VOS Telecommunications PMOIII 23 24 March 2006 Hamburg

VOS - Telecommunications PMO-III, 23 -24 March 2006, Hamburg, Germany.

PMO-III, 23 -24 March 2006, Hamburg. VOS - Telecommunications § Inmarsat Communications § § § Code 41 procedures Costs Future Trends - decompression AWS Other Comms Systems § § Globalstar Argos Iridum Meteosat

PMO-III, 23 -24 March 2006, Hamburg. Inmarsat Communications § Derived from SOLAS/GMDSS requirements and applicable to all ocean going ships (depending upon the sea areas they operate in) § Ship Observations are transmitted by Inmarsat C (using Special Access Code 41) to National Met Services that host Inmarsat Land Earth Stations (LES) § A list of LES stations accepting Code 41 messages is maintained on the WMO website at http: //www. wmo. ch/web/aom/marprog/Operational-Information/inmarsat-code 41 -stations. htm However there is no clear ownership of this list or responsibility to maintain it up to date § Suitable LES are also detailed in Turbo. Win

PMO-III, 23 -24 March 2006, Hamburg. Inmarsat Communications

PMO-III, 23 -24 March 2006, Hamburg. Inmarsat Communications Land Earth Stations Accepting Code 41 observations Aussaguel Goonhilly Southbury Station 12 Thermopylae Arvi Perth Sentosa Yamaguchi Santa Paula France United Kingdom USA Netherlands Greece India Australia Singapore Japan USA Land Earth Stations Operators France Telecom Stratos (Xantic) Telenor & Singapore Telecom, VSNL, OTE SA, KDDI

PMO-III, 23 -24 March 2006, Hamburg. Inmarsat Communications § Inmarsat communications has proven to be very reliable with approximately 95% of ship observations being received within HH+120 ( based on UK/Goonhilly figures) § However there have been problems with some LES e. g. § Arvi LES § (Regional restrictions will accept code 41 reports from within Metarea VIII (N) only) § Eik LES

PMO-III, 23 -24 March 2006, Hamburg. Inmarsat Communication Costs § Costs for all ship observations sent using Special Access Code 41 are borne by the small number of National Met Services that host Inmarsat Land Earth Stations (LES) - The shipowner does not pay. § These NMS bear the costs irrespective of the country that recruited the ship to do weather observations § The problem appears to be of greater concern to European NMS § As the majority of VOS observations (SHIP and TEMP Code) are still sent via Inmarsat this imposes an unreasonable cost burden on those countries that host LES which accept Code 41 observations

PMO-III, 23 -24 March 2006, Hamburg. Inmarsat Communications The cost burden has been amplified in recent years by. . . … § Closure of some LES and mergers between Inmarsat Suppliers § Relocation of shipowners with large fleets to other countries, and consequent re-routing of observations. § The increase in the number of ASAP TEMP messages being sent via LES § The increasing use of shipborne AWS sending hourly observations

PMO-III, 23 -24 March 2006, Hamburg. Inmarsat Communication Costs § Various proposals were made to resolve the cost burden problem at an international level, including. . § developing a Global Cost Sharing Scheme with costs paid through a central (WMO) Fund § Appoint an Accounting Authority to oversee the payment of Code 41 transmission costs § However these proposals were not successful, although the WMO continues to keep the matter under review

PMO-III, 23 -24 March 2006, Hamburg. Inmarsat Communication Costs It was therefore left for solutions to the VOS cost burden problem to be developed on a regional basis. Within Europe this has been addressed through the E-SURFMAR programme § Since 2005 the E-SURFMAR Programme has compensated its member countries for their VOS data transmission costs. Member countries contribute to this programme based upon their GNI and are then compensated according to the number of SHIP code reports received via their GTS originating centres. The compensation is therefore mainly directed to those members that host LES i. e. France, Netherlands, Greece and UK § Compensation for each report is expected to be in the order of 0. 187 Euros in 2006 (compared with the current cost of approx 1 Euro/observation) § However it must be remembered that a significant percentage of the costs borne by European LES continues to be generated by non-E-SURFMAR Ships and that not all European VOS fleets use code 41 for their Inmarsat transmissions

PMO-III, 23 -24 March 2006, Hamburg. Inmarsat Communication Costs Bilateral Agreements § Whilst the E-SURFMAR agreement goes some way to resolving the problem within Europe, it was recognised that bilateral agreements would also be needed to lessen the burden in some cases. For example agreements have been reached between DWD in Germany the UK Met Office to reimburse the cost of German ship observations sent via Goonhilly. A similar bilateral agreement was established between KNMI and DWD for the cost of observations sent via Burum/Station 12 following the closure of Raisting LES E-ASAP § Arrangements were agreed under the E-ASAP Programme in 2005 for the UK Met Office to be reimbursed for the cost of upper air TEMP messages that are sent through Goonhilly LES by ASAP ships that participate in programme

