WORLD CLIMATE RESEARCH PROGRAMME
 TROPICAL OCEAN AND GLOBAL ATMOSPHERE |
 WORLD OCEAN CIRCULATION EXPERIMENT |
TROPICAL OCEAN AND GLOBAL ATMOSPHERE |
WORLD OCEAN CIRCULATION EXPERIMENT |
TOGA/WOCE XBT/XCTD PROGRAMME
PLANNING COMMITTEE
Report of the Third Meeting
(TWXXPPC-3)
and
WOCE UPPER OCEAN THERMAL DATA ASSEMBLY
CENTRES COORDINATION GROUP
Report of the Fifth Meeting
(UOT/DAC-5)
RSMAS, University of Miami,
Miami, Fl, USA
25-29 April 1994
WOCE Report No. 121/94
August 1994, WOCE International Project Office, Wormley
WOCE is a component of the World Climate Research Programme (WCRP), which was established by WMO and ICSU, and is carried out in association with IOC and SCOR. The scientific planning and development of WOCE is under the guidance of the JSC Scientific Steering Group for WOCE, assisted by the WOCE International Project Office. JSC is the main body of WMO-ICSU-IOC formulating overall WCRP scientific concepts.
BIBLIOGRAPHIC CITATION
WOCE INTERNATIONAL PROJECT OFFICE 1994 Report of the third meeting of the TOGA/WOCE XBT/XCTD Programme Planning Committee (TWXXPPC-3) and Report of the fifth meeting of the WOCE Upper Ocean Thermal Data Assembly Centres Coordination Group, UOT/DAC-5,RSMAS, University of Miami, Miami, Fl, USA, 25-29 April 1994
WOCE International Project Office, WOCE Report 121/94, 36pp.
CONTENTS
PART I - TWXXPPC-3
2 Review of Action Items from TWXXPPC-2
6 Analyzing the Present and Future Network
6.1 Atlantic Ocean
6.2 Indian Ocean
6.3 Pacific Ocean
6.4 TRANSPAC
7 Prioritizing the Network
7.1 WOCE
7.1.1 Previous and Existing Priorities
7.1.2 Future Directions
7.2 TOGA
8 Using TAO Data to Analyze Sampling Strategy
9 Data Management
9.1 TAO Data
10 The Workshop on the Use of Sub-Surface Thermal Data for Climate Studies
11 Post WOCE and TOGA
11.1 GOOS and GCOS
11.2 Future of the XBT Pool
12 The Future of the Committee
13 Next Meeting
Appendix A: Agenda
Appendix B: Participants
Appendix C: 1993 Observations
Appendix D: Using TAO Time Series to Estimate Errors due to Sampling Strategies
PART II - UOT/DAC-5
Action Items
Report of Meeting
PART I - TWXXPPC-3
1. GSDC prepare an annual summary for 1985-1995 of XBT distribution in the "dot" format where the size of dot represents the number of months for which profiles were archived for each 2 degree square.
2. Operators continue to provide the IGOSS Operations Co-ordinator with six-monthly reports on their activities 3 months after the end of each six month period.
3. Japan (JMA/STA) to consider using T7s on PX5 as it has been suggested that the deeper probes are required in the Western Pacific.
4. All XBT lines should be sampled along the entire section (coast to coast), even if that means the whole line is less intensively sampled than some portions are at present.
5. France to alter the sampling pattern of AX5, AX20 and AX11, to extend them beyond 25 degrees N where the present sampling ceases. This could be accommodated by having fewer sections per year (perhaps asking the ships to sample only on one direction of the round route).
6. A. Lusquinos and M. Szabados to contact G. Siedler (BSH, Germany) to discuss options for possible vessels for AX17 and AX18.
7. France to liaise with South Africa, UK and USA to help reduce logistical problems they currently face with AX15.
8. Chairman to inform Core Project 2 of the possibility of a joint Drake Passage project with Argentina using their Antarctic supply vessel which makes two crossings per year four months apart. They presently deploy one XBT per day on the crossings and would be willing to deploy more as a joint project with another country if the second country could provide probes.
9. France and Japan maintain their monthly programme along IX6 but USA (NOS) redirect resources planned for IX6 elsewhere. Should the French programme cease in 1995, then NOS should begin sampling along this line. The Committee envisaged this would be possible because the relevant vessels will continue to sample in the Pacific all the time.
10. IX10: WOCE Core Project 1 to note that although there is no high density planned at the moment, the present level of high frequently repeated low density sampling may come close to simulating high density. Thus the low density should continue as it is now. However Japan has proposed, and the USA may propose to do some high density sampling, and if such a project is funded, then USA, France and Japan should co-ordinate their efforts to avoid redundancy of sampling.
11. G. Meyers to write to chairman of French National WOCE Committee informing him of the recommendation that French sampling continues at least to the end of the WOCE field programme (1997).
12. France to use probes planned for PX2 and PX43 in 1994 elsewhere as these lines are already fully sampled by the USA.
13. USA and France liaise with each other to ensure PX9/13/31 continue to be fully sampled to at least at the low density rate.
14. ORSTOM Noumea to recognize the lower priority of PX51, PX52 and PX53 in the WOCE network when preparing plans for future sampling. (These are not TOGA lines).
15. France to consider using resources identified from other lines to sample PX4 in particular beyond 25 degrees N.
16. WOCE Core Project 1 10 note that PX5 and PX12 have no high density planned at the moment, but the present frequent sampling at low density rate may come close to simulating high density. Thus the low density should continue as it is now.
17. WOCE Core Project 1 also to note that high density lines require the identification of a PI who has time and money for each section. This can be as big a stumbling block as finding probes and vessels.
18. Future programmes such as CLIVAR, GOOS or climate prediction projects to maintain long time series including sampling in the TRANSPAC region.
