India is a South Asian country situated between the Himalayas in the north and the Indian Ocean in the south and flanked on either side by Pakistan and Burma. India is a federal republic made up of 25 states and covers a total area of 3287728 km². One-half of India's national territory lies above the Tropic of Cancer and the rest is within the tropics. India's population of over 900 million is the second largest in the world and the population density is also very high with more than 70% of the people living in rural areas. The economy is agriculture-based. Although rich in natural resources and self-sufficient in food grains, India has many unresolved problems relating to poverty, unemployment and malnutrition.

India is a land of diversity. The climate ranges from tropical heat in the south to temperate in the north. The landscape includes towering mountains, extensive alluvial plains, riverine wetlands, plateaus, deserts, coastal plains and deltas. The main soil types are alluvial, deep and medium black, red and yellow, laterite, saline and desert, and forest and hill. Almost all conceivable forms of vegetation, including tropical evergreen, littoral and swamp, tropical moist deciduous, tropical thorn, montane subtropical, Himalayan and Alpine are present in various parts of the country. The major physiographic divisions are the Himalayas, the Indo-Gangetic plains, the Vindhyas, the Satpuras, the Western Ghats, the Eastern Ghats, coastal plains, deltas and the riverine wetlands.

With a long 8129-km coastline, 0.5 million km² of continental shelf and 2.02 million km² of exclusive economic zone (EEZ), India is a major marine fish producer ranking seventh in the world. The warm, fertile inshore waters of India are among the most productive fishing grounds in the world, yielding shrimps, sardines, mackerels, Bombay ducks, carangids, croakers, soles, and a variety of other marine fish. Marine products are a major component of India's overseas trade. However, inland fisheries resources are equally rich and varied.

Fish and fisheries form important economic activities in India. The total fish production in the country increased sixfold from 0.75 milliont in 1950-51 to 4.95 in 1994-95. During the same period, inland fish production increased from 0.22 milliont to 2.09 milliont registering a near tenfold increase (Table 3.1). Over the last four and one-half decades, the annual growth rate in inland fisheries was 11% and that of the marine segment was 9.0%. The contribution from inland fisheries to national fish production has been consistently increasing. While in the 1950s it supplied less than one-half the volume of marine catch, it now contributes almost an equal quantity. Despite being a marine-fish producer traditionally, India is one of the few countries in the world which has substantially exploited its inland fisheries potential.

India's inland fisheries resources are as diverse as they are plentiful, comprising rivers, floodplains, estuaries, mangroves, estuarine impoundments, lagoons, upland lakes, reservoirs and ponds (Table 3.2). In India, inland fisheries is classified as follows: freshwater aquaculture, including the pond culture of carp; brackishwater aquaculture, involving mostly shrimp culture; and capture fisheries in rivers, estuaries, lakes, reservoirs, etc.

The country's major and minor rivers along with their tributaries, minor streams, creeks and all other microlotic systems have an estimated combined length of 45000 km. These along with the numerous man-made canals have a combined length of 0.17 million km. The Ganga river system and its tributaries have a combined length of 12500 km and the Brahmaputra is 4023 km long. The peninsular rivers, Mahanadi, Godavari, Krishna and Cauvery, cover 6437 km while the west-flowing Narmada and Tapti of Western Ghats have a combined length of 3380 km. The snow- and rain-fed Himalayan rivers are characterized by complicated flood regimes, a heavy silt load, marked seasonal variability in volume, course shifting and heavy bank erosion. In contrast, the rain-fed peninsular rivers are torrential and have well-defined stable courses, a lighter silt load and, except in the deltas, rocky terrain.

The catch from rivers does not contribute significantly to the total inland fish production in terms of volume, although a large number of traditional, artisanal fishers make a living on it. A substantial part of the Ganga and its tributaries flow through the Indo-Gangetic plains providing the richest source of freshwater capture fisheries in India. The Ganga is also the original habitat of the prized Indian major carp and a major source of riverine carp spawn, which meets 30% of the carp seed requirements of the aquaculture sector.

Commercial fishery in the upland waters of the Ganga system is virtually non-existent because of exploitation problems, low quantity of small-sized commercial species, inaccessible terrain and poor communication links. The upper reaches of Brahmaputra cross through inaccessible terrain so fishing activity is limited to the middle and lower stretches. Similarly, the headwaters of peninsular rivers pass mainly through rapids, gorges and terrains where fishing activity is limited. Commercial fishing is largely restricted to the middle and lower stretches of these rivers. For centuries, all of the river systems of the country have provided a means of livelihood to thousands of fishers. The riverine setting, however, is a complex mix of artisanal, subsistence and traditional fisheries with a highly dispersed and unorganized marketing system which frustrates all attempts to collect regular data on fish yield. Without a firm database on production trends, one has to depend on information collected by the Central Inland Fisheries Research Institute in the 1960s from selected stretches of the rivers Ganga, Brahmaputra, Narmada, Tapti, Godavari and Krishna. Fish yield from these rivers ranged from 0.64 to 1.64t per km, with an average of 1t per km. The number of fishers per km of river stretch varied between 3.2 in Narmada and 7.8 in Ganga, the average being 6.5. It is estimated that 190000 fishers are presently engaged in riverine fisheries and the average catch per fisher amounts to a meagre 150 kg/year.

