AGENDA

1. ORGANIZATION OF THE MEETING
1.1 Opening of the meeting
1.2 Adoption of the agenda
1.3 Working arrangements
2. WIS PART A IMPLEMENTATION AND TRAFFIC ANALYSIS
3. REVIEW OF THE MONITORING RESULTS OF THE OPERATION OF THE MTN
4. PROCEDURES FOR THE COLLECTION, ROUTEING AND DISTRIBUTION OF DATA AND PRODUCTS
5. OPERATIONAL INFORMATION SERVICE (RELATED TO ISS)
6. QUANTITY MONITORING OF THE OPERATION OF THE WWW
7. WIS PLAN AND ROADMAP
8. ANY OTHER BUSINESS
9. CLOSURE OF THE MEETING

Executive Summary

The meeting of the CBS Expert Team on GTS-WIS Operations and Implementation was held in Geneva from 22 to 24 September 2010 under the chairmanship of Mr Kelvin Wong (Australia).
The participants in the meeting made presentations on the current status of implementation of their centres. The documents of the meeting and presentations are available from the WMO website:
http://www.wmo.int/pages/prog/www/WIS/wiswiki/tiki-index.php?page=et-oi-meeting-geneva
The meeting reviewed the status of the Regional Telecommunication Hubs (RTH), including their traffic volumes, progress on implementation of the WMO Information Systems (WIS) and procedures for the collection, routeing and distribution of data and products on the Global Telecommunication System (GTS). The meeting included a joint session with the CBS Expert Team on Communication Techniques and Structure (ET-CTS) on 24 September during which time it reviewed some common issues, in particular the drafts for; the amendment to the Technical Regulation WMO 49 Section A3 to incorporate WIS, a draft Manual on WIS, a draft Guide to WIS, and amendments to the Manual on the GTS. The meeting also reviewed the status reports of the Operational Information Service (OIS).

The outcomes of the meeting will be submitted to the meeting of the Implementation Coordination Team of the Information Systems and Services (ICT-ISS) of the CBS OPAG-ISS, to be held on 27-30 September 2010 in preparation for the extraordinary session of CBS in November 2010.

1. ORGANIZATION OF THE MEETING

1.1 The CBS Expert Team on GTS-WIS Operations and Implementation (ET-OI) was held in Geneva from 22 to 24 September 2010 under the chairmanship of Mr Kelvin Wong (Australia). Nine RTH experts from nine countries participated in the meeting. The complete list of participants is included in this report.

1.2 Mr Wong opened the meeting and welcomed participants. He gave a brief summary of the agenda noting that there will be a half day joint session with the Expert Team on Communication Techniques and Structure (ET-CTS) to discuss some important issues of traffic analysis for GISC network bandwidth requirement and matters of common interest to the two ET groups such as meteorological bulletins submitted via email input systems in MIME encoded messages, updates to file naming convention and GTS headers, as well as a review of the technical regulations affecting WIS and in particular the GTS. Mr David Thomas, Project Manager for WIS of the Observation and Information Systems department, spoke on behalf of Mr Michel Jarraud, Secretary-General of WMO. He welcomed the participants and recalled the importance of this group in maintaining operational systems that form the basis of WMO operations for all Members. He noted that WIS is a collective system based on components owned and operated by WMO Members working collaboratively. The timely implementation of WIS is dependent on Members systems providing the infrastructure and new services.

1.3 The outcomes of the meeting will be submitted to the meeting of the Implementation Coordination Team of the Information Systems and Services (ICT-ISS) of the CBS OPAS-ISS, to take place in Geneva the following week (27-30 September).

1.4 The meeting adopted the agenda as reproduced at the beginning of this report.

2. WIS PART A IMPLEMENTATION AND TRAFFIC ANALYSIS

2.1.1 Part A of WIS includes the ongoing support of the GTS and associated data management. Reports were received from 13 RTHs and included references to traffic volumes on GTS, Internet links, and where relevant, satellite distribution systems. The participants in the meeting made presentations on the current status of implementation of their centres. Many of the reports also provided details of the number and volume of messages and files switched across individual circuits which when combined could give an overview of the traffic between regions. Key issues highlighted included the successful move to MPLS, the important role of satellite distribution systems and the contribution of message switching systems in supporting plans to implement the new functionality of WIS and in supporting the migration to table driven code forms (TDCF).Summaries of the key points for each report are included in the annex to this paragraph. Meeting documents and presentations are available from the WMO Website:

http://www.wmo.int/pages/prog/www/WIS/wiswiki/tiki-index.php?page=et-oi-meeting-geneva

2.1.2 It was evident that RTHs are providing more consistent monitoring of links volumes with better traceability of loading on individual links. This was a significant improvement since 2007 largely facilitated by applications on routers and MSS, and RTHs making use of available reporting tools that produce similar types of reports. The meeting noted that in addition to the volume monitoring, a need remains to identify performance monitoring for inside MSSs, particular delta time in and out of the RTH. It was agreed to take prime examples and define a common set of monitoring for GISC, DCPC and NC. This will be especially important for establishing the required monitoring of an all hazard warning network within WIS to ensure WIS will meet the 2 minute global distribution requirement.

2.1.3 TIGGE remains a massive user of broadband and demonstrates well the role of the Internet to shed load off operation critical circuits such as the core network of WIS between GISCs. New systems such as the European radar service also contributed to significant regional increases in traffic. In addition to utilising the Internet, MPLS and multicast could facilitate reduced volume loads on dedicated networks by reducing the outflow of traffic. However, considerable redesign will most likely be needed to MSS and application layers to manage multicast. Also network level performance is not fully understood for multicast. At this time, satellite distribution through the IGDDS remains the most effective multicast tool. However, the meeting also noted that IGDDS may also need to utilise the terrestrial components of WIS for passing information between upload sites, especially where it is not possible to have overlapping footprints necessary for turnaround uplink stations. Furthermore, participants warned of another big increase in volume with next generation multi channel satellites and resultant products which will also need to be sent to global modelling centres.

2.1.4 The meeting noted that the rate of growth of traffic is close to that predicted in ICG-WIS 2007 and ET-OI 2008 meetings, and roughly doubling each year, With regard to Internet volumes, the report from WMC Melbourne highlighted that Internet growth analysis at the WMC showed that the national Internet usage may be swamping the figures for WIS AMDCN and Core network Internet usage. Some participants agreed that this was the case in their centres too, although they had not reported on it. The chair suggested that WIS centres need to monitor their internal and WMO related Internet usage and take steps to prevent adverse effects on international exchange mechanisms. Two key solutions to this problem were identified in different reports. The most used approach is to utilise a second ISP and load share, another idea is to separate WMO traffic or local public access to a mirror site.

2.1.5 After reviewing the reports, the chair highlighted the need to consider how best to depict AMDCN and RMTN networks. Several participants liked Russia’s presentation due to traceability of application based topology over MPLS networks. The meeting noted the need for depiction of IGDDS distribution and receiving points to complement the terrestrial and two way satellite communication systems diagrams.

2.1.6 The Internet is being used increasingly to supplement or replace dedicated GTS circuits. This is evident in RA I, III and V. In particular, the Internet provided significant improvements across Africa and provided backup to VSAT links, which experienced problems with their X25 protocols around August 2010 which led to many VSAT links moving from X25 to V34 protocol. Note also that the bandwidths on the Internet when used are orders of magnitude larger than available through dedicated links. The use of the Internet was also demonstrated with the demise of BT FR across the Asia Pacific. Buenos Aires also highlighted that its regional move to VPN over Internet is a first step to MPLS.

2.1.7 Reports show MSS are increasingly able to utilise the Internet, e-mail and support national web based services. MSS are also playing a major role in supporting the migration to TDCF. The meeting noted that the migration to TDCF is proceeding well in many regions but that the parallel transmission of TAC and TDCF will be needed for some time.

2.1.8 Centres also reported on their preparation for the new functionality of WIS. CMA, DWD and JMA GISCs are now all preoperational. UK, MF, and BoM reported they will be using OpenWIS. All report to be progressing on target, although there are still some problems to address. INMET and Buenos Aires reported that they are reviewing DWD MSS and Geonetwork. It was clear from reports that work being done by MSS suppliers for major centres is spreading to other RTH and NMCs.

2.2 Traffic analysis for GISC network bandwidth requirement

The meeting reviewed a paper by Mr Kenji Tsunoda. It noted that this issue had been referred to the ET-CTS. Criteria for their study included; what kind of data had to be shared between GISCs, handling of duplicate traffic, and peak traffic load (Rx and Tx). The paper reviewed the situation for RTH Tokyo and provided estimates of the above. The results and trends in the analysis were consistent with the information provided in the RTH reports under section 2.1.