19. TWXXPPC Chairman and IPO to help organize the creation of a multinational TRANSPAC Pool.
20. WOCE IPO to inform WOCE SSG and Core Project 1 of the Committee's findings on prioritizing the network.
21. The Committee encourages study (by the PMEL group and others) of problems in the use of the complementary thermal data sets (XBTs, TAO) in interpolating between observations to construct research-quality maps of fields.
22. WOCE IPO and the ITPO to contact originators of real-time data not submitted in delayed mode yet, in order to encourage the submission of the delayed mode data.
23. The GSDC in Brest to investigate the possibilities for TOGA hydrographic data being merged with the XBT and other data as soon as possible.
24. TAO data not to be mixed with the XBT, MBT data in the GSDC and NODC files, but that a separate TAO file should be maintained.
25. At convenient intervals (perhaps monthly) the Brest Centre should get the updated, complete, clean TAO data set from PMEL.
26. An overview article of all the work being done with WOCE and TOGA data to be published and hence widely distributed to the scientific community. G. Meyers to canvass Workshop authors for their plans regarding submitting papers, and their opinions on producing a collection of published papers.
27. Chairman to send summary of the findings of the workshop to the planning committees of GCOS and GOOS.
28. A multinational follow-on TOGA-type XBT Pool to be created for post-TOGA and post-WOCE resources. The global Pool will need to have 42,100 probes per year to maintain the WOCE and TOGA low density networks. Chairman to write a document outlining the benefits of the TOGA Pool and a future Pool, including consequences of not having a Pool, for the GCOS, GOOS, CLIVAR etc. planning committees.
29. The WOCE SSG/JSC to consider maintaining the TOGA/WOCE Committee until such a time as CLIVAR plans are more developed.
30. Future committees should be organized around fields to be observed rather than the instruments used to make the measurements.
31. G. Meyers will step down as chairman of TWXXPPC in 1994. WOCE and TOGA SSG to consider the proposals for the new Chairman (Rebert) and Vice-Chairman (Dobson).
The third meeting of the TOGA/WOCE Programme Planning Committee (TWXXPPC) was opened by the chairman, G. Meyers. The meeting was hosted by R. Molinari at the Rosenstiel School of Marine and Atmospheric Sciences, University of Miami, Florida, USA. Apologies for absence were received from T. Ando (JMA), A. Sy (BSH), V. Detemmerman (ITPO) and D. Roemmich (SlO). Apart from Ando, Roemmich and Sy, all committee members were present and several invited experts also attended. The full participants list is in Appendix B.
2. REVIEW OF ACTION ITEMS TWXXPPC-2
Following are the latest status of the action items resulting from the last TWXXPPC meeting in October 1992. The items are numbered as per the report of the meeting (WOCE Report No. 99/93).
1 & 2. T. Wright explained that GTS code changes and other technical modifications were in progress and the new codes will be ready to be put into practice in November 1995 (the original target date was January 1995). A completely new code with the same header format but with additional information tables will be implemented. The paper by the members of the IGOSS Task Team for Quality Control of Automated Systems (TTQCAS) which describes the new fall-rate equations, has not yet been published. M. Szabados reported that the paper has been submitted to Deep-Sea Research and reviews are expected back around mid-May.
3, 4 & 5. J-P. Rebert reported that the changes in the data centres were being examined under the framework of the GTSPP. The technical question of including information on the fall-rate equation used, the probe type and other information such as flags and codes arising from scientific quality control were being considered together. It is technically feasible to make these modifications but the data centres are awaiting a final decision from scientists about the way to include the information (via the UOT/DAC Co-ordination Group; see Section 13 for the report of that meeting).
6, 7 & 8. T. Wright reported that he had not yet received information from Sparton Inc. concerning users of Sparton probes. This presents an ongoing problem as data from Sparton probes continue to get into the data centres without being identified as such. J-P. Rebert described the ORSTOM intention to compare data from 5 boxes of Sparton probes deployed on the WOCE cruise CITHER-1 with the CTD data once they are calibrated.
9 & 10. This information was brought to the attention of the IOC member states at the fourteenth session of the IODE committee (December 1992).
11. A letter expressing the scientific importance of the continuation of the TRANSPAC sampling after 1993 was sent to the Commanding Officer of the FNOC from the Director of the WOCE IPO. No response was received.
12. The issue of high density sections was discussed again at this meeting. See section 5.
13. The Chairman of TWXXPPC consulted G. Siedler of IFM Kiel regarding high density sampling along AX17 or 18 in the South Atlantic. Siedler informed the Committee that some previous problems with the vessels have been resolved, but further help with logistics is still required. The intention is to carry out high density sampling on either AX17 or AX18.
14. WOCE Core Project 1 designated AX3 as a high density line but stressed that the study of heat flux in the North Atlantic still required high density sampling along AX7. AOML had planned to start high density sampling on AX7 in May 1994 but the vessel sank. Now plans are to start in June 1994.
15. Articles describing the availability of climatologies and other products from the Brest Centre have appeared in both TOGA Notes and the International WOCE newsletter.
16. NODC will shortly begin to produce line by line reports of real-time data.
17. GSDC in Brest has not yet produced any dot maps as described. The Committee requests that GSDC prepare an annual summary for 1985-1995 of XBT distribution in this format. (Action Item 1)
18. A complete list of number of probes required per TWI line can be seen in Appendix G of WOCE Report No. 99/93.
19. IX24 is no longer a route.
20. Not yet implemented.
21. Chairman did forward Committee comments on GCOS/OOSDP XBT paper.
22. Workshop was held in September 1993 (see Section 10).
P. Holliday presented a summary of the 1993 observations according to reports received from the operators directly or via the IGOSS Operations Co-ordinator. The report divided the lines into ones not sampled at all, those sampled but at less that 50% of the requirements, those sampled at between 50 and 90% of the requirements, and those sampled to at least 90% of the requirement. The requirements were taken as the values shown in Appendix G of WOCE Report No. 99/93 (number of observations per year). Low density lines were considered separately from TOGA high frequency and WOCE high density lines. The full report is shown in Appendix C and Figures 1-3 show the summary in graphical forms.
