1. The Fourth Session of the Committee for Inland Fisheries of Africa (CIFA) Sub-Committee for the Development and Management of the Fisheries of Lake Victoria was held from 6 to 10 April 1987, in Kisumu, Kenya.
2. The Session was attended by representatives of Kenya and Tanzania. Observers from the: European Economic Community (EEC), Uganda Freshwater Fisheries Research Organization (UFFRO), Kenya Marine and Fisheries Research Institute (KMFRI), African Development Bank (AfDB), Haplochromis Ecology Survey Team (HEST), and the International Agricultural Centre, The Netherlands were also in attendance. The list of delegates and observers is given in Appendix A of the report.
3. The Session was opened on behalf of the Honourable Andrew Omanga, Minister for Tourism and Wildlife of Kenya by his Assistant-Minister, the Honourable Moody Awori, who welcomed the participants. He emphasized the importance of international collaboration for proper management of the fish resources of the lake and expressed the wish that the Sub-Committee produce practical recommendation to help bring this about.
4. The FAO Senior Regional Fisheries Office for Africa, Dr. W.Q.B. West, welcomed the participants on behalf of the Director-General of FAO, Dr. E. Saouma; the Assistant Director-General a.i. (Fisheries Department), Dr. A. Lindquist, and the FAO Regional Representative for Africa a.i., Mr. T. Taka.
5. Mr. S.O. Allela, Director of the Kenya Marine and Fisheries Research Institute and Mr. J. Yonazi, Fisheries Officer of the Fisheries Division, Tanzania, were elected Chairman and Vice-Chairman, respectively.
6. The Annotated Agenda and Timetable were adopted as attached in Appendix B. The documents presented to the Session are shown in Appendix C.
Composition of the Catch
7. The two member countries present at the meeting reported very similar trends in their fisheries as determined from catch records and in the composition of the fish stock as determined by experimental fishing (see Appendix D. 1–2). A paper available to the meeting confirmed these trends for Uganda (Appendix D.3). These were:
the continued increase in abundance of Lates in both catches and experimental trawl samples. The species contributed 56 percent of the Kenyan catch and 38 percent of the Tanzanian catch in 1985. Kenya noted some changes in the distribution of Lates with a diminution of the population in the Nyanza Gulf and an increase in the more open waters of the lake;
the complete disappearance of haplochromine cichlids from the catch. Both Kenya and Tanzania, however, had remarked on the persistence of haplochromines in certain shallow water habitats (see Annex). In other areas where Nile perch densities have fallen, populations of haplochromines have reappeared;
increase in catches of Oreochromis niloticus: catches have increased especially in shallow areas associated with papyrus swamps and now contribute about 10 percent of the catch in both Kenya and Tanzania;
increase in the quantity of Rastrineobola: both Kenya and Tanzania remarked upon the increase in numbers of Rastrineobola argentea. This pelagic species is playing a more important role in the fishery having risen in Tanzania and attained 29 percent of the catch in Kenya. Rastrineobola has also increased in importance as an item in the diet of Nile perch;
increase in numbers of other species: some indications of an increase in the numbers of some other species, including Labeo, Schilbe and Barbus have been noted in all those countries with good runs into rivers this year, and
increase in total catches: total catches have continued to rise in both Kenya and Tanzania with 89 000 t being recorded for Kenya and 99 000 t for Tanzania. Some doubts were expressed about the reliability of these figures due to possible errors in collection of data.