Over the last few decades, marked alterations in the riverine systems have been detrimental to the fish stocks. The once remunerative capture fisheries no longer exists. Fish yield in middle stretches of the Ganga, which was 50.3 kg/ha/yr during the 1960s, declined to 20.0 kg/ha/yr in 1972 and 6.5 kg/ha in the mid 1980s. The average yield of Indian major carp declined from 13.3 to 4.6 kg during the same period. Fishing intensity along river courses varies from stretch to stretch owing to various factors such as:

• seasonality of riverine fishing activity;
• inconsistent catch composition;
• conflicting multiple uses of river water;
• aesthetic objections owing to nutrient loading and pollution;
• lack of understanding of the fluvial system and unreliable database;
• fragmentary and outmoded conservation measures;
• lack of enforcement machinery;
• inadequacy of infrastructure and supporting services;
• defective marketing and distribution systems;
• demand according to availability, affordability and palatability; and
• socio-economic and socio-cultural determinants.

India has more than 200000ha of floodplain lakes (locally called mans, chaurs, beels, jheels, and pats), especially in the States of Assam, Bihar, and West Bengal (Table 3.3). These take the form of typical oxbow lakes, sloughs, meander scroll depressions, backswamps, residual channels or tectonic depressions, but it is very often difficult to establish their identity because of natural and man-made modifications. The evolution of river beds and the effects of extensive flood control and irrigation works have reduced the fish production levels of many of these lakes through siltation, habitat destruction, heavy weed infestation and isolation from seasonal floods and natural sources of fish stock. Heavy weed infestation, eutrophication and accumulation of bottom muck are the greatest problems, further aggravated by the indiscriminate killing of brood fish during recruitment phase.

The current yield of floodplain wetlands is not precisely known. Nonetheless, important scientific studies have been made on the limnology and productivity of these ecosystems and a production potential ranging from 1000 to 2000 kg/ha/yr has been estimated. They also provide the ideal location for pen and cage culture operations. Pen culture of major carp has indicated a production potential of up to 4 t/ha in six months according to the results obtained in a man in the Gandak basin. The rearing of Clarias batrachus and Heteropneustes fossilis in a weed-choked Assam beel, over a period of 90 days in 2 km² cages, obtained production results ranging from 0.190 kg to 4.8 kg.

The various estuarine systems of India have a combined area of 2.7 millionha with an estimated yield of 45-75 kg/ha. The Hooghly Matlah estuary covering 234000ha is the largest and most important from a fisheries point of view. This is connected with the Sunderbans, the only major mangroves in India which cover an area of 365000 ha. The three largest brackishwater lagoons of the country, the Chilka, Pulicat and Vembanad, have a combined area of 190000 ha. The upland lakes, mostly situated in the sub-Himalayan ranges, do not make a large contribution in terms of fish production, though they harbour a variety of native ichthyofauna.

Ponds are among the most important inland fisheries resources of India, in terms of both surface area and the volume of fish produced. The country is believed to have a surface area of more than 2.21 millionha that can be developed for freshwater aquaculture, only a fraction of which is currently utilized. About 0.4 millionha are being supervised by government-sponsored aquaculture promotion agencies. Carp culture, traditionally restricted to the States of West Bengal, Bihar and Orissa, has recently spread to almost all the states of the country. Private fish farmers of the southern state of Andhra Pradesh have only recently engaged in carp culture and have made great progress. The growth of brackishwater aquaculture has been most successful in the maritime states. Focusing on export-oriented shrimp culture, industrial giants of the corporate and multinational sectors are also involved.

A more detailed account of reservoir fisheries and freshwater aquaculture is given in the following paragraphs.

In India, reservoirs are considered the prime resource as regards capture fisheries and extensive aquaculture. The rivers, estuaries and other natural water bodies are threatened by increasing environmental degradation. With the great emphasis on conservation, there is little scope for a substantial increase in yields. The reservoirs, which already cover 3 millionha of surface area, are a growing resource with enormous potential for yield augmentation. The available estimates made by various agencies are conflicting and inaccurate. The most recent study (Sugunan, 1995) estimated the combined surface area of all reservoirs, irrespective of size, as 3135366 ha. Enumeration of the medium and large reservoirs in India is a relatively easy task as there are fewer of them and the details are readily available from the irrigation, power and public works authorities. However, compilation of data on small reservoirs is a tedious task as they are ubiquitous and too numerous to count.

There are number of ambiguities in the definition of reservoir and nomenclature and classification adopted by some of the states. The word "tank", for example, is loosely defined and often used in common parlance to refer to some of the small irrigation reservoirs. Thus, a large number of small man-made lakes are called tanks in Tamil Nadu, Andhra Pradesh and Karnataka and are thereby excluded from reservoir estimates. Tanks in Andhra Pradesh are further classified as perennial and long seasonal. In the Srikakulam, East Godavari and Krishna districts, 1804 of the 4604 perennial tanks, with an average size less than 10 ha, are not considered reservoirs. The remaining 2800 tanks, which cover a total area of 177749 ha, are regarded as reservoirs. In Tamil Nadu, the tanks are further classified as short seasonal and long seasonal. The latter, also known as major irrigation tanks, have an average size of 34ha and retain water for 9 to 12 months a year. Major irrigation tanks of Chengalpattu MGR and Salem districts are larger, with an average area of 222 and 156ha respectively. A total of 8837 major irrigation tanks of Tamil Nadu, with a combined water surface area of 300278 ha, have been classified as small reservoirs. Similarly, 4605 large perennial water bodies in Karnataka listed as major irrigation tanks are considered reservoirs.