Levels of coverage have remained similar over the last three years, but with noticeable changes in the Pacific in 1993. Sampling in TRANSPAC is severely reduced due to the ending of the FNOC COOP programme; the area is now sampled at less than 50% of the requirement. Other FNOC North Pacific lines such as PX20, PX37, PX38 and PX39 have also been terminated and PX7/13 will cease late in 1994 when stocks of probes run out. The coverage in the Southeast Pacific is improving; sampling on PX14 has begun though is low in this summary because the programme started late in the year. NOS has however ceased sampling on PX25 because of problems with the vessel.
The low density coverage in the Indian Ocean has improved and will remain at a good level for the major WOCE survey in 1995. See Section 5 for more detailed information about the Indian Ocean.
Sampling in the Atlantic remains at a low level except for a few lines. Many tropical lines are not sampled coast-to-coast (sampling ceases at 255N) and several occupied lines do not have the required 4 obs. per day.
The Southern Ocean remains sparsely sampled, but XBTs are beginning to be deployed. The French and Australian Antarctic supply vessels have begun the Southern Ocean sections south of Hobart and in the Southern Indian Ocean, and XBTs have been deployed from several UK research cruises south and south-east of Africa.
The Committee recommended operators continue to provide the IGOSS Operations Co-ordinator with six-monthly reports on their activities, and that they should endeavour to provide them in the time frame agreed (three months after the end of the six month period). (Action Item 2). The Committee thanked P. Holliday for assembling information on the status of the network and summarizing it in a useful format.
Representatives from individual countries summarized their intended sampling for 1994 (Table 1). The USA plans are likely to be the same for 1995 except where indicated in the table; the commitment has been made to continue the TOGA sampling after 1994 and WOCE lines will be occupied to the end of WOCE. France is not yet in a position to say whether sampling will continue after TOGA finishes. Germany has indicated the WOCE resources currently used for XBTs in the Atlantic will end after 1995. Sampling by JMA and STA in the Western Pacific and Indian Ocean may continue beyond 1995. Australia will continue the present level of sampling until the end of 1995.
Figure 1. 1993 Low density coverage
Figure 2. 1993 WOCE high density coverage
Figure 3. 1993 TOGA frequently repeated coverage
Table 1. Sampling Plan for 1994
| Line | Country | Low D Sects/year |
High D Sects/year |
Notes |
| AX1 | UK | less than 2 | ||
| AX2 | Canada / UK / USA | 2 | ? | HD start in Jan 1994 |
| AX3 | Germany | 12 | 12 | |
| AX4 | USA | 12 | ||
| AX5 | France | 28 | 2 obs/day, sampling stops 25 deg N | |
| AX6 | 0 | |||
| AX7 | USA | 12 | 4 | HD start in June 1994 |
| AX8 | USA | 12 | Wide band of sampling | |
| AX9 | 0 | |||
| AX10 | USA | 10 | ||
| AX11 | France | 18 | ||
| Germany | 18 | |||
| AX12 | Germany | 1-2 | ||
| UK / USA | 12 | |||
| AX13 | 0 | |||
| AX14 | USA / Nigeria | 12 | Start mid-1994 | |
| AX15 | France | 8 | Some logistical problems; France seeking assistance with support. | |
| AX16 | USA / Nigeria | 12 | Start mid-1994 | |
| AX17 | Germany | ? | ? | Have resources but seeking vessel and local support. HD planned for either AX17 or 18. |
| AX18 | Germany | ? | ? | As for AX17 |
| AX20 | France | 10-12 | With thermosalinograph | |
| AX21 | 0 | |||
| AX22 | 0 | |||
| AX25 | Germany | ? | Intermittent | |
| AX26 | France | 12 | Logistical problems. | |
| AX27 | 0 | |||
| AX29 | USA | 12 | ||
| Line | Country | Low D Sects/year |
High D Sects/year |
Notes |
| IX1 | Australia | 18 | 4 | HD: 1 in 1994, 2/yr. after. Research vessels will make 3 additional HD sections in 1994. |
| IX2 | 0 | 0 | ||
| IX3 | France | 12 | 0 | |
| IX6 | France / Japan | 8 | 0 | |
| USA | 12 | 0 | ||
| IX7 | USA | 12 | Start mid 1994 | |
| IX8 | India | Several | 4 ? | Propose 4 HD per year |
| IX9 | Australia | 6 | Mostly north part only. | |
| Japan | 12 | North part only | ||
| IX10 | France | 10 | 0 | |
| USA | 12 | 0 | ||
