AN ASSESSMENT OF THE ECONOMIC BENEFITS FROM STOCKING SEASONAL FLOODPLAINS IN BANGLADESH

Md. Liaquat Ali
Department of Fisheries, Dhaka, Bangladesh
Md Zahirul Islam
Third Fisheries Project, Department of Fisheries, Dhaka, Bangladesh

ABSTRACT

Floodplain stocking with carp fingerlings undertaken under two Development Projects with two different concepts has been found both technically and economically viable and effective. Under a project named Third Fisheries Project (TFP) carp fingerlings were directly stocked in mainly seasonal floodplains in western Bangladesh in 6 growing seasons (total cumulative area of 149,500 ha) over a period of six years (1991–96). Production increased by 1.5 to 16 times with an average of 8 times the stocked quantity of fingerlings. The tenfold increase or more has been found to be economically viable. The economic rate of return (ERR) of the TFP programme ranged from 28.6% to 122.44% with an average of 38.1%. Income of the fisherfolk increased 2 to 6 times in three floodplains in three years starting in 1992. Stocking in floodplains having some perennial water area in the north-east of Bangladesh by fry/fingerlings raised in low submersed dyked nurseries under the other project, Second Aquaculture Development Project (SADP), has helped to enhance floodplain fisheries and incomes of fisher persons.

1. BACKGROUND

Situated in the deltaic plain of the three river systems namely the Ganges, the Brahmaputra and the Meghna, Bangladesh has vast inland open water resources in the form of rivers, canals, beels, floodplains, etc., having a total area of 4,047,316 ha (rivers and estuaries 1,031,563 ha, beels, i.e. deeper part of floodplains 114,161 ha, floodplains 2,832,729 ha, Kaptai Lake 68,800 ha) with high potential for fish production. In addition, there are 260,658 ha of closed water bodies (ponds 146,890 ha, oxbow lakes 5,488 ha, shrimp farms 108,280 ha) suitable for aquaculture (DoF, 1986, 1997). Inland fisheries resources in Bangladesh are among the richest of the world, being second after China as of 1992 (FAO, 1995). The world inland capture fisheries production as of 1992 was 6.5 million tonnes of which China contributed 1.23 million tonnes, and Bangladesh 0.48 million tonnes (FAO, 1995). Inland open waters have been the major source of fish production in Bangladesh from time immemorial. In the 1960s about 90% of the country's fish production came from inland open water fishery (DoF, 1997). But due to man-made causes such as overfishing in absence of appropriate fisheries management and conservation measures, implementation of flood control and drainage programmes, construction of roads and dams, indiscriminate use of insecticides and pesticides for crop production, fish production has declined significantly in inland waters during the last three decades (Fig. 1).

Figure 1. Fish production trends in Bangladesh.

The contribution of inland open waters to the country's fish production has continued to decline to about 50% at present (DoF, 1997). Source-wise fish production over the period since 1983–84 is shown in Table 1. This decline has been comparatively high in the case of important and valued fish species such as major carps rui (Labeo rohita), catla (Catla catla), mrigal (Cirrhinus mrigala), kalibous (Labeo calbasu), and ghonia (Labeo gonius). The major carps earlier contributed about 30% of the total production but this has now dropped to 5–6% (DoF, 1997). The species composition in open waters has been out of balance because of the disturbance in natural reproduction of fish by overfishing and other causes. In the open water production system, floodplains play an important role for fish production. Floodplains are the low-lying areas which are flooded by rivers and rain waters during monsoon (June to December). About one third of the total area of Bangladesh is flooded every year and remains under water for 4–6 months. The floodplains are naturally very rich in nutrients and fish food and thereby are suitable breeding and grazing grounds of almost all inland fish and other aquatic animals. During the flood season (June-December) fish and shellfish grow in the floodplains and are harvested by the rural people. With receding water, fish accumulate in the deeper part of the floodplains, locally called beel, or they migrate to rivers which retain water throughout the year. Due to the decline of fish population in inland open waters and to obstructions of their migration routes, recruitment of fish, particularly carps, in the floodplains has substantially declined and natural resources available in some ecosystem niches of the floodplains remain vacant and unused.

Table 1. Fish Production in Bangladesh.

Various measures have been taken to increase the fish production in the country such as exploitation of marine fisheries, development of aquaculture and stocking of inland open waters. A major focus has been on inland aquaculture development. Aquaculture production has increased considerably during the last one and a half decades. Stocking of oxbow lakes in Bangladesh with carp fingerlings was an age-old practice but due to the lack of scientific considerations in respect of species composition, stocking density and size of fingerlings, the production level was very low, most of it contributed by the natural fish stocks. However, a systematic and sustainable stocking programme in six oxbow lakes in southern Bangladesh (Jessore district) was initiated under a World Bank financed project during the second half of the 1980s by introducing polyculture of local major carps and exotic carps (silver carp, grass carp, big head, common carp, etc.) at a density of 20–25 kg/ha. The proper species composition and a good size of fingerlings (12 cm+) stocked increased the production up to 950 kg/ha/yr, with an average production of 500 kg/ha/yr in oxbow lakes. The success in oxbow lake stocking formed a basis for starting open water stocking in Bangladesh and in the mid-1980s inland open water stocking with carp fingerlings was initiated.

