tropical cyclone responsibilities

REGION I Centre Status Models Resolution Levels Range Dissemination ACMAD Special Centre Access to GM LAM (RSM) 50 km 28 36h NCEP (USA) HARARE Special Centre LA REUNION TCRSMC Full access to GM and MSM ALGIERS Geo. RSMC LAM (ETA) 55 km 24 72h NCEP (USA) MSM (ETA) 25 km 24 48 h NCEP (USA) CAIRO Geo. RSMC MSM (ETA) 35 km 32 120 h NCEP (USA) CASABLANCA Geo. RSMC MSM (ALADIN) 16 km 31 48 h ARPEGE (France) DAKAR Geo. RSMC NAIROBI Geo. RSMC Access to GM LAM (RAMS) 50 km 28 72h NCEP (USA) PRETORIA Geo. RSMC GM (NCEP Anal.) Tl62 L28 7 days GM Ens 16 members BT T62 L28 14 days GM (COLA) T30 L28 8 months LAM (ETA) 48 km 38 48 h NCEP (USA) TUNIS Geo. RSMC MSM (ALADIN) ARPEGE (France)

REGIONll ", Centre Status Models Resolution Levels Range Dissemination BANGKOK NMC GM 100 km 19 72h LAM 50 km 19 72h MSM 17 km 31 36h HANOI NMC MSM (HRM) 28 km 20 72h GME (Germany) HONG KONG, NMC • LAM (ORSM) 3 D-VAR 60 km 36 48 h GSM Gapan) CHINA MSM(ORSM) 20 km 36 24h GSM Gapan) OMAN NMC MSM (ORM28) 28 km 20 48h GME (Germany) MSM (ORM07) 7km 20 48 h GME (Germany) PYONGYANG NMC Hemispheric model (HM) T42 14 96h LAM-RSM 100 km 14 48 h LAM 50 km 18 24 h SEOUL NMC GM (GDAPS) T213 L31 240 h MSM (RDAPS) 30 43 48 h GDAPS MSM (RDAPS) 10 43 24h GDAPS MSM (RDAPS) 5 43 24 h GDAPS Typhoon (KTM) 1/6" 18 72h GDAPS GM Ens. 32 members, BT Tl06 L21 10 days GM Ens. 32 members, BTg Tl06 L21 180 days TEHRAN NMC LAM ULAANBAATAR NMC LAM NCMRWF-INDIA Special Centre GM T80 118 5 days LAM 50 km 18 5 days BEl]ING Geo. andTM GM (3 D-VAR) Tl06 L19 10 days RSMC LAM (HLAFS) OS 20 48 h GM LAM (MTTP Tropical) 50 km 15 48 h GM GM Ens. 31 members SV Tl06 L19 10 days GM Ens. 12 members T63 116 1 month Time lagged over 3 days JEDDAH Geo. RSMC LAM 48 km 48 48 KHABAROVSK Geo. RSMC LAM 75 km 11 48 h NOVOSlBIRSK Geo. RSMC Hemispheric (01) T40 15 48 h TASHKENT Geo. RSMC LAM 250 km 10 36 h Fax NEW DELHI Geo. and TM Full access to GM (NCMRWF) RSMC LAM (LAFS) 3 D-VAR 10 12 48 h NCMRWF TOKYO RSMCGTCTM GM (GSM0103) 3 D-VAR T213 40 216 h GM Ens. 25 members BGM Tl06 40 9 days GM Ens. 26 members BGM Tl06 40 34 days MSM (RSM0103) 3 D-VAR 20 km 40 51 h GSM MSM (RSM0103) 10 km 40 18 h GSM GM Coupled (GCM/OGCM) T42 21 18 months Typhoon (TYM0103) 24 km 25 84 h GSM 36 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION

REGIONllI Centre Status Models Resolution Levels Range Dissemination BOGOTA NMC MSM (MM5) 25 km LIMA NMC LAM (ETA) 48 km 36 120 h AVN MSM (ETA) 25 km 36 48 h AVN CCM3 Coupled Ens 6 months SST, USA 12 members, perturbed SST QUITO NMC MSM(MM5) 25 km SANTIAGO NMC LAM (MM5) 60 km 60 h AVN MSM (MM5) 20 km INPE/CPTEC - Special Centre GM CPTEC/COLA Tl26 28 7 days SAO PAULO LAM (ETA) 40 km 38 60h GM Coupled, Ens. 25 T62 28 6 months members (Random perturbations) BRASILIA Geo. RSMC fun access to GM MSM (MBAR-HRM) 25 km 35 48 h GME (Germany) BUENOS AIRES Geo. RSMC LAM (ARPE) 150 km 10 36 h

REGION IV Centre Status Models Resolution Levels Range Dissemination MONTREAL Geo. and T.M. GM (GEM) 3 D-VAR 0.9° (-100 km) 28 240 hand RSMC 360 h GM Ens. 16 members T95 28 240h Random perturbation and 2 models modified 8 times GM (GEM Regional) Variable mesh 28 48h 0.22° (-24 km) over North America MSM (HIMAP) 10 km 35 30h GM Ens. 5 members T63 23 1 month (24-h time lagged) GM EPS 12 members T63 23 100 days (24-h time lagged, T32 10 two models) MIAMI Geo. and T.M. Fun access to GM RSMC and LAM HCN (hurricane) 0.16° 18 72h WASHINGTON WMCGeo.and GM (AVN) (3 D-VAR) T170 42 7 days T.M.RSMC T62 42 7 to 16 days LAM (RUC) 40 km 40 12 h AVN LAM (NGM) 85 km 16 48 h AVN MSM (HiRes) 3 D-VAR 12 km 60 84 h AVN (Meso-ETA) over Alaska MSM (HiRes) 3 D-VAR 10 km 28 48 h AVN (Meso-ETA) over Hawaii Ens. 10 (SREF) (North 48 km 60 48 h America) (Breeding) Ens. 20 members (Breeding) Tl26 28 84 hours T62 85 hours to 16 days Ens. 20 members T62 28 7 months THE GLOBAL DATA-PROCESSING SYSTEM 37

REGION V Centre Status Models Resolution Levels Range Dissemination SINGAPORE NMC and ASEAN GSM 1.875· 16 240h Special Specialized LAM (FLM) 127 km 12 48 h Meteorological LAM (VFM) 63 km 13 48 h Centre (ASMC) DARWIN Geo. RSMC Full access to GM LAM (TLAPS) 0.375· 19 48 h NADI TC RSMC Access to GM WELLINGTON Geo. RSMC LAM 150 km 8 48 h MELBOURNE WMCGeo. and GM (GASP) T79 29 240h GTS; Fax; SAT; T.M. RSMC OI-lD-VAR LAM (LAPS) 0.375· 29 48 h GASP LAM (TLAPS) 0.375· 29 48 h GASP MSM (SE and SW MESO-LAPS) 0.125· 29 36 h LAPS MSM (Sydney) 0.05· 29 36h LAPS TCLAPS 0.15· 19 48 h GASP Ens. 32 members T119 19 10 days