| Japan | 12 | 4 ? | East part only. Propose 4 HD per year. | |
| IX11 | 0 | |||
| IX12 | Australia | 16 | 0 | |
| IX14 | India | ? | ||
| IX15 | USA | 12 | 1 | HD 1 in 94, 2 in 95, and 4/yr. after. |
| IX18 | 0 | |||
| IX19 | France | ? | Intermittent, depending on supply vessels. | |
| IX21 | USA | 12 | 1 | HD 1 in 94, 2 in 95, and 4/yr. after. |
| IX22 | Australia | 6 | ||
| IX23 | 0 | |||
| IX25 | USA / Mauritius | 12 | Start late 1994 | |
| IX26 | 0 | |||
| IX28 | Australia | 3 | 1 | At least 3/yr. LD proposed, plus 1 or more HD. |
| Line | Country | Low D Sects/year |
High D Sects/year |
Notes |
| PX1 | 0 | |||
| PX2 | Australia | 14 | ||
| France | ? | Planned to start in 1994 | ||
| PX3 | Australia | 6 | ||
| PX4 | France | 6 | ||
| PX5 | France | 20 | 0 | |
| Japan | 18 | 0 | ||
| PX6 | USA | 0 | 4 | |
| PX8 | USA | 12 | ||
| PX9 | USA | ? | 4 | FNOC LD will cease in 1994. |
| PX10 | USA | 12 | 2 | |
| PX11 | Australia | 6 | ||
| PX12 | France | 12 | 0 | |
| USA | 18 | 0 | ||
| PX13 | USA | 12 | FNOC LD will cease in 1994. | |
| PX14 | USA | 18 | Logistical problems. | |
| PX15 | USA | 12 | ||
| PX16 | USA | 12 | Not occupied as of April 94 | |
| PX17 | France | 17 | ||
| PX18 | USA | 18 | ||
| PX20 | 0 | FNOC sampling ceased 1993 | ||
| PX21 | 0 | |||
| PX22 | 0 | |||
| PX23 | 0 | |||
| PX24 | 0 | Occasional Japanese research vessel. | ||
| PX25 | USA | 0 | Interest but no vessel. | |
| PX26 | USA | ? | 2000 obs/yr. due to end in 1994 | |
| Japan | 18 | May end in 1995 | ||
| PX29 | USA | ? | ||
| PX30/31 | Australia / USA | 0 | 4 | Brisbane to Fiji. |
| PX31 | 0 | France sampling ceased 1993 | ||
| PX33 | 0 | |||
| PX34 | Australia | 12 | 4 | |
| PX36 | 0 | |||
| PX37 | USA / ROC | 0 | 2 | FNOC LD sampling ceased 1993 |
| PX38 | USA / ROC | 0 | 2 | FNOC LD sampling ceased 1993 |
| PX39 | 0 | FNOC LD sampling ceased 1993 | ||
| PX43 | USA | 12 | ||
| PX44 | USA / ROC | 0 | 2 | |
| PX49 | Japan | 12 | ||
| PX50 | USA | 12 | 2 | |
| PX51 | France | 12-18 | ||
| PX52 | France | 12-18 | ||
| PX53 | France | 12-18 |
G. Meyers reviewed the commitment of resources to the Indian Ocean lines in view of the main thrust of WOCE observations in 1995. Table 1 shows the commitment to the low density sampling is now in good shape, with many lines spinning up in 1994 to be sampled at the full rate in 1995 and beyond. In response to a request from Core Project 1 for high density sampling on several lines to substitute for repeat hydrography, G. Meyers surveyed possible PIs for their plans. The response was good and several lines have high density sections proposed or funded. The USA are carrying out 2 repeats of the WOCE Repeat Hydrography line IR3 which runs from the Persian Gulf to Madagascar, and might partially replace high density sampling on IX3. The plans for high density sampling in the Indian Ocean are summarized in Table 2.
Table 2. High density sampling in the Indian Ocean (not all proposals are funded yet)
| Line | Country | PI | 94 | 95 | 96 | Later | Notes |
| IX1 | Australia | Meyers | 1 | 2 | 2/yr. | In 1995, 3 hydrographic sections also. | |
| IX2 | No Commitment | ||||||
| IX3 | France | Fieux, Donguy | Interest | ||||
| IX6 | Japan | Mizuno | 4 | 4 | Proposed | ||
| IX8 | India | Gopalakrishna | 4 | 4 | 4/yr? | Interest | |
| IX10 | Japan | Mizuno | 4 | 4 | 4 | Interest: eastern half only | |
| USA | Interest | ||||||
| IX12 | No Commitment | ||||||
| IX15/21 | USA | Talley | 1 | 2 | 4 | 4/yr.? | |
| IX28 | Australia | Rintoul | 3 | 3 | 3 | 3 | Using supply vessel. |
| France | Donguy | Plans for sections in early, middle and late summer. | |||||
| IX??? | India | Gopalakrishna | 4 | 4 | 4/yr.? | Funding is available. Looking for direct line from India to Australia. |
6. ANALYZING THE PRESENT AND FUTURE NETWORK
Having looked at what had been achieved in 1993 and previous years, and also at what was planned now for the next year or two, the Committee then attempted to identify those areas which appear to have excess sampling and those key areas which are undersampled. The aim of the exercise was to see if any existing resources could be redirected from one area to another. Each ocean basin was considered in turn and several recommendations resulted.