PMO-III, 23 -24 March 2006, Hamburg. Inmarsat Communication Costs How else can Inmarsat C VOS costs be reduced ? Data compression techniques; BATOS AWS coded observations Binary compression was successfully implemented by E-SURFMAR from a BATOS AWS in March 2006 using Inmarsat data reporting service. The cost for one report is now ~0. 145 € …. 7 times less than the cost of a standard VOS observation sent via Code 41 savings will allow greater volume of data to be sent with same budget Conditional upon being visible to only one Inmarsat satellite Turbo. Win coded observations It may be possible to ‘half’ compress Turbo. Win messages ( as 2 x 256 bit blocks will be needed) Could reduce costs by more than half but has implications for the Code 41 system

PMO-III, 23 -24 March 2006, Hamburg. Inmarsat – other factors § § § E-MAIL transmission – as ship-based broadband communications gets cheaper ship-owners may be increasingly willing to submit observations by email and absorb the costs themselves. Observations can be compressed SMART transmission – AWS software adapted so that observations are only sent when needed i. e. reduce transmission intervals when weather conditions are sensed to be within certain parameters BUFR code transmission – Migration from SHIP to BUFR to take place 2007 -2012. Risks increasing costs and introduces problems with GTS exchange during transition

PMO-III, 23 -24 March 2006, Hamburg. Telecommunications – Other Systems ARGOS § § § Used by MINOS AWS systems and drifting buoys Polar orbiting satellites (6 satellites at present) Doppler positions One way system Data timeliness can be a problem The constallation Low Cost ~ 0. 4 Euros/observation (based on 19 reports a day) GLOBALSTAR § § Being tested for E-ASAP ships to reduce current Inmarsat costs (~12 Euros/TEMP message) Used on Norwegian weather ship Mike Uses 50 satellites worldwide Less than 10% of Inmarsat costs

PMO-III, 23 -24 March 2006, Hamburg. Telecommunications – Other Systems IRIDIUM § § § Under study for use on Moored buoys to replace analogue Meteosat DCP Short Burst Data cost reduced - possibilities for AWS under study Two way – email communication METEOSAT § Costs absorbed by Eumetsat membership § Used by German AWS VOS & E-SURFMAR Moored buoys

PMO-III, 23 -24 March 2006, Hamburg. Globalstar Satellite coverage Cortesy - http: //savi. sf. net/

PMO-III, 23 -24 March 2006, Hamburg. Iridium Satellite Coverage Cortesy - http: //savi. sf. net/

PMO-III, 23 -24 March 2006, Hamburg.

PMO-III, 23 -24 March 2006, Hamburg. v Constraints for the future II. 1. Communication costs must be saved. II. 2. Migration to TDCF (BUFR) is required. v Observations v v v III. 1. AWS are strongly desirable for a better quality of basic measurements and to minimize observer's work. v III. 2. Their deployment should significantly increase in the coming years. v III. 3. They can easily report hourly data but they are restrained by communication costs. Fortunately, the cheapest communication system can be chosen since the transmitter is a part of the AWS.

PMO-III, 23 -24 March 2006, Hamburg. Inmarsat Communications ATLANTIC OCEAN REGION-EAST (AOR-E) Name of station. Country. ID number. Aussaguel. France 121 Goonhilly. United Kingdom 102 Raisting. Germany 115 Southbury. USA 101 Station 12 Netherlands 112 Thermopylae. Greece 120 ATLANTIC OCEAN REGION-WEST (AOR-W)Name of station. Country. ID number. Goonhilly. United Kingdom 002 Southbury. USA 001 Station 12 Netherlands 012 I NDIAN OCEAN REGION (IOR)Name of station. Country. ID number. Arvi. India. Aussaguel. France 321 Perth. Australia 322 Raisting. Germany 333 Sentosa. Singapore 328 Station 12 Netherlands 312 Thermopylae. Greece 305 Yamaguchi. Japan 203 PACIFIC OCEAN REGION (POR)Name of station. Country. ID number. Perth. Australia 222 Santa Paula. USA 201 Sentosa. Singapore 210 Yamaguchi. Japan 203 Note: Arvi will accept code 41 reports from within Metarea VIII (N) only.

PMO-III, 23 -24 March 2006, Hamburg. v IV. 1. The transmission of BUFR reports from ships to VOS NMSes has two drawbacks at least: • • v v v the data volume remains high compared to FM-13 (no savings expected); the transmission of binary data through Inmarsat-C GMDSS beacons seems compromised[1]. IV. 2. NMSes receiving FM-13 or proprietary format reports, may code the data in FM-96 BUFR before GTS data transmission. I recommend not to request the transmission of BUFR messages by the ships. [1] This must be verified. It seems that, although binary data can be transmitted through Inmarsat-C transmitters, this can be done through another mode, called “data reporting” (see below). The possibility to switch the transmitters from a mode to another is not obvious and could be not allowed on GMDSS beacons.