The issue was also reviewed in the ET-CTS run in parallel to the ET-IO meeting and then finding discussed in a joint session. At this joint session participants agreed that there were many issues with trying to simplify the traffic analysis based on present exchange rates between the regions and past trends. However, all accepted the findings of the ET-CTS and agreed that the traffic would be dependent on the number of GISCs, and that based on a full mesh topology with all GISCs sending information from their AMDCN to all other GISCs would provide a linearly increasing volume load. As such, the number of GISCs in a system would multiply their bandwidth requirements as each existing GISC would have to set up arrangements to transmit the data to a new GISC. The group did not set a maximum number of GISCs but noted that the costs of bandwidth and administering connectivity would be limited by financial and overhead aspects rather than technical.

2.3 Management and operational aspects of GISC metadata and data synchronisation

Mr Kenji Tsunoda presented a paper on some key issues that arise from the implementation of a WIS DAR (data Discovery, Access and Retrieval) metadata catalogue. He noted that there were several metadata catalogues already in use by RTH that describe various data streams, including the Volume C1, Resolution 40 additional data, the TIGGE catalogue and other similar catalogues for differing data streams. He noted that Vol. C1 includes all WWW data and products with GTS headings but new types of data/products without GTS headings such as ATOVS-RARS files are not included. The DAR catalogue will be a superset of metadata that includes all of Vol.C and Resolution 40 additional data but may only contain subsets of Vol. A or TIGGE catalogue.

The meeting reviewed the maintenance process for Vol.C1, its central reference repository and its METNO notification of changes to Vol. C1 were important tools for all RTH. It agreed that eventually Vol. C1 could be replaced by the DAR catalogue as could many of the other catalogues that are subsets of the DAR catalogue, however, the METNO notification process would also need to be replaced by something that has the same outcome of notifying interested users of changes to the catalogue. Metadata for other data streams such as TIGGE and RARS could also be incorporated into the DAR catalogue. However, Vol. A contains information that may not necessarily be in the DAR catalogue so would not be able to be fully generated from the DAR, and would have to continue in its own right. The meeting addressed these key issues and summarised them in the following table.

Table-1: Differences between Volume C1 and DAR
	Volume C1	DAR (pre-operational)
Who makes update information?	Associate RTHs	Each NC and DCPC
Who manages the catalogue?	WMO secretariat	Each GISC manages its own AMDCN component
Is there an ORIGINAL one?	YES	NO, each GISC has to maintain it individually and synchronises using OIA-PMH with all other GISCs to create a virtual centralised database.
How to advise users of updates?	METNO bulletin	Could RSS feed or equivalent notifying subscribers of updates (Note that this feature has not been defined in WIS)
Complete revision periodically?	Yes, semi annual update	No
Regulations?	Defined in the Manual on the GTS	Defined in the Manual on WIS, but to what extent?
Publish?	a revised edition issued yearly on CD-ROM.	Users can subscribe to GISC online, order a CD or possibly have WMO or someone set up a flat file storage area of XML metadata files.

The meeting also noted that the notification of WMO Secretariat of changes to Resolution 40 additional data is a manual process involving the PR notifying of new data and products and that they are to be listed as “Additional Data”. Under WIS, this would be by a centre's metadata manager entering in the access conditions in the appropriate metadata field when the metadata is created. This may have political ramifications as it may affect the formal notification process, so needs to be studied further.

Given the importance of catalogues like Vol. C1 and Res. 40 Additional Data to operations, and that it will take a few years for the DAR metadata process to settle in, the meeting decided it did not want to have the DAR catalogue take over the role of these for at least the next four years. This will give time for GISCs, DCPCs and NCs to establish their processes. As such it is recommended that the current update processes continue in parallel to the population and maintenance of the DAR catalogue until Congress XVII. It noted that this does not exclude some GISCs from implementing systems where the Vol C1 is updated by extraction from their DAR catalogue earlier if they have operationally robust solutions. This decision will need to be passed to the other expert teams also investigating the transition of Vol C1 to DAR catalogue (eg IPET-MDI)

2.4 Management and operational aspects of DCPC and NC in areas of GISC

The meeting reviewed a case study of the relationship and processes between the Hong Kong Observatory’s candidate DCPC the “World Weather Information Service (WWIS)” and the candidate GISC Tokyo. It noted that Tokyo utilises OAI-PMH to harvest metadata from WWIS daily so that users searching the Tokyo DAR catalogue can find weather forecasts for any city available at the WWIS. A user searching the GISC (http://www.gisc.kishou.go.jp/) will receive relevant metadata that match his search criteria. This metadata will direct the user to the WWIS (http://worldweather.wmo.int/) to access the product. It noted that a user can also use the metadata server at the Hong Kong Observatory directly and be steered to the product.

3. REVIEW OF THE MONITORING RESULTS OF THE OPERATION OF THE MTN

3.1 Monitoring results of AGM, SMM and IWM

The analysis prepared by the Secretariat on the availability of SYNOP, TEMP and CLIMAT reports from RBSN/RBCN stations during the July 2009 to April 2010 AGM/IWM/SMM exercises are available on the WMO FTP server as follows:
ftp://ftp.wmo.int/GTS_monitoring/AGM-IWM-SMM/From_WMO/200907_201004/
The diagrams displaying the percentage of SYNOP/TEMP/CLIMAT reports available at MTN centres in comparison with the reports required from the RBSN/RBCN stations for the 2009 AGM/IWM are available on the WMO FTP server as follows:
ftp://ftp.wmo.int/GTS_monitoring/AGM-IWM-SMM/From_WMO/AGM/200910/ANALYSIS/maps
KML files of the diagrams which can be visualized using a kml browser such as Google Earth are also available

3.2 Differences in the availability of reports at MTN centres during monitoring exercises

The differences in the availability of SYNOP, TEMP and CLIMAT reports between MTN centres during the 2009 AGM/IWM and SMM exercises are presented in the meeting. Details of the diagrams and tables are available on the WMO FTP server as follows:
ftp://ftp.wmo.int/GTS_monitoring/AGM-IWM-SMM/From_WMO/AGM/200910/ANALYSIS/comparisons/diff/
The meeting noted the following reasons for differences in the availability of data between centres:
(a) Differences of requirements in the reception of data,
(b) Data not monitored,
(c) Shortcomings in the relay of the data on the GTS,
(d) Differences in the implementation of the monitoring procedures at centres.
The meeting acknowledged the monitoring results are very useful for the detection of deficiencies at MTN centres and strongly encouraged MTN centres to review those differences, in particular those showing low availability of data at their centres, and to take actions to reduce the differences.

3.3 Timeliness of reception of bulletins at MTN centres

The diagrams in Annex to this paragraph displaying the percentage of SYNOP and TEMP bulletins (Part A) received by MTN centres within 1 to 15 minutes at the July 2010 SMM were presented indicating some systematic delays in data reception at RTH Toulouse in comparison with other centres. The meeting requested RTH Toulouse to investigate the source of the delay and rectify the problem.

3.4 Routeing and monitoring of CLIMAT bulletins and GCOS GSNMC monitoring reports

GCOS Secretariat Richard Thigpen provided the meeting participants with a summary of the GCOS Surface and Upper-air Network, CBS Lead Centres for GCOS and their activities, and the various GCOS programs for equipment replacement and technical support projects including the GCOS Cooperative Mechanism which provides source of funding of the projects.
The purpose of this session is to present the findings of the investigation conducted on the data receipt anomalies as noticed in the monthly data monitoring products compiled by the GCOS Monitoring Centre in DWD, Germany. The problem was first brought to the notice of ET-OI chair in May 2009. A report compiled by ET-OI chair was presented to the Eleventh Session of the CBS Management Group meeting (MG-11) in March 2010. The report is available at the CBS MG-11 meeting page
http://www.wmo.int/pages/prog/www/CBS/Meetings/MG-11/documents/ET_OI_report_on_CLIMAT_201003_v0.3.pdf
The report is to seek the cooperation of RTH Focal Points to coordinate with the NMCs involved in their zone of responsibility and to rectify the problems as identified for the CLIMAT bulletins originated at their centres.