REGION VI Centre Status Models Resolution Levels Range Dissemination ANKARA NMC Full access to GM (ECMWF) MSM (NCEP/ETAP 27 km 32 48 h ATHENS NMC Full access to GM (ECMWF) LAM (ETA-NMC) 0.21· 32 48 h MSM (nested) 0.07· 32 24 h BELGRADE NMC LAM (ETA 95) 3D-VAR 46 km 32 120h NCEP (USA) MSM (ETA) 23 km 64 48h ETA BETDAGAN NMC MSM (HRM) Bkm 30 48 h GME (Germany) BRATISLAVA NMC MSM (ALADIN/SLOVAKIA) 7.18 km 31 48 h BRUSSELS NMC Full access to GM (ECMWF) MSM (ALADIN) BUCHAREST NMC MSM (ALADIN) 10 km 31 48 h ARPEGE (France) BUDAPEST NMC MSM (ALADIN) 8km 31 48 h ALADIN/LACE (Prague) COPENHAGEN NMC Full access to GM (ECMWF) HIRLAM-G 3D-VAR 0.45· 31 60h ECMWF MSM (HIRLAM-N) (Greenland) 0.15· 31 36 h HIRLAM-G MSM (HIRLAM-E) 0.15· 31 54 h HIRLAM-G MSM (HIRLAM-D) 0.05· 31 36h HIRLAM-G DE BILT NMC Full access to GM (ECMWF) HIRLAM OS 31 48 h DUBLIN NMC Full access to GM (ECMWF) LAM - MSM (HIRLAM) 0.3· 24 48 h HELSINKI NMC Full access to GM (ECMWF) LAM (ATA) 3D-VAR 0.4· (44 km) 31 54 h MSM (ENO) 0.2· (22 km) 31 54 h ATA KIEV NMC LAM LISBOA NMC MSM (ALADIN-PortuguaI) 12.7 km 31 48 h ARPEGE (France) LJUBLJANA NMC MSM (LACE) 11.6 km 31 48 h MADRID NMC Full access to GM (ECMWF) LAM (HIRLAM) OS 31 48 h MINSK NMC LAM 300 km 6 36h MSM (HIRLAM) 0.2· 31 24 h 38 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION REGION VI Centre Status Models Resolution Levels Range Dissemination NORRKOPING NMC Full access to GM (ECMWF) LAM (HIRLAM) 0.4 0 31 48 h MSM (HIRLAM) 0.20 31 36 h OSLO NMC Fun access to GM (ECMWF) LAM OS 31 48 h MSM 0.1 0 31 48 h PRAGUE NMC MSM (ALADIN/LACE) 12 km 31 48 h RIGA NMC LAM (HIRLAM 2) 55 km 16 36 h SOFIA NMC MSM 9km 31 48 h WARSAW NMC MSM (ALADlN) 17 km 31 38 h ARPEGE (France) ZURICH NMC Full access to GM (ECMWF) MSM (LM) 7km 35 48 h GME (Germany) ECMWF RSMC for GM T511 60 240h Medium- GM Ens. 50 members SV T255 40 240h Range GM EPS 30 members Tl59 40 6 months time-lagged over a month coupled OBNINSK TMRSMC Full access to GM, HM and LAM (for RA II) TOULOUSE TMRSMC GM (ARPEGE) Variable mesh 41 96 h (4 D-VAR) T293C3.5 19 km to 230 km GM (ARPEGE) TC T299 41 72h Indian Ocean) MSM (ALADlN) 7.5 km 41 48 h ARPEGE (France) MSM (ALADINTC Indian Ocean) 31 km 41 72h ARPEGE (France) GM (ARPEGE) T63 31 125 days Ens. 10 members OFFENBACH Geo. RSMC GM (GME) 60 km 31 174 h 3 D-VAR MSM (LM) 0.0625 0 (7 km) 35 48 h ROME Geo. RSMC full access to GM (ECMWF) MSM (HRM) 15 km 30 72h GME (Germany) BRACKNELL Geo. and T.M. GM (Unified Model) 3D-VAR 60 km 30 120 h RSMC MSM 0.11 0 38 36 h GM GM Ens. 9 members, 2.5 0 19 4 months 6 hours time-lagged GM Ens. 9 members, 2.5 0 19 2 years 6 hours time-lagged MOSCOW WMC, Geo. GSM 3D-VAR T85 31 240 h RSMC LAM 75 km 10 48 h GSM MSM 10 km 15 36 h LAM GROUP OF COUNTRIES MSM (ALADlN) 12 km 31 48 h ARPEGE (France) (LACE project): Austria, Croatia, /LACE project Hungary, Czech Republic, run in Prague Slovenia, Slovakia THE GLOBAL DATA-PROCESSING SYSTEM 39

Table IV-3 Main and secondary computers used for data processing at RSMCs and NMCs (based on the latest information from 2000 to 2002)

REGION I Centre Mainframe (number cruncher) Secondary computer(s) Work stations ACMAD INTEL-based servers (AMEDIS system) - PCs SUNSPARC LAGOS pes (AFDOS system) HARARE IBM PSs - PCs ALGIERS CDC DYBER 932 Data General MV7800 3 SUN (Ultra30, SpareS), 10 PCs CAIRO IBM S/390 12 IBM PC 300 GL, 18 PC Pentiurn CASABLANCA J916 CRAY CS6400 - SUNSPARC 1000 SGI - 3 DEC Alpha - Motorola DAKAR PCs NAIROBI IBM VAX3900 - VAX 11/750 SGI - PCs LA REUNION Work stations PRETORIA CRAY J916, CRAY SV1 2SGI Origin 200, 2 SGI Indigo - SUN PCs Enterprise 3000 TUNIS BULL DPS 7000/40 2 MINI - 20 MICROs

REGIONll Centre Mainframe (number cruncher) Secondary computer(s) Work stations ALMATY - - PCs BANGKOK IBM SP2 DATA GEN. MV7800 - 2 HP9000 DECAlpha HONG KONG, SGI CRAY SV1-1A 3 IBM RS/6000 590, SGI Origin 2000, WSs CHINA (16 processors) 2 SUN E450, 2 SGI 02 KARACHI - - PCs OMAN SUN E450 HPC 12 processors - - PYONGYANG - Pentiurn III PC/AT -PS/2 SEOUL NEC SX-5/16A, SX-4/2A HP V2500 (48PE) SUN 2000 TEHRAN - - PCs 486 ULAANBAATAR - - MICRO VAX 3400 NCMRWF-INDIA CRAY XMP-216 - DEC Alpha WSs, SUN Ultra Spare II WSs, SGI Origin 200 and 02 WSs BEI]ING IBM SP, CRAY C92/1-128, , IBM SP2/32 WSs SW1, YN3 JEDDAH - CDC CYBER 962 - 2 CDC 910 3 SG - 4 VAX - 3 CDC KHABAROVSK - COMPAREX 8/830 PCs NOVOSIBIRSK - COMPAREX 8/830 PCs TASHKENT - IBM 370 (2 4381 P13) PCs NEW DELHI CDC CYBER 2000U CDC 4680 2 VAX 11/730, WS 4 CYBER 910-485, VAX 3400, 5 Pentium II TOKYO HlTACHI S3800/480 HITACHI 3500 3 HlTACHI 3050RX WSs

REGION III Centre Mainframe (number cruncher) Secondary computer(s) Work stations LIMA IBM XP 1000 6 DECAlpha SANTIAGO HP E800 8 Sun Ultra 1/40 INPE(CPTEC- NEC SX 4/8A, NEX SX 3/12R DEC Alpha 3000/500, 2 DEC 62 WSs (DEC, Compaq), 41 PCs SAO PAULO) Alpha 4100 5/300, 7 DEC Alpha 3000/4000 BRASILIA 2 DEC Alpha 3000-300 WSs (10): DECs, SGls BUENOS AIRES SGI Impact 10000 3 SGI INDIGO - IRIS - INDY 40 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION

REGION IV Centre Mainframe (number cruncher) Secondary computer(s) Work stations SAN JOSE 2SUNSPARC VSAT/STAR-HP WS - 12 IBM PCs MONTREAL 2 NEC SX-5/16, NEC SX-4/32 4 SGI Origin 2000 - 10 SGI Origin 200 5 SGI Indigo - 12 SGI-Indy, 35 SGI 230, 36 HP - 45 NCD MIAMI WSs WASHINGTON IBM RS/6000 SP (2336 processors) 2 CRAY J916

REGION V Centre Mainframe (number cruncher) Secondary computer(s) Work stations SINGAPORE NECSX-4 2 FUJITSU M1600, FUJITSU DS90s, SUN Sparcs, SGI Octane/02/ SGl 2000, SUN, VAX Indy, PCs DARWIN WSs NADI 2 IBM RS6000/J50 WSs, PCs WELLINGTON ALPHA- 9 PARALLEL PROCESSORS. WSs MELBOURNE 2NECSX -5 11 HP 9000/800

REGION VI Centre Mainframe (number cmncher) Secondary computer(s) Work stations ANKARA SGI Onyx 2, DEC station 5000/200, Dell Pentium PCs 2VAX3100,8750 ATHENS CONVEX SPP1600-8 SGI: 2 INDIGO - 3 INDY - 1 ULTRA 8 WSs - 1 HP WS - PCs BELGRADE SGI POWER INDIGO RSOOO SGI INDY, 550, 2x02 - 6 PCs BETDAGAN SGI Origin 2000, 2xSGI Origin 200 8 SGI WSs BRATISLAVA HP 9000/720, SUN Sparc HS21 HPs BRUSSELS CRAYJ916 HP servers WSs BUCHAREST DEC PCs BUDAPEST IBM p690 (32 processors) SGI Origin 2000 (16 procesors) HP L2000 SUNWS - PCs K250, K200, C200, D280, B180, J210, 755, 715, 710, DEC 600 COPENHAGEN NEC-SX4 (16 processors) 2 SGI Origin 200 WSs DE BILT SGI Power Challenge, SGI Origin 2000 Compaq clusters - WSs DUBLIN 2 SGI Origin 200, SGI CHALLENGE L- 2 VAX 4200 - VAX 3100 HELSINKI (512 RISC processors) VAX 6240 VAX Clusters - WSs (shared)- SGI Origin 2000 KIEV EC-1061 PCs LISBOA 2 DEC Alpha 2000 4/275 LJUBLJANA VAX 3100/10 PCs MADRID CRAY C-94 VAX - HP/HPUX and SUN/SOLARIS SUNWSs MINSK 2 Intel Celeron 600 3 Intel PIlI, 2 Intel P-Il NORRKOPING CRAY T3E (232 processors) SGI 3800 DEC Alpha servers 29 VAX (Clusters) - 7 DEC - SUN WSs (shared) OSLO CRAYT3E 2 IBM RS6000 3 SGI Origin 2000, 200 VAX 4000-200/3300 DEC3100 Alpha-ZOO PRAGUE NEC SX-4/3A 3 ICL DRS6000 SUN SPARC WKs - 2 TWO - 10 ONE RIGA VAX 3100-40 Pentium 90 SOFIA MOTOROLA SYSTEM WARSAW CRAY YMP4E(shared) IBM AS/400 - RS/600 PCs ZURICH CRAY SV 1-B (shared) SUN (servers) SUNWSs ECMWF IBM Cluster 1600 - 1408 4 IBM RS 6000/R40, IBM SP 4R50, SGls indigo - indy, PCs processors -pSeries 690 3 HP9000 Kseries, 4 SGI Origin 2000 TOULOUSE FUJITSU VPP 5000 (31 processors) HP N4000, HP T600, HP D370, HP C180 OFFENBACH IBM RS/6000 (80 x 16 processors), CRAY J932 - 4 ORIGIN 2000 WSs CRAY T3E 1200 (816 processors) ROME IBM 3090/12E - 438/13P VAX 2 8250 Cluster - 4000 BRACKNELL NEC SX6, CRAY T3Ea (880 2 IBM z800 processors)-CRAY T3Eb (640 processors) MOSCOW CRAYY-MP8E CRAY Y-MPEL 98 - COMPAREX 8/83 2 HP 735 WSs THE GLOBAL DATA-PROCESSING SYSTEM 41