Recognizing the value of coast-to-coast sampling, the Committee recommended that all lines in the tropical Atlantic should be sampled along the entire section, even if that means the whole line is less intensively sampled in some portions than at present. In particular the Committee noted that French occupied lines of AX5, AX20 and AX11 should be extended beyond 25 degrees N where the present sampling ceases, which could be achieved by having fewer sections per year (perhaps asking the ships to sample only on one direction of the round route). (Action Items 4 and 5)
Several operators expressed difficulties with logistics in the South Atlantic where the sampling is sparse. The intention to sample is there, and in many cases the resources are identified but the operators have difficulty servicing the routes. These difficulties need to be overcome as soon as possible and the meeting provided a forum for starting discussions on possible co-operation. Germany has the resources to sample AX17 and AX18 but has no vessel identified. A. Lusquinos and M. Szabados reported they both may know of suitable vessels. (Action Item 6)
France is currently operating AX15 but continue to have logistical difficulties. It is possible that South Africa may be able to help support the ships at the southern end of the line, and UK may be able to help with the northern end. The USA have had considerable help from South Africa with ships of opportunity and M. Szabados offered to help France make contact with South Africa. (Action Item 7)
Argentina has an Antarctic supply vessel which makes two Drake Passage crossings per year, four months apart. They presently deploy one XBT per day on the crossings and would be willing to deploy more as a joint project with another country if the second country could provide probes. The Committee recommended the Chairman should inform Core Project 2 of this possibility. (Action Item 8)
As sampling in the Indian Ocean spins up in preparation for the major WOCE effort in 1995, it appears that more than one operator may have plans to begin sampling on the same line, while others remain out of the plans altogether. In particular, the Committee recommended that France and Japan maintain their monthly programme along IX6 but that USA (NOS) redirect resources elsewhere. Should the French programme cease in 1995, then NOS should begin sampling along this line. The Committee envisaged this would be possible because the relevant ships of opportunity will continue to sample in the Pacific all the time. (Action item 9)
IX10 has France, Japan and USA all doing low density sampling throughout the year. The Committee would like to bring to the attention of WOCE Core Project 1 that although there is no high density committed at the moment, this frequently repeated low density sampling may come close to simulating high density. In that case the low density should continue as it is now. However, Japan has proposed and the USA may propose to do some high density sampling, and it such projects are funded, then USA, France and Japan should co-ordinate their efforts to avoid oversampling. (Action Item 10)
France (J-R. Donguy) has begun to implement sampling in the Southern Ocean using French and Australian Antarctic supply vessels, but future funding for this project is by no means secure. Recognizing that the WOCE Core Projects 1 and 2 request sampling in the Southern Ocean, and recognizing that the French infrastructure for sampling presently exists, the Committee recommends that French sampling continues at least to the end of the WOCE field programme (1997). (Action Item 11)
Recent changes in operators' plans (partly as a result of previous meetings which identified areas of high sampling) need to be co-ordinated to ensure an even spread of resources. The Committee tried to identify lines which had a redundancy of sampling so resources could be directed to undersampled lines. This proved to be very difficult especially as the network has already been downgraded considerably by the termination of the FNOC SOOP sampling. However the Committee made the following recommendations.
Australia presently samples PX2 to the full requirement so the Committee recommended that the French probes planned for this line in 1994 should be directed elsewhere. Similarly, the USA already sample PX43 to the full requirement so the French probes could be used to more effect elsewhere. (Action Item 12)
Both France and USA have plans to cease or have already ceased low density sampling on the central Pacific lines PX9/13/31. This band has great importance to WOCE and TOGA and is one of the longest XBT time series. A large number of scientific articles have been published on the basis of historical data here. The Committee strongly recommends that USA and France ensure this line continues to be fully sampled at least at the low density rate. (Action Item 13)
PX51, PX52 and PX53 were added to the WOCE network at the request of ORSTOM Noumea who were sampling these lines. However WOCE Core Project 1 agreed with the caveat that these lines should not take resources away from pre-existing WOCE/TOGA lines. The Committee recommended these lines be given low priority when ORSTOM Noumea prepares plans for future sampling. (Action Item 14)
The recommendation that all lines be sampled coast to coast applies to the Indian and Pacific Oceans as well as the Atlantic. The Committee requested that France consider using resources identified on other lines to sample lines (PX4 in particular) beyond 25 degrees N. (Action Item 15)
Both PX5 and PX12 have been identified as WOCE high density lines but have no commitment for that sampling. However both lines have multiple operators and are repeated frequently at the low density rate (in the same way as IX10). The Committee would like to bring this to the attention of WOCE Core Project 1 when they are considering the state of the high density network. The Committee felt that because of the particular value of frequent sampling along these lines, they could not recommend directing any of the resources elsewhere. (Action Item 16)
SOOP should note that a major problem with implementing high density sampling is identifying a PI to co-ordinate the sampling on each line, to bring some finance for travel etc. and to actually be on the ship for each section. It is not always easy to identify such a PI and this can be as big a stumbling block as getting the probes or vessel. (Action Item 17)
As described in past reports (TWXXPPC-2, WOCE Report No. 99/93; Workshop on the Use of Subsurface Thermal Data, WOCE Report No. 110/93, ITPO Report 9) TRANSPAC is a very important part of WOCE and has provided an extremely valuable data set. The TRANSPAC data set has allowed descriptions of subsurface thermal variability on time scales from annual to interdecadal, of long term current variability, and of heat storage and heat fluxes in the north Pacific (with high density lines). The Workshop report noted that allowing TRANSPAC to degrade will make these descriptions and analyses impossible in the future. Yet TRANSPAC has already been degraded with the termination of the FNOC COOP sampling. Japan also operates ships of opportunity in the TRANSPAC areas and the future of that sampling after 1995 is uncertain. The Committee recommended that immediate action must be taken to restore the level of sampling in the TRANSPAC area to at least the 1991/92 level (around 3000 observations per year). The approximate WOCE requirement in this region is 5000 observations per year and in 1993 it had already dropped to 1875 obs.
The Committee noted that national and international ocean observing system planning documents continually stress the importance of time series and their long term maintenance. TRANSPAC is an existing time series that will provide valuable data for climate research as long as it is continued at the required level of sampling. The Committee recommends that future programmes such as CLIVAR, GOOS or climate prediction projects maintain sampling in the TRANSPAC region. (Action Item 18)
However, to meet the WOCE requirement that TRANSPAC sampling be maintained now and for the remainder of the WOCE observation period (to the end of 1997), the Committee agreed a good approach would be to form a TRANSPAC "Pool" along the same lines as the TOGA Pool. Participating countries and organizations would contribute probes, shipboard equipment and shore support in varying amounts according to the resources available. In this way even small contributions could be used to maximum benefit. The Committee suggested interested parties could include USA (NOS and the Consortium for Oceans Role in Climate), Canada, Japan, Korea, PICES etc. The TWXXPPC Chairman and IPO agreed to try to organize the creation of this multinational TRANSPAC Pool. (Action Item 19)
The WOCE Core Project 1 and the SSG have asked that the Committee attempt to prioritize the sampling network to ensure resources are being directed to the areas where they are most needed. Such an exercise would also provide valuable information for planning of long term observing systems.
