______
REGIONAL ASSOCIATION V
IMPLEMENTATION - COORDINATION MEETING ON THE GTS AND ISS IN RA V
WELLINGTON, 1 - 4 DECEMBER 2003 (tent.)
NEW DELHI, 7 - 10 JANUARY 2002 / RA V/ICM-GTS 20023/Doc. 3.2(1)
(231.IXII.20013)
______
ITEM 3.2
ENGLISH only
RMTN operational procedures
(Submitted by the Secretariat)
Summary and Purpose of DocumentThe document presents the outcome and recommendations of the Commission for Basic Systems (CBS) at its extraordinary session (Cairns, Australia, 2002)Implementation Co-ordination Meeting on the Main Telecommunication Network (Geneva, June 2001) as regard GTS operatiooperation and information exchangenal matters. The document includes follow-up action proposed by the Secretariat.
ACTION PROPOSED
The meeting is invited to:
- Review the outcome of CBS-Ext. (02) (Cairns, Australia, 2002) and recommendations of the Implementation Co-ordination Meeting on the Main Telecommunication Network (Geneva, June 2001) as regard GTS operation and information exchange;al matters
- Consider their implementation in the RMTN follow-up action proposed by the Secretariat.
Appendix:A.Follow-up action proposed by the Secretariat as regard the recommendations of the Implementation Co-ordination Meeting on the Main Telecommunication Network (ICM-MTN –2001)(Geneva, June 2001)
B.Outcome relevant to the GTS of first meeting of the Expert Team on Migration to Table Driven Code Forms (ET/MTDCF) (Geneva, May 2001)
RA V/ICM-GTS 20023/Doc. 3.2(1), p. 1
DISCUSSION
1.An Implementation Co-ordination Meeting on the Main Telecommunication Network (ICM-MTN-2001)(Geneva, June 2001) reviewed matters related to the operation of the GTS.
Test of the proposed "relay-to-all" procedure for routeing data on the MTN
2.The former CBS-ISS Implementation Co-ordination Team on Information Exchange Management (ICT-IEM) recommended that all the observational data (excluding radar and satellite data) for global exchange received by a MTN centre from an adjacent MTN centre be relayed to all other adjacent RTHs located on the MTN. The ICT-IEM re-emphasised the importance of the implementation of the procedures to detect and eliminate duplicated bulletins at MTN centres to avoid loops in the transmission of the bulletins on the MTN.
3.With a view to assessing the impact of such a routeing procedure on the exchange of data on the MTN, centres tested it on the same set of bulletins. For this purpose, the Secretariat identified the following lists of SM bulletins, based on the results of the July 2000 SMM exercise, and posted them on the WMO FTP server
(see GTS_monitoring/AGM/TEST_IEM/headings_1.txt and headings_2.txt):
A list of 343 SM bulletins with ii<20, corresponding to the list of bulletins for global exchange,
A sub-list of 35 SM bulletins with ii<20, corresponding to the following areas for which anomalies were detected in the analysis of the AGM or SMM exercises: Southern part of Africa, part of Region II covering Iran, Pakistan, India and Myanmar, Philippines in Region V and parts of Region VI.
MTN centres were invited to implement the test for the 343 SM bulletins, or at least for the list of 35 SM bulletins. The test started in July 2000.
4.CBS-XII noted that the test had some positive impact on the exchange of bulletins involved in the test, and had also revealed some operational difficulties, in particular as regards the elimination of duplicated bulletins. CBS-XII agreed that the new proposed procedure needed further development. It noted with appreciation that a spontaneous group of experts from RTHs started to exchange views on the test. It invited this ad-hoc group to continue working on the test and related proposed changes to routeing procedures under the co-ordination of Mr F. Branski (USA) and to report to the chairman of the OPAG-ISS. Relevant proposals would then be submitted to the relevant OPAG-ISS team.
5.The test showed that bulletins with the same abbreviated headings but with different contents were relayed on the GTS. This was in particular due to the distortion of the bulletins when transmitted on circuits without error detection/correction protocols. This was also due to corrections made by some RTHs to the bulletins, when relaying the bulletins. Since the contents were different, the GTS centres could not detect these bulletins as duplicated bulletins and could not eliminate them. It had not been possible to identify the centres and circuits, which were the sources of the problems. The ICM-MTN-2001 stressed that these shortcomings in the operation of the GTS, as revealed by the test, should be mitigated with a view to ensuring an efficient operation of the GTS. Due to the difficulties experienced during the test, several RTHs stopped their participation in the test.
6. The ICM-MTN-2001 invited the MTN centres and other RTHs to provide the following information to the Secretariat, which will post it on the WMO server:
Information on the procedure implemented at their centres for the detection and correction of errors in the bulletins issued from their zones of responsibility (as an RTH in their respective Region and as an RTH located on the MTN) and in the bulletins issued from outside their zones of responsibility; this information should help in identifying the causes of the problems revealed by the test.
Information on the impacts of the test on the operation of their centres.