4. PROCEDURES FOR THE COLLECTION, ROUTEING AND DISTRIBUTION OF DATA AND PRODUCTS

4.1 Meteorological bulletins

4.1.1 Formats of meteorological bulletins submitted via email

This various data format issues of meteorological bulletins submitted via email were discussed in the joint session with members of the ET-CTS. The matter is related to the problems caused by the conversion of MIME encoded messages when the reports are transmitted via email. The provision of guidelines in Attachment II-16, Manual on GTS, on the procedures for transmitting and collecting meteorological bulletins is considered to be insufficient. The chair of ET-CTS was requested to review the guidelines and provide clarifications or recommend amendments to the Manual on GTS.

4.2 Abbreviated headings

4.2.1 Incorrect use of abbreviated headings

The meeting was informed of the incorrect use of abbreviated headers for some BUFR SHIP, BUFR BATHY/TESAC/TRACKOB and BUFR BUOY bulletins by some RTH centres. The appropriate abbreviated headers to be used were clarified to be “ISS” for BUFR SHIP, “IOR” for BUFR BATHY/TESAC/TRACKOB and “IOB” for BUFR BUOY bulletins. The RTH centres using incorrect headers were requested to take actions to change the headers according to Table C6 and Table C7 of Attachment II-15, Manual on the GTS and notify WMO Secretariat accordingly.

4.2.2 Allocation of abbreviated headers for oceanographic data

In addition to the existing data designators for oceanographic data, the meeting has agreed to allocate a new heading T1T2A1 = IOZ with data category/sub-category 031/007 for the deep ocean tsunameter in table C6 of Attachment II-5 in the Manual on the GTS.
Draft amendment to TABLE C6 as follows:

T1T2A1	ii	Data type	TAC correspondence	Data Category/Sub-Category Common Code Table C13
IOZ		Deep ocean tsunameter		031/007

4.2.3 Allocation of abbreviated headers for ozone data

The meeting was provided with a brief from IPET-DRC regarding the terminology of ozone data for the following three types of measurements:
- Total ozone (amount of ozone in an entire atmospheric column [DU]),
- Ozone vertical profiles from ozone soundings (partial pressure of ozone at particular levels [Pa]),
- Surface ozone (concentration of ozone in the atmospheric boundary layer [molmol-1]).
However, the data sub-categories in Common Code Table C-13 and the abbreviated headers in Tables C6 and C7 are not sufficient to distinguish these types of observations.
The meeting agreed to amend the allocation for ozone data in Tables C6 and C7 and replace the following allocations:

	T1T2A1	Data type	Data category/subcategory
Table C6 - BUFR	ISE	Ozone measurement at surface	008/000
Table C7 - CREX	KSE	Ozone measurement at surface	008/000
Table C6 - BUFR	IUE	Ozone vertical sounding	008/001
Table C7 - CREX	KUL	Ozone vertical profile	008/001

by the following new allocations (the modifications/additions are marked in blue):

	T1T2A1	Data type	Data category/subcategory
Table C6 - BUFR	ISE	Measurement of surface ozone	008/000
Table C7 - CREX	KSE	Measurement of surface ozone	008/000
Table C6 - BUFR	IUE	Ozone vertical sounding	008/001
Table C7 - CREX	KUE	Ozone vertical sounding	008/001
Table C6 - BUFR	IUL	Total ozone	008/002
Table C7 - CREX	KUL	Total ozone	008/002

4.2.4 Delivery of corrected and delayed TAC and BUFR bulletins

The meeting was provided with a submission from IPET-DRC on the matter of delivery of corrected and delayed TAC and BUFR bulletins. The intention is to give BUFR bulletins the flexibility to include reports (subsets) of all stations in the next updated bulletins. The general practice for TAC bulletins is to resend only the corrected or delayed reports in the subsequently delivered bulletins. The meeting has no objections for BUFR bulletins to have this flexibility if there is a requirement for the user or the data provider. In fact some TAC bulletins are also resending all stations scheduled to be in the bulletin whenever a subsequent bulletins is re-compiled for distribution due to some stations being received late or corrections being made.

4.2.5 Changing message contents and the impacts on the GTS and users

The incidents of messages with contents being modified by a receiving RTH centre and re-transmitted on the GTS with the abbreviated header unchanged have raised some concerns. According to the Manual on the GTS, Part I, para. 2.1 this is not allowed. The report submitted to the meeting indicated that most of the changes made to the body of the message are basically related to error corrections in the text of the OPMET data bulletins such as METAR, TAF, SIGMET, etc.
A proposal to insert a 5-digit numerical group in the message for the calculation of a hash function at the message generating centre was discussed but it was not convinced this method will solve the problem without creating other problems. The meeting referred the matter to members of the group for further investigation. The preferred solution should always be coordinated between the message generating centre and the message receiving centre following the guidelines of the Manual on the GTS.

4.2.6 Abbreviated headings for satellite data bulletins

Satellite data centres have long been using various designators other than those described in Attachment II-5 of the Manual on GTS and the situation is rather chaotic especially when the data centres do not inform WMO Secretariat and provide information of their bulletins for publication in WMO Volume C1, the Catalogue of Meteorological Bulletins. Despite the efforts of a few satellite operators including EUMETSAT, ESA-ESRIN and JMA to provide detailed information of the data in their products web page and also notify WMO Secretariat through RTH centres in Offenbach, Rome and Tokyo with the necessary information required for WMO Volume C1 there are still many satellite data bulletins delivered by other data centres with completely no information in WMO Volume C1. In particular the complete lack of updates in Volume C1 for bulletins delivered by the US data centres since 2002 has made it very difficult to find information for some of the bulletins.

The meeting examined a range of TAC and BUFR satellite data bulletins now being distributed on GTS. There have been insufficient designators or no provisions in Attachment II-5 of the Manual on GTS for the allocation of abbreviated headers for the BUFR satellite data bulletins whereas designators for TAC satellite data bulletins are under utilised or in some cases misused. The Expert Team is invited to examine the issues of an increasing number of BUFR satellite data bulletins now being distributed on GTS for which the provisions of abbreviated headings described in Attachment II-5 of the Manual on GTS are insufficient to cover the ever increasing satellite data and products.

It is recommended to formalise the designators in Attachment II-5 of the Manual on GTS for some of the well described designators that follow the good practice and sustainable use of designators.

The use of new A1 designators to prescribe specific satellite data products similar to A1 = A, B, H, M representing AMSU-A, AMSU-B, HIRS and MHS would be required. Some general practice following EUMETSAT would be appropriate:
For polar systems, e.g. EUMETSAT Polar System (EPS)
• A1 = G for GRAS
• A1 = I for IASI
• A1 = O for Soil Moisture
• A1 = S for ASCAT

Existing provisions in Attachment II-5 include A1 = C, R, T. The data type may need to be more broadly re-defined for A1 = C and T as both refer to some TAC equivalents for SAREP/SATOB, and SATEM,/SARAD/SATOB. The higher resolution AMV products usually represented by A1 = V are gradually replacing SATOB and it may be appropriate to include the high density AMV including visible, infrared and clear sky water vapour winds in the category using A1 = C. Cloud analysis (CLA) could also be classified under the broad definition. A1 = T is subject to further discussion with other data centres as considerable usage under this designator referred to GPS radio occultation including GRACE/GRAS, COSMIC and the now defunct CHAMP.

For geostationary systems, e.g. Meteosat, MTSAT-2
• A1 = C for cloud analysis (CLA), AMV (VIS, IR, WVW)
• A1 = F for All Sky Radiances (ASR)
• A1 = H for Tropospheric Humidity (TH)
• A1 = L for Total Ozone (TOZ)
• A1 = R for Clear Sky Radiances (CSR)
• A1 = V for Atmospheric Motion Vector (AMV)
• A1 = T for GPS Radio occultation

The classification of T2 designators by satellite systems (polar system or geostationary) and mode of operation (rapid scanning or backup satellite system) is an option to conserve designators and it can also be associated with multiple CCCC indicators defined to indicate further information of data coverage (global or regional) or other details for easier identification of the data and products delivered by the same data centre.

The option of re-defining T1 = T for use in both TAC and BUFR satellite data bulletins as described in para. 3 is also a possible solution to relieve the shortage of designators in BUFR encoded observational data (T1 = I).

Because of the technical nature of satellite products it is recommended the method of classification of the products and the best practice for applying designators to represent the data products or message bulletins should be further discussed both within OPAG-ISS groups particularly IPET-DRC and also WMO Programmes especially the Space programme and satellite expert groups including the CGMS.