F. Dobson outlined the consensus of WOCE priorities from the SSG, the Core Projects 1, 2, and 3, and the Brest Workshop. The major issues so far defined by WOCE are as follows:
7.1.1 Previous and Existing Priorities
As outlined in Section 6.4 TRANSPAC is too valuable a research resource for understanding climate-related ocean physics to lose it at this juncture. In fact, if either the TRANSPAC or TOGA sampling programmes are seriously degraded, the whole climate programme will be affected. Even if CLIVAR, GOOS or some other future programme were to implement sampling in TRANSPAC, there will be a gap in the time series between now and when such a programme begins (1996/97?). The Committee recommended that funding for TRANSPAC and other Pacific lines, which are being degraded by the termination of the FNOC SOOP, be urgently sought from other sources such as a multinational effort. (See Section 6.4).
The WOCE and TOGA XBT programmes should be incorporated into GOOS/GCOS plans.
The XBT surface layer programme provides information on the air-sea fluxes, heat transport and storage, on variability (and hence on forcing), on ocean dynamics, and on validation and representativeness of the WOCE hydrographic and satellite sampling programmes.
Resources should be assigned to data-sparse areas only if an adequate level of sampling can be achieved, and without detracting from well sampled regions such as TRANSPAC, the tropical oceans and the North Atlantic. In the Southern Ocean the emphasis was placed on "choke point" lines which are critical for heat flux computations.
XBT sampling resources should be allocated preferentially to the Indian Ocean for 1996 and the North Atlantic in 1996-97. (The TWXXPPC recognizes the importance of observations in different regions at different times, but also noted that this kind of instruction was difficult to follow because it takes time to set up a network of greeters and vessels. Switching from one ocean to another makes it very difficult to return to the original region once the network has been destroyed.)
Ships of opportunity should be used to obtain as broad a range of data as possible, so should be equipped with thermosalinographs and hull contact sensors etc.
Specific priorities for lines as defined by WOCE committees in the past are as follows:
In the Indian Ocean several lines should be upgraded to high density to replace repeat hydrography which is lacking. IX1, 3, 6, 2, 15/21 should have two repeats per year, and IX10 should have four per year. According to Core Project 1 these would then form the primary repeat hydrography component of Indian Ocean WOCE. Section 5 of this report describes the response to this CP1 recommendation.
In the Atlantic Ocean, it was recommended that some sections have increased frequency and density to map the heat content of the subtropical gyre. Important sections through the centre of the basin are AX7, AX8, AX13, AX14, AX16 and AX18. The subpolar gyre is poorly covered because traffic lanes are parallel to the major ocean features. In addition, too many shallow XBTs are used along lines that do traverse the features, and XBTs alone are insufficient because salinity is a controlling variable in convective cooling.
In the Pacific Ocean, the recommendations were that TRANSPAC be continued through WOCE and that the high density lines which partition the basins into regions for heat transport/flux calculations are high priority. All lines presently sampled only in the tropics should be sampled polewards of the tropical regions too.
The Committee considered the future directions for the XBT network with these previous recommendations and the state of the sampling at present in mind. The main priorities they see are:
- To broaden the range of measurements made by ships of opportunity. Vessels collecting XBTs should all be WMO selected ships for meteorological observations (VOS). A subset of vessels should also incorporate surface temperature (hull contact sensor) and surface salinity (thermosalinograph) devices. Air-sea flux measurements (solar and infrared radiation incoming and outgoing, air-sea temperature difference, rainfall, wind stress and so on) should be implemented on carefully selected vessels. New profiling systems such as ADCP and profiling ALACE floats should also be included in the network, particularly in those areas which are sparsely sampled.
- Notify the research community that XBT lines are in place now or are being planned or are being encouraged which could provide thermal upper ocean structure measurements in the Antarctic Circumpolar Ocean. An example of operations in place are the French-Australian sections, and an example of infrastructure in place awaiting resources is the Argentine supply vessel in the Drake Passage.
- Continue studies using XBT measurements of the representativeness of the WOCE one-time and repeat hydrography lines for estimation of climatically important processes (heat storage, meridional heat transport and so on). Objectives should be to optimize the XBT sampling to complement the WOCE hydrographic and satellite sampling systems. Working with the numerical modeling community is a key part of these studies.
- Continue the development of autonomous sampling systems and of ways of keeping them well-calibrated, to ensure higher density, more accurate, more reliable measurements.
The goals of the XBT programme in TOGA are ( i ) to describe the tropical ocean thermal structure and the major ocean currents as elements of a coupled ocean-atmosphere system for 10 years, and ( ii ) to help design a sampling system for input to the interannual climate prediction system. Sampling consists of low density which will define a minimum level of 2 degree lat. by 15 degree long. by two month time scales, as well as the frequently repeated sections.
There has been substantial progress in describing the time-variable thermal structure and currents of the tropical ocean and in developing models and observing systems for predicting variability of the coupled ocean-atmosphere system, as documented in detail in the report of the Workshop on the Use of Subsurface Thermal Data in Climate Studies (WOCE Report No. 110/93, ITPO Report 9). In particular, forecasts of climate variability directly linked to events in the tropical Pacific Ocean are now made routinely and have a high skill factor (up to correlation >0.8 in some areas) in predicting seasonal rainfall anomalies in some continental areas. The committee felt the economic value of these forecasts should be recognized when making decisions about the future of subsurface sampling. New avenues of research should be opened, directed toward learning how to make economic use of the forecasts.