Stocking of carp fingerlings in floodplains to (i) increase fish production by utilising the under-utilised resources, (ii) enhance the income of fishers, and (iii) create employment opportunity was started in 1989. In the beginning, stocking in open waters with carp fingerlings (5–8cm) produced in the Government Fish Seed Farm was initiated by the Department of Fisheries and continued for three years. As there was no production monitoring system its impact could not be assessed directly. However, an upward trend in fish production in open waters, mainly in floodplains, was observed. A large-scale stocking programme of floodplains with carp fry/fingerlings under two development projects, namely the Third Fisheries Project (IDA, ODA and UNDP assisted), and the Second Aquaculture Development Project (ADB financed) has been going on since 1991.

Under the Third Fisheries Project (TFP) mainly the seasonal floodplains in western Bangladesh were stocked with carp fingerlings (7–15 cm) through contractors, while under the Second Aquaculture Development Project (SADP) hatchery produced carp hatchlings (3–5 days old) were stocked in submersible low-dyked nurseries in beels (perennial water bodies in the deeper floodplain basin) and reared up to 5–10 cm size or until the nurseries overflooded and the fish dispersed in the floodplains. This paper describes in brief the floodplain stocking programme and reviews the economic benefits of the programme including the national objectives, methods, and management of the programme.

2. STOCKING UNDER THE THIRD FISHERIES PROJECT (TFP)

2.1 Objective

The objective of the stocking programme under this project is to develop sustainable management model(s)/system(s) to adopt a long-term strategy for enhancement of floodplain fisheries through application of a culture-based stocking programme to increase fish production in order to meet nutritional demands of the rural poor, in particular, increase the income of fishermen and create new employment opportunities for alleviating poverty of the local community.

2.2 Assumptions

2.2.1 Socio-economic assumptions

According to the Staff Appraisal Report (SAR) of TFP 1990 financial and economic rate of return should not be less than 25% and 12% respectively (IDA, 1990) for achieving sustainable performance. Stock enhancement would improve the income of fishermen in the floodplain area and induce changes in income distribution and assist in poverty alleviation in the floodplain community. The floodplain stocking component would allow genuine fishermen greater access to floodplain fishery. A large number of additional jobs would be created for producing and marketing of hatchlings, fingerlings and food fish in the private sector due to this programme.

2.2.2 Institutional assumptions

The system of leasing of public water bodies in the floodplains would be gradually replaced by a licensing system to establish direct access of the fishermen to the fishery. The private sector nursery operators would provide incremental production of fingerlings. The private sector traders and contractors, with administrative and financial skills, would organise nursery operators. Managerial capability of the DoF to manage large-scale stocking should be developed through technical and management assistance. The Fish Act would be strictly enforced to control the early catching of stocked fingerlings, and other conservation measures would be undertaken.

2.2.3 Technical assumptions

Floodplains are rich in nutrients and fish food and thus are the most suitable areas for breeding, nursery and growth of fish which migrate into and out of the floodplain with rising and falling floods. Therefore, a planned and systematic programme of stocking with suitable fish species in appropriate balance has been conceived to utilise the under-utilised nutrients and food available in the floodplains during flood season (June-December). This intervention, if properly applied, is expected to result in an incremental yield of 300 kg/ha in the major floodplains against the present low level of production of 70–100 kg/ha. A 12% retrieval rate of stocked fingerlings and a certain percentage of fingerling escapement from stocked floodplains were assumed. Most importantly, the assumption is that vacant ecological niches would be properly used and external stock enhancement with proper species mix, density, etc. would have no negative impact on biodiversity. Restocking would compensate for a reduction of natural fish recruitment and the decrease of fish habitats without affecting the rate of growth of non-stocked species. Baseline data on gear selectivity and fish food availability/limnology would be generated by a research component.