Changes in the Fishery
8. The fishermen around the lake have responded to the presence of Nile perch by a rapid adaptation to the new species.
| 9.(i) | Pronounced shifts in the mesh sizes of gillnets in use have been noted with the emergence of a range of large mesh sizes (150–250 mm) hitherto absent. It was noted that the current legislation regulating fishing was now outdated and applied to a fish community structure which no longer exists. The difficulties of establishing mesh size regulations that would allow for three diverse fisheries, each of which was selected for by very different size ranges of fish, was recognized. Furthermore, the desirability of establishing regulations in a rapidly changing fisheries situation was questioned. The fishery had shown itself flexible enough to respond to successive overfishing and also to the emergence of a new species by changes in mesh size and would presumably continue to do so if only for economic reasons. Besides such regulations had historically proved difficult to enforce. The problems created by different standards existing in the member countries was also recognized. It was therefore proposed that an attempt be made to harmonize fishery regulations among the member countries of the Sub-Committee by reviewing the existing situation prior to proposing suitable revisions to the legislations of each country. It was recommended that the Directors of Fisheries transmit a precis of current regulations to the Secretariat for dissemination. These could then be discussed in detail at the next Session of the Sub-Committee; |
| (ii) | acceptability of Nile perch and its various products had emerged in nearly all regions around the lake with the development of more appropriate means of cooking and preservation. Of these, frying the fish in its own oil has proved to be most widely adopted, and |
| (iii) | although many problems remain to be solved with marketing, sale of fish for export to urban centres or even overseas had increased producing a price structure more attractive to the fisherman. There is, however, a need for harmonization of quality control among the member countries and it was recommended that member countries seek assistance in food technology to enable them to standardize product development and standards for export and internal consumption. Furthermore, with the development of new types of fishing and fish products and the growth of a more export-oriented fishery economy, old marketing patterns were changing. There is now little knowledge of the fate of the fish caught in the lake basin and a systematic study is needed to clarify this. |
Feeding Habits of Nile Perch
10. Scientists from all three member countries have made similar observations on shifts in the diet of Lates. The early reliance on Haplochromis has diminished considerably as this prey species had disappeared. There has been a corresponding increase in the appearance of Rastrineobola in the diet and a very rapid growth in the percentage of fish feeding on Caridina nilotica. The capacity of these two organisms to support massive and sustained predation is one of the most important single factors in determining the eventual stabilization yield and population structure of the Lates.
Fish Kills
11. An increasing number of cases of fish kills have been noted both in Tanzanian and Kenyan waters of the lake (see paragraph 18).
12. A number of nationally and internationally financed projects for the investigation and development of Lake Victoria fisheries were reviewed.
Improved Utilization of Nile Perch
13. This Dutch-funded project is investigating the current utilization of Nile perch and possible ways in which it might be improved. The major factor limiting the benefits derived from the catches of Lates is the lack of facilities for low temperature storage of fish. Insulated cold stores would alone prevent much of the qualitative wastage now current and would extend the life of the products so that they may receive wider dissemination and marketing.
14. Nearly all parts of the fish are being used at present with a filleting industry providing material for export. The quantity of fish available to the filleting plants is limited by the quality of the landed product which is frequently unacceptable for other than local consumption.
15. Additional fish products might be envisaged which utilize other parts of the carcass. For instance, swim bladders have a high value for export. Fish skins are also potentially valuable although the development of an industry to prepare these would need expertise.
16. Traditional processing methods are being reviewed by the project.
17. A seminar or workshop on food technology had been included in the project document but progress so far did not justify organizing this. It was seen as desirable that the workshop, which could include Tanzanian and Ugandan participants, should be provided for under an extension phase of the project.
Studies of Fish Kills
18. Fish kills in the Kenyan waters of the lake have been studied by KMFRI (Appendix E). Although records of fish kills date back to the 1920s, there is evidence that they have been increasing in recent years. Large fish are mainly affected and although a number of causes have been suggested, including poisoning by heavy metals or by pesticide residues, discharge of deoxygenated water from river mouths, changes in the circulation pattern after the construction of the Mbita causeway and local algal blooms, the pattern of kills would suggest massive deoxygenation through overturn as being one of the major factors. This pattern of kills, however, merits further study on a collaborative basis and the Sub-Committee recommended the research institutes in the three member countries to include the item in their programmes.
19. Fish kills also raise serious concern about the environment of Lake Victoria and the impacts of development activities within the lake basin. Pollution from industrial, agricultural and urban sources is growing and physical alterations to the lake shore through constructions may well also increase in the future. It was felt that the Sub-Committee should concern itself with the assessment of such impacts and that mechanisms should be sought within the member countries whereby proposed developments could be examined for their possible effects on the aquatic environment. Rwanda and Burundi should be invited to participate in this matter.
Haplochromis Ecology Survey Team (HEST)
20. Observers from HEST presented three papers describing work on the artisanal fisheries for Nile perch of southern Lake Victoria, recent development in the fishery for haplochromine and Nile perch in Lake Victoria and a comparison of trawl surveys in the Tanzanian part of the lake (see Appendix F.1–3). The findings of these various activities supplement and confirm many of the conclusions described in paragraphs 7–10.
21. In response to the changing situation in the lake the HEST team has been reorienting its programme toward Nile perch stock assessment. Research also includes investigations into Caridina and Rastrineobola. A training course in stock assessment and species identification for statistical purposes to be held in Mwanza is under consideration. Participants will come from all three member countries.
