5. Sri Lanka

5.1 The country

Sri Lanka is a small continental tropical island situated southeast of the Indian peninsula covering an area of 65000 km². The island, believed to be an extension of the Indian shield, is mostly made up of pre-Cambrian crystalline rock; rudimentary Miocene limestone is found in the north and the northeast. Most of Sri Lanka is low-lying plain country 150 m below sea level. The highlands are restricted to the south central region where the mountain range rises up to 2 524 m above sea level. Three main morphological zones are recognized, i.e. the coastal lowlands (270 m above MSL), the uplands (270 to 1060 m above MSL) and the highlands (1060 to 2524 m above MSL).

The soil types, generally poor in mineral nutrients, have low humus content and are in many cases acidic. Fourteen to 17 soil types are recognized, of which reddish brown earth is the most common. The monsoons of the southwest (active from June to September) and the northeast (active from December to February) play a vital role in the climatic patterns of Sri Lanka. The three principal climatic divisions of the country and their average monthly rainfall are: the dry zone (125-187 cm); the intermediate zone (187-250 cm); and the wet zone (>250 cm). Most of the wet zone is situated in the southwest corner of the island.

More than one hundred rivers branch out from the central highlands flowing in all directions and carrying an annual discharge of 39 032x10⁶m³. Twenty-eight big rivers with basin areas exceeding 500 km² account for 92.5% of the total surface flow. Mahaweli, the biggest river of Sri Lanka, has a catchment area of 10237 km², which is equivalent to one-sixth of the country's geographical area. With a population of more than 16 million, Sri Lanka is one of the most densely populated countries in the world, and in the Asian region perhaps only second to Bangladesh. In ancient times, according to reports, the population was concentrated in the dry zones. This is substantiated by the preponderance of ancient reservoirs in the dry zones. Presently, however, 65% of the population lives in the wet southwest region.

Although the fisheries sector contributes only 2% to the Gross National Product (GNP) of Sri Lanka, it plays a vital role in nutrition as well as in the socio-economic factors of the country. Employing 120000 people directly in fishing and another 20 000 in ancillary activities, the fisheries sector has a return of Rs. 2144 million in foreign exchange through export of marine products. According to one estimate 65% of the animal protein consumed in Sri Lanka in 1991 was derived from fish. With a coastline of 1585 km and a continental shelf area covering 26000 km², the primary focus of fishery development in the country is on marine fisheries.

The inland fishery and aquaculture resources of the island comprise rivers, estuaries, lagoons and major and minor reservoirs including village tanks, brackishwater ponds and floodplain lakes. The inland fish fauna of Sri Lanka includes 51 to 55 species, of which two, Labeo dussumeiri and Puntius sarana, are primary food fish (De Silva, 1988). Thirty species belonging to 11 families and six orders are reported to be in the reservoirs. There is no aquaculture activity in the state except for the recent spurt in prawn culture in the coastal belts and lagoon margins. There is no riverine fishery worth mentioning. The floodplains, on the other hand, are reported to be productive ecosystems. Hettiarachchi (1983) estimates that the size of 26 floodplains in the Mahaweli basin ranges from 9 to 796 ha with an annual fish yield varying from 18 to 284 kg/ha. Habitat loss in floodplains owing to river valley projects and other developmental activities resulted in a sharp decline in their fishery potential. Thus, the main share of inland production is from the reservoirs.

The contribution of inland fisheries to the total fish production in the country is increasing steadily. In 1970 it constituted 8.3% of the total fish production, the remainder coming from coastal and deep-sea fishing. By 1989 the national fish production had doubled to 205000t and the percentage of inland fish had grown to 19.3%. There has been a steep fall in inland fish production since 1990, when state support for inland fisheries was withdrawn (Table 5.1).