Table IV-4 300 -.------, ECMWF products on the GTS 250 +------j GLOBAL DOMAIN (900N-90oS ): E 200 +------'="-'-"'=~~'------j Mean Sea Level Pressure, 850 hPa Temperature, .s 500 hPa Height, u, v (850, 700, 500, 200 hPa), g 11l 150 +------~~"'--j Relative humidity (850, 700 hPa), iii 1l Analyses and forecasts every 24 hours up to day 7 (168 hours) 11l l5 100 +------:::004~.--=------j "- TROPICAL BELT (3S0N-3S0S): 50 +------~6~=--=------I Divergence and vorticity (700 hPa), Analyses and forecasts every 24 hours up to day 6 (144 hours) 0+----1!~_--___,---...---_--__;r__-____j Oh 12 h 24h lime 36 h 48h 72h BASED ONEPS Figure IV-5 - RSMC Miami official average track errors during Probability of: 2001 for a homogenous sample in the North Atlantic and in the Precipitation >10mm, NH, SH at H+72, 96, 120, 144 Eastern North Pacific Precipitation >20mm, NH, SH at H+72, 96, 120, 144 l Wind gusts >15 ms- NH, SH at H+72, 96, 120, 144 600 ,------..., Wind gusts >25 ms-l NH, SH at H+72, 96, 120, 144 500 +------,P\----:-----j Products are in format FM 47 GRID with SOx5° resolution and in format FM 92 GRlB with 2Sx2S resolution ~ 400 +---~--____:------__J'-----'l_--I-'...--1 g -t6 300 T------''lr--...,...... ;;;;;~-~F~~~-J 1l ~ 200 24-hour 300,------, I.L

100 +------1 250 -t------''a------j

E 200 +------~~~::"'":...... ____-----l 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 ~ Year g T+12 11l 150 +---~------1 1ii Figure IV-6 - RSMC Tokyo mean forecast errors 1l (typhoon positions) of operational forecasts ~ 100 -t---~------~...------j

T+O 50 +------=------1 North Pacific basins. The 2000 official errors for the ...-- . . • • • North Atlantic were nearly 16 per cent smaller than the 0, lO-year average at all time periods, except 72-hour. The g,'O ~ g,'O g,Q> I:>~ 'o ~ 1:>" 1:>" average 2000 forecast for the Eastern North Pacific ~"q; ~"q; ,~ g,'O,,,q; ~ g,~ I:>~ I:>~ pf" 'O ~ ifJ>fJ' R)<:l ~ '),<:l demonstrate skill at all forecast periods and improve­ "q; "q; " "q; "q; 'V Season ment over their long-term average. RSMC Tokyo runs a Typhoon Track Model and also uses its Global Spectral Figure IV-3 - RSMC Nadi position forecast verification statistics Model (GSM) for typhoon tracking. Forecast verification for official warnings issued for typhoons over the South Pacific statistics for RSMC Tokyo are shown in Figure IV-6. Annual mean position errors in 2001 were 155 km for 24-hr forecast, 274 km for 48-hr forecast and 411 km for 72-hr forecast. The position error for the 24-hr forecast 300 -.------, in 2001 is nearly equal to that of 2000. The 48-hr fore­ 250 -t-----=~_=_------j cast was slightly better than the 282 km error in 2000, although the n-hr forecast in 2001 was slightly worse ~2oo -t------...... ,------j than the 404 km error in 2000. Figure IV-7 shows the e RSMC La Reunion mean forecast position errors for trop­ :;; 150 +------'...------1 ical cyclones in the South-West Indian Ocean during the I 1999-2000 cyclone season. ~ 100 +------_r-----I 9. Steady progress has been achieved in forecasting 50 -t------j the tropical cyclone track positions, but work still has to be done with respect to forecasting their intensities. The O~--.__-_--_--_--_--_-___l dynamical model tropical cyclone tracks have become 1995 1996 1997 1998 1999 2000 2001 Year the primary guidance for track forecasting at a number Figure IV-4 - RSMC New Delhi mean forecast position errors for of forecast centres. An example of a product generated tropical cyclones over the Bay of Bengal and the Arabian Sea by a SO-member EPS system giving track forecast and based on RSMC forecast models striking probabilities is given in Figure IV-8. 42 WORLD WEATHER WATCH - TWENlY-FIRST STATUS REPORT ON IMPLEMENTATION

20020827 12 UTC 300 ,------=-----, Probability that ZJW will pass within 65 nm radius during the next 120 hour tracks: black=OPER, green=cmL, blue=EPS 100 250 -1------1

90 ;[200 80 g '" 150 -1------­ Vi 70 ~ 4O'N ~ 100 80

50 50

3O"N 40 o . 12-hour 24-hour 48-hour 30 Time step Figure IV-7 - RSMC La Reunion mean forecast position errors (in 20 kilometres) for tropical cyclones in the South-West Indian Ocean 2O'N during the 1999-2000 cyclone season c.1. =Current Intensity 10 (Dvorak scale) 5 120'1: 14O'E PROVISION OF TRANSPORT MODEL PRODUCTS Figure IV-8 , Example of application of Ensemble Prediction FOR ENVIRONMENTAL EMERGENCY RESPONSES System to tropical cyclone forecasting: Strike probability (within 65 nm) of Typhoon Rusa over the next 120 hours. 10. There are eight RSMCs designated for the provi­ Starting time of the forecast is 27 August 2002 12 UTC. sion of transport model products in cases of man-made Full dots give the observed position over the and other environmental emergencies, in particular period 27 August to 1 September 2002. nuclear emergencies, haze from wild land fires and Substantial progress can be noted in RA III, where chemical incidents. All centres have implemented the Bogota, Lima, Quito and Santiago run mesoscale models. regional and global arrangements for the provision of NMC Lima also runs the CCM3 coupled model in ensem­ products as defined by WMO in the Manual on the ble mode for six-month forecasts. There is still a lack of Global Data-processing System (WMO-No. 485). They real-time data-processing capabilities in the NMCs of provide, upon request, specialized transport/disper­ many developing countries. However, the GTS and satel­ sion/deposition model products in accordance with the lite-based dissemination systems enable NMCs to receive Manual, Appendices I.3 and 11.7. The centres are shown products directly and reliably from WMCs and RSMCs. in Table IV-5. Many centres now have Internet access to selected prod­ ucts that are made available by the GDPS centres. NMCs AND CENTRES WITH SIMILAR FUNCTIONS VERIFICATION OF NUMERICAL WEATHER 11. The location of sites running numerical models at PREDICTION (NWP) NMCs are shown in Figure IV-I. Adetailed listing of these centres, the models they run and the computer equip­ 12. The monthly exchange of verification scores using ment available are given in Table IV-2 and Table IV-3. standards and procedures agreed upon by CBS has Many NMCs, particularly in Regions Il, III and VI, have continued among the centres in Bracknell, Melbourne, well-developed computer capabilities and are able to run Montreal, Moscow, Offenbach, Tokyo, Toulouse, mesoscale models with a resolution of 35 km or finer. Washington and ECMWF. 13. A limited set of basic scores is included in the Table IV-S annual WWW Technical Progress Reports on the GDPS. Centres responsible for provision of Mean annual RMS errors of 500 hPa height against Transport Model Products observations for 72- and 120-hour ranges over Asia, Australia/New Zealand, Europe and North America are Centre Area ofresponsibility plotted in Figure IV-9 for the years 1991 to 2001. The RMS scores for global models running at Bracknell, RSMC Bracknell RA VI and RA I ECMWF, Montreal, Tokyo, Toulouse and Washington RSMC Toulouse RA VI and RA I forecasting centres are shown. RSMC Montreal RA 1II and RA IV 14. All forecast centres continue to show a general RSMC Washington RA 1II and RA IV trend towards improved forecasts over all four areas and RSMC Beijing RAIl for both time ranges. However, the degree of improve­ RSMC Obninsk RAIl ment is not completely uniform. Most noticeable is a RSMC Tokyo RAIl decrease of forecast skill for a number of areas after 1996, RSMC Melbourne RAV in particular the skill decreases over North America and Europe in 1999. Taken collectively, the degradation in THE GLOBAL DATA-PROCESSING SYSTEM 43

90 90

70 70

",80 ",80 ~ ~ ...