According to the records of the Government of India, reservoirs are generally classified as small (<1000 ha), medium (1000 to 5000 ha) and large (>5000 ha). All man-made impoundments, created by erecting a dam of any description in order to obstruct the surface flow of a river, stream or any water course, have been classified as reservoirs. However, water bodies of less than 10ha in area, being too small to be considered lakes, are excluded.

For purposes of clarity, the anomalies in nomenclature, especially as regards small reservoirs, can be ignored by simply classifying large (above 10 ha) irrigation tanks as reservoirs. According to this system of grouping, India has 19134 small reservoirs with a total water surface area of 1485557 ha. Similarly, 180 medium and 56 large reservoirs of the country have an area of 527541 and 1140268ha respectively. Thus, the country has 19370 reservoirs covering 3153366ha (Table 3.4).

The State of Tamil Nadu has the greatest surface area of small reservoirs in the country (315941ha), followed by Karnataka (228657 ha) and Utlar Pradesh (218 651 ha). Medium reservoirs constitute less than 1% of the total number of reservoirs and 17% of the total surface area. Andhra Pradesh, Rajasthan, and Gujarat have more medium reservoirs than Madhya Pradesh, though the water surface area in these states is much less. Karnataka has a preponderance in number (12) of large reservoirs. Nevertheless, the seven large reservoirs in Andhra Pradesh are much larger and have a total surface area of 190151 ha. The concentration of reservoirs in the peninsular states, i.e. Tamil Nadu, Karnataka, Andhra Pradesh, Kerala, Orissa and Maharashtra, is illustrated in Figures 6 to 9. More than 56% of the total reservoir area in the country is found in these six states. Of the 19134 small reservoirs, 17989 (94%) are located there, contributing 63% of the total water surface area. Similarly, 34% of the medium reservoirs are distributed in these states.

A reliable estimate of fish production from Indian reservoirs is not available. Compared with the impressive volume of data on limnological aspects of reservoirs collected by individual researchers and various institutions, the estimates of fish catch remain grossly inadequate. Furthermore, the production data available are at times inaccurate and unreliable. Following are the main reasons for this deficiency:

• complications in the collection of data in some states because of the multiplicity of agencies owning fishing rights;
• widely dispersed and unorganized market channels, mostly controlled by unauthorized money lenders;
• an ineffective cooperative setup;
• diverse licensing/royalty/crop sharing systems practised by different state governments; and
• inadequate and poorly trained workforce at the disposal of the states to collect catch data by following statistically sound sampling procedures.

Fish production data from 291 small reservoirs, based on the records available from various state governments, are presented in Table 3.5. Fish production figures of Andhra Pradesh as given by the State Department of Fisheries are the highest in the group, followed by Kerala, Tamil Nadu, Madhya Pradesh, Rajasthan and others ranging from 3.91 kg/ha (Bihar) to 188.0 kg/ha (Andhra Pradesh). On average, the 291 small reservoirs of India yield fish at the rate of 49.9 kg/ha. The yields from medium and large reservoirs in the various states are shown in Table 3.6. The average estimated yields of small, medium and large reservoirs are 49.9, 12.30 and 11.43kg/ha respectively. By applying the average yield of small reservoirs to their total surface area of 1485557 ha, their current production rate can be estimated at 74129 t. In the same way the production of medium and large reservoirs is estimated at 6488 and 13033t respectively (Table 3.7).

3.4.3 Fisheries management practices

Fisheries are under individual state control in India and there is a great deal of variation ranging from outright auctioning to almost free-fishing in the management practices followed by different states. Cooperative societies and the state-level fisheries development corporations are also involved in fishing and marketing operations. The nature of their involvement and their role in fishery and market interventions often varies from one reservoir to another within the same state. Reservoirs in India, with very few exceptions, are public water bodies owned by government departments such as those responsible for irrigation and power generation. In many cases the management of their fisheries is handed over to the fisheries departments.

The fishers are required to obtain a licence from the local office of the state fisheries department. The local assistant director or deputy director of the department of fisheries determines the number of fishers and fishing units that can be operated in a reservoir. Since the mesh regulations and the decisions regarding closed seasons are policy matters, they are determined by the directorate at the state capital. The public sector companies, floated by many of the states and styled as fisheries development corporations, have not functioned effectively and showed losses. In some states such as Madhya Pradesh and Himachal Pradesh, the fisheries development corporations are responsible for managing reservoir fisheries. These corporations act as overseeing bodies for the numerous cooperative societies that are working at the reservoirs, and they undertake marketing functions to ensure that the fishers get the right price for their catch. However, their market interventions have not markedly benefitted the fishers.

Reservoir fishers are among the weakest groups in Indian society and are heavily assisted by the government. Normally, the department of fisheries stocks the reservoirs free of charge and offers a number of loans and subsidies to the fishers for the procurement of nets and boats. The amount of subsidies and the nature and terms of the loans varies from state to state. The states are generally found to be quite liberal so that fishing is made to be virtually free for all. This is one of the reasons why accurate statistics are impossible to obtain.