Sri Lanka has inherited an extraordinary legacy of water resource management systems dating back to the fifth century BC. The ancient hydraulic civilization harnessed river water, using it to its full potential for irrigation and other human purposes. The arrangement of small tanks in a cascading sequence within microcatchments was one of the strategies followed in ancient times which ensured very efficient water management. Parakrama Bahu, the ruler of Polonnaruwa, was responsible alone for the construction and restoration of at least 165 dams, 3910 canals and 2539 reservoirs. Most of the present day irrigation reservoirs of Sri Lanka are the ancient ones, restored and repaired from time to time. Even today the density of reservoirs (230 ha for every 100 km² of land area) in Sri Lanka is the highest in the world. There are only two reservoirs which are over 5000 ha in area and only 12 are larger than 1000 ha, mostly comprising the recently formed storage reservoirs in the Gal Oya, Kirindi Oya and Mahaweli basins. Thus, an overwhelming majority of the reservoirs in the country can be classified as small water bodies.

Estimates of the surface area of ancient reservoirs – many of which are in derelict condition – vary from 3500 to 20000 in number. Most of the discrepancies are related to the number and surface areas of microcatchments, over which earthen dams have been constructed to store water. Sometimes, the word anicut is used to distinguish minute impoundments from the small and minor irrigation reservoirs. Minor irrigation reservoirs are also referred to as tanks, village tanks, etc. Seasonal tanks are the small reservoirs (<80 ha) that retain water for six to ten months a year. The register of irrigation projects in Sri Lanka (1975) provided a list by district of large (total irrigated area >1000 acres), medium (total irrigated area <1000 acres) and minor irrigation schemes of the country, along with large and small anicuts. According to this document, there are 72 large, 160 medium and 3296 minor irrigation projects, as well as 66 major and 1575 minor anicut schemes, covering a total area of 436385 ha. Thayaparan (1982) established the number of functional minor reservoirs in the country to be 3279. A more recent estimate of various categories of reservoirs is summarized in Table 5.2.

Considering that reservoirs form the bulk of inland fishery resources and a vast majority of them have an area of less than 1000 ha, reservoirs as such are considered small water bodies for the purpose of this study. This includes large, medium and small reservoirs, tanks and seasonal tanks.

5.3 Reservoir fisheries

Sri Lanka has the largest number of reservoirs for a country of its size and almost the entire inland fish production, which accounts for 20% of the total fish produced in the country, comes from capture and culture-based fisheries of the reservoirs. The available fish yield records from the reservoirs are equally impressive. The present high yield is a rather recent phenomena, generally attributed to the introduction of exotic cichlid (Oreochromis mossambicus) in 1952. It was realized as early as the 1950s that Sri Lanka did not have any indigenous species capable of contributing to a sustainable fishery and attempts were made to overcome this shortcoming by stocking the reservoirs with exotic species. Over the years, 22 species have been introduced in the reservoirs, of which 19 were exotic fish and three were translocated within the country (Table5.3).

The first fish to be introduced in a reservoir was Osphronemus gourami, a species which had already been brought into the country. Large quantities of the fish were collected from the lower reaches of the Mahaweli river and stocked in irrigation reservoirs during the early 1950s. During the same period Oreochromis mossambicus was introduced into the country. This fish was bred in government fish farms and its seed was distributed all over the country. Within a few years of its introduction, O. mossambicus was able to establish naturalized populations in a number of reservoirs, leading to steep increases in fish yields. Since then the reservoir fishery has ceased to be a subsistence activity of the local populations. It attracted the more professional migrant fishers from the coastal areas who undertook commercial fishing especially during the seasons when they were unable to fish at sea owing to adverse weather conditions. Subsequently, other African cichlids such as O. niloticus and Tilapia rendalli were also introduced; O. niloticus has performed very well. Over the years, attempts were also made to introduce other exotic species such as Indian major carp, Chinese carp and common carp. However, carp did not succeed in forming breeding populations in reservoirs and therefore did not contribute substantially to the catch.

The success of tilapia had a socio-economic impact in rural areas. The villagers accepted this alien species as food and reservoir fishing became a professional activity. A new trade network has gradually developed for procurement, transport and marketing of fish, with a corresponding growth in the trade of fishing equipment and implements. The Department of Fisheries popularized reservoir fishing by supporting the fishers through stocking the reservoirs and subsidizing fishing crafts and gear. It also provided training to fishers on various fishing techniques and fish curing. Owing to these factors, the inland fisheries sector performed well from the 1950s to the 1980s both in terms of production and creation of awareness. The contribution of the inland subsector (mainly reservoir fisheries) to the country's overall fish production was 20% by the end of the 1980s. One of the most significant benefits derived from this development is the increase in fish consumption in the hinterland and the concomitant improvement in the protein-intake levels.