90 -,------, 90 -,------, (c) Australia/NZ 72 hours (d) Australia/NZ 120 hours 80 +------1

70 +------1 70~----

'" 60 +------_____j '" 60 +------\---,f-----'''------j ~ ~

40 -(------.::::'~t;::;~

30 +------1

20 -!----,..-----,----,------.------j 1993 1995 1997 1999 2001 1991 1993 1995 1997 1999 2001 Year Year

90 90-,------, (e) Europe 72 hours 80 80 P"'~.,.,q."""':a::~lI<_-----__c::_c_---1

70

",60 '" 60 +------~::,,;::;!_:~.J.t ~ ~

20 2OL------~--___4 1991 1993 1995 1997 1999 2001 1991 1993 1995 1997 1999 2001 Year Year 90 90-,------, (g) North America 72 hours (h) North America 120 hours 80 80+------1

70

",60 '" 60 1--~;:2~s:~~~~~~~~~E~t~t ~ ~ t . ~ 40 :::!'-..... 40 ~------__1 -"- ...... ~ ------...::: 30 -- - 30+------1

20 2O+------~--__I 1991 1993 1995 1997 1999 2001 1991 1993 1995 1997 1999 2001 Year Year ___ Bracknell --+- Montreal -()- Toulouse ---.-ECMWF ---iIE- Tokyo --+-- Washington

Figure IV-9 - Mean annual SOD hPa height errors over Asia, AustralialNew Zealand, Europe and North America 44 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION

9 9

(a) 72 hours 850 hPa (b) 120 hours 850 hPa 8 8 -g7 . 8 ~ ~ ~ 3l a~ ....-:-~ "'- .A l- 8.6 ~ ~ 'l<" -...... f"" -- ... --=:; '" ~ ~ '%.: -; ...... ::!5 - --~ ..... JE: 4 4

3 3 1991 1993 1995 1997 1999 2001 1991 1993 1995 1997 1999 2001 Year Year

13 -,------, 13 -,------,

(c) 72 hours 250 hPa (d) 120 hours 250 hPa 12 -j------j 12 -j---""""""------j

----'8--1---"0 11 +:==t~======1------g 8-t---11 -1___-----'__,-----~~...... ~ m 8. 10 8. 10 -1------"'1f;::;;;;:~~~::::n~'*'=:i-"',-~;:::__j ~'" ~'" ~ 9 ::! 9 -t------~I?'_~==i

8+------~=------~:_;_--~ 8+------1

7-j------,-----,------,-----,----~ 7 -1------,----.------,----.------1 1991 1993 1995 1997 1999 2001 1991 1993 1995 1997 1999 2001 Year Year ___ Bracknell ___ Montreal -4-Toulouse ...... ECMWF """*- Tokyo ___ Washington

Figure IV-lO - 72- and l20-hour mean annual RMS tropical wind speed errors for 8S0 and 250 hPa

80 -,------,

~Usingcomputers for NWP 70 +------1 -a-Global

601------=:::;;;::::;;:;;;;;;;ii#I'~--1.....Hemispheric

~5Or------=:::.-~~~;:_------1_Limited area Q) _Mesoscale -5 40 +------:::;~....~~~=------~~.,___------~~~-1 .8 E ~30I----=------c..tJIL------!

201------="""------1

Figure IV-ll - GDPS centres running 1990 1992 1994 1995 1996 1997 1998 1999 2000 2001 NWP models, and showing the various Year models run in those centres

scores over North America in 1999 is significant. now also very similar, with the most improvement over However, forecast skills over Europe and North America the past 10 years being in the 120-hour forecasts over have shown an improving trend since 1999. Europe. The spread of scores from all centres is now very 15. Asia has traditionally shown the best verification similar for all forecast areas. The spread of scores over scores from all centres, both for the 72- and 120-hour Australia/New Zealand has decreased over the past forecast ranges, but the scores over the Australia/New several years and is now similar to the spread of the Zealand area are now comparable. Three centres have scores over the other forecast areas. shown a decrease in forecast skill for 120-hour forecasts 16. Model forecasts from all centres have shown over Asia in 2001. The height errors over Europe and significant improvements over the lO-year period of North America, both for 72- and 120-hour forecasts are record. The Tokyo 72-hour and 120-hour forecasts over THE GLOBAL DATA-PROCESSING SYSTEM 45

Australia/New Zealand showed marked improvement range predictions, especially over the tropical regions. since 1992, even allowing for the anomalous scores over Global models now run at 16 GDPS centres and 10 of that area in 1996. In general, the increase in forecast skill these disseminate their products on the GTS or through by all centres over the past 10 years is very apparent, satellite systems. Twelve centres now run EPS, which are with the possible exception of a leveling out of skill in applied to all forecasting ranges. the forecasts over Asia. 20. Continuous operations are performed by the 17. Figure IV-lO shows the RMS for tropical winds at Association of South-East Asian Nations (ASEAN) 850 hPa and 250 hPa, respectively, for 72- and 120-hour Specialized Meteorological Centre (ASMC) in Singapore ranges. In general, forecasts of the tropical winds at 850 and by emerging centres, such as the African Centre of hPa are about the same from all cc;ntres, with a spread of Meteorological Applications for Development (ACMAD), less than two metres per second at both 72- and 120­ and the Drought Monitoring Centres (DMCs) in Nairobi hour ranges. The wind error at 850 hPa has decreased by and Harare. These centres have plans to further improve an average of one metre per second over the lO-year their equipment and to train staff at various levels. period of record, whilst at 250 hPa the error has ACMAD disseminates guidance to NMCs that is aimed at decreased by over two metres per second over the same medium-range forecasting, monthly climate bulletins period. The spread of errors is a bit larger at 250 hPa and experimental seasonal outlooks. ACMAD is also than at 850 hPa. working on the development of a regional model using 18. CBS has defined and recommended standard scores boundary conditions obtained from an advanced GDPS to be attached to long-range forecasts and to be used for centre. DMCs Nairobi and Harare proVide decadal quality assessing. The core standardized verification climate diagnosis information with seasonal outlook for system also provides a means of exchange between GDPS East Africa and Southern Mrica. centres, of standardized verification statistics. 21. Sixty-seven GDPS centres currently run NWP models operationally; 35 run LAMs with resolution TECHNICAL DEVELOPMENT AT GDPS CENTRES coarser than 35 km and 44 run mesoscale models with a resolution 35 km and finer. As shown in Figure IV-!! the 19. Advanced GDPS centres have implemented data number of centres running mesoscale models increased assimilation systems with 3-DVAR and even some with dramatically by the end of the last decade. Also note­ 4-DVAR, including direct analysis of satellite radiance worthy is that the number of global models increased and humidity data. Physical parameterization schemes while the number of hemispheric models decreased to (such as convection, cloud and radiation) are constantly two. Some centres have started running high-resolution improved, leading to better very-short-range and short- non-hydrostatic models operationally. 22. Twelve centres prepare operational EPS products Table IV-6 for short-range, medium-range, extended range or long­ Centres producing EPS products - short, medium & range (see Table IV-6 and Table IV-7). For example, extended range Washington runs operationally a 20-member ensemble for medium-range and long-range predictions and a 10­ Centre Number Range Comment member ensemble for short-range predictions. A list of members verification products to be exchanged between centres producing EPS data, both short-range and medium­ ECMWF 50 10 days Singular vectors technique for perturbation of the range, has been developed. A subset of these scores will analysis Table IV-7 Montreal 16 10 days Random perturbations and 2 models Centres producing long-range EPS products 5 1 month 24 hour time lagged Pretoria 16 14 days Breeding technique Centre Number Range Comment members Beijing 32 10 days Singular vectors technique for perturbation of the ECMWF 30 6 months Time lagged analysis Montreal 12 12 1 month Average forecasts lagged 100 days 24 hour time lagged, two over 3 days models Seoul 32 Seoul 32 10 days Breeding technique 6 months Breeding technique Tokyo 25 9 days Breeding of growing modes Sao Paulo 26 1 month technique (BGM) INPE/CPTEC 25 6 months Random perturbations 12 Washington 10 3 days Breeding of growing modes Lima 6 months Perturbed SST 11 14 days technique (BGM) Toulouse 10 4 months - BracknelI 9 4 months 6-hour time lagged Washington 20 6 months - Melbourne 33 10 days Singular vectors technique Bracknell 9 2 years 6-hour time lagged for perturbation of the analysis Melbourne 10 8 months - 46 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION be published in the annual GDPS Technical Progress GDPS centres are actively using NWP results in creating Report. A list of EPS producers and their products is long-range forecasts, including seasonal to interannual available on the GDPS Web site at http://www.wmo.ch/ outlooks. Using ensembles combined with statistical web/www/DPS/EPS-HOME/eps-home.htm. linear regression is the method used in most cases for 23. Several GDPS centres are engaged in climate diag­ monthly and seasonal outlooks. Several centres make nosis activities at global, regional or national levels. available on the Internet extended and long-range fore­ Several centres, including Tokyo, Lima, INPE/CPTEC, casts including national and even global seasonal Melbourne and ECMWF, now operationally run coupled forecasts. Some centres, such as the International ocean-atmosphere models. These produce useful long­ Research Institute for Climate Prediction (IRI) and range forecasts up to seasonal and multi-seasonal NOAA-NCEp, offer open access and others, including prediction periods, for both SST and some atmospheric ECMWF and INPE, restrict access by password. WMO parameters. Nine centres are using EPS for long-range Members can request the password for accessing forecasting as shown in Table IV-? Nine ACMAD ECMWF's extended tropical belt products and to centres, DMCs Nairobi and Harare and a few major ECMWF global products. CHAPTER V STATUS OF THE AVAILABILITY OF OBSERVATIONAL DATA FROM WWW