7.The ICM-MTN-2001 recommended that the functions and responsibilities of RTHs included in the Manual on the GTS related to the correction of bulletins by RTHs should be reviewed in order to limit correction to the originating NMC and possibly the associated RTH, with a view to avoiding the transmission of multiple bulletins with the same abbreviated headings and different contents. In this respect, the Secretariat prepared a proposal to amend the Manual on the GTS, which is included in paragraph 1.3 of Appendix A. The meeting is invited to consider this proposal.
8.The ICM-MTN-2001 was of the opinion that the test should be continued with the participation of a limited number of centres (e.g. Bracknell, Moscow, New Delhi, Tokyo and Washington) on a limited number of bulletins. The test should then be extended as soon as the existing problems are solved. The meeting recommended that:
The circuits without error detection/correction protocols should not be used in the test;
The bulletins received from outside the zones of responsibility of the RTHs should not be corrected; the bulletins received from their zones of responsibility should be corrected by the RTHs as much as possible in co-ordination with the associated NMCs; a group BBB (CCx) should be inserted into the abbreviated headings of the corrected bulletins in case that these bulletins were already inserted into the GTS through another route.
The bulletins should not be transmitted back to the centres, which originated or compiled them.
Use of different location indicators CCCC to define new Tables in the Attachment II-5 of the Manual on the GTS
9.The increasing number of processed products and data types versus the inherent limitations of the T1T2A1A2ii is a problem that will persist until file exchange, including sensible filenaming and metadata conventions, has superseded the current GTS messages. Any significant change in the composition of the abbreviated heading would affect more than 200 centres that currently operate with T1T2A1A2ii, including for national operation. Changes in the format would lead to large implementation costs that are not worthwhile with a view to future file exchange.
10.Taking into account these implementation constraints, the ICM-MTN-2001 considered a proposal to somewhat extent the possibilities of the current procedures, while the T1T2A1A2ii reaches its limit. A possible extension could result from the use of different CCCC for a WMC, RTH or an RSMC, preparing a large number of bulletins, and to index different tables for T2 - and possibly A1 A2 - to the different CCCC for new types of data and products. This would open a large number of new abbreviated headings, fully compatible with the current implementation at GTS centres and enabling routeing as required. In view of the limited number of RTHs or GDPS centres having these requirements (max. a dozen, ECMF, KWBC, AMMC, EGRR, LFPW, EDZW, RJTD), there should not be difficulty in finding additional CCCCs by changing the last letter only of their current CCCC.
11.As an example, ECMWF expressed the need to allocate T2 for GRIB bulletins containing vorticity and divergence data for the four tropical belt areas, products that are not covered in the current Table B2. ECMWF notified the Secretariat that as from 2 May 2001, ECMWF would continue to use the current T1T2A1A2ii allocations with its CCCC=ECMF, and would use T2 = D for divergence and T2 = V for vorticity with CCCC=ECMW in order to meet this urgent requirement.
12.The Secretariat had invited the RTH focal points to comment on this proposal to use different CCCC for a WMC, RTH or an RSMC, preparing a large number of bulletins, and to index different tables for T2 - and possibly A1A2 - to the different CCCC for new types of data and products.
13.There was a general consensus in the ICM-MTN-2001 that the use of different CCCC for centres (WMC, RTH, RSMC or Meteorological Satellite centre) preparing a large number of bulletins would extend the current provisions. RTH Offenbach expressed difficulties in implementing the procedure of indexing different tables for T2 - and possibly A1A2 - to the different CCCC; these difficulties are not met at the telecommunication level but at the level of the processing of the bulletins when decoding them. The meeting recommended that this procedure should be further reviewed by the ICT-ISS, in light of further experience.
14.The meeting noted that different CCCC are used by WMC Washington to differentiate the bulletins containing processed information issued from different models; in that case, the Tables for T1T2A1A2ii are not indexed to the different CCCC but are those included in the Manual on the GTS. The meeting noted that such a procedure could also be used to differentiate bulletins presented in the code forms GRIB and GRIB2.
15.The ICM-MTN-2001 noted that current Table B2 covers both GRIB/GRID and pictorial information; As proposed by RTHs Melbourne and Toulouse, it could therefore be split into two Tables, a revised Table B2 for GRIB/GRID bulletins (= D, G, H, X or Y) and a new Table B6 for pictorial information (T1 = P, Q). The letters C, F, G, I, L, M, S, X and Y for T2 are only used for pictorial products, and can be allocated to GRID/GRIB bulletins containing new products, such as vorticity and divergence data. The meeting agreed on the principle of using two Tables, a revised Table B2 for GRIB/GRID bulletins (T1 = D, G, H, X or Y) and a new Table B6 for pictorial information (T1 = P, Q). The meeting recommended that the ICT-ISS further consider the detailed allocation of the data type designators T2 in these two Tables.
16. As a follow-up action of the ICM-MTN-2001, the Secretariat proposes the action given in paragraph 2.3 of Appendix A. The meeting is invited to consider that action.