Stock Assessment Using Length Frequency Analysis
22. Assessment methods developed for tropical fish stocks and used in the DANIDA/FAO training course have been applied to Nile perch. This study is appearing in an FAO Technical Report. The method extracts growth and mortality parameters from samples analysed for length frequency which when extrapolated to total catch can be used in a length-based cohort analysis to estimate total biomass independently of trawl survey. The Sub-Committee recommended the general adoption of this methodology by the Institutes working on the lake and further recommended that funds be sought to organize a training course to promote understanding of the method in all such institutes. The course should be attended by training staff as well as research scientists.
Current and Potential Socio-Economic Effects of Nile Perch
23. While considerable attention has focused on biological and technical issues related to the rapid spread of Lates in Lake Victoria, socio-economic aspects of the situation have been relatively neglected. Past observation of human welfare impacts and potentialities largely have been based on indirect and informal evidence rather than thorough study and documentation. It is recognized therefore that a pressing need exists for information on the Nile perch fishery with reference to its current and possible future effects upon local and regional communities. Such information would provide a sound foundation for practical development initiatives.
24. The Sub-Committee noted the lack of knowledge of socio-economic implications of recent changes in the lake fisheries and recommended that a project be prepared to carry out studies to resolve this.
EEC Project
25. The Sub-Committee was informed that the EEC had now submitted a formal proposal to the member countries and was awaiting a request from the respective Government. The project could be implemented within a short period of receipt of these requests.
Short and Medium-Term Priorities for Research
26. The standing Working Party on Research Coordination for Lake Victoria established by the Third Session of the Sub-Committee was not convened during the intersessional period. However, a meeting of the Directors of Research of the three countries in Arusha, provided the opportunity for a review of the progress of existing projects and the priorities for future research.
27. Many of the topics for research defined by the Third Session were being pursued by national or international projects. Unfortunately, limitations on the resources available did not permit work on all topics. The Sub-Committee reviewed the needs for research in the light of the past priorities and the changing ecology and fisheries of the lake and defined the following:
research on Nile perch biology, ecology and population dynamics: although progress had been made on this research capacity was still limited. A comprehensive research project on all aspects of the biology of Lates and on the dynamics of its principal prey species (Caridina and Rastrineobola) is the subject of a proposal to IDRC by Tanzania and Uganda;
research on Oreochromis niloticus: with the re-emergence of this tilapine species as a significant element in the catches, and its ability to co-exist with Lates research on this species was assigned high priority. No external funding has been envisaged for this topic as yet although assistance might be sought to resolve any doubts as to whether or not existing stocks are hybrids of O. niloticus and O. esculentus;
research on Rastrineobola argentea: in addition to its role as prey for Lates, R. argentea forms the basis for a growing fishery. The distribution of this species in the lake is not fully known, particularly with regard to the central area of the lake. Studies of the distribution, and the essential population parameters of this species are of high priority to predict stabilization levels of Lates, possible levels of offtake for a fishery and possible trade offs between fishery and Lates abundance. This topic is so far not funded;
periodic stock assessment: a need was foreseen for periodic (five yearly) stock assessment surveys of the whole lake. Such surveys would require resources beyond those available under normal natural projects and would require location of external donors;
fish kills: in accordance with the observation made in paragraph 18, coordinated research on this topic was assigned a high priority;
research on anadromous species (Barbus, Labeo, Schilbe): these species still form the basis for small but locally significant fisheries. Their recent reappearance after a period of decline give some hope that these fisheries can be retained. Research was felt to be necessary in order to define the management measures, both on the fishery or by stocking programmes, that might be necessary for their maintenance. It was noted that Kenya had already established a riverine laboratory at Sangoro and could therefore take a lead role in the topic, and
standard reference collection of fish: none of the research institutes around the lake possesses an adequately catalogued and complete collection of lake fishes and difficulties had already been experienced in fish species recognition for statistical purposes. It is felt that conformed collections in all three of the main national research institutes should be established. Assistance should be sought in this. The Secretariat agreed to contact possible sponsors.
28. A workshop on fisheries statistical systems of the riparian states funded by the FAO Regional Office was held under this item of the Agenda. This was attended by participants drawn from all three member countries.
Review of National Statistical Services
29. The systems for the collection and compilation of fish production data in the riparian states were described.
30. In Tanzania the catch assessment which was established in 1972 covers selected beaches known as recording stations chosen according to the importance of the station, but also with a view to administrative consideration, each district having at least one station. At each recording station catches by canoes selected by random sampling but ensuring that each type of gear has a chance of being sampled, are reported four days per week. In addition, information on the type of boat/canoe and the number of fishermen per unit is also collected.