4.3 Table driven code form (TDCF) step-by-step migration by zones

4.3.1 Monitoring of the migration to TDCF for RBSN stations

The meeting was presented with a summary of the status of the migration for the RBSN stations based on the results of monitoring statistics of Integrated WWW Monitoring (IWM), Annual Global Monitoring (AGM) and the Special MTN Monitoring (SMM) conducted in July/October 2009 and Jan/April 2010. The Team analysed the diagrams displaying the distribution of RBSN stations for which BUFR surface and radiosonde reports in ISM, IUS, IUK bulletins were included in Volume C1 or received by MTN centres during the four monitoring periods. Discrepancies were identified and the Team agreed to follow up with the responsible RTH Focal Points to rectify the problems, notably the stations for which BUFR reports were received but the information is not provided in Volume C1.

The relevant information extracted from Volume C1 and the AGM/IWM/SMM statistics is presented at the level of the station in tables available from the WMO server as follows:

http://www.wmo.int/pages/prog/www/WMOCodes/MigrationTDCF/status/SUMISM_097-104.xls
http://www.wmo.int/pages/prog/www/WMOCodes/MigrationTDCF/status/SUMIUK_097-104.xls
http://www.wmo.int/pages/prog/www/WMOCodes/MigrationTDCF/status/SUMIUS_097-104.xls

4.3.2 Status of the national plans for the migration to TDCF and step-by-step migration by zone

A summary report on the status of the national plan for the migration to TDCF as of June 2010 was presented in the meeting. The results were based on the responses received from 57 countries in January 2010 in the survey conducted by WMO Secretariat in November 2009. The analysis also included the results of an earlier survey in 2008 in which 9 countries provided information on the national migration plan of their country.

It was noted that 11 countries expressed the need for assistance for the migration and a lot of countries did not respond to the survey

(a) Continue reviewing the analysis of the monitoring reports on the migration, and, when a deficiency is identified, ensure that the TDCF bulletins are distributed on the GTS and that Volume C1 is updated;
(b) Continue ensuring that they provide monitoring results for the monitoring exercises co-ordinated by the Secretariat, in particular the Integrated WWW Monitoring (IWM), and post the monitoring results on the WMO server as soon as possible after the end of the monitoring period,
(c) Support the step-by-step migration before stopping the parallel distribution of TAC and TDCF data ;
(d) Convert TAC data into TDCF data when the parallel distribution of TAC and TDCF data will be stopped and if an associated NMC is still not in a position to send TDCF data,
(e) Convert TDCF data into TAC data when the parallel distribution of TAC and TDCF data will be stopped and if an associated NMC still need receiving TAC data.

4.3.3 Outcomes of the second meeting of the IPET-DRC related to the migration to TDCF

A key recommendation on MTDCF by IPET-DRC was not to extend the target dates in the original migration matrix. However, to acknowledge the reality that many WMO member states will not meet the target dates for Category 1 data (SYNOP, TEMP, PILOT and CLIMAT), the group agreed that dual dissemination of TAC and BUFR could pass the original target dates and agreed to add the deadline of November 2014 to stop dual dissemination of Category 1 TAC and BUFR data. The deadline of November 2014 will also apply to Category 4 (maritime) TAC and BUFR data. The purpose of this decision is to urge those Members who have not progressed or implemented their national migration plans to do so as soon as possible in the next four years. The meeting agreed with the migration matrix being considered by IPET-DRC and supported working collaboratively with the RTH centres in the step-by-step migration before stopping the parallel distribution of TAC and BUFR data. The situations where one NMHS needs to assist another NMHS in the conversion between TAC and TDCF data or vice versa are quite complicated. Appropriate arrangements should be made between the NMHSs concerned when the TAC to TDCF conversion or vice versa of one NMHS are generated by another NMHS and insert the converted data into the GTS on behalf of the other NMHS. It was agreed that formal procedures and practices should be developed for each scenario such as the following:

(a) Conversion of TAC to TDCF for NMHS1 TAC bulletins by NMHS2 on behalf of NMHS1 at the request of NMHS1
In this case the NMHS1 are countries which are unable to meet the TDCF migration plan of WMO. NMHS2 is the country offering the assistance to NMHS1 countries for the conversion of NMHS1 TAC bulletins to TDCF bulletins. This is most likely to happen for CAT1 (Common) TAC bulletins after November 2010 and for CAT4 (Maritime) TAC bulletins after November 2012 when the migration plans for each of the data types are scheduled to have completed.

The arrangements should include the definition of the abbreviated headings of the bulletin(s) comprising the converted data. It is recommended that the abbreviated headings (TTAAii) to be those that the NMHS1 issuing the original data would have used, but preferably selecting a specific location indicator CCCC of that NMHS1 to mark that the data were converted by NMHS2. Both centres should ensure that the bulletin(s), comprising the converted data, are added to Volume C1 and that the information, that the data were converted by the NMHS2, is included in the column “remarks” of Volume C1 for each relevant bulletin.

(b) Conversion of TDCF to TAC for bulletins generated in other zones by NMHS1 for use in NMHS2 on bilateral arrangement
In this case the NMHS2 countries are unable to receive or decode any TDCF. NMHS1 is the country offering the assistance to NMHS2 countries to convert TDCF to TAC for bulletins originated in other zones for use in NMHS2 countries. This is most likely to happen when the respective migration plans for the data types have completed and parallel distribution of TAC and TDCF bulletins have stopped by the third party countries in other zones. As the converted TAC bulletins are redundant to countries outside the zone of NMHS2 countries and they contain incomplete or less information than their corresponding TDCF bulletins the distribution of the TAC bulletins should be restricted to the zone of NMHS2 countries and routeing of the converted TAC bulletins to other centres should be suppressed.

The arrangements in this case should include the defined minimum number of stations required and the definition of the abbreviated headings of the bulletin(s) comprising the converted data. It is recommended that the TAC bulletins converted from TDCF should only be distributed to the NMHS2 and other NMHS under this bilateral arrangement within the zone to avoid the bulletins returning on the GTS and overwriting the corresponding reports in TDCF. A specific location indicator CCCC of NMHS1 should be selected to indicate that the TAC bulletin is generated by NMHS1 and its distribution should be restricted by RTH/MTN centres.

(c) Conversion of TDCF data to TAC for exchange on GTS
In the event that a third party country has stopped the parallel distribution of TAC and TDCF before the TAC termination schedule as given in the WMO TDCF migration plan and has agreed another NMHS to provide the TDCF to TAC conversion on their behalf for exchange on GTS, it is agreed that the NMHS providing the conversion can do so with appropriate arrangements to fully describe the TAC bulletins in Volume C1 in the "remarks" column indicating that the TAC bulletins are converted by them. A specific location indicator CCCC should also be used to identify the TAC bulletin in case some Member states do not want to receive the bulletins.

4.3.4 Monitoring of data in TDCF

In view of the impending termination of parallel distribution of TAC and BUFR data in November 2014 the meeting agreed to extend the monitoring of the availability of other types of TDCF data to include CLIMAT, upper-air and marine data. Melbourne and Tokyo have commenced the collection of BUFR CLIMAT and marine data, Beijing has agreed to start soon and Brasilia also agreed to consider joining as SMM raw data collection centre. Melbourne is able to serve as SMM pre-analysis centre for BUFR surface, marine and CLIMAT data.

4.4 Delivery of CAP messages over GTS

The meeting was informed of the plan of EUMETSAT for CAP messages to be distributed on the GTS. The intended abbreviated header EUMETSAT arranged with RTH Offenbach for the CAP messages was tentatively allocated for T1T2 = WO. Due to the extensive use of T1T2 = WO for ‘Other Warnings’ including gale warning and aviation warning it was determined an exclusive designator for the CAP messages would be more appropriate. The use of designators either T1 = ‘M’ or T1 = ‘X’ would be more suitable.

The meeting recommended the allocation of T1=X to “Common Alert Protocol (CAP) messages” instead of using T1T2=WO as suggested by some other centres to avoid conflicts with existing warning messages and in anticipation of the widespread application of CAP encapsulated messages in the near future. The current allocation of T1=X for “GRID products for regional use” is very uncommon since GRID products were terminated several years ago. The meeting suggested any products currently using T1=X should be investigated and new allocations of T2A1A2 to be further discussed.

There was also a consensus among members that CAP messages should be delivered as binary files instead of alphanumeric messages to avoid corruption to the text of the CAP messages as they go through switching centres.