Licences for gillnets are issued by the Fisheries Department. Normally the fishers are organized into cooperatives and assisted by the government through stocking and subsidizing of fishing crafts. The government offers a subsidy of up to 90% for the purchase of fishing boats. Up until 1990 the total number of boats distributed within the scheme was 4519. The mesh size limit is fixed by the Ministry of Fisheries at 7.5 cm in order to prevent the capture of young tilapia. There is also a limit determined by the department on total fishing effort. The monitoring of fisheries and enforcement of restrictions imposed by the ministry are inadequate and as a result there is no effective control over fishing activities. There are reports of uncontrolled growth of fishing units in many reservoirs (Amarasinghe, 1991). One of the reasons for the undue increase in fishing effort is the demographic pressure from refugees. Conflict between the local fishers, who are the members of the cooperative societies, and the "outsiders" (who are settlers from disturbed areas) are common. The use of trammel nets is popular in many reservoirs. The cooperative societies are not very effective in Sri Lanka and there is an active presence of intermediaries at all levels.

In 1990 the government abruptly announced the withdrawal of all support and patronage for inland fisheries and aquaculture, causing a major setback to the inland fisheries development process. Despite the shift in policy, no restrictions were imposed on the private sectors and NGOs. The government agencies also continued their extension activities and the organization of cooperative societies. A disturbing consequence of the new policy was the discontinuance of stocking in reservoirs. Under the changed setup, government nurseries were leased to private operators who became responsible for production and stocking of the water bodies. This measure did not achieve the desired goal and many water bodies remain understocked. The government has recently reversed its policy and decided to resume support to inland fisheries. However, it will be some time before discontinued activities are reorganized.

On the basis of their size and fishery management norms, the reservoirs of Sri Lanka can be conveniently grouped under three broad categories, i.e. the large and medium reservoirs (capture fisheries), small reservoirs (culture-based fisheries) and seasonal tanks (culture fisheries), each deserving separate consideration. The size limits of the three categories are approximately more than 1000 ha, 80 to 1000 ha, and less than 80 ha, respectively.

While most of the large reservoirs in Sri Lanka are recently constructed storages for hydroelectric power generation, the medium ones are ancient. Medium and large reservoirs account for most of the inland fish in the country. In both cases the fishery is managed according to capture fishery norms, depending on the self-propagating stocks of the introduced African cichlids, Oreochromis mossambicus or O. niloticus. The management measures in these reservoirs are limited to the prescription of minimum mesh size for gillnets and the establishment of the fishing effort to its optimum level. Fish yields from 19 selected reservoirs in the Anuradhapura, Ampari, Hambantota, Moneragala, Polonnaruwa and Ratnapura districts are known to vary from 40 to 650 kg/ha with a mean of 244 kg/ha (Table 5.4).

The total annual catch from reservoirs, estimated at 27000 to 30000t, in a yield range of 270 to 300 kg/ha with an average of 300 kg/ha (Pet, 1995), is quite high for tropical lacustrine fisheries. The high yields of Sri Lankan reservoirs can be linked with three main features of the reservoirs as follows: (i) water rich in ionic concentration as evidenced by its high specific conductivity; (ii) low mean depth; and, (iii) the presence of a species successful in utilizing the niches and propagating itself.

A comparison of the situation with other Asian countries reveals that in many cases such high productivity is lacking despite the first two conditions being met. Consequently, it would seem that the success obtained in Sri Lanka is directly related to the ability of tilapia populations to establish themselves in reservoirs and utilize the main food supply of the ecosystem. Fernando and Holcik (1982) indicated the importance of species to the success of reservoir fisheries, especially in the context of Sri Lanka, and classified inland fish into two categories, the riverine and the lacustrine. They claimed that few species which evolved as lacustrine species succeeded in establishing themselves in the reservoirs and forming a sustainable fishery. The authors classified clupeids as lacustrine species, citing the example of kapenta (Limnothrissa miodon) in Lake Victoria. They are, however, not explicit about the status of tilapia, although they suggest that tilapia is a lacustrine species. The evolution and existence of specialized lacustrine species for freshwaters need further elucidation.