1. The availability of data on the MTN of the GTS is (ii) The percentages of SYNOP reports received monitored through two types of exercises: and percentages expected according to (a) The Annual Global Monitoring (AGM) of the Weather Reporting (WMO-No. 9) Volume A operation of the WWW is carried out from 1 to 15 with reference to reports required from October. Thirteen RTHs located on the MTN RBSNs are shown in Figure V-3; provided global statistics for the 2002 AGM; (iii) The number of SYNOP reports received during (b) The Special MTN Monitoring (SMM) is carried out the October AGM for the period 1998-2002 from 1 to 15 February, April, July and October. Several by WMO Region is shown in Figure V-4; RTHs located on the MTN participate in the SMM. (iv) A global map showing the location from 2. The reports from fixed stations (SYNOP and Part A which SYNOP reports from RBSN stations of TEMP) given in this chapter result from the analysis of were received for 0000, 0600, 1200 and the AGM statistics. The reports from mobile stations 1800 UTC during the 2002 AGM period is (SHIP, TEMP SHIP, BUOY, AIREP and AMDAR) result from shown in Figure V-5; the analysis of the SMM statistics. (v) Figure V-6 provides information on the number of surface stations (SYNOP) in the RESULTS OF THE ANALYSIS RBSNs for the period 1998-2002; (c) The following five figures provide detailed infor­ 3. The analysis of the AGM statistics for SYNOP mation on Part A of TEMP reports: reports and Part A of TEMP reports, together with the (i) The number of Part A of TEMP reports analysis of the SMM statistics for mobile stations, is received during the October 2002 AGM, given in the following Figures and Tables: shOWing also the number of reports required (a) SYNOP and Part A of TEMP reports received during from RBSNs and reports expected according the lO-year period 1992-2002 are shown in Figure to Weather Reporting (WMO-No. 9) Volume V-I as a percentage of the number of reports A is shown in Figure V-7; required from the RBSNs; (ii) The percentages of Part A of TEMP reports (b) Five figures provide detailed information on received and percentages expected according SYNOP reports: to Weather Reporting (WMO-No. 9) Volume A (i) The number of SYNOP reports received with reference to reports reqUired from TEMP during the October 2002 AGM, showing also stations in RBSNs are shown in Figure V-8; the number of reports required from RBSNs (iii) The number of Part A of TEMP reports and reports expected according to Weather received during the October AGM for the Reporting (WMO-No. 9) Volume A is shown period 1998-2002 by WMO Region is shown in Figure V-2; in Figure V-9;

100%

90%

CD Cl 80% ca C ... CD SYN~ l:! ... ~ CD no 70% ..-

.... ~ ".------60% - -- Figure V-I - Percentage of ------....-.--- Part A of TEMP SYNOP and Part A of TEMP reports received during 50% October AGM periods with 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 reference to reports Year required from RBSNs 48 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION

18000 -r------,

16000 +------==------1

8000 +------

4000 Figure V-2 - Number of SYNOP reports received 2000 during the 2002 AGM, showing also the number 0 of reports required from RAI RAil RA IV RAV RA VI Total RA '" RBSNs and reports expected IIIRe uired 2352 4936 1740 2048 1576 3076 15728 according to Weather • Expected 2123 4698 1349 1532 1363 2911 13976 Reporting (WMO-No. 9) o Received 1179 4118 1110 1746 1089 2876 12117 Volume A

100% -,------,

80% -

60%

40%

Figure V-3 - Percentages of 20% SYNOP reports received and percentages expected according to Weather 0% Reporting (WMO-No. 9) RAI RAil RAil' RA IV RAV RA VI Total Volume A with reference to • Expected 90% 95% 78% 75% 86% 95% 89% reports required from 50% 83% 64% 85% 69% 93% 77% OReceived RBSNs

5000 4500 4000

3500 14000 3000 12000 2500 10000

2000 8000 1500 6000

1000 4000

500 0 0 RAI RAil RA III RA IV RAV RA VI All Regions 1111998 1245 3763 1005 1724 857 2688 .1998 11282 .1999 1230 3899 1110 1697 1008 2628 .1999 11572 02000 1167 3848 1193 1728 1001 2647 02000 11583 02001 1221 4110 1191 1722 1023 2622 02001 11888 .2002 1179 4118 1110 1746 1089 2876 .2002 12117

Figure V-4 - Number of SYNOP reports received during October AGM for the period 1998 to 2002 by WMO Region STATUS OF THE AYAILABILllY OF OBSERYATIONAL DATA FROM WWW 49

• • •

• 90% to 100% (2456) • 50% to 90% (745) 1% to 50% (442) • silent stations (361)

Figure Y-5 - Availability of SYNOP reports from RBSN stations during the 1 to 15 October 2002 monitoring period (the percentage of reports received is based on the main synoptic hours 0000, 0600, 1200 and 1800 UTC)

1400

1200 5000 1000 4000 800

3000 600

400 2000

200 1000

0 0 RAI All Regions 1i!11998 621 1998 3922 .1999 593 .1999 3881 02000 593 02000 3876 02001 593 02001 3916 _2002 588 .2002 3932

Figure Y-6 - Number of surface stations (SYNOP) comprising the RBSNs for the period 1998 to 2002 by WMO Region and globally

(iv) A global map showing the locations from (f) A global map showing the locations from which which Part A of TEMP reports were received TEMP SHIP reports were received during the for 0000 and 1200 UTC during the October October SMM is shown in Figure V-14; AGM is shown in Figure V-IQ; (g) A global map showing the locations from which (v) Figure V-ll proVides information on the AIREP and AMDAR reports were received during the number of upper-air stations (TEMP) in the October SMM is shown in Figure V-IS; RBSNs for the period 1998 to 2002; (h) The daily average number of mobile station (d) A global map showing the locations from which reports received by SMM centres since 1996 is SHIP reports were received for 0000, 0600, 1200 shown in Figure V-16; and 1800 UTC during the October SMM is shown (0 The daily average number of AMDAR reports received in Figure V-12; by SMM centres since 1996 is shown in Figure V-I?; (e) A global map showing the locations from which (j) The timeliness of reception of reports for SYNOP BUOY reports were received during the October and Part A of TEMP are given in Tables V-I and V­ SMM is shown in Figure V-13; 2 respectively; 50 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION

1800 -r------,

1600 +------

1400 -/------

1200 -j------

1000 -j------

800 -j------

600 -/------Figure V-7 - Number of Part Aof TEMP reports 200 received during the 2002 AGM, showing also the o number of reports required from RBSNs and reports I!I Re uired 1616 expected according to • Ex ecled 1317 Weather Reporting (WMO­ o Received 1028 No. 9) Volume A

100% .------...,

80% -1------

60% -1------

40%

Figure V-8 - Percentages of 20% Part Aof TEMP reports received and percentages expected according to 0% Weather Reporting (WMO­ RAI RAil RA III RA IV RAV RA VI Total No. 9) Volume A with .Ex ected 52% 88% 56% 93% 75% 91% 81% reference to reports required o Received 27% 64% 38% 87% 64% 73% 64% from RBSNs

450

400

350

300 1200

250

200 800

150 600

100 400

50 200

0 0 RAI RAil RA III RA IV RAV RA VI All Regions 1!11998 54 389 39 238 101 205 1111998 1027 .1999 56 333 39 239 106 199 1111999 972 02000 59 382 44 245 108 206 02000 1043 02001 56 371 48 247 111 208 02001 1041 .2002 50 376 44 247 112 199 .2002 1028

Figure V-9 - Number of Part Aof TEMP reports received during October AGM for the period 1998 to 2002 by WMO Region STATUS OF THE AVAILABILITY OF OBSERVATIONAL DATA FROM WWW 51

• 90% to 100% (409) • 50% to 90% (143) 1% to 50% (112) • silent stations (156)

Figure V-1O - Availability of TEMP reports from RBSN stations during the 1 to 15 October 2002 monitoring period (the percentage of reports received is based on Part A for 0000 and 1200 UTC)

400 -

350

300 1000 250 - 800 200 600 150 - 400 100

50 200

0 0 RAI All Regions IJ 1998 101 1998 864 .1999 96 .1999 852 02000 96 02000 852 02001 96 02001 823 .2002 91 .2002 808

Figure V-11 Number of upper-air stations (TEMP) in the RBSNs for the period 1998 to 2002 by WMO Region and globally

(k) Table V-3 provides information on the silent (64 per cent) and in Region V (69 per cent) as shown in stations, both for SYNOP and Part A of TEMP; Figure V-3. 5. Table V-I shows that the availability of SYNOP ANALYSIS OF THE AVAILABILITY OF SYNOP reports was particularly low for Region I for 0000 UTC REPORTS (36 per cent) and for Region III for 0600 UTC (31 per cent). This shows deficiencies in the preparation and 4. The daily average number of SYNOP reports has collection of surface reports during night time in these increased from 11 282 in 1998 to 12 117 in 2002, an Regions. increase from 72 per cent to 77 per cent of the reports 6. The apparent decrease in the number of SYNOP required from the RBSNs (Figures V-2 to V-4). The reports from Region V from 1995 to 1996 was a conse­ increase was particularly significant for stations located quence of a change in the method of counting the in Regions III and V; however, there were still reports introduced in 1996 by MTN centres. In 1996, the deficiencies in the availability of SYNOP reports, MTN centres started disregarding the reports not made particularly in Region I (50 per cent), in Region III at the main synoptic hours. The AGM statistics show S2 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION

Table V-I Timeliness of reception of SYNOP reports at MTN centres - monitoring period 1 to 15 October 2002

Percentage ofSYNOP reports received in comparison with the reports required from the RBSNs I Number of WMO stations Observation: 0000 UTC Observation: 0600 UTC Observation: 1200 UTC Observation: 1800 UTC Total Region comprising received within received within received within received within received within the RBSN

1 h 2h 6h 1 h 2h 6h 1 h 2h 6h 1 h 2h 6h 1 h 2h 6h Region I 588 29 32 36 45 52 58 49 56 58 42 45 47 41 46 50 Region 11 1234 81 83 85 82 83 85 81 82 83 78 80 81 81 82 83 Region III 435 67 68 68 28 29 31 76 78 79 75 77 77 62 63 64 Region IV 512 85 86 86 79 80 80 85 86 87 87 88 88 84 85 85 Region V 394 68 71 73 69 70 72 65 66 67 58 61 64 65 67 69 Region VI 769 87 88 89 91 92 93 91 92 92 93 93 93 90 92 92 Antarctic 72 44 63 72 38 52 56 51 73 76 52 72 72 46 63 69 TOTAL 4004 72 74 75 70 72 74 76 79 80 74 77 77 73 75 77

Table V-2 Timeliness of reception of TEMP (Part A) reports at MTN centres - monitoring period 1 to 15 October 2002

Percentage ofTEMP reports received in comparison with the I I reports required from the RBSNs Number of WMO stations Obsen1ation: 0000 UTC Observation: 1200 UTC Total Region comprising received within received within received within I the RBSN I 2h 12 h 2h 12 h 2h 12 h Region [ 91 26 28 I 24 27 25 27 Region 11 294 57 67 55 61 56 64 Region III 58 13 15 49 61 31 28 Region IV 142 85 85 89 89 87 87 Region V 87 70 77 51 52 61 64 Region VI 136 75 78 68 69 72 73 Antarctic 12 54 71 61 67 58 69 TOTAL 820 60 65 59 62 59 64

Table V-3 Number of stations included in the RBSNs from which no SYNOP or TEMP (Part A) reports were received at MTN centres - monitoring period 1 to 15 October 2002

I SYNOP TEMP (Part A) I WMO Region Silent stations included Silent stations but Silent stations included Silent stations but ill the RBSN implemented in the RBSN implemented (reference: Weather Reporting (reference: Weather Reporting (WMO-No. 9) Volume A) (W,\tfO-No. 9) Volume A)

Region [ 99 96 47 32 Region 11 84 78 SS 47 I Region 1II 60 54 20 IS Region IV 25 21 9 6 Region V 66 41 14 10 Region VI 15 13 11 8 Antarctic 12 11 0 0 TOTAL 361 314 156 118 STATUS OF THE AVAILABILITY OF OBSERVATIONAL DATA FROM WWW S3

Figure V-12 - SHIP reports made at 0000, 0600, 1200 and 1800 UTe from 1 to 15 October 2002

Figure V-13 - BUOY reports made from 1 to 15 October 2002

that stations in Region V implemented observation the data collection systems, in particular HF radio systems programmes for the main synoptic hours as from 1998, (SSB) in Region [ and ARGaS systems in the Antarctic. with a corresponding increase in the number of reports 8. [n 2002, 361 stations (9 per cent of the RBSN received. Figure V-I shows this activity reflected in the stations) were silent for SYNOP reports required from global percentage of SYNOP reports received during the RBSNs, down from 555 in 1998 and 438 in 2000 (Table period 1992-2002. V-3). The reports missing from these silent stations 7. During the 2002 AGM period, 73 per cent of the represented in 2002 almost one half of the total missing required SYNOP reports were received within one hour reports from the RBSN stations. after the observation time, and an additional 4 per cent were received between one hour and six hours after the ANALYSIS OF THE AVAILABILITY OF PART A OF observation time (Table V-I). The most worrying delays in TEMP REPORTS the reception of SYNOP reports were found for reports issued from Region [ and the Antarctic. One of the reasons 9. The daily average number of Part A of TEMP for these delays is the limitation in the techniques used for reports received during AGM periods decreased from 54 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION

Figure V-14 - TEMP SHIP reports made from 1 to 15 October 2002

Figure V-IS - AIREP and AMDAR reports made from 1 to 15 October 2002

1145 to 972 during the period 1992-1999, i.e. from 10. As can be seen in Table V-2 the availability of Part 65 per cent to 57 per cent of the reports required from Aof TEMP reports was particularly low for Region III (15 the RBSNs (Figure V-I). This decrease was mainly due to per cent) for 0000 UTe in comparison with the avail­ a deterioration of the operation of the upper-air obser­ ability for 1200 UTe (61 per cent). vation network in the northern part of Region 11. An 11. During the 2002 AGM period, 59 per cent of the improvement can be noted in the period since 1999 and required Part AofTEMP reports were received within two has increased to 1028 reports in 2002, which is 64 per hours after the observation time, and an additional 5 per cent of those required (Figures V-7 to V-9). The avail­ cent were received between two hours and 12 hours after ability of TEMP reports is relatively satisfactory for the the observation time (Table V-2). eastern and southern parts of Region I1, the western part 12. In 2002, 156 stations (19 per cent of the RBSN stations) of Region Ill, the northern part of Region IV, some coun­ were silent for Part Aof TEMP reports required from RBSNs, tries in Region V and the western part of Region VI. The down from 197 in 1998 and 181 in 1998 (Table V-3). The availability of TEMP reports is generally insufficient for reports missing from these silent stations represent more than most of the other parts of the world. half of the total missing TEMP reports from the RBSN stations. STATUS OF THE AVAILABILITY OF OBSERVATIONAL DATA FROM WWW ss

25000,------,

20000 +- :<<:-"""'i"--'\';;;-__-;:-;; --.:!

15000 +------.,d'------'-"------j

Oct· Apr- Ocl- Apr· Ocl· Fel>- Apr· Fel>- Apr· Ocl· Feb· Apr· Oct- Feb· Apr· Jul-98 Jul-99 Jul-OO Jul-01 Jul-02 Ocl- 96 97 97 98 98 99 99 00 00 00 01 01 01 02 02 02 ~SH'P reports (00, 06,12,18 UTC) 2705 2612 2489 2529 2598 2563 2484 2527 2703 2801 2777 2719 2793 2691 2631 2513 2870 2549 2603 2797 2742 ~TEMPSHIP reports 16 17 16 21 25 15 19 13 25 18 16 18 14 19 19 17 18 18 17 19 16 ~BUOY reports 6046 5934 6878 7010 7256 6906 6573 5715 7628 8094 8405 9215 9542 10436 9613 10374 10919 11337 11620 11924 10841 ··"""_AMDAR reports 9211 9526 10044 9522 9468 12164 13813 14361 16801 16073 16642 15531 17252 19728 19164 21385 17886 19441 18651 18358 19514 ~AIREPreports 4064 4210 4514 4421 4673 4573 4144 4309 4573 4031 4294 4484 4278 3935 3751 4128 3475 3149 3270 3561 3592

Figure V-16 - Daily average number of mobil station reports received by SMM centres during the period 1996-2002

25000.,------,

20000 +------a~F-___'\:__-"'---.,.j

15000 -}------:;,,..------''''------{

Fel>- Fel>- Fel>- Fel>- Ocl-97 Apr-98 Jul-98 Ocl-98 Apr-99 Jut-oo Oct-OO Ocl-01 Oct-96 99 Jul-99 00 Apr·OO 01 Apr·01 Jul-01 02 Apr-02 Jul-02 Oct-02 ...... From Australian zone 4284 5546 4118 3982 4268 4533 4933 4948 4248 3976 4476 5112 4262 4502 4812 1255 1404 971 999 1306 ...... From European zone 2448 2100 2586 2990 5188 6516 6203 8504 8085 8107 6854 7414 9557 8595 10445 10956 12768 12378 11159 11972 W'~':;~rom outside Australia and Europe 2480 2398 2818 2496 2707 2764 3225 3349 3740 4559 4200 4725 5908 6067 6129 5675 5269 5302 6200 6236 =-Allreports 9211 10044 9522 9468 12164 13813 14361 16801 16073 16642 15531 17252 19728 19164 21835 17886 19441 18651 18358 19514

Figure V-l? - Daily average number of mobil station reports received by SMM centres during the period 1996-2002

ANALYSIS OF THE AVAILABILITY OF REPORTS reports was not balanced between the hemispheres FROM MOBILE STATIONS (Figure V-12), as during the October 2002 SMM 2375 SHIP reports were issued on average each day from the 13. During the October 2002 SMM, centres received northern hemisphere while only 367 reports were issued on average 2742 SHIP reports each day. The availability from the southern hemisphere. of SHIP reports has not significantly changed during the 14. During the October 2002 SMM, centres received period 1996-2002 (Figure V-16). The distribution of SHIP on average 10841 BUOY reports each day. The availabil- S6 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION ity of BUOY reports has been increasing regularly since 16. During the October 2002 SMM, centres received 1996, and has almost doubled dUring the past six years on average 3592 AIREP and 19 514 AMDAR reports daily (Figure V-16). The distribution of BUOY reports (Figure (Figure V-16). The daily average number of AIREP reports V-B) is more balanced between the hemispheres than decreased from 4064 in 1996 to 3592 in 2002. The daily for the SHIP reports, with 6607 BUOY reports issued on average number of AMDAR reports doubled during the average each day from the northern hemisphere and period 1996-2002; two-thirds of the AMDAR reports were 4234 reports issued from the southern hemisphere. issued from Europe (Figures V-IS and V-17). The aircraft 15. During the February, April, July and October 2002 reports presented in BUFR code are not yet monitored by SMM exercises, centres received on average each day 17 the SMM centres. TEMP SHIP reports (Figure V-16). Most of the TEMP SHIP reports were issued from the Atlantic Ocean and the western part of the Pacific (Figure V-14). CHAPTER VI