31. The information collected at the beach is later transferred to a daily record form (form FSF 21B) at the end of the day. All the daily record forms are sent to the regional office for processing and onward transmission to Fisheries Headquarters.
32. The second part of the statistical system is the annual fishing village survey conducted on all the landing beaches of the country. The information collected by this survey is: (a) number of fishing vessels by type; (b) number of fishermen, residents and non-residents, and (c) number, type and size of fishing gear owned.
33. In 1971/72 a special survey was conducted to determine the number of fishing days per month. It was found that on average a fishing canoe operated for 20–25 days per month and hence about 250 days per year.
34. The canoe or boat is considered the main component of the fishing unit and hence the basic raising factor for the estimation of the total catch. The average catch per boat is calculated by dividing the total recorded landing by the number of recorded boat landings. This is then multiplied by 250 to give the average catch per boat per year. Then using the number of the fishing vessels obtained, from the frame survey the annual production for each station is calculated and these summed to obtain total production.
35. In Kenya the system was also established in 1972 and some 12 out of about 200 landing sites on Lake Victoria are sampled. Each beach has two enumerators (fish scouts) supervised by a district fisheries officer. Data collection by enumerators includes catch by weight and species per individual canoe; the number of boats at sampling beaches is known but the number of fishermen or gear is not recorded.
36. Raw data are forwarded by the DFOs (who are responsible for validating the data) to the Kisumu Regional Fisheries Office. Monthly summary sheets are sent to Nairobi for processing to produce the annual production estimates according to the following:
where: A = Estimated monthly production in metric tons
B = Number of canoes operating in the lake
C = Number of fishing days per year
D = Average catch per canoe per day
E = Quantity in kilograms estimated as unrecorded
37. In Kenya no frame survey has been carried out since 1971 when the number of canoe operating on Lake Victoria was recorded as 4 000. However, registrations showed that the number of boats operating had increased to 5 000 by 1983.
38. An additional survey is carried out which collects rather more data of a biological nature, including length frequencies.
39. In Uganda landings at all major beaches are recorded and estimates made for the minor ones. All enumerators (who have 0 level secondary schooling and 2–3 years training in fisheries training institutes) analyse their own data and write monthly reports which are sent to the Regional Fisheries Office, who in turn sends a monthly statistics compilation and report to Entebbe Headquarters. (In the absence of a delegation from Uganda this paragraph was summarized from information given in CIFA:DM/LV/87 Inf.4, which contains detailed statements of all three countries' fishery statistical systems.)
40. The importance of adequate statistics for rational management of the fish stocks of the lake was stressed and in this connexion several short-comings of the systems presently in operation were discussed as outlined below:
species identification was felt to be poor with the enumerators often having to rely on the fishermen. Not only was there a significant quantity of unidentified fish, but species were still being recorded from areas where they had disappeared. When subjected to checks, enumerators were unable to identify fish although they were appearing in the records. To improve the situation in this respect the Sub-Committee recommended the production of a field guide for the fishes of the lake (identification sheets) together with basic training for all enumerators;
it was noted that the enumerator was the key element in the system, and it was recognized that incentives were needed to produce accurate records. In particular, comprehensive training of enumerators in fishery matters was urged;
field level supervision was most important, but was often inadequate with infrequent visits to the beaches by supervisory staff. This was mainly due to the large areas supervised which were difficult to oversee due to problems of transport;
apart from transport, equipment in general was noted to be lacking in particular scales for weighing fish and calculators were in short supply, and including, where necessary, sheds for the enumerators;
from the point of view of methodology the systems used by Kenya and Tanzania were noted as being too rigid. In particular beaches where catches were recorded had remained largely unchanged since the inception of the system some 16 years ago, in spite of known changes in the pattern of landings. Continuous monitoring of the coastline was seen as essential, but was difficult as enumerators, particularly in Tanzania where their number had recently been reduced from 24 to 18, would have well over 100 km of shoreline to cover without transport. It was also noted that within 100 km considerable variety can be found and that within such a large area it would often be difficult to find a “representative” site, and
although Tanzania carries out annual frame surveys, no such surveys have been held in Kenya or Uganda since 1971, and many of the basic parameters used in the calculation of monthly and annual catch data need updating. (Statement on Uganda derived from CIFA:DM/LV/87 Inf.4.)