A survey of the experiences of many countries, especially India, reveals that the major commercial species of the parent rivers, usually the cyprinids, require more exacting breeding conditions. This, coupled with their food preferences, leads to their failure to find a foothold in the newly created reservoir environment. On the other hand, minor cyprinids, some catfish and clupeids, with their flexible breeding and feeding habits, adapt well to the new habitat and propagate. Thus, it is the adaptability of some of the riverine species to the new circumstances that facilitates their success in reservoirs. Tilapia basically belong to the rivers of Africa and it is its ability to breed in the reservoir environment that facilitates its autostocking.

Nevertheless, the positive role of tilapia in augmenting the fish production in Sri Lanka is undeniable. In Parakrama Samudra, the annual fish production of a 2662 ha reservoir, which was 10t before the introduction of tilapia, increased to 1000t in the 1970s when 80% of the catch consisted of Oreochromis mossambicus. With a similar rate of increase in many other reservoirs, the national inland fish production increased from 3000t in the beginning of the 1950s to 40000t in the late 1980s. Subsequently, O. niloticus and Tilapia rendalli were introduced in some of the reservoirs, where the first species performed very well.

The performance of Oreochromis mossambicus in the Sri Lankan reservoirs is remarkable. The fish was also introduced in India in 1952, the same year in which it was brought into Sri Lanka. Subsequently, it was stocked extensively in reservoirs of all types in the southern Indian States of Tamil Nadu and Kerala. However, the fish could neither establish itself successively in the reservoir, as it did in Sri Lanka, nor did it contribute to any substantial hike in production. Many theories have been postulated to explain the remarkable performance of tilapia in Sri Lanka, attributing it, for example, to the shallow margins which facilitate nest building activities and frequent level fluctuations. But it is interesting to note that similar situations in India failed to produce high production rates.

Limnological data of 21 reservoirs (De Silva, 1988) indicate that the low mean depth and high specific conductivity (93 to 847 micro mhos) of Sri Lankan reservoirs are conducive to high primary production. High chlorophyll concentration and phytoplankton counts also substantiate this notion. In the absence of a rich indigenous fish fauna, the tilapia takes full advantage of this food resource. The two important indigenous cyprinids are Labeo dussumeri and Puntius sarana. Tilapia, which breeds readily in the reservoir, has an advantage over these two carp which are potamodromous and dependent on the availability of suitable breeding grounds for their propagation.

Indian reservoirs have a rich fish fauna comprising a number of carp minnows, clupeids and a variety of predators including catfish, which readily take advantage of the recruitment failure of major carp (either stocked or present in the parent river) and proliferate. The minnows become forage for the small predators which in turn sustain larger predators such as Channa marulius, C.striatus, Mystus aor, M. seenghala , M. cavasius, M. punctatus, Wallago attu, Silondia silonia, Notopterus chitala, N. notopterus, Clarias batrachus and Heteropneustes fossilis. This long grazing food chain involves tremendous energy loss leading to poor productivity. There are very few predators in Sri Lankan reservoirs and the predatory catfish, especially the freshwater shark (Wallgo attu), are very few or absent in many reservoirs. Low predator pressure is a major factor in the success of tilapia. Moreover, in India the tilapia stocked in the reservoirs had to compete with a number of indigenous species for food and breeding space. This explains the difference in performance of the species in the two countries.

As in the case of all the other Asian countries, capture fisheries management in Sri Lanka is either absent or very inadequate. The reservoirs in Sri Lanka, though producing high fish yield, can be considered unmanaged, since aside from the initial stocking with O. mossambicus (and later with O. niloticus and T. rendalli), there was very little effort on the part of the government to manage the open waters. The stocked fish propagated and supported good fisheries in some reservoirs. There are a number of statutory restrictions on the use of gillnets such as mesh size regulation, number of fishing units, etc., but their enforcement is very poor. This problem seems to be common to all Asian countries.