OPERATIONAL INFORMATION SERVICE (OIS)

1. The purpose of the WWW Operational Information manually into Volume A. This is not efficient and is a Service (OlS) is to collect from and distribute to WMO possible source of errors. Procedures for updating and Members and WWW centres detailed and up-to-date distributing Volume A in quasi-real-time by using elec­ information on observing stations and programmes, facil­ tronic media were part of a study carried out recently by ities, services and products made available in the the Rapporteur on Improvement of Volume A. day-to-day operation of WWW. OIS publishes this infor­ 4. Information on the Data-processing and Forecast mation in printed form, on computer diskettes, on Systems is provided on a yearly basis in the WWW CD-ROM and increasingly on the Internet, where it can Technical Progress Report on the GDPS. Further infor­ be accessed on the OlS home page (http://www.wmo.ch/ mation on the processed information exchanged on the web/www/ois/ois-home.htrn). Table VI-1 shows the infor­ GTS is available in Volume Cl - Catalogue of mation currently available on this Web page. Meteorological Bulletins. It is also planned to distribute 2. Updated versions of Volumes A, Cl, C2 and D of the WWW Technical Progress Report on the GDPS on a Weather Reporting (WMO-No. 9) are now being prepared CD-ROM instead of in paper format. and dispatched on CD-ROM, as requested by the CBS 5. With respect to Volume Cl - Catalogue of Management Group in December 2001. This has proven Meteorological Bulletins, the MlN centres Bracknell, to be more cost-effective than prodUcing it in paper Melbourne, Moscow, Nairobi, Offenbach, Sofia, Tokyo format, and is becoming popular with Members. and Toulouse have implemented new database proce­ However, paper copies are also available to Members if dures for maintenance of their own parts of Volume Cl. so reqUired. It is anticipated that paper format and When all MlN centres implement these procedures, there diskette distribution of many publications will be super­ will be a complete real-time online catalogue of bulletins. seded by distribution on CD-ROM and particularly with 6. Volume C2 - Transmission Schedules contains the interactive online access. transmission programmes of the distribution systems of 3. Most changes to Volume A- ObserVing Stations the GTS (satellite distribution systems, radioteletype are currently sent to the Secretariat on paper and entered broadcasts (RTT) and radio-facsimile broadcasts). In order to avoid unnecessary duplication of information, in particular with Volume D and routeing catalogues of Table VI-1 RTHs, it has been recommended that Volume C2 should Information available on the OIS Home Page contain the identification and technical specifications of each data distribution system, and a summary of the Weather Reporting (WMO-No. 9) Volume A- Observing Stations transmission programmes. Catalogue of Radiosondes and upper-air wind systems in use by Members AVAILABLE INFORMATION Regional Basic Synoptic Networks (RBSN) and Regional Basic Climatological Networks (RBCN) Weather Reporting (WMO·No. 9) Volume A- Observing Weather Reporting (WMO-No. 9) Volume Cl - Catalogue of Stations Meteorological Bulletins 7. Volume A contains a list of 11 000 surface and Weather Reporting (WMO-No. 9) Volume C2 - Transmission 1000 upper-air stations, which are used for synoptic programmes purposes. The following information for each station is Routeing catalogues of bulletins kept on record: WMO index number, latitude and longi­ Monitoring Reports tude, elevation, pressure level, surface synoptic Weather Reporting (WMO-No. 9) Volume D- Information for observations, hourly and half-hourly observations, Shipping upper-air observations and other observations and remarks. It is published twice a year. Intemational List of Selected, Supplementary and Auxiliary Ships (WMO-No. 47) Weather Reporting (WMO·No. 9) Volume C1- Catalogue Operational Newsletter on the World Weather Watch and of Meteorological Bulletins Marine Meteorological Services Additional data and products as defined in Resolution 40 (Cg-XII) 8. Volume Cl contains the list of meteorological bulletins being transmitted for global, inter-regional and S8 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION

regional exchange. An abbreviated heading (TTAAII Council Working Group on Antarctic Meteorology is in CCCC) uniquely identifies each bulletin. Each bulletin charge of reviewing the Antarctic basic synoptic network. records the code form used, time group, contents of These lists are updated regularly as required. bulletins and remarks. New editions are published twice a year. Routeing catalogues of bulletins

Weather Reporting (WMO-No. 9) Volume C2 ­ 14. The routeing catalogues of the GTS centres provide Transmission Schedules on the relay of the bulletins on the GTS pOint-to-point circuits and their distribution on the pOint-to-multipoint 9. Volume C2 contains the transmission programmes circuits, such as the satellite distribution systems. The of data distribution systems of the GTS. Induded are the routeing catalogue of a GTS centre comprises the following transmission content and transmission schedule. information for each bulletin received or transmitted by Supplements are published four times a year. the GTS centre: the GTS circuit on which the bulletin is received by the GTS centre and the list of the GTS circuits Weather Reporting (WMO-No. 9) Volume D- Information on which the bulletin is sent by the GTS centre. They are for Shipping regularly updated and accessible on the OIS home page.

10. Volume D comprises information on the meteoro- Monitoring Reports logical broadcasts by radiotelegraphy, radlOteIephony and radio-facsimile, global maritime distress and safety system, 15. Three types of quantity monitoring exercises are coastal radio stations and Inrnarsat land earth stations coordinated by the Secretariat within the framework of accepting ships' weather and oceanographic reports, the WWW Programme: AGM, SMM and the speCific marine meteorological services available at the main ports, monitoring on the exchange of Antarctic data. The ship weather routeing services, and visual storm warning reports are distributed on the OIS home page. signals. Supplements are published bimonthly. MontWy Operational Newsletter on the Operation of the International List ofSelected, Supplementary andAuxiliary WWW and Maritime Meteorological Services Ships (WMO-No. 47) 16. The operational Newsletter provides a summary of 11. This list contains information about the approxi­ the latest operational information on the GOS, GTS, mately 7000 ships participating in the WMO VOS GDPS, DM induding WMO code forms, and the Marine programme. Information is provided on details of vessels Meteorological Services Programme. It is pUblished and their routes and technical information on meteoro­ bimonthly and distributed electronically. logical equipment and methods used on board, induding exposure, location and instrument height/depth. This Additional data and products according to Resolution 40 information has been supplied by the countries which (Cg-XII) have recruited the ships. Updates are issued quarterly and a new consolidated edition is pUblished annually. 17. This is a regularly updated list of additional data and products according to Resolution 40 (Cg-XII), WMO Policy Catalogue of radiosondes and upper-air wind systems in and Practice for the Exchange of Meteorological and use by Members Related Data and Products Induding Guidelines on Relationships in Commercial Meteorological Activities. 12. This catalogue provides information on the type According to the practice adopted by Congress, all mete­ of radiosonde, windfinding equipment, ground systems, orological and related data and products reqUired to fulfil radiation corrections applied to temperature observa­ Members' obligations under WMO Programmes should be tions and other local practices for the radiosondes and encompassed by the combination of essential and addi­ upper-air stations. It is updated annually. tional data and products exchanged by Members. Members are urged to make known to all Members, Regional Basic Synoptic Network (RBSN) and Regional through the WMO Secretariat, those additional meteoro­ Basic Climatological Network (RBCN) logical and related data and products on which Members placed conditions related to their re-export for commercial 13. These lists contain the observing stations of the purposes outside the receiving country (or group of coun­ regional basic observing networks. The WMO Regional tries forming a single economic group). Associations define regional basic networks of surface and upper-air stations adequate to meet the requirements of METNO messages Members and of WWW. The Regional Assodations also define regional basic climatological networks necessary to 18. The weekly METNO (Vol. A and Vol. Cl) notifica­ prOVide a good representation of climate on the regional tions transmitted on the GTS carry the most up-to-date scale, in addition to global scale. The WMO Executive information on operational changes. ANNEX I ACRONYMS