4.5 Implementation of the file naming convention

The meeting discussed a number of file naming conventions put forward by ET-CTS and IPET-MDI. There were no objections to the metadata file name extensions suggested by ET-CTS. The recommendation by IPET-MDI to allocate a new file name extension ”iso19139.xml” was rejected as it conflicts with the existing guidelines in the Manual on the GTS.

5. OPERATIONAL INFORMATION SERVICE (RELATED TO ISS)

5.1 Operational information Services related to ISS

The objective of the Operational Information Services (OIS) is to collect and distribute to WMO Members and WWW Centres detailed and up-to-date information of the facilities, services and products made available in the day-to-day operation of the WWW. A range of the operational information including WMO Volume A – Observing Stations, WMO Volume C1 – Catalogue of Meteorological Bulletins, Routeing Catalogues of Bulletins, etc. are key information for the operation of the GTS/MTN centres. In particular the responsibilities for maintaining the Catalogue of Meteorological bulletins by WMC and RTH centres on the MTN are standard (mandatory) practices and procedure under the Technical Regulations of WMO. It is therefore mandatory that Members shall do their utmost to implement the standard practices and procedures.

5.2 Status of updates of the catalogue of meteorological bulletins and routeing catalogues

The lack of maintenance by five MTN centres particularly RTH Washington since 2002 on their parts in WMO Volume C1 – Catalogue of Meteorological Bulletins was discussed in the meeting and also in the joint session with ET-CTS. In the absence of representatives from NWS in the ET-OI meeting and a formal replacement of RTH Focal Point for Washington the chair sought the cooperation of Mr Ashgar Noor, member of ET-CTS during the joint session of the two groups. It is very important to have the support of all the RTH/MTN centres in operational and implementation matters especially during the period of WIS migration. The level of cooperation of some RTH centres in matters related to OIS has been disappointing. The complete lack of information on bulletins delivered from RTH Washington and the NMCs in its area of responsibility in WMO Volume C1 since 2002 has a serious impact on the results of various monitoring exercises coordinated by WMO Secretariat.

5.3 GeoNetwork and WIS DAR interface

One of the requirements of WIS is the implementation of a Discovery Access and Retrieve (DAR) catalogue for the searching and accessing of data and products. The catalogue needs to be compliant with the international standards of ISO 19115 (Geographic information-Metadata) and ISO 19139 (Geographic information-Metadata-XML implementation) and has to provide searching via ISO 23950/SRU (Search and Retrieval by URL). In addition, the catalogue must be able to harvest and be harvested by nodes using the Open Archives Initiative Protocol for Metadata Harvesting (OAI-PMH). This protocol will be used for metadata synchronisation between GISCs and also for uploading of metadata from DCPCs and NCs to a GISC.

The meeting noted that GeoNetwork is a standards-based, open source catalogue application for managing spatial metadata and data and it provides a web based interface for creating, importing, validating, harvesting and editing metadata records. Searching can be done on local and distributed catalogues based on a variety of criteria such as titles, keywords, spatial and temporal extents. ISO profiles are supported and scheduled harvesting and synchronisation of metadata records from other distributed catalogues is available.
The ability to use GeoNetwork to satisfy the WIS requirement of a DAR catalogue was trialled at a number of WIS centres where WMO Volume C1 metadata was populated and synchronisation of the metadata was conducted between WIS centres. The reports on installation and testing of the GeoNetwork software indicated that it was able to fulfil most of the requirements of WIS.

6. QUANTITY MONITORING OF THE OPERATION OF THE WWW

6.1 Quantity monitoring of the operation of the WWW

The Secretariat provided a status of the implementation of the Integrated WWW Monitoring (IWM). As shown in the following table, ten MTN centres provided IWM RTH/MTN reports during the monitoring period October 2009 – July 2010. However, not all of the centres provided reports for each of the monitoring exercise.

IWM monitoring period	MTN centres provided IWM RTH/MTN reports
October 2009	Algiers, Beijing, Cairo, Melbourne, Prague, Tokyo, Toulouse
January 2010	Melbourne, Moscow, Tokyo, Toulouse
April 2010	Brasilia, Melbourne, Tokyo, Toulouse
July 2010	Brasilia, Dakar, Melbourne, Tokyo, Toulouse

The IWM leads to the integration of the Annual Global Monitoring (AGM) and the Special MTN Monitoring (SMM) into one monitoring scheme. The support of the RTH centres in IWM is important on the sharing of the responsibilities of the monitoring between the WWW centres and the Secretariat. The meeting urged all RTHs:

(a) To coordinate the implementation of the IWM with their associated NMCs;

(b) To provide IWM RTH reports, including the monitoring reports of their associated NMCs;

(c) To examine the analysis of the monitoring exercises as provided and posted by the Secretariat on the WMO server, and to take further action to mitigate the deficiencies observed, in particular within their zone of responsibility.

6.2 Development of Integrated Quantity WWW Monitoring Application (IQWMA)

The meeting was presented with a summary of the development of the Integrated Quantity WWW Monitoring Application (IQWMA). Reference is made to the document describing the monitoring tasks of the Quantity Monitoring of the Operation of the WWW prepared by the Secretariat:

ftp://ftp.wmo.int/GTS_monitoring/JENN_monitoring_datasets/qty_monitoring_tasks_T1-to-T10.doc

The meeting noted that China Meteorological Administration (CMA) has kindly offered the assistance for the development of the IQWMA which is expected to be completed by the end of this year. IQWMA is a desktop application operating in Windows XP, Windows Vista, Windows 7 and SUSE Linux using MySQL, Java and open source tools. The functions being developed are based on the monitoring tasks outlined in the document. The Secretariat thanked CMA for their contribution to the quantity monitoring activities.

7. WIS PLAN AND ROADMAP

7.1 This section was focused on current work being undertaken in the preparation of an amendment to WMO 49 Technical Regulation (Volume 1, Section A3) to incorporate WIS, and to introduce the new draft Manual on WIS and Guide to WIS. These draft documents are available on the WIS Wiki at http://www.wmo.int/pages/prog/www/WIS/wiswiki/tiki-index.php?page=WIS-manual. In the joint session of the ET-CTS and ET-OI, the meeting was briefed by the Secretariat (Eliot Christian and Jean Michel Rainer) on the status of the documents, and the timeline for their final presentation to Congress in June 2011. The meeting noted that Section A3 of the Technical Regulations had changed little with respect to its description of the GTS but that it had been reworked to incorporate the key aspects of WIS and to introduce the Manual on WIS. It noted that the Technical Regulation’s amendments would be presented to CBS Ext 2010, along with the new Manual on WIS. Both documents would be in all official WMO languages. The Guide will hopefully also be provided in the six languages, or at least English and French.

7.2 The draft documents have been available on the WIS Wiki since version 0.1 and that all input has been incorporated. The current versions of the documents are Version 0.4 for the WMO No.49 and Version 0.5 for the Manual and Guide. A news item is published on the WIS web page (www.wmo.int/wis) announcing each release. This has an RSS feed so it can be subscribed to. Current versions will be presented to the ICT-ISS (27-30 September 2010) and to the ICG-WIS (1 October 2010) for review, and then updated and sent to the WMO Language Service Program for translation and preparation for CBS Ext 2010 to be held in Namibia 17 to 24 November. Because of the tight translation schedules, the key deadlines for further input will be up until 1 October 2010, then at CBS and finally at Congress XVI in June 2011.

7.3 The joint session of the ET_CTS and ET-OI also reviewed proposed amendments to the Manual on GTS and Attachment II-15 and Attachment II-16. Details of these are discussed under the relevant sections above and will be presented to the ICT-ISS September meeting. The joint session also covered operational procedures and performance monitoring.

8. OTHER BUSINESS

None.