In some states the departments owning the reservoirs have not handed over the fishing rights to the fisheries department, in which cases they directly manage the fisheries. Sometimes the reservoirs are owned by forest and wildlife or tourism departments.

In Tamil Nadu, the long seasonal tanks (irrigation impoundments retaining water for more than eight months) and the short seasonal tanks (retaining water for four to six months) are under the control of panchayats (village administration) and are leased out to local cooperative societies. The fisheries department does not keep any record of the fish catch or revenue obtained from these water bodies. The fishing in medium and large reservoirs in the state is done, with very few exceptions, directly by the department. Similarly, in the States of Karnataka and Maharashtra, the zilla parishads (the district administration) controls irrigation tanks of up to 40ha in size.

The commercial exploitation systems followed in different states can be summarized as:

• departmental fishing;
• lease by auction;
• issuance of licences for fishing; and
• fishing on royalty (crop sharing).

Direct departmental fishing, being uneconomical, is practised in very few reservoirs. In some reservoirs such as Hirakud, Nagarjunasagar, and DVC, this practice is partially resorted to for experimental and exploratory purposes. In most cases, the department exerts its control over the exploitation by acting as a marketing link and controlling the fishing effort. In Rajasthan, Madhya Pradesh and Uttar Pradesh, the small reservoirs are mostly auctioned on an yearly basis. In a number of large reservoirs, free unlimited licences are issued to fishers. This virtually free-for-all system has been found to be detrimental to the ecosystem and fishers in Nagarjunasagar, Yerrakalava and a large number of reservoirs in Andhra Pradesh.

Crop sharing is a very popular mode of exploitation in Tamil Nadu, where the fishers are provided with all fishing implements, in exchange for royalty (sometimes up to 50% of the catch) paid to the government. A short-term leasing policy, adopted for small reservoirs by some states, is not in accordance with the norms of sustainable development.

In many cases, licences are issued either free of cost or for a very small nominal fee for political reasons or because of other considerations. Reservoir fisheries development is often linked with many social welfare and relief activities under the village and district administration. In such cases, immediate social goals are pursued overlooking long-term interests and tenets of capture fishery norms.

A socio-economic evaluation of reservoir fisheries is very tedious owing to the multiplicity of agencies involved in reservoir management. Reservoir fisheries are basically developed along the lines of capture fisheries, following the common property norm. As in the case of rivers, lakes and seas, biological wealth is considered nature's endowment and state intervention in developmental activities benefits the poor fishers. The investment made in developing reservoir fisheries should be evaluated in the light of the social benefits in the form of:

• rehabilitation of displaced population;
• improvement of the living conditions of fishers; and
• provision of employment opportunities.

In aquaculture, it is estimated that 77.23% of the price paid by the consumer is received by the producer (Paul, 1990). In reservoir fisheries, however, a large portion of the price is taken by wholesalers and other market intermediaries. A study of seven reservoirs (Paul and Sugunan, 1990) over a six-year period revealed the most significant factors determining the remunerativeness of fishing in reservoirs. Reservoir fisheries is a sector in which the chief input is labour. Even if the costs of stocking and other developmental measures as well as a marginal depreciation of crafts and gear are taken into account, this area does not call for heavy investment as in the case of pond culture.

There is a need to join the twin objectives of conservation and yield optimization in reservoir fisheries management. While the profit-oriented fishers and fish merchants strive to increase production, it is the responsibility of the state to ensure that economic expediency of development does not interfere with ecological considerations. Virtually free fishing, as followed in Andhra Pradesh, is counter-productive in terms of conservation and yield optimization. Although there are reasonable possibilities of linking reservoir fisheries development to poverty alleviation programmes, the progress made in this direction so far is not very encouraging. In the majority of Indian reservoirs the chances of creating additional employment are minimal. However, many reservoirs have more fishers than needed who can be diverted to those reservoirs able to absorb more human resources without eroding the income level of the their own fishers (Paul and Sugunan, 1990).

Stocking of reservoirs, except those leased out to private pisciculturists, is done departmentally. All of the states have nurseries to raise fish fingerlings for stocking their reservoirs. However, most of the reservoirs remain understocked because of the inability of departmental farms to produce the requisite number and species of fish seed. Scientific norms are seldom followed in selecting the species for stocking. The policies hitherto adopted in Indian reservoirs emphasize mainly stocking fingerlings of a species or a combination of species without taking into consideration density levels or ratios based on the biogenic capacity of the reservoir. Species used and the rate of stocking are often determined by availability.

The existing pathways of energy transfer in reservoirs can be an important criterion for guiding stocking requirements. Indian reservoirs, by and large, have a wide spectrum of phytoplankton comprising Cyanophyceae, Chlorophyceae, Dinophyceae and Bacillariophyceae which dominate the zooplankton such as copepods, cladocerans, rotifers and protozoan. Benthos is represented by insect larvae and nymphs, oligochaetes, nematodes and molluscs. There is a rich growth of periphyton on submerged objects. The magnitude of water level fluctuations does not favour the establishment of aquatic macrophytic communities. Significantly, many of the above niches, with the exception of insects, Cyanophyceae and molluscs, are shared by Indo-Gangetic major carp and trash fish, which emphasizes the need to control carp minnows and weed fishes.