3-DVAR Three-dimensional Variational Analysis DCS Data Collection System 4-DVAR Four-dimensional Variational Analysis DM Data Management ACARS Aircraft Communications Addressing and DMC Drought Monitoring Centre Reporting System DMSP Defence Meteorological Satellite ACMAD African Centre of Meteorological Programme Applications for Development DVB Digital Video Broadcast ADS Automated Dependent Surveillance DWD Deutscher Wetterdienst AFS aeronautical fixed service (ICAO) ECMWF European Centre for Medium-Range AFS area forecast system (WMO) Weather Forecasts AGM Annual Global Monitoring EMWIN Emergency Managers Weather AHRPT Advanced high-resolution picture Information Network transmission EPS Ensemble Prediction System AIREP Aircraft weather report ERA Emergency Response Activities AMDAR Aircraft Meteorological Data Relay ESCAP Economic and Social Commission for Asia AMSU advanced microwave sounding unit and the Pacific APT .Automatic picture transmission EUME1NET European Meteorological Services ARGOS Data relay and platform location system Network ASAP Automated Shipboard Aerological EUMETSAT European Organization for the Programme Exploitation of Meteorological Satellites ASDAR Aircraft to satellite data acqUisition and FAO Food and Agriculture Organization of the relay (aircraft report) United Nations ASEAN Association of South-East Asian Nations FGGE First GARP Global Experiment ASECNA Agency for Air Safety in Africa and FNMOC Fleet Numerical Meteorology and Madagascar Oceanography Center (USA) ASMC ASEAN Specialized Meteorological Centre FSS Fixed Ship Stations ATOVS Advanced TIROS Operational Vertical FTP File Transfer Protocol Sounder FY-l, -2 Feng-Yun series of geostationary AWS Automatic Weather Station meteorological satellites (China) BIPM International Bureau of Weights and GARP 'Global Atmospheric Research Programme Measures (Original: French) GATE GARP Atlantic Tropical Experiment CBS Commission for Basic Systems GAW Global Atmosphere Watch CEOS Committee on Earth Observation GCOS Global Climate Observing System Satellites GDPS Global Data-processing System CGA Coordination Group for the Composite GMDSS Global Maritime Distress and Safety Observing System for the North Atlantic System CGMS Coordination Group for Meteorological GMS Geostationary meteorological satellite Satellites Oapan) CHRPT China high-resolution picture GOES Geostationary operational environmental transmission satellite (USA) CIMO Commission for Instruments and GOMS Geostationary operational meteorological Methods of Observation satellite (Russian Federation) CIR Committed Information Rate GOS Global Observing System CUPER Climate and Persistence Model GPS Global Positioning System CMC Canadian Meteorological Centre GSM Global Spectral Model CNES Centre nationale d'etudes spatiales GSN GCOS Surface Network (France) GSNMC GCOS Surface Network Monitoring Centre CPTEC Centro de Previsao de Tempo e Estudos GTOS Global Terrestrial Observing System CIimaticos (Brazil)/Centre for Weather GTS Global Telecommunication System Prediction and Climate Studies GUAN GCOS Upper-air Network CTBTO Comprehensive Nuclear Test Ban Treaty HF High frequency Organization HR High resolution DAB Digital Audio Broadcasting HRPT High-resolution picture transmission DBCP Data Buoy Cooperation Panel IAEA International Atomic Energy Agency DCP Data Collection Platform ICAO International Civil Aviation Organization 60 WORLD WEATHER WATCH - TWENTY-FIRST STATUS REPORT ON IMPLEMENTATION

ICSU International Council for Science OIS Operational Information Service IGBP International Geosphere-Biosphere OMEGA Radio navigation system Programme (ICSU) OPAG IOS Open Programme Area Group on IGOSS Integrated Global Ocean Services System Integrated Observing Systems IMOP Instruments and Methods of Observation OPMET operational meteorological information Programme (Code form) INMARSAT International Mobile Satellite PUMA Preparation for use of MSG in Africa Organization PWSP Public Weather Services Programme INPE Instituto NacionaI de Pesquisas Espaciais R&D Research and Development (Brazil)jNational Institute for Space RA Regional Association Research RANET Radio and Internet INSAT Indian national satellite (India) RASS Radio-Acoustic Sounding System IOC Intergovernmental Oceanographic RBCN Regional Basic Climatological Network Commission RBSN Regional Basic Synoptic Network IODC Indian Ocean Data Coverage RESURS Environmental satellite (Russian IRI International Research Institute for Federation) Climate Prediction (USA) RETIM Reseau europeen de transmission ISCS International Satellite Communications d'information mHeorologique (par System satellite) ISO International Organization for RMC Regional Meteorological Centre Stamicm:ttzatioll RMBeN-----Regional-Mereorologieal-B-atil ITU International Telecommunication Union Communication Network JMA Japan Meteorological Agency RMS Root-mean-square KNMI Royal Netherlands Meteorological RMTC Regional Meteorological Training Centre Institute (Utrecht) RMTN Regional Meteorological LAM Limited Area Model Telecommunication Network LRIT Low Rate Information Transmission RSMC Regional Specialized Meteorological LRPT Low Rate Picture Transmission Centre MDCRS The Meteorological Data Collection and RTH Regional Telecommunication Hub Reporting System SADC Southern Africa Development Community MDD Meteorological Data Distribution SADIS Satellite Distribution System (ICAO) METEOR Meteorological satellite (Russian SMM Special MTN Monitoring Federation) SPECI Aviation selected special weather report METEOSAT EUMETSAT series of meteorological SSA System Support Activities (WWW) geostationary satellites SSB Single sideband (transceivers) METNO Advance telegraphic notification relating SST sea-surface temperature to the operation of WWW (Code form) TAC traditional alphanumeric codes METOP Meteorological Operational Satellite TCP Tropical Cyclone Programme (EUMETSAT) TCPjIP Transmission Control Protocol/Internet MMOP Marine Meteorology and Oceanography Protocol Programme TDCF Table Driven Code Forms MMS Marine Meteorological Services TIROS Television infrared observation satellite MSG METEOSAT second generation (United States) MSLP minimum sea level pressure TOGA Tropical Ocean and Global Atmosphere MTN Main Telecommunication Network Programme MTSAT Multifunctional transport satellite Oapan) UKMO United Kingdom Meteorological Office NCDC National Climatic Data Center (United UKSF United Kingdom Satellite Facility States) UNEP United Nations Environment Programme NCEP National Centres for Environmental UTC Universal time coordinated Protection (USA) VISSR Visible and Infrared Spin Scan Radiometer NMC National Meteorological Centre VOS Voluntary Observing Ship NMHS National Meteorological and Hydrological VOSClim vas Climate Project Services VSAT Very small aperture terminal NMOC National Meteorological Operations WAFC World Area Forecast Centre Centre WAFS World Area Forecast System NMS National Meteorological or WCP World Climate Programme Hydrometeorological Service WCRP World Climate Research Programme NMTN National Meteorological WDM WWWData Management Telecommunication Network WEFAX Weather facsimile NOAA National Oceanic and Atmospheric WHO World Health Organization Administration (USA) WMC World Meteorological Centre NPOESS National Polar Operational Environmental WMO World Meteorological Organization Satellite System (USA) WRC World Radiocommunication Conference NWP Numerical Weather Prediction WWW World Weather Watch ANNEX 11 REFERENCES

The 2000 Annual Global Monitoring of the Implementation of the WWW Commission for Basic Systems, Extraordinary Session, 2002 - Abridged final report with resolutions and recommendations (WMO-No. 955) Fifth WMO Long-term Plan (2000-2009) (WMO-No. 908) Annual Report ofthe World Meteorological Organization, 2001 (WMO-No. 935) World Disaster Report, Red Cross, 2002 Improvements in the Skill ofNumerical Weather Prediction with Global Models, Hollingsworth, et al., 2002 Cummulative Results of Extended Forecast Experiments I. Model Performance for Winter Cases, K Miyakoda, GD Hembree, RF Strickler and 1Shulman, (1972), Mon. Weath. Rev. lOO, 835-855 Some Aspects of the Improvement in Skill of Numerical Weather Prediction, Simmons A.]. & A. Hollingsworth (2002), Quart J. Roy. Meteor Soc. 128, 647-677 Guide on meteorological observation and information distribution systems at aerodromes (WMO-No. 731) Guide on the automation ofdata-processing centres (WMO-No. 636) Guide on the Global Data-processing System (WMO-No. 305) Guide on the Global Observing System (WMO-No. 488) Guide on World Weather Watch data management (WMO-No. 788) Guide to agricultural meteorological practices (WMO-No. 134) Guide to climatological practices (WMO-No. 100) Guide to hydrological practices (WMO-No. 168) Guide to marine meteorological services (WMO-No. 471) Guide to meteorological instruments and methods Ofobservation (WMO-No. 8) Guide to moored buoys and other ocean data acquisition systems (WMO-No. 750) Guide to practices for meteorological offices serving aviation (WMO-No. 732) Guide to public weather services practices (WMO-No. 834) Guide to qualifications and training ofmeteorological personnel employed in the provision ofmeteorological services for international air navigation (WMO-No. 114) Guide to the applications ofmarine climatology (WMO-No. 781) Guide to the provision ofmeteorological service for international helicopter operations (WMO-No. 842) Guide to wave analysis and forecasting (WMO-No. 702) International list ofselected, supplementary and auxiliary ships (WMO-No. 47) Manual on the Global Data-processing System (WMO-No. 485) Manual on the Global Observing System (WMO-No. 544) Manual on the Global Telecommunication System (WMO-No. 386) Manual on codes (WMO-No. 306) Weather reporting (WMO-No. 9) The World Weather Watch Programme 1996-2005: Fourth WMO Long-term Plan, Part 11: Volume 1 (WMOjTD-No. 700) WWW Technical Progress Report on the Global Data-processing System: 2000 (WMOjTD-No. 1061) WWW Technical Progress Report on the Global Data-processing System: 2001 (WMOjTD-No. 1115)