41. As to harmonization of statistical systems, internationally it was recommended that use be made of the newly designated documentation centre at Mombasa. It was noted that UFFRO Jinja had been designated as coordinating centre for stock assessment for Lake Victoria.
42. The Tanzanian delegation noted that the harmonization of data systems should not present a serious problem as they were already rather similar and only the raising factors differed. A possible area of conflict was that apart from being a member of the Lake Victoria Sub-Committee, countries were members of other (different) groupings, e.g., Tanzania is a member of SADCC and SWIO which recommends different statistical systems. While it was noted that a common system was desirable, it was also the case that different fisheries often required different systems and that a common methodology was a less critical factor than ensuring the efficient working of whatever system was in operation.
43. In view of the need for improvement in the statistical systems presently in operation for the collection of fisheries data on Lake Victoria, the Sub-Committee approved the following resolution:
The Sub-Committee recognizing the importance of the fisheries resources of the lake for the riparian populations.
Emphasizing the necessity of international cooperation for effective management of these resources.
Aware of the vital importance of reliable statistics in the management process, and
Convinced of the importance of standardized systems for the collection and analysis of data recommended the establishment of a Regional Fisheries Statistics Programme, including the following elements:
Appreciating the general interest of AfDB in assisting fishery development in the region and stressing the urgency of carrying out a frame survey and a stock assessment survey, the Sub-Committee recommends that this interest be explored as a matter of urgency.
44. It was, however, stressed that approaches should not be limited to AfDB. FAO was, therefore, requested to formulate a project on the basis of the above resolution and to submit this project to donors (including AfDB) for financing as soon as possible.
45. The Sub-Committee expressed its appreciation to the Regional Office for Africa for funds made available for the preparation of a study on fisheries statistical sampling in the riparian states and hoped that further assistance in this sector would be made available.
46. Six species exotic to Lake Victoria have been introduced to the basin. Of these, one, Cyprinus carpio, has not so far appeared in the lake; three, Tilapia zillii, T. rendalli and O. leucostictus are present but are insignificant, O. niloticus can possibly be seen as a benefit and Lates niloticus is highly controversial. Plans for further introduction into the basin are probable both for capture fisheries in the lake or for aquaculture within the lake basin. In this respect two countries, Rwanda and Burundi, which lie within the basin are not members of the Sub-Committee and yet introductions made into their waters could also penetrate the lake.
47. The Sub-Committee concluded that as a principle further introductions of fish species into the Lake Victoria basin should only be carried out after careful study on the biology and ecology of the candidate species according to a procedure similar to that of EIFAC and ICES Code of Practice and after international consultation under the provisions of a convention which should be defined. The countries of the basin, Rwanda, Burundi, possibly Sudan, Egypt and Ethiopia should be invited to participate in any eventual convention and the consultations leading up to it. To assist Member Governments in formulating such a convention, it was recommended that a workshop be convened to discuss the topic of introductions and formulate a suitable mechanism for control and regulation of them. Other CIFA member countries could be invited to participate in the workshop and especially the members of the CIFA Sub-Committee on Lake Tanganyika.
48. The Sub-Committee expressed its concern as to the often alarmist and ill-informed articles that have appeared in the popular and scientific press as to the effects of Nile perch on the fish and fisheries of the lake. At present it is far too early to express a definite opinion as to the eventual balance which will emerge between the advantages and disadvantages of this introduction. However, from the evidence presented at this meeting there are now indications that adjustments are being made both by the fish community and the fishermen.
49. Because of the international nature of the lake and its resources the Sub-Committee felt it essential that a more formalized mechanism be established for the joint management of the lake fisheries by the three member countries. Present mechanisms including the Sub-Committee itself were advisory in nature, but the present Session has shown that the needs for collaboration in management of the lake now go beyond its terms of reference and exceed its response time. Recommendations have been made which call for policy decisions and possibly legislative action on the part of the Member Governments which will have to be coordinated and administered. An international body founded on local institutions is therefore needed for the lake and it was suggested that a body similar to the Lake Victoria Fisheries Commission could be re-established to fulfil these factors. The Sub-Committee recommended that the competent Government authorities work toward the establishment of such a Commission.
50. Tanzania offered to host the Fifth Session of the Sub-Committee in 1989. This offer was gratefully accepted by the Sub-Committee. The exact date and place of the Session will be decided by the Director-General of FAO in consultation with the Chairman and authorities of the host country. The Session should be preceded by a meeting of the Working Party on Research.