Changes to Tables A, B1 and B2
17.A designator T2 = L was added to table B1 (Attachment II-5, Part II, Volume I, Manual on the GTS) for experimental international exchange of ozone data presented in CREX code form. Canada, Czech Republic and Japan joined this project using this designator. Subsequently, a provision for the allocation of abbreviated headings for data (T1 = K), including ozone data (T2A1 = UL), presented in the CREX code form was introduced into the Manual on the GTS on 1 November 1997. On 5 May 2000, the code form FM 95-XI Ext. CREX entered into force. Noting the present duplication of the allocation of abbreviated headings for ozone bulletins, the meeting recommended that the designator T2 = L be deleted from table B1.
18.The allocation of the designators T2A1A2 is the same for GRID bulletins with T1 = D and T1= G. Some RTHs are still transmitting GRID bulletins to NMCs which are not yet in a position to receive and process binary data (GRIB bulletins). The designators T1 = D and T1= G are used to differentiate the bulletins containing processed information issued from different models or presented with different resolutions. The transmission of GRID bulletins should progressively stop, as the NMCs are able to receive and processed binary data. The meeting recommended to follow the evolution of the remaining requirements for the transmission of GRID bulletins and to delete the designators T1 = D and T1 = G from Table A when possible.
19. At present T2 = O is allocated to oceanographic data presented in the code form FM 63-XI Ext. BATHY, FM 64-XI Ext. TESAC and FM 65-XI Ext. WAVEOB. The meeting considered a proposal to allocate a specific data type T2 for the data presented in the code form FM 65-XI Ext. WAVEOB. The meeting recommended to invite the ICT-ISS to further consider this question, in particular to identify the requirements and possible issues and impacts of such an allocation.
20.As a follow-up action of the ICM-MTN-2001, the Secretariat proposes the action given in paragraphs 3.2.1, 3.4.1 3.6.1 and 3.8.1 of Appendix A . The meeting is invited to consider that action.
Migration to Table Driven Code Forms (TDCF)
21.The main conclusions of the first meeting of the Expert Team on Migration to Table Driven Code Forms (ET/MTDCF) are presented in the Annex to this paragraph.
22.The migration would lead to an increase in the volume of the data exchanged on the GTS and to an increase in the number of bulletins to be switched by the GTS centres. The ICM-MTN-2001 underlined the need for a detailed assessment of the impact. The impact should be reduced by limiting the double transmission of data presented in Traditional Alphanumerical Codes (TAC) and in Table Driven Code Forms (TDCF), and by avoiding as much as possible the conversion from one code to another. In this regard, the ICM-MTN-2001 stressed the serious issues associated to conversion between TAC and TDCF in terms of meteorological and data processing required and possible significant delays incurred in distributing observational data. It was of the opinion that conversion of observational data reported in TAC into TDCF was neither required nor advisable for data dissemination, and should be performed by GDPS centres for their own requirements. With respect to TDCF into TAC, the ICM-MTN-2001 noted that there are still NMCs that are not currently able to receive and process binary data and that they should be served by their associated RTH. The ICM-MTN-2001 expressed the opinion that cooperation assistance should focus on assisting the upgrade of the connection of such NMCs and their data-handling systems (e.g. PC-based systems) to enable the reception and use of BUFR data at the centres, when migrating to TDCF.
23.The ICM-MTN-2001 noted that the data type designators T1 = I and T1 = K are allocated to observational data presented in TDCF (BUFR and CREX). The allocation of geographical area designators is more limited in Table C3 used for TDCF than in Table C1/C2 used for TAC. However, the use of CCCC should be sufficient to differentiate the bulletins issued by the different countries. The ICM-MTN-2001 was of the opinion that a revision of Tables B3, C3, C6 and C7 was needed to facilitate the migration. The ICM-MTN-2001 recommended that the ICT-ISS review the Tables on the basis of consolidated requirements.
Use of the group ii for the global exchange of the observational data
24.CBS-XII agreed to review the principles for the exchange of observational data on the MTN to better take into account the requirements of GDPS centres as well as the current capacity of the GTS. It agreed that all observational data, with the exception of radar and satellite data, available from the WMO Member countries should be globally exchanged on the MTN.
25.CBS-XII noted that such a change in the principle for routeing data on the MTN would need a modification of the definition and use of the group ii of the data designator of abbreviated headings in order to allocate the series 01-39 for global distribution. CBS-XII requested the OPAG-ISS to review procedures with a view to facilitating the implementation of the new principles for routeing data on the MTN. In the meantime, the centres located on the MTN should strive to exchange the newly defined global data set.
26.The former CBS/ ICM-IEM confirmed that the reports prepared at the main synoptic hours at the stations included in the RBSNs shall be compiled within bulletins with ii in the series 01 to 19. In many instances, the same reports are compiled within bulletins with different abbreviated headings in the series 01-39. With a view to avoiding several retransmissions of the same reports on the same MTN circuits, the ICT-IEM recommended that GTS centres, which recompile reports into bulletins for specific purposes, shall use ii above 39. CBS-XII The meeting confirmed these recommendations of the ICT-IEM.
27.With respect to the related procedures for the exchange in bulletins of “essential data”, as defined in Resolution 40 (Cg-XII), the meeting noted that CBS-XII agreed that “essential data” shall be compiled into bulletins with ii in the series 01 to 19 and that other types of data, including "additional data", should be compiled in the ii series above 19, as a goal for the future.