The ecosystem-oriented management policy places due emphasis on trophic strata in terms of shared, unshared and vacant niches. Two main pathways, through which primary energy finds its way to fish flesh, are the grazing chain and the detritus chain. The contribution of both pathways to the total availability of energy needs to be assessed in order to determine the species combination most suited to the ecosystem. A large number of Indian reservoirs exhibit the detritus chain of energy transfer.

Prior to the development of carp seed production technology in India, natural spawn collected from rivers was stocked in reservoirs. Thus, the seed of Puntius spp., Cirrhinus spp. and Labeo spp. collected from the Cauvery were extensively stocked in the reservoirs of Tamil Nadu along with euryhaline species such as Chanos chanos, Etroplus suratensis and Megalops cyprinoides. Labeo fimbriatus, Cirrhinus cirrhosa, tilapia and Etroplus suratensis were the common stocking material in Kerala during the early years. Owing to the arrival of induced breeding, most of the states in India were able to raise carp seed in large quantities by the 1970s which resulted in a shift in species-mix in favour of catla (Catla catla), rohu (Labeo rohita) and mrigal (Cirrhinus mrigala). Today, reservoir fisheries in India largely centre on development of carp fisheries. The importance of their role has been demonstrated in the Gangetic as well as peninsular reservoirs.

Indian major carp, are indispensable in reservoir management, by virtue of their feeding habits and fast growth rate. However, they are ill-suited for the utilization of phytoplankton, the most dominant fraction of plankton. The remarkable ability of silver carp in efficiently converting phytoplankton into fish flesh has been demonstrated in Kulagarhi and Getalsud reservoirs, despite persisting doubts about the digestibility of Microcystis. However, the introduction of exotic fish to open waters is still a subject of controversy because of the potentially deleterious effects on indigenous populations.

Development of endemic species as stocking material has not made much progress in the country although some of them have a proven track record in ensuring an efficient energy transformation rate. The opinion of Indian experts on the subject runs along the following lines.

P. pangasius, subsisting on a molluscan diet, is a species to be considered in the detritus-based, mollusc-rich reservoirs of the country. Puntius pulchellus, a peninsular species, is a well-known macrophyte feeder and Thynnichthys sandkhol consumes Microcystis, the common alga in Indian waters. Diversification of stocking material is essential for establishing multi-species fish stock which utilizes all food niches of the ecosystem. In reservoirs, where annual drawdown is not pronounced and water level fluctuations are not steep, phytobenthos and macrovegetation develop in various degrees. Grass carp (Ctenopharyngodon idella) can be considered for such water bodies. Common carp is being stocked in many reservoirs. This torpid fish does not normally survive in the warm, deep-basin reservoirs of the south especially when these are infested with predators. However, being a prolific feeder it could establish a niche for itself in the reservoirs of the northeast, in Gobindsagar and in some of the peninsular reservoirs such as Krishnarajasagar. The fish, being a mud-stirrer, is considered to be unsuitable for waters which are already turbid.

Oreochromis mossambicus, owing to its records of rapid proliferation and consequently stunted growth in pond ecosystems, is not favoured by many fishery managers, although in Amaravathy and Malampuzha it has performed well. T. putitora, L. dero, and exotic species such as mirror carp, silver carp, grass carp Tinca tinca and Carassius carassius are sometimes promoted for high altitude reservoirs.

Based on Indian experience, the stocking programme in medium and large reservoirs can be termed successful only when the stocked fish breed in the reservoir and contribute toward autostocking. In many cases, despite persistent stocking, the transplanted species did not appear in the catch, therefore the expenses incurred in stocking were lost. There were a few cases where the resources mobilized for stocking purposes were compensated for by income generation through the recapturing of stocked fish. Sreenivasan (1984) reviewed the impact of stocking in ten reservoirs of Tamil Nadu. The stocked catla built up a naturalized population in the Mettur reservoir. Only 10 000 fingerlings were stocked between 1922 and 1935; they formed the nucleus of self-propagating stock and dominated the catch during the 1960s. Catla fisheries, however, suffered periodic setbacks because of breeding failures. The current contribution is as low as 10%. The recapture of two other stocked fish, L. rohita and L. calbasu, is reported to be adequate. However, stocking of L. fimbriatus (over 2 million), common carp (1million), L. kontius (0.4million), P. carnaticus (0.4 million), C. reba (several hundred thousand) and P. dubius (several hundred thousand) is believed to be wasteful, since they were never recaptured in any significant quantity.

In the Bhavanisagar reservoir of Tamil Nadu, the transplanted catla appeared in the catch, but could not make an impact on the catch structure. However, L. calbasu, another transplant, found the environment congenial for breeding and established itself in the reservoir, supporting a lucrative fishery over the last few decades. More than 2million common carp fingerlings, of which only a few hundred were recaptured, were stocked in the reservoir. Intensive stocking of L. fimbriatus did not have any effect on the fisheries despite the fact that the fish were native to the system.

In Sathanur, a breeding population of catla has been successfully established through stocking and this species constitutes more than 80% of the total catch. This species has also surpassed the indigenous (L. fimbriatus) by reducing its presence in the catch from 36 to 1%. It is pertinent to note that the stocking done over 12 years involving 2 million fingerlings of indigenous species, such as C. cirrhosa, L. kontius, C. reba and L. fimbriatus, could not contribute to the restoration of their fisheries, primarily because of the inability of the fish to breed and propagate themselves.