9. CLOSURE OF THE MEETING

9.1 The meeting closed on Friday 24 September 2010 at 16h30.

Annex to paragraph 1.1

LIST OF PARTICIPANTS

Mr Kelvin WONG (Chair)	Bureau of Meteorology G.P.O. Box 1289 MELBOURNE, VIC 3001 Australia tel: +(61 3) 9669 4227 fax: +(61 3) 9669 4128 k.wong@bom.gov.au
Mr Leonid BEZRUK (Co-chair)	Main Radiometeorological Center B. Predtechensky st. 13 123242 Moscow m/651 Russian federation Tel: +74997952260/2551477 Fax: +74992525504 bez@mecom.ru
Mr José GIANNI	Servicio Meteorológico Nacional 25 de Mayo 658 1002 BUENOS AIRES Argentina tel: +(54 11) 5167 6767 ext. 18243 fax: +(54 11) 5167 6711 gianni@smn.gov.ar
Ms Xiang LI	China Meteorological Administration 46 Zhongguancun Nandajie Haidian District BEIJING 100081 China tel: +(86 10) 6840 6275 fax: +(86 10) 6217 3225 lixiang@cma.gov.cn
Mr Jacques ANQUETIL	Météo-France (DSI) Direction des systèmes d'information 42, avenue Gaspard Coriolis F-31057 TOULOUSE Cédex 1 France tel: +(33 5) 6107 8572 fax: +(33 5) 6107 8109 jacques.anquetil@meteo.fr
Mr Bernd RICHTER	Deutscher Wetterdienst Frankfurter Str. 135 D-63067 OFFENBACH Germany tel: +(49 69) 8062 2559 fax: +(49 69) 8062 3559 bernd.richter@dwd.de
Mr Kenji TSUNODA	Japan Meteorological Agency Information and Telecommunication Division 1-3-4, Otemachi Chiyoda-ku TOKYO 100-8122 Japan tel: +(81 3) 3218 3825 fax: +(81 3) 3211 8404 tsunoda@met.kishou.go.jp
Mr Henry KARANJA	Meteorological Department Dagoretti Corner Ngong Road P.O. Box 30259 00100 NAIROBI Kenya tel: +(254 722) 364465 fax: +(254 20) 3876955 hkaranja2001@yahoo.com hkaranja@meteo.go.ke
Mr Chris LITTLE	Met Office FitzRoy Road EXETER United Kingdom of Great Britain and Northern Ireland tel: +(44 1392) 88 62 78 fax: +(44 1392) 88 56 81 chris.little@metoffice.gov.uk

WMO Secretariat
Mr David THOMAS	WIS Project Manager WMO Information System Branch Observing and Information Systems Department World Meteorological Organization Tel: +(41 22) 7308241 Fax: +(41 22) 7308021 dthomas@wmo.int
Mr Pierre KERHERVE	Consultant, Data Representation, Metadata and Monitoring Division WMO Information System Branch Observing and Information Systems Department World Meteorological Organization 7 bis, avenue de la Paix CH-1211 Genève 2, Switzerland Tel: +(41 22) 7308218 Fax: +(41 22) 7308021 pkerherve@wmo.int

Annex to paragraph 2.1

Summaries of RTH reports presented to the meeting.

Status of RTH Beijing - Item 2.1(1)

Ms Li Xiang presented document 2.1(1). She addressed the status of GTS links and provided details of the type of connectivity including five MPLS circuits to Offenbach, Tokyo, Moscow, New Delhi and EUMETSAT, six leased lines to Bangkok, Pyongyang, Ulaanbaatar, Hanoi, Hong Kong and Macau, and one frame relay to Seoul and two Internet links to Melbourne and Almaty. Four circuits were backed up utilising the Internet. Key recent improvements included upgrading Hanoi and Mongolia in 2009 and 2010 to digital 64K. Also, the MPLS links to Offenbach, Tokyo, Moscow, New Delhi and EUMETSAT were upgraded from 2Mbps to 3Mbps in April 2010. The Internet is playing an important role with FTP services running in parallel or as back up of GTS circuits. The Internet is also being used to exchange large volume model and satellite data, including TIGGE exchange with NCAR and ECMWF. Ms Li Xian showed that CMA’s GTS system was collecting 490GB per day and distributing 91GB per day. The meeting noted that of the 490GB of data collected via the GTS and Internet per day, that 450GB of traffic was TIGGE data coming from 10 centres.

Ms Li Xiang provided an update on CMA’s satellite distribution systems PCVSAT (Ku-band) and FengYunCast (C-band) and briefed the meeting on their new system CMACast. She noted that all PCVSAT and FengYunCast systems will be migrated to CMACast which has a capacity of transmitting more than 400GB daily. The meeting noted that as well as being a core component of WMO’s IGDDS, CMACast (as has FENGYUNCast and EUMETCast) will be a core system of GEONETCast complementing NOAA’s GEONETCast Americas.

Ms Li Xiang also briefed the meeting on CMA’s progress at implementing GISC Beijing. In its goal to be a GISC in WIS, CMA has implemented the key features of a GISC including metadata search, data subscription and data request. CMA will be demonstrating their GISC at CBS Ext 2010 and plan to run operationally in 2011.

Status of RTH Brasilia - Item 2.1(2)

As Mr José Mauro de Rezende was unable to attend the meeting, Brasilia’s report was presented by the Chair. The report highlighted that INMET’s Linux based IBL message switching system (MSS) is planned to be updated along with the installation of a GISC solution. It noted that the traffic volume being switched to and from WMC Washington showing about 350Mbytes inwards a day and 1.8Gbytes outwards per day. The meeting noted that INMET’s plans to connect to the MPLS core network of WIS, and that although there are some issues to be resolved first, this plan combined with the recent improvements in the Internet bandwidth to 16Mbits/s and the planned upgrade of the IBL MSS to enable full WIS functionality are progressing INMET’s goals to implement a GISC. The meeting also noted Mr Mauro’s efforts to increase Brazil’s participation in monitoring of the WWW and willingness to participate in WIS monitoring activity.

Status of RTH Buenos Aires - Item 2.1(3)

Mr Jose Luis Gianni reported on the status of RTH Buenos Aires, which is in the process of renovating its telecommunication infrastructure. Work to date has seen the replacing the national VSAT Satellite Network with MPLS. VSAT units will remain on site as back up in most major sites and the Internet will also be used in some cases. He reported that VPN over the Internet has also been established across all 13 National Meteorological centres in RA III as a first stage to moving to MPLS over the region. The plan is to have Buenos Aires and Brasilia as the AMDCN connection points to the WIS core MPLS network with Brasilia taking on the role of the GISC node and Maracay and Buenos Aires continuing as RTH DCPCs.

Buenos Aires is assessing GeoNetwork and IBL Moving Weather MSS to support its new WIS functionality and network topology, and last year increased its Internet bandwidth from 17 to 24 Mbits/s and ultimately to 30 Mbits/s along with a change of internal topology and hardware. It does have concern about the reliability of the Internet and have observed system failures under heavy load. To address this, it is investigating adding a second ISP 2Mbps to improve reliability and stability.

Unfortunately, the Buenos Aires office will be moving next year to a location which is not presently equipped with the required communication facilities putting pressure on available funds and resources for completing present and future projects related to WIS and the GTS. The link to Buenos Aires – Brasilia will rely on their VPN over the Internet during the move of the Buenos Aires office and until the connection to the Core WIS MPLS network is enabled. Even without the office relocation, there are problems with access to staff with the required ICT skills, so the combination of these issues is adversely affecting the implementation of WIS in the region. A particular issue of concern from the meeting was the lack of updates to Vol C1 coming from RA III, which is affecting operations in other regions.

In addition to his report, Mr Gianni reflected that WIS implementations in RA III are very slow. NMCs are depending heavily on RTHs (also with their own problems). As a reference to keep in mind, the workshop about VPN implementation in RA III held last year in Brasilia was a very good and successful idea. It allowed to address and solve such a difficult issue in "a reasonable time". It was planned to do the same with TDCF/TAC migration which could be extended to WIS/GTS generalities. He also noted there should be a mechanism to push forward the RA III PRs concerns about the slow implementation of WIS in the region.

Status of RTH Exeter - Item 2.1(4)

Mr Chris Little presented the report on RTH Exeter addressing the connectivity, bandwidth and volumes of messages and files from Exeter. He noted that applications supporting the 17 GTS circuits to RTHs and NMCs and other links to major centres such as NESDIS and ECMWF consisted of a message switching systems (Frost), file switching systems (DART) and email gateway systems (IronPort). He also noted that the figures presented for the Internet did not separate their large national distribution from international WMO traffic, although this could be done in the future. He showed the MSS receives 3.6GB and sends 6.4GB a day on the GTS circuits while the file switch receives 11GB and sends 5GB a day. Non-GTS connections receive 14.3GB and send 54GB a day. Internet traffic receives 7.2GB and sends 100GB a day. Recent trends, including the addition of the exchange of European radar images (Opera) shows these figures are likely to double in the next one to two years.

Mr Little highlighted that the availability of the RTH switch had been 100% for the FTP switch and email gateway, and that the WMO message switch had been 99.95% between September 2009 and August 2010.