Trends of production loss owing to improper gear selection and high fishing pressure because of overexploitation have been indicated by various authors. Persistence of this situation can cause undesirable consequences leading to a decline in yield. Some of the existing regulations, such as the minimum mesh size of 7.5 cm, need revision. Amarasinghe (1988) and De Silva and Sirisena (1987), on the basis of gillnet selectivity studies on O.mossambicus in a number of reservoirs, concluded that the mesh size suitable for removing the fish at a mean size acceptable to consumers (24 cm) was 10 cm. In Parakrama Samudra, where O. mossambicus account for 80% of the catch, the average size has fallen below 20 cm, leading to overexploitation and poor marketability. De Silva (1991) estimated the maximum sustainable yield (MSY) of Sri Lankan reservoirs as a whole at 256 kg/ha/yr and optimum fishing effort unit area at 3.2 boats/km², which was below the present average yield of 283-307 kg/ha and the actual fishing intensity is 3.1boats/m². Amarasinghe (1991) found that when the reservoirs were considered separately, most of them appeared to yield fish at suboptimal levels, but when all the reservoirs were considered as a single unit, there seemed to be overexploitation. These trends were all the more evident in reservoirs with reduced catch per unit of effort (CPUE) and reduced landing size of O. mossambicus.

Many earlier research workers appeared not to have been concerned by the stunting of tilapia in Sri Lankan waters, mainly because it did not grow stunted during the last 35 years of its existence in the reservoirs (De Silva, 1988). However, recent observations suggest the need for more caution. In Parakrama Samudra the landing size of O. mossambicus is decreasing every year owing to a lack of enforcement of mesh size regulations and the uncontrolled increase of fishing effort. Pet (1995) has found that the size at maturity of O. mossambicus has decreased from 16 to 13 cm during recent years, and a danger of growth overfishing was clearly present. Tilapia, at its present landing size in the reservoir, is still marketable albeit below the consumer preference level. However, if the decline in size continues it may affect the marketability of fish and cause the fisheries as a whole to malfunction. Thus, there is an urgent need to implement the existing regulations in fishing activities.

O. niloticus is normally believed to be immune to stunting. However, this species hybridizes freely with O. mossambicus and the hybrids are known to grow stunted.

At present, most of the attention is focused on tilapia but there is room for the exploitation of more fish species. In Sri Lanka some attempts have been made to estimate the yield potential of fish populations other than tilapia. The minor cyprinids, Amblypharyngodon melettinus, Barbus dorsalis, B. chola, B. filamentosus and Rasbora daniconius, do not presently figure in commercial fishery as they are smaller than the minimum mesh sizes prescribed. In many reservoirs there is an active fishery of minor cyprinids accounting for more than 50% of the total catch. However, they are not reported as they are caught illegally in the small-meshed nets.

De Silva and Sirisena (1989) estimated the yield potential of minor cyprinids as 35 to 2191kg/ha/yr in reservoirs, depending on their hydrological and hydrobiological characteristics, with a mean of 907 kg/ha/yr. There is a case for introducing a subsidiary gillnet fishery in the major perennial reservoirs, employing gillnets of 15, 25, or 30 mm stretched mesh, without unduly affecting the main cichlid fisheries. This, however, is based on the premise that the small-meshed nets hung in the middle region do not catch small tilapia. Some ambitious estimates (Amarasinghe, 1989) suggest that the reservoirs could improve their total catch tenfold, which would result in an expected total annual national inland production increase of 90700t, on account of this supplementary fishery (De Silva and Sirisena, 1989). The above estimates, though perhaps overly generous, highlight the importance of these species which have hitherto not been considered seriously.

The importance of smaller clupeids in the reservoir capture fisheries has been emphasized by Fernando and Holcik (1982). Ubolratna in Thailand and Lake Kariba in Africa attest to the role of clupeids in enhancing the fish yield in reservoirs. In Sri Lanka, the importance of Ehirava fluviatilis is not properly recognized. Preliminary studies (Newrkla and Duncan, 1984) suggested that they could be as important as cichlids in their biomass production.