In Krishnagiri, although the increase in percentage of stocked major carp has resulted in some initial increment of their presence in the catch, the recapture was not commensurate with the stocking rate. The stocking of Gangetic carp, common carp and L. fimbriatus in Vaigai has been described as wasteful, where presently, catla contribute 20% to the catch.

In Malampuzha and Peechi, the two medium-sized reservoirs in Kerala did not register any substantial increase in yield rate despite sustained stocking with Gangetic major carp, mainly because of their failure to breed. In Malampuzha, despite a sharp increase in stocking density from 52 per/ha during the 1970s, and a corresponding increase in the percentage of major carp stocked, the yield rate remains at a low level of 5.0 kg/ha. Although catla grows to an impressive size, the contribution of major carp never exceeded 20% of the total catch. Considering the low yield rate, the quantity of major carp harvested does not commensurate with the stocking effort. Similarly, in Peechi, 90% of the fingerlings stocked are of Indian major carp, especially L. rohita, but this is not reflected in the catch and the yield rate remains at a low level of 4.5 kg/ha.

In Nagarjunasagar, Andhra Pradesh, during the 1970s, regular annual stocking of 50000 to 833000 catla, rohu and mrigal fingerlings had little impact on the catch as none of the stocked fish could breed and contribute to recruitment. Similarly, stocking as a management option has failed in the Tungabhadra reservoir situated in the same basin and Krishnarajasagar in the Cauvery basin. In Tungabhadra, the stocking rate over 17 years ranged from 1 to 11/ha, while it varied from 0.15 to 67/ha in Krishnarajasagar. In both reservoirs, the selection of species was arbitrary, the stocking density was inadequate and the stocked fishes failed to breed in the reservoir.

In Gandhisagar in Madhya Pradesh, stocked fish contributed to the catch in a sustained manner through breeding and recruitment. From the 1950s to the 1970s, because of the steady stocking of catla (2 million), rohu (1.3 million), and mrigal (1.1 million), the fish yield rose from 0.51 kg/ha to 20.33 kg/ha. Among the stocked fish, catla contribute 60 to 70%, while mrigal and rohu contribute only 1 to 20%. Mrigal, despite being a part of the indigenous ichthyofauna, is on the decline. Ravishankarsagar, in the same state, was stocked on average with 1.77 million fingerlings every year and yet the yield rate could not be raised above 8.3 kg/ha primarily because the carp were not established. The amount spent on stocking has been much higher than the cost of recaptured fish. Rihand reservoir in Uttar Pradesh managed to build up a breeding population from the initial stocking. Although the yield rate of 0.58 kg/ha is not impressive, 73 to 99% of the catch is made up of C. catla.

The success of stocked Indian major carp in the Ukai reservoir in Gujarat can also be attributed to their breeding in the reservoir. Aside from augmenting the reservoir fisheries through recruitment, young Indian major carp are also reported to escape through the outlet of the dam and contribute to stocking of downstream impoundments. In all the DVC reservoirs in Bihar, i.e. Konar, Tilaiya, Maithon and Panchet, stocking did not have any impact. Some catla and mrigal breeding activity has been reported, though the low survival rate of the spawn and frequent breeding failure, owing to erratic monsoons, prevented proper recruitment of the stocked species.

Stocking has been much more effective in improving the yield from small reservoirs than it has been in large and medium reservoirs. This is because success in the management of small reservoirs depends more on recapturing the stocked fish than on their building up a breeding population. The smaller water bodies have the advantage of easy stock monitoring and manipulation. Thus, the smaller the reservoirs, the better the chances of success in the stock and recapture process. In fact, an imaginative stocking and harvesting schedule is the main concern of fisheries management in small, shallow reservoirs. The basic principles of such a system involve:

• selection of the right species, depending on the fish food resources available in the system;
• determination of stocking density on the basis of production potential, growth and mortality rates;
• proper stocking and harvesting schedule including staggered stocking and harvesting, allowing a maximum growth period, and taking into account the critical water levels; and
• in the case of small irrigation reservoirs with open sluices, the seasonal overflow and the possibility of water levels falling too low or completely drying up are also to be taken into consideration.

The success attained in Aliyar attests the efficacy of management based on staggered stocking. The salient features of the management options adopted in Aliyar are:

• limiting stocking to Indian major carp (earlier, all indigenous, slow-growing carps were stocked);
• increasing the size at stocking to 100 mm and above;
• reducing the stocking density to 235-300/ha (earlier rates were erratic ranging between 500 and 2 500/ha);
• staggering the stocking; and
• regulating mesh size strictly and banning the catch of Indian major carp <1 kg in size.