In summarising, Mr Little highlighted that the WMO GTS traffic, including some FTP Internet links, represents only 15 to 20% of total MSS activity. He noted that the RMDCN access is presently 2 by 2Mbps links with no load sharing, with a plan to upgrade to 2 by 4 or 6Mbps to support Opera Radar, and possibly upgrade again for more WIS traffic. A link between Exeter and Tokyo had been established in 2009 and they are hoping to establish a link to New Delhi this year. It was noted that there has been no recent progress on connecting Pretoria to the RMDCN MPLS. Other links include a special 40Mbps link to ECMWF, a 1Mbps trans-Atlantic NESDIS connection and 2 by 100 plus Mbps links to the Internet.

Status of WMC Melbourne - Item 2.1(5)

Mr Kelvin Wong provided the WMC Melbourne report. He addressed the improvement of MTN and regional circuits in RA V including the network migration of WMC Melbourne. RA V has three MPLS carriers. OBS MPLS replaced the BT Frame Relay (FR) circuits in 2009 connecting Melbourne to Tokyo, Washington and Exeter. SingTel MPLS replaced BT FR to Singapore and Jakarta, and NOAANet MPLS connecting USA territories and the NE Pacific. FR circuits remain for connecting to Nadi and Noumea. Many other centres use VPN over the Internet, including those NMCs connected via Wellington as well as many centres in other regions.

Mr Wong provided daily traffic information for the various managed GTS circuits totalling about 1.7 GB into Melbourne and 0.9GB out. He also described in detail GTS traffic on VPN over the Internet circuits of 10GB inwards and 1.8GB out of Melbourne. With regards to the Internet circuits, he demonstrated the significant improvement in traffic management in Melbourne made possible by the segregation of its public Internet services from its services to WMO Member countries and registered users since October 2009. He also provided a comparison between the UK report and Melbourne. The comparison highlighted the highly un-predictable demands on Internet load driven by national services such as satellite and radar displays and other web services had the potential to swamp and overload Internet based GTS circuits. This is evident in that the combined Internet traffic of national and GTS came to 165 GB inbound and 532GB outbound. Thus he strongly recommended that GISCs supporting public access should closely monitor the data traffic from the public Internet service and consider data segregation if necessary.

Mr Wong also reported on the progress of the implementation of WIS in Melbourne noting that Australia had utilised the WMO secretariat Jump Start offer. This allowed the Bureau to get a better understanding of WIS components and use its existing facilities to build new systems that meet GISC requirements. Activities included installation of GeoNetwork in their DMZ and demonstrating metadata synchronisation with CMA using OAI-PMH. In addition, the Bureau was able to populate the WIS DAR metadata catalogue with GTS data and product metadata records obtained from CMA and generate new metadata records from its nationally integrated system of clients and product database management system thus putting the Bureau in a better position to optimise the exploitation of OpenWIS to their existing systems when it becomes available.

Mr Wong provided some details in the method of analysis to determine the usage of bandwidth and traffic volumes necessary to manage WIS requirements into the future. He showed that analysis needed to resolve the loads on circuits that would be hidden if the sampling period is too long. Circuits can look healthy and satisfactory when analysing average daily statistics available from the router monitoring software (12 hour sampling), but that when resolving to 2 minute sampling it can be seen that circuits become quite saturated at model and data release times. Such sampling can have significant affects on planning performance for the proposed WIS All Hazards Warning network which requires warnings to be distributed globally within 2 minutes.

In addition to his presentation on the status of WMC Melbourne, Mr Wong took the opportunity to discuss the depiction of RMTNs and AMDCNs. He highlighted the network diagrams from Buenos Aires, Moscow and Melbourne. The meeting agreed that although the cloud presentation used for RA III was effective, for more complex regions, the depictions of network clouds showing logical links as per the Moscow diagrams provided the maximum usage of the diagrams and were an improvement over the traditional spaghetti diagrams shown in Melbourne’s presentation. However, the discussion was not conclusive and suggests wider investigation into this be undertaken.

Status of WMC Moscow - Item 2.1(6)

Mr Leonid Bezruk reported on the status of Moscow. He noted that the MSS in Moscow was modernized in 2010 and is based on TCP/IP procedures. Moscow is supporting 21 GTS circuits, both regional and inter-regional. Most of these utilise MPLS, digital lines or the Internet, although there are a limited number of asynchronous connections remaining. The Russian satellite broadcast system DVB-S “MITRA” is also integrated into the MSS. Messages can also be received or transmitted via e-mail. A/N messages and fax charts can be received via a web-interface. Planned upgrades include increasing the RMDCN access circuit up to 10Mbps as well as increasing the bandwidth to Khabarovsk, Minsk and Kiev. They also plan to migrate Moscow-Tbilisi, Moscow-Baku, Moscow-Erevan, Moscow-Kishinev links to RMDCN or establish digital lines. The meeting noted the diagram in the report for depicting network topology (slide 9), which was thought to represent well the new cloud structure of networks, while maintaining the features necessary to understand the application layer connectivity between centres.

Mr Bezruk provided details of the daily number of messages and files and their volumes for each circuit. For GTS circuits, incoming totalled to 2.1GB of messages and 0.6GB of files. Outgoing messages totalled 1.2GB. Internet incoming was 78GB and outgoing 44GB. In addition to the 21 GTS circuits, the Moscow to Washington link passes 50GB of NCEP products, 4.7GB of COSMIC data and 5MB of POLAR WIND data.

Mr Bezruk also described WMC Moscow’s contribution to the IGDDS MITRA © (Multi-address Information TRAnsmission) along with the products it now distributes. This is a main system of ROSHYDROMET (Russia) for large volume of meteorological data delivery. The MITRA technology uses DVB-S standard as a base for data transmission via satellite channels, and specialized application software for data loss protection and provision of guaranteed content delivery.

The Moscow report included information on Monitoring for internal performance and in support of the WWW AGM, SAM and IWM. Mr Bezruk also provided an update on the progress of WIS implementation in GSIC Moscow and the various DCPCs that have been identified. The contract for establishing the GISC and related software has been awarded and the date of completion of the project is in the first quarter of 2011.

Status of RTH Nairobi - Item 2.1(7)

Mr Henry Karanja presented the report for Nairobi providing details of the RTH’s national and international roles including network connectivity, interfaces, details on the MSS and number of messages switched, future plans and the results of Nairobi’s monitoring processes. He also reported on the migration to TDCF. The meeting noted the effective use of the WMO FTP over the Internet in the region which was complemented by the use of satellite broadcast systems. Some participants were very concerned about the ongoing problems with NMHSs in getting messages to Nairobi for distribution on the GTS, but appreciated Nairobi’s efforts in making significant improvements in many areas.

Mr Karanja reported that the 4Mbps Internet link provided the main link to the MTN via Offenbach as well as supporting the operational regional links to Entebbe, Dar-Es-Salaam, Addis Ababa, Pretoria and Reunion. He noted that the 9.6kbps X25 link to Cairo was not operational and is to be migrated to the Internet. Other regional links to be migrated to Internet links are Kigali, Bujumbura, Djibouti and Mogadishu, all of which presently use 50 Baud AFTN links. In addition to the use of the Internet for national links, 23 centres report via GPRS. Traffic through the MSS was reported in messages received via ftp per hour, showing traffic more than doubled from around 1600 to nearly 4500 at the synoptic reporting hours.

Mr Karanja reported that data distribution in Kenya is supported by satellite based stations, in particular EUMETCAST receivers, RANET and RETIM-Africa. The meeting noted that RETIM-Africa is moving to be transmitted over EUMETCAST so will now be able to use the same EUMETCAST receiver.

The meeting noted Mr Karanja’s report on the migration to TDCF including training programs and software development. He reported that the RTH is presently receiving TDCF (BUFR and CREX) data from the stations for testing at National level, and that regional exchange was scheduled to start testing in September 2010, with operational parallel exchange of TAC and TDCF starting in October 2010.

Status of RTH Tokyo - Item 2.1(8)

Mr Kenji Tsunoda presented the report on RTH Tokyo showing the progress in upgrade of GTS circuits in RA II and on the MTN. A highlight was the replacement of MTN BT frame relay circuits in 2009 with the new OBS MPLS so that Tokyo now has MPLS connections to Beijing, Exeter, Melbourne, Washington and India through the RMDCN. This move was a major step in the merging of the two IMTN clouds into one. The OBS MPLS has two 3Mbps load balancing connections to Tokyo. Bangkok, Hong Kong and Manila are connected to Tokyo via a 2Mbps NTT MPLS cloud. Khabarovsk and Seoul remain connected to Tokyo via dedicate 64 and 128Kbps lines respectively. Mr Tsunoda reported that the total volume of traffic (Text, Binary, Fax, Message and File) is presently 3.5GB outwards and 2.8GB inwards. This compares to 0.6GB outwards and 1.1GB inwards in 2006. He showed that more than half the traffic in and out of Tokyo was binary. Files represent about one third of the traffic and text less than 10%. Fax’s account for only 1% of the traffic.