The small reservoirs of Sri Lanka, including tanks, village tanks and seasonal tanks, are ideally suited for promoting culture-based fisheries. Unlike the culture fisheries of the larger water bodies, which depend mainly on self-propagating populations, these water bodies require that emphasis be on stocking and recapture. Indian major carp, especially rohu, is best suited for this stocking and recapture system by virtue of its fast growth and increase in biomass. In 1990 the government prepared a master plan for stocking the seasonal water bodies.

Initially, 340 small reservoirs were selected from 24 districts for stocking with Indian major carp (Table 5.5). This was in addition to the 233 seasonal tanks in 14 districts. However, the scheme was not put into action owing to the policy changes at the highest level of government and cessation of all state-sponsored fishery development plans. Consequently, the two government-run nurseries in the country, Udawalawe and Dambulla farms, were handed over to the private operators. Recently, the government has reversed its policy and decided to resume state support of fish culture activities. Fish breeding activities in the two farms are expected to begin soon.

Most of the detailed scientific investigations on fish population dynamics have been conducted on the capture fisheries of the large reservoirs such as Parakrama Samudra and Udawalawe. These studies were limited to the tilapia species. Yield predictive models involving Indian major carp are conspicuously absent. Various options involving different species, stocking rate, mesh selection and fishing effort are to be based on population dynamics in order to derive the best management options. Since small cyprinids are an important component of fisheries in some of the small reservoirs, the validity of the models involving single species is to be reconsidered with respect to these reservoirs. In the 220-ha Tissawewa reservoir, Pet (1995) demonstrated that a major part of the biomass (54%) was constituted by the minor cyprinid (Amblypharyngodon melattinus) which lived directly on phytoplankton and detritus.

In response to the new policy shift by the government in favour of supporting inland fisheries activities, the Ministry of Fisheries drew up a strategy for inland fisheries development in 1995. The highlights of the new policy are as follows.

This new division has been established and its major functions are seed production, stocking of water bodies, training of personnel in aquacultural practices, development of adaptive research, creation of awareness in environmental aspects and transfer of technology to the private sector.

The state will assist and provide incentives to the private sector including the purchase of fish seed from private farmers. The government will also promote the participation of rural communities in seed production activities to ensure self-employment and additional income to the rural people. As a part of the new strategy, the government has already taken steps to make the aquaculture centres at Udawalawe and Dambulla functional. Indian major carp will be bred in these stations and the fry of these species will be handed over to the private sector for further rearing. The estate management in the country will be encouraged to produce fingerlings in the estate tanks through active participation of estate workers. The necessary inputs such as fish fry, training facilities and financial assistance will be provided by the state. Part of the plan is also to encourage the NGOs and agricultural farmers to produce fish fingerlings on a small scale and establish community-based fish seed production centres. All necessary assistance will be provided by the government.

Stocking of water bodies

Fish seed required for the stocking programmes will be purchased from the farmers. Participation of fishers cooperatives in the production of seed in the major tanks will be encouraged.

Considering the common property nature of the reservoirs, active individual participation will be ensured by cooperative societies which will be organized where they do not already exist. Similarly, in the case of seasonal tanks, appropriate organizations will be formed to involve the participation of local inhabitants in resource management.

Appropriate incentives such as tax holidays, duty free imports and proportionate reduction in lease rentals, are envisaged for the promotion of aquaculture ventures.

The three main components of the fisheries development strategy are:

1. Conservation of the tilapia capture fishery in large reservoirs. It has been suggested that the fishery is near MSY level and there is no room for substantial increase from this type of fishery. The main emphasis should be on strict adherence to mesh regulations and fishing effort controls.

2. The development of the subsidiary fisheries of minor cyprinids and clupeids which have the potential to increase productivity substantially.

3. The development of culture-based fisheries of Indian and/or Chinese carp, based on the population dynamics models already developed or to be developed by the scientists of Sri Lanka.