The Committee also noted deficiencies in the forecasts and the potential for improving them through a better understanding of the variations in other tropical oceans. For example, rainfall in Australia is strongly influenced by a pattern of SST anomalies in the Indian Ocean (Nicholls, 1989, J. Climate for example). The ocean mechanisms that are (or may be) active in controlling the SST anomalies have only recently begun to be studied. Models which assimilate subsurface thermal data of the Indian and Atlantic oceans similar to those existing for the Pacific are being developed.
Decadal variations in the thermal structure of the Pacific Ocean have only recently been documented. The Committee noted the need to explore the global nature of this signal, and to understand the background annual variation throughout the oceans, estimated from synoptic global data collected over a considerable period of time.
The low density network throughout the tropical oceans is finally coming close to full implementation in 1994 after 10 years of development. While a line by line may appear to be useful or needed by some, the Committee believes that there is no scientific basis on which this can be done at the present time. Such decisions might be possible in the future by working closely with the modeling community to determine what data are required for model-data-assimilation. The ability to make such decisions require that the networks as presently implemented is maintained for some finite period of time (e.g. for two ENSO cycles).
Finally the Committee felt that it is essential at this time to re-emphasize the importance of preserving long time-series of systematic XBT observations. The loss of PX20, PX37, PX38 and PX39 is a matter of urgent concern along with the forthcoming cut of PX7, PX13 and PX26 (TRANSPAC). The loss of the central tropical Pacific lines (PX9/13/31 ) is also a matter of urgent concern requiring immediate coordination among the relevant agencies to restore them. (Action Item 20)
8. USING TAO DATA TO ANALYZE SAMPLING STRATEGY
W. Kessler presented a summary of an analysis the PMEL group has been doing using TAO data. The results are consistent with and expand on earlier sampling analyses presented to the Committee. The highly resolved TAO time series allow the precise quantification of aliasing errors associated with sparse sampling. The study has found that the signal to noise ratio varies according to the region of sampling, and according to the time period of the signal being resolved. The technique allows predictions of the level of sampling required to resolve a particular signal with an acceptable signal/noise ratio in various regions, or to estimate the errors due to existing sampling strategies. The PMEL group will continue the study, expanding to regional and temporal differences in the signal/noise ratios. A full description of the work is given in Appendix D.
The Committee encourages study of problems in the use of the complementary thermal data sets (XBTs, TAO) in interpolating between observations to construct research-quality maps of fields. (Action Item 21)
J-P. Rebert gave an overview of the data management structure including the activities of the Global Subsurface Data Centre (GSDC) at Brest. A full report from the GSDC was circulated to the meeting participants after the meeting. Many of the data management issues were discussed in the meeting of the Upper Ocean Thermal Data Assembly Centres Co-ordination Group which met in the two days following this meeting. The findings of that meeting are presented in Part ll.
The GSDC is now collecting global subsurface temperature (from XBTs), and surface salinity data sets for WOCE and TOGA, although real time data is being received only from the tropical oceans. A new data processing system, which includes a continuously updated database, became operational in 1993. Procedures have been revised concerning data flow and additional information related to the data and their qualification can be stored. All data received are archived on optical disk at their original resolution. However, because of space constraints they are stored in the continuously updated database reduced to a maximum of 300 points, if their depth resolution exceeds 2 metres.
The GSDC still holds data received in real-time which were never replaced by the delayed mode higher resolution profiles. The Committee recommended that originators of that data be contacted by the WOCE IPO and the ITPO in order to encourage the submission of the delayed mode data. (Action Item 22)
The issue of merging the XBT and buoy data with TOGA hydrographic data was raised. WOCE is already discussing ways in which to make all the different types of WOCE data available to the community, and one option is an Integrated Data Set (the data files being connected by software for easy access and searching). The Committee recommended the GSDC in Brest investigate the possibilities for TOGA hydrographic data being merged with the XBT and other data as soon as possible. (Action Item 23)
At present the TAO data which are received over the GTS are captured by the GTSPP and the GSDC and retained in their subsurface thermal data sets. The Committee thought it inappropriate that the XBT data be mixed with the TAO data in the data base, since the sampling characteristics of the two data sets are so different. The Committee recommended that the TAO data not be mixed with the XBT, MBT data in the GSDC and NODC files, but that a separate TAO file should be maintained. (Action Item 24)
The final TAO dataset in the data centres should be the complete, checked set, which is best obtained from PMEL in Seattle, rather than relying on the GTS. At convenient intervals (perhaps monthly) the Brest Centre should get the updated TAO data set from PMEL. There is no benefit in adding TAO data to the Brest database on a daily basis. Users who require daily-updated TAO data, as for model/data assimilation, are able to get the data either from the GTS or from PMEL over the Internet. (Action Item 25)
10. THE WORKSHOP ON THE USE OF SUB-SURFACE THERMAL DATA FOR CLIMATE STUDIES
G. Meyers noted the success of the Workshop held in Brest in September 1993 which summarized for the TOGA and WOCE community the science that was being done with the XBT data. The report of the Workshop has been sent to the usual TOGA and WOCE report distribution lists, but the question for this meeting was how to publicize the results more widely.
As many authors have already submitted their papers to scientific journals, the Committee agreed an appropriate action would be to publish an overview article of all the work being done with WOCE and TOGA data. G. Meyers also agreed to canvass authors for their plans regarding submitting papers, and their opinions on producing a collection of published papers. (Action Item 26)
The Committee also agreed a summary of the findings of the workshop should be written for the planning committees of GCOS and GOOS. (Action Item 27). As an initial step, Meyers, Molinari and White summarized applications of subsurface thermal data presented at the Workshop, and the sampling density required for each application (Table 3).
W. Woodward presented the US plans for long term observing systems. Planning continues for the implementation of the US contribution to the international Global Ocean Observing System (GOOS). GOOS is an operational system for collecting ocean observations for the benefit of society. It is not a research programme but it provides data for research and operational purposes. GOOS is partitioned into five application areas or modules. These are Climate, Health of the Ocean, Living Marine Resources, Marine Weather and Operational Oceanographic Services, and Information for Coastal Zone Management.