A direct result of the above management practice was an increase in fish production from 1.67 kg/ha in 1964-65 to 194 kg/ha in 1985. Effective recapture of the stocked fish renders the stocking more remunerative in small reservoirs. Successful stocking has been reported in a number of small reservoirs in India. In Markonahalli, Karnataka, on account of stocking, the percentage of major carp has increased to 61% and the yield increased to 63 kg/ha. Yields in the Meenkara and Chulliar reservoirs in Kerala have increased from 9.96 to 107.7 kg/ha and 32.3 to 275.4 kg/ha respectively through sustained stocking. In Uttar Pradesh, Bachhra, Baghla and Gulariya reservoirs registered a steep increase in yield through improved management concentrating on stocking. An important consideration in Gulariya has been to allow a maximum growth period between stocking and final harvesting, i.e. before the levels go below the critical mark. The possible loss owing to the small size of the fish at harvest was compensated for by the amount. Bundh Beratha in Rajasthan, stocked with 100000 fingerlings a year (164/ha) yielded 94 kg/ha of fish, 80% of which were catla, rohu and mrigal (Table 3.8)

The fish faunistic spectrum of India is very rich. More than 400 species of freshwater fish, many of which are economically important, have been identified in Indian rivers. The Gangetic system alone harbours at least 265 species of fish. One hundred and twenty-six species belonging to 26 families have been identified in the Brahmaputra system. The peninsular rivers have been reported to harbour at least 76 fish species. Despite the faunistic changes associated with impoundments, Indian reservoirs harbour a rich variety of fish species. On the basis of studies conducted so far, large reservoirs, on average, harbour 60 species of fish, of which at least 40 contribute to commercial fisheries. The fast growing Indo-Gangetic carp are among the most commercially important fish. More recently, a number of exotic species have contributed substantially to commercial fisheries. Following is a broad categorization of the species:

Indian major carp: Labeo rohita, L. calbasu, L. fimbriatus, Cirrhinus mrigala, Catla catla

Mahseer: Tor tor, T. putitora, T. khudree, Acrossocheilus hexagonolepis

Minor carp, snow trout and peninsular carp: Cirrhinus cirrhosa, C. reba, Labeo kontius, L.bata, Puntius sarana, P. dubius, P. carnaticus, P. kolus, P. dobsoni, P. chagunio, Schizothorax plagiostomus, Thynnichthyes sandkhol, Osteobrama vigorsii

Large catfish: Aorichthys aor, A. seenghala, Wallago attu, Pangasius pangasius, Silonia silondia, S. childrenii

Featherback: Notopterus notopterus, N. chitala

Air-breathing catfish: Heteropneustes fossilis, Clarias batrachus

Murrel: Channa marulius, C. striatus, C. punctatus, C. gachua

Weed fish: Ambassis nama, Esomus danrica, Aspidoparia morar, Amblypharyngodon mola, Puntius sophore, P. ticto, Oxygaster bacaila, Laubuca laubuca, Barilius barila, B. bola, Osteobrama cotio, Gadusia chapra

Exotic fish: Oreochromis mossambicus, Hypophthalmichthys molitrix, Cyprinus carpio specularis, C. carpio communis, Gambusia affinis, Ctenopharyngodon idella.

Most of the catfish, featherback, air-breathing fish, murrel and weed fish are distributed countrywide, while the distribution of major carp, minor carp and mahseer varies according to the river basins. Indian major carp are the main native ichthyofauna of the rivers of the Gangetic system. These rivers also harbour Labeo bata, P. sarana, P. chagunio, and C. reba. Tor putitora, Labeo dero and snow trout (Schizothorax spp.) are the dominant riverine fish fauna of Indus system. Mahseer, especially chocolate mahseer (Acrosscheilus hexagonolepis) are also found in the streams of all the major river systems of the country. Indigenous fish of the peninsular rivers include Cirrhinus cirrhosa, C. reba, Labeo kontius, L. fimbriatus, P. dubius, P. sarana, P.carnaticus, P. kolus, P. dobsoni, T. tor, T. sandkhol and O. vigorsii.

The three types of Indian major carp have been stocked extensively in reservoirs all over the country for many decades and, in many cases, they have established themselves in reservoirs which are far away from their original habitat. Sathanur reservoir in Tamil Nadu has a naturalized population of catla which constitute 80 to 90% of the total catch. This Gangetic carp has prevailed over indigenous fish fauna including Labeo fimbriatus, which made up 36% of the catch during the mid 1960s. Similarly, introductions of silver carp in Gobindsagar, common carp in Krishnarajasagar and tilapia (Oreochromis mossambicus) in Amaravathy are examples of man-induced changes in fish communities.

The formation of reservoirs has particularly affected the following indigenous fish stocks:

• the mahseer, snow trout and Labeo dero, and L. dyocheilus of the Himalayan streams;
• the anadromous hilsa, catadromous eel, and freshwater prawn of all major river systems;
• P. sarana, T. tor mahanadicus, T. mosal, L. fimbriatus, L. calbasu, and Rhinomugil corsula of the Mahanadi river;
• P. dobsoni, P. dubius, P. carnaticus, C. cirrhosa and Labeo kontius of the Cauvery basin;
• P. kolus, P. dubius, P. sarana, P. porcellus, L. fimbriatus, L. calbasu, L. pangusia and Tor kudree of the Krishna river system; and
• the mahseer, eel and Osteobrama belangiri of the northeast (see Figure 10).

A vast majority of the 300 exotic species introduced into the country are ornamental fish remaining more or less confined to aquaria. Others have been introduced in aquaculture and open water systems with varying degrees of success. Three larvicidal fish, i.e. Lebistes reticulatus, Nothobranchus sp. and Gambusia affinis, were introduced to contain insect larvae in confined waters. Silver carp and the three varieties of common carp were brought into the country with the objectives of broadening the species spectrum in aquaculture and increasing the yields through better utilization of trophic niches. In recent years, the bighead carp (Aristichthys nobilis) and O. niloticus have been reported from the culture systems of eastern India. Having been introduced without authorization, these two species are becoming popular among the aquaculturists of the region. While the introduction of a few of the species proved to be a boon in aquaculture and optimized the yield from ponds, the accidental and deliberate introduction of some of the exotic fish into open waters has generated much debate in recent years.