PMO-III, 23 -24 March 2006, Hamburg. v v V. 2 Processing before GTS data transmission Compressed data (“proprietary” format) are received at Météo. France through emails without any delays. They are uncompressed and then coded in FM-13. Individual ships have their metadata centralized. They can be used for BUFR coding. Outputs can be issued for Pub No 47. Metadata can be entered through a Web interface but plans exist to enable the BATOS systems to send compressed metadata reports through the same way they send observation data (during PMOs visit for instance). These metadata reports will be processed in order to replace old metadata. Possibility exists to correct systematic biases as well as to switch on/off any parameters before GTS transmission.

PMO-III, 23 -24 March 2006, Hamburg. v v v Proposition V. 1. Open compressed data format and software The compressed data format as well as the processing software may be given to any NMS which would be interested in. They can be used for AWS using other commercial satellite links than Inmarsat. The principle being flexible, variants of the format could be easily developed.

PMO-III, 23 -24 March 2006, Hamburg. Background At SOT 2 the Task Team presented 3 scenarios; 1. Continue existing system with costs borne by just a few NMS 2. Share transmission costs among VOS operators in proportion to their fleet sizes 3. Develop a new Global Cost Sharing Scheme with costs paid through a central (WMO) Fund 3 a. Totally independent of Inmarsat Provider 3 b. Single Provider Solution

PMO-III, 23 -24 March 2006, Hamburg. Background -ctd After SOT 2 the Task Team proposed that Scenario 3 a should be supported as • • It offered a fair method of sharing the cost burden It was independent of the service provider It would encourage future transmission methods (e. g. email) It avoided any changes to the established Code 41 principles This proposal was then put to the JCOMM Management Committee in March 2004 and was subsequently raised at the Executive Council by the JCOMM Co-President

PMO-III, 23 -24 March 2006, Hamburg. Background - ctd The Executive Council…. ‘………requested the SOT to proceed with the preparation of a proposal on the issue for consideration by the Executive Council. The Council recognized that the problem was not necessarily a global one, but might best be addressed on a regional basis, and that in any case more detailed information was required before any decisions could be considered. It therefore requested that such detailed information, together with possible options for solutions, should be made available for consideration by EC-LVII in 2005’

PMO-III, 23 -24 March 2006, Hamburg.

PMO-III, 23 -24 March 2006, Hamburg. The latest proposal of the Task Team v Appoint an Accounting Authority to oversee the payment of Code 41 transmission costs v Accounting Authority would act as the billing intermediary between the LES Service providers and the NMSs that operate VOS and who host LES Stations v Accounting Authority role would ideally be performed by a Responsible National Met Service

PMO-III, 23 -24 March 2006, Hamburg.

PMO-III, 23 -24 March 2006, Hamburg. Advantages Disadvantages v Costs fairly distributed v v Code 41 system maintained (no disruption to observers) Limited scope for encouraging competition between LES v Large Administrative effort for Accounting Authority (although electronic bills linked to databases of Inmarsat numbers could reduce this, in time) v Possibility that some VOS operators may reduce or withdraw their VOS in order to avoid costs v Start up and ongoing running costs will be incurred v Possibility of bad debtors v Scheme could be enhanced to include new data streams (email) v Scheme could regional solutions v encompass Administrative effort concentrated at one location (avoids duplication of effort between NMSs)

PMO-III, 23 -24 March 2006, Hamburg. Some Questions that need to be addressed v v v v Who should ‘own’&’control’ the list of Code 41 LES ? Should geographic limits imposed by Code 41 LES be resisted ? How should the Accounting Authority be reimbursed ? How much would the start up costs be ? How much would the administration of the system cost and how much effort is likely to be involved ? How would bad debtors, or unknown Inmarsat numbers, be dealt with ? How should costs be collected from VOS operators – will contractual agreements need to be established with the Accounting Authority ? How can VOS Inmarsat station numbers be related back to a VOS operating country (should they be included in Pub 47) ? Which/How many countries should be involved in the scheme (all VOS operators) ? How would the scheme deal with regional groupings like E-SURFMAR ? Which Inmarsat Suppliers should the Accounting Authority act for ? Should new transmission methods be addressed (e. g. email) ? Should the cost of TEMP messages be included ? Should costs be decided on the basis of individual ship observations - or based on the volume of SHIP/TEMP code messages sent through the GTS Collecting Centres ?

PMO-III, 23 -24 March 2006, Hamburg. Questions for SOT 3 1. Do you agree with the Accounting Authority Approach ? 2. Do you agree that an NMS should act as the Accounting Authority ? 3. How should the Task Teams work be progressed from here ?

PMO-III, 23 -24 March 2006, Hamburg. § The cost burden is likely to increase because of … § The migration to BUFR coded messages § The Inmarsat suppliers that can accept code 41 messages from any part of the globe (even if they are not in sight of the primary Inmarsat satellite)