The report highlighted that TCP/IP is now used on all Tokyo’s GTS links. Applications are using simple FTP put and get to exchange files. Some file renaming takes place on most circuits, utilising the GTS file-naming convention.

RTH Tokyo’s report included information on the use of the Internet for services such as JMA's DDB (Distributed Data Base), NEAR-GOOS (North East Asia Regional - Global Ocean Observing System) Regional Real Time Data Base and WDCGG (World Data Centre for Greenhouse Gases). It highlighted that it has two permanent connections to ISPs, which provides a more stable and reliable access to the Internet than just one ISP.
Mr Tsunoda also summarised Tokyo’s role in the routine monitoring of the World Weather Watch, in particular the IWM (AGM/SAM) and the SMM.

Status of RTH Toulouse - Item 2.1(9)

Mr Jacques Anquetil reported on the status of RTH Toulouse and included Noumea and Papeete. It has 12 GTS links via the RMDCN, which is connected at 8Mbps. Three links are via VSAT (ASECNA) providing 64Kbps connectivity and two links via leased lines at 64Kbps. Morocco is also connected by leased line at 128Kbps, but as Morocco has now joined the RMDCN, this link will be moved to be via RMDCN. The protocol is FTP on all circuits, mostly WMO FTP, but some simple FTP get or put are also used. Permanent backup is provided via the Internet for the VSAT links to Dakar and to Niamey. RTH connection to the Internet is 150Mbps. In addition to supporting research traffic and web services for Météo France, the Internet provides backup for the RMDCN in pull mode. Mr Anquetil showed that the RMDCN link was approaching 8Mbps inwards usage several times per day. He noted that there is a plan to upgrade to 10Mbps in 2011.

The MSS for RTH Toulouse is known as Transmet. Although it is being planned to be upgraded in 2011 to merge processing of Alpha and graphic data, at present these are handled separately. For Alpha, traffic is about 600MB inwards per day and 7GB outwards. Graphics traffic is about 74GB inwards per day and 450GB outwards per day. Included in Transmet’s dissemination lings are email and fax. 4000 emails are sent to any of 1200 distinct addresses. For fax, about 300 faxes are sent per day, while a further 900 are sent on “in calling”. A file switching service (Difmet FSS) has been in operation since around June 2010. This provides a more flexible file distribution service and supports ftp, e-mail, fax, SMS and voice.

The report noted that Météo France satellite distribution system RETIM has now completed its migration to be transmitted over EUMETCast and is fully operational. The old RETIM 2000 system was stopped on 16 August 2010. RETIM on EUMETCast receivers are standard PCs under linux with PCI DVB card. The system is compatible for a reception on all EUMETCast receivers and there is no change required for the end user applications. Mr Anquetil described how the data is uploaded for RETIM Europe and re-uploaded at a turnaround station in Italy for RETIM Africa.

Mr Anquetil described the Opera Data Centre that will be supplying high quality radar composites for Europe. This centre distributed between Exeter and Toulouse is planned to be a DCPC in WIS. He also provided details on the progress of the VGISC development and operational schedule he noted that the core applications being developed is OpenWIS in collaboration with the Met Office, KMA and BoM. OpenWIS will be applicable to GISC, DCPC and NC functions. It is presently in phase 1 of development until November 2010, phase 2 of development for the first quarter of 2011, then acceptance testing before going operational early in the second half of 2011.

Status of RTH Algiers - Item 2.1(10)

Mr Henry Karanja provided a report on RTH Algiers. The report describes the national circuits in Algiers building on 64Kbps leased digital lines and a message switching system called MESSIR-COM. This MSS is also described and plays a major role in managing the collection of observations from their national network, reception of satellite products from RETIM and EUMETCast as well as aviation products from SADIS. International GTS circuits include Digital 64Kbps links to Toulouse, Cairo, Jeddah and Casa. The Toulouse link is planned to upgrade to 256Kbps. Other links include an X25 2.4Kbps link to Tunis supplemented by VPN over the Internet. The Internet also connects to Tripoli, while Niamey and Dakar are connected through VSAT at 9.6Kbps. The report notes that connecting to the RMDCN is planned but that the costs and benefits are less advantageous than existing links.

The report also addresses Algiers role in monitoring of the WWW as well its progress on migrating to TDCF, in particular noting that the RTH has been able to encode and decode BUFR since October 2008

Status of RTH Dakar - Item 2.1(14)

Mr Saidou Diene, Dakar RTH focal Point, submitted a report on RTH Dakar. He noted that Dakar had one MTN circuit (Dakar/Toulouse), one interregional circuit (Niamey/Toulouse), five main regional circuits (Dakar/Algiers, Niamey/Algiers, Dakar/Niamey, Brazzaville/Niamey and Dakar//Brazzaville) and 26 regional circuits. Centres included the RTH Dakar MTN centre plus two RTHs out of the MTN and 28 NMCs. The MTN link was via VSAT (28Kbps) with backup provided by VPN over the Internet. Both the primary and backup use FTP over TCP/IP. Interregional (9.6Kbps) and regional (2.4Kbps) circuits are all via VSAT. Protocol over VSAT was X25, and although this remains the case for 11 links, several have migrated to V.24 protocol. The migration has been driven by problems with X25 service since July 2010.

The region has a strong reliance on satellite distribution systems including EUMETCast, the ICAO Satellite Distribution System 2nd Generation (SADIS 2G) and/or the Météo France Satellite distribution system for Africa (RETIM-Africa). Many of the receiving stations have been supplied through the PUMA project and by ASECNA. The report includes details of receiving stations at all centres, and includes reference to recent training and technical issues. The meeting noted that RETIM-Africa has switched to being broadcast via EUMETCast since August 2010.

Details are also provided for MSS, Internet and e-mail services. All the western Africa meteorological services have Internet access, but many have not developed a web server and/or VPN services, etc. Message switching for the RTH and Abidjan, Douala, Malabo and Libreville utilise Corobor’s MESSIR-COM. Dakar and the RTHs use more powerful versions, while smaller versions are equipped at the NMCs. The MSSs are ready for the Table Driven Code Forms (TDCF) and WIS migration: full support of migration from Traditional Alphanumeric Codes (TAC) to TDCF including support for Cat1, Cat2, Cat3 codes and WMO – DR&C templates. Not all NMCs have MSS. More details are provided in the report.

The report notes that the lack of coordination between RTHs, RTHs and NMCs in particular for the GTS plan implementation, links upgrade and for resume during failure, constitutes a big deficiency in the area. The lack of well trained operational and technical staff adversely affects the operation, the coordination, the implementation and the maintenance of GTS systems. Use of AFTN is limiting the improvement of data exchange in the area, both for national collection and regional observational bulletins exchanged from NMCs to associated RTHs.

Status of RTH Offenbach - Item 2.1(23)

Mr Bernd Richter presented a report on the status of RTH Offenbach. As an RTH on the MTN, Offenbach’s RTH responsibility includes Israel, Jordan and Switzerland. It also is a gateway to the GTS for EUMETSAT and IAEA. It has 14 GTS links via its 2Mbps connection to the RMDCN with an inbound traffic of 1.4GB daily and outbound 16GB daily. It also has 10 links via the Internet to various RA I, II and VI NMCs, but each of these account for less than 1% of its daily traffic. He provided details of the RTH operations centre, noting the 24/7 hours service includes on duty staff, stand by support and incorporates a help desk which supports German and English languages. Including internal national traffic, he reported that the MSS handles 14GB per day inbound messages and files and sends 40GB per day of messages and files. Traffic flow runs at about 2-6 Mbps outbound spiking routinely at 6am and 6pm to 12Mbps.

Mr Richter reported on the implementation of GISC Offenbach (http://gisc.dwd.de) that was launched in pre-operational mode in May 2010. He noted this had completed the software integration into the GISC environment (AFD and Moving Weather) and had integrated GISC operations into the RTH operations. He listed the internal and external DCPCs that will be supported through the GISC.

Annex to paragraph 3.3

Comparison in the times of reception of SYNOP and TEMP bulletins at SMM centres during the July 2008 SMM