2.3 The programme

As planned under the programme 100,000 ha of floodplains, particularly in the major depressions in the Khulna-Narail, Gopalgonj-Madaripur and Chalan beel of Rajshahi-Pabna region, were scheduled to be stocked with 6–12 cm fingerlings of major carps with stocking density of 20–30 kg/ha in phases starting with 29,000 ha in 1992. By stocking carp fingerlings at a rate of 30 kg/ha in 100,000 ha of floodplains a total incremental production of 30,000 tonnes of fish was expected to be obtained at 300 kg/ha, i.e. 10 times the weight of stocked fingerlings. A pilot stocking programme was initiated in 1991 in two minor floodplains, namely Garalia beel (1000 ha) in Jessore district and Hilna beel in Naogaon district (2000 ha), with a total of 73 tonnes of fingerlings released. The encouraging results of this stocking programme provided guidelines in respect of species composition for the major stocking programme in 1992. In 1992 three major floodplains, namely poldered BSKB beel (area 13,000 ha, out of which effective stocking area is 6,000 ha) in Khulna Narail region, Chanda beel (open system with 6000 ha) in Gopalgonj and Halti beel (semi-closed with 10,000 ha) in Rajshahi-Natore region and one previously stocked beel were stocked (total area of 23,000 ha) with 392 tonnes of fingerlings. The stocking results of 1991 and 1992 provided a valuable technical basis for improving stocking of floodplains in 1993. In 1993, stocking was undertaken in BSKB, Chanda, Halti, Garalia and with varying stocking densities and species compositions based on the productivity of the beels and past experiences, and a total of 417 tonnes of fingerlings were released. The stocking performance in all the beels except Halti was found biologically and economically viable although the social objective was not achieved as per target in all areas of stocking. The programme for Halti in 1992 and 1993 was not economically viable because of the low productivity of the beel and poor ecological conditions, and as such, this beel was dropped from the programme in 1994. In subsequent years new floodplains were identified for stocking based on the selection criteria. In 1994, 1995 and 1996 floodplains having total areas of 24,000 ha, 32,000 ha, 43,500 ha were stocked with 417,595 and 628 tonnes of fingerlings respectively, with a stocking density of 13–26 kg/ha. A summary of stocking and results including area, fingerling quantity, cost, production and value are furnished in Tables 2 and 3.

2.4 Management

The success of the stocking programme as indicated by incremental production and its economic benefits depend on many factors like productivity of the floodplain, flooding period, selection of species, species composition of fingerlings stocked, stocking density, mortality after stocking (due to transportation stress, predation, disease, escapement, catching undersized fish before attaining marketable/legal size), quality of fingerlings (genetically sound), conservation measures, community participation, etc. Economic benefits were also contingent upon the expenditure of stocking including cost of fry, management, harvesting and marketing of the fish, etc. The above-mentioned parameters and issues of the stocking programme are briefly discussed as follows.

2.4.1 Preparation of the annual programme

The stocking programme in different years including issues such as selection of beel, species mix, density, species size, stocking points, etc. was prepared by exchanging and sharing views and ideas amongst field officers of DoF, Fisheries Research Institute (FRI), fishermen and NGOs and specialists.

Table 2. Incremental yield, value and cost of production of stocked species. Third Fisheries Project.

Note: Weight per fish at harvest: 400 g
Price per kg = Tk. 30.00
* Parenthesis indicates the average cost at harvest of 5 years stocking
** Because of unusal drought in 1992 (1 in 20 uears) production of both stocked and nonscoked species was remarkably low during that year.

Table 3. Third Fisheries Project: summary results of floodpolain stocking.

Note: Fish price per kg = carp Tk. 30.00; non-stocked species Tk. 25.00
Other cost includes DoF's administration, fishing and NGOs (only 1996) cost.

2.4.2 Selection of floodplains

The criteria followed for selection of the floodplains were: (i) physical and biological conditions, (ii) inundation period, (iii) depth and area of water, (iv) vegetation coverage, (v) accessibility to stocking points, (vi) community participation and scope for sustainability of the programme. Most of the floodplains were regulated by flood control measures and partially or completely dyked with sluice gates. Some floodplains were almost free of vegetation after harvesting Boro paddy and some were covered with Amon paddy. The floodplains should have a minimum water area of 1000 ha during the peak flooding for a period of 4–6 months (June-November) and a minimum water depth of 1 metre, and should not be covered by water hyacinth and other thick vegetation that hampers light penetration in the water. The stocking point should be accessible by water and road for transportation of fingerlings to the water body. The flood water should have such a point clearly marked and should have a good connection with river and other water bodies.