In the US an Interagency Working Group for GOOS has been established with NOAA, Navy, NSF, NASA, DOE, EPA and State as charter members. Elements of the US GOOS strategy include selecting observations that return significant economic/social benefits, forging academic-federal-private partnerships for GOOS implementation and continuing/improving existing time series of ocean observations. US planning has been the most active for the climate module and a zero order or base-line GOOS that includes the present research funded TOGA observing system, is being planned in the US. Because the initial focus of the US climate module is on observations for forecasting at seasonal to interannual time scales, this module is building directly upon the successes of TOGA. New operational funding for ocean observations beginning in FY 1996 is being requested by NOM under an emerging Seasonal-to-lnterannual Climate Prediction Program and a Program for Prediction and Assessment on Decadal to Century Time Scales. If successful, the new funds will provide long-term support of the TOGA observing system as well as NOAA's Global Sea Level System and components of a global surface drifter programme.
US GOOS Climate Module planners are stressing the importance of identifying the oceanographic fields to be observed and are adopting a three pronged approach to observations: (i) stabilize existing components; (ii) optimize present system through a strong programme of evaluation; and (iii) develop criteria for expansion/modernization through process experiments. As GOOS matures and operational sources of money become available, choices will be required on which types of observing system components should be sustained and how they should be blended. this will demand the strong focus on evaluation and trade-off analyses mentioned above among the various contributing observing systems.
The Committee noted the publication of the report entitled Ocean-Atmosphere Observations Supporting Short-Term Climate Predictions" (the "Knox Report") prepared by the US TOGA Advisory Panel's Panel on Near-Term Development of Operational Ocean Observations. The Committee warmly welcomed the emphasis on the contribution of XBT data, the generous acknowledgment of the work of the Committee in attempting to optimize the observing strategy and the encouragement to the Committee to continue with the present thrust of its activities.
The Committee regretted the wording of one statement about the (alleged) redundancy of XBTs which "overlap the TAO array", quoted from a letter to the author by E. Sarachik. This statement is open to misinterpretation because it does not make clear that a potential for redundancy exists only in the rare event of an XBT being dropped in the same location as a mooring. Overall, however, the thrust of the report is positive concerning the importance of the role of XBT data (and of the Committee) in supporting short-term climate predictions.
The TOGA Pool has been the main tool for successfully implementing the TOGA XBT network to the required level as described by the TOGA Implementation Plan. The Pool arose as a result of signed agreements between USA, France and Australia which allowed for the distribution of the shared resources in an efficient and scientifically beneficial way. The Committee felt that a similar Pool with several countries contributing will be the best way to distribute resources within future programmes such as CLIVAR, GOOS or climate prediction projects. The Committee recommended that a follow-on Pool be created for post-TOGA resources.
In terms of numbers of XBTs, the requirement for WOCE and TOGA according to their Implementation Plans is a total of 42,100 probes per year for low density, plus 14,360 probes per year for high density sections. The Committee recommends that to maintain the WOCE/TOGA network the global Pool should consist of 42,100 probes for operational, low density sampling. High density lines are used primarily for research on heat transport, and they require principal investigators. The Committee felt that the pool should not include the HD probes at this time. (Action Item 28)
12. THE FUTURE OF THE COMMITTEE
It has been suggested that the TWXXPPC change into the CLIVAR-WXXPPC (CWXXPPC) on 1 January 1995 when TOGA ends. The Terms of Reference would remain the same (substituting CLIVAR for TOGA throughout) for at least the first year. During 1995 the CLIVAR and WOCE SSGs would develop an appropriate committee structure to meet their goals beyond 1995, and in consultation with the CWXXPPC, develop terms of reference for the relevant follow-on group(s).
The Committee expressed concern about the stage of planning and development that CLIVAR has reached as of April 1994, and the fact that CLIVAR was not yet in a position to define detailed requirements for ocean observations. The Committee therefore felt that it might be more appropriate to remain a TOGA/WOCE Committee until such a time as CLIVAR plans are more developed. (Action Item 29)
The Committee noted that there is a lot of scope at this time for consolidating international committee work. At least four WMO and/or IOC-sponsored committees are concerned with the collection and use of XBT data. The Committee urged the CLIVAR and WOCE SSGs to take into account the overall existing committee structure when making plans for future committees, and that these plans should be developed in close coordination with the IOC, WMO, GCOS and GOOS. Noting that, for instance the TOGA TAO Implementation Panel and the WOCE/TOGA Surface Velocity Programme Planning Committee (SVPPC) are also concerned with upper ocean thermal data, the TWXXPPC felt that future committees should be organized around fields to be observed rather than the instruments used to make the measurements. (Action Item 30) One Committee could deal with the scientific issues related to upper ocean temperature and salinity fields, taking into account data collected by XBTs, moored and drifting buoys, profiling floats, satellite measurements, thermosalinographs, etc. One or more implementation groups would then be required to oversee the logistics of collecting the data, implementing the sampling strategy and developing new instrumentation.
Because the life of the Committee in its present form is to be extended until the end of 1995, it was recommended that some rotation of members occur if necessary. G. Meyers expressed a wish to step down after serving as chairman of TWXXPPC for three years and another two years before that as chairman of the TOGA XBT committee. Committee members thanked G. Meyers for the work he has done as chairman and proposed that he remain an ex-officio member of the Committee. The Committee proposed J-P. Rebert be the new Chairman and that F. Dobson be Vice-Chairman. (Action Item 31)
The next meeting of the committee should be October or November 1995. P. Holliday offered to host the meeting in Southampton and C. Henin offered to host the meeting in Noumea. The location which is most economical in terms of travel should be chosen.