2.4.3 Species composition, stocking density and size of fingerlings

As mentioned above the success of the oxbow lake model of enhancement programme was the basis for undertaking floodplain stocking; the same species composition and stocking density as in the oxbow lakes was planned, with only the size of the fingerlings differing. As the oxbow lakes are perennial with an abundance of predators, they were stocked with largesized (12 cm and above) fingerlings, while the seasonal floodplains would initially have fewer predators and were stocked with smaller-sized fingerlings (7–15 cm). The species composition in oxbow lake stocking was: silver carp 50%, bighead carp 20%, common carp 10%, rui 10%, catla 5%, mrigal 5%. In the first year of the project in 1991 two floodplains were stocked with the above species composition at the rate of 24 kg fingerlings/ha, but the survival of silver carp was very poor while common carp did well. Subsequently, this species composition was adjusted and the rate of stocking of silver carp was drastically reduced or even cancelled for most of the floodplains, while the rate of common carp was increased. Another new species ‘Thai sarputi’ (Puntius gonionotus) was introduced. Stocking density was kept within 13 to 26 kg/ha. The species composition and density varied from floodplain to floodplain depending upon the physical and biological conditions. The species composition/mix, all cyprinids, after adjustment in the programme was silver carp (Hypophthalmichthys molitrix): 0–5%, common carp (Cyprinus carpio): 25–30%, rui: 16–30%, catla: 20–29%, mrigal: 9–16%, kalibous (Labeo calbasu): 0–1.6% and Thai sarputi: 10–13%.

2.4.4 Procurement of fingerlings

Fingerlings were procured and stocked through contractors prequalified by international bidding. The contractors entered into sub-contracts with fingerling raisers for supply. Initially contract lots were comparatively large and competition was restricted to contractors and supplier/traders only. Lot sizes were then reduced gradually. Since 1994 lot sizes were broadly divided into two categories, one of 5 to 10 tonnes restricted to fingerling raisers/nursery owners, and the other of 15 tonnes and above (up to 60 tonnes) for contractors and traders. Specifications and conditions for supply and delivery of live carp fingerlings were very rigid. These rigid conditions and penalty clauses were to some extent responsible for higher prices of the fingerlings. This trend has been reduced gradually. In 1993 the fingerling price was rather high (average Tk. 121/kg) due to time constraints and price rigging by the contractors. In the subsequent years fingerling price per kilogram gradually decreased to Tk. 99/60, Tk. 93/73, and Tk. 77/54 in 1994, 1995 and 1996 respectively due to better management of procurement.

2.4.5 Inspection and delivery

There was provision for inspection of nurseries from which the fingerlings were scheduled to be supplied by the selected suppliers. For quality control the stocking period was restricted to a maximum of 30 consecutive working days for lot sizes 15 tonnes and above, while it was 15–20 days for smaller lots during the period from 15th June to 31st August every year, although there was slight deviation in some cases. Fingerlings were carried by fish carrying drums and/or tanks oxygenated by aerator or by hand and then loaded, after conditioning, in a truck. Fingerlings were stocked and kept in hapas (specification designed by the project) at least for one hour for acclimatisation. Then fingerlings were released to the floodplains after being checked and sampled by a Government-nominated departmental committee including representatives from the local Government institution of Union Parishad and Thana Parishad and the fisher community. In addition, in 1992 and 1993, there was an arrangement for external checking by an independent certifying firm and later by the joint effort of DoF, NGO and fishermen representatives.

2.4.6 Conservation

Conservation of stocked fingerlings is one of the most crucial issues for the success of the programme. Enforcement of the Fish Protection and Conservation Act 1950 was strengthened by increasing manpower and logistic support. During initial years of stocking DoF had to face social and moral problems by prohibiting local fishermen to catch undersized fish as well as to use contravened gears that are harmful for stocked fingerlings. Size of released fingerlings of different major carps ranged from 7 to 16 cm and of Thai sharputi from 5 to 11 cm, although size range was changed and adjusted every year. Released fingerlings require at least 3 months to become marketable size. So, special arrangements were made to conserve these undersized fish of stocked species by enforcing the Fish Protection and Conservation Act. Gradually, the target groups and local community were motivated and actively involved in the conservation of fisheries resources including stocked fingerlings and in preventing the use of banned gears. NGO interventions in community participation have been found to be useful in this respect due to the fact that DoF has limited manpower, as well as limited knowledge and experience with social issues and mobilisation.

2.4.7 Production monitoring

In order to assess the impact of stocking in each floodplain prior to stocking, data were gathered on the types of fishermen, fishing gear, species composition and socio-economy of different categories of fishermen. A local consulting firm (Bangladesh Centre for Advanced Studies, BCAS) monitored the fish production for 3 years in three major floodplains (Chanda, BSKB and Halti) using household based survey system, gear dependent sample survey and FAD (fish aggregating devises). The remaining floodplains and the above three were covered by DoF's own manpower through gear dependent sample survey system developed jointly by DoF and BCAS.

2.4.8 Sustainability

Project design, in general, did not provide any means for sustainability of the stocking programme, except in terms of cost recovery from the beneficiaries. In order to address this vital and crucial sustainability issue, 19 NGOs were deployed in 16 floodplains in the last year of the project. Beneficiaries were to be organised and motivated by the NGOs and on smaller floodplains the beneficiaries would initially share the cost of stocking. Gradually they would take over the full responsibility of stocking and management. For larger floodplains public intervention with community participation might be an alternative option.