Oreochromis mossambicus, Hypophthalmichthys molitrix, Ctenopharyngodon idella, Cyprinus carpio communis, C. carpio specularis and C. carpio nudus have entered the reservoir ecosystem through accidental or deliberate stocking. Among them, tilapia, silver carp and common carp could be detrimental to the fisheries in various reservoirs in the country. Cases of naturalization of Gambusia affinis in reservoirs are rare. In Markonahalli reservoir, the species has developed as a breeding population and is reported to be affecting the larval stages of commercially important fishes.

The tilapia (O. mossambicus) was first introduced into the pond ecosystem of the country in 1952 and it was shortly thereafter stocked in the reservoirs of south India. By the end of the 1960s, most of the reservoirs in Tamil Nadu and those in the Palakkad and Trissur districts of Kerala were regularly stocked with tilapia. The performance of tilapia in south Indian ponds has been discouraging mainly because of early maturity, continuous breeding, overpopulation and dwarfing of the species. According to reports, it matures at an age of 75 days when it is 6 cm in length, and breeds at monthly intervals under tropical conditions. The warm waters of the tropical reservoirs in India provide an ideal habitat for tilapia and it has secured a niche in a number of south Indian reservoirs. Concerns regarding its stunted growth have been allayed as the average size of tilapia did not decline as much as it did in ponds.

The size of tilapia caught commercially in reservoirs has been substantial, as opposed to the unmarketable size reported from the ponds. During the 1960s the average weight of tilapia from Tamil Nadu reservoirs was 1.5 kg, with a minimum weight of 500g. During the same period, tilapia weighing 2.5 kg was common in the Malampuzha reservoir in Kerala, where the average weight ranged from 1.50 to 1.75 kg. The present average weight of 0.50-0.70 kg in Malampuzha reservoir and the average of 0.68 kg in Tamil Nadu reservoir are well within the limits of market preference. However, their continuous decrease in size over the years is cause for concern as it is feared that the tilapia may become unmarketable.

O. mossambicus was introduced in India at a time when inland fisheries contributed negligibly to the total fish production in the country and at the outset of the ecosystem management. Today, a number of indigenous species are available for stocking in order to broaden the species spectrum, bridge the gaps in niche utilization and increase the yield. With the exception of very few reservoirs, tilapia-dominated fishery invariably leads to low yields. In many reservoirs such as Krishnagiri and Vaigai, production has been erratic because of the unpredictable behaviour of tilapia owing to competition with other fish. Fishery managers of India are striving to change the dominant position of tilapia wherever it occurs. Given the present situation, tilapia is not among the species preferred for stocking in Indian reservoirs.

O. niloticus has not yet been introduced to the reservoir ecosystem in India. Confined to the estuarine and freshwater wetlands of eastern India, the species has registered an impressive growth of 250 g in 6 months. Since this species is not reported to have problems of stunted growth and prolific breeding, it will probably play a more positive role than that of O. mossambicus in Indian reservoirs.

Silver carp (Hypophthalmichthys molitrix) was introduced in India in 1959 and, unlike tilapia, did not stray into many reservoirs. However, silver carp has attracted more attention from ecologists and fishery managers, generating a more animated debate. The significance of silver carp in reservoirs is mainly because of:

• its reported ability to utilize Microcystis;
• its impressive growth rate; and
• its propensity to affect indigenous species, especially Catla catla.

The idea of stocking silver carp in reservoirs was abandoned following experimental introductions in the Kulgarhi (Madhya Pradesh) and Getalsud (Bihar) reservoirs. The most impressive reported performance of silver carp was in the Gobindsagar reservoir, where, following an accidental introduction, the species formed a breeding population and brought about a phenomenal increase in fish yields. Silver carp was instrumental in incrementing the fish production from the reservoir from 160t in 1970-71 to 964t in 1992-93 (Figure 11). It is significant to note that despite its entry into a number of Indian reservoirs, by accident or otherwise, silver carp failed to become naturalized anywhere except Gobindsagar. The reservoir, with its temperate climate, is closer to the original habitat of the species and has a distinctly cold water hypolimnion owing to the discharge from Beas. In this reservoir, silver carp seems to have found a congenial habitat for growth and propagation. Although the introduction of silver carp was never cleared by the Committee of Experts constituted by the Government of India, the fish is being stocked in a number of reservoirs in the country, but so far has not had the same success in any other reservoir as it did in Gobindsagar. Therefore, fears regarding the threat of catla extinction in the Gangetic and peninsular areas of India because of silver carp are possibly groundless.

The three varieties of the Prussian strain of common carp, i.e. the scale carp (Cyprinus carpio communis), the mirror carp (C. carpio specularis) and the leather carp (C. carpio nudus), were introduced in India in 1939. During the 1950s they were stocked in several high altitude ponds and lakes. In 1957 the Chinese (Bangkok) strain of the common carp was brought into the country, primarily for aquacultural purposes, because of its warm water adaptability, easy breeding, omnivorous feeding habits, good growth and hardy nature. Like tilapia, common carp soon found its way to all types of reservoirs in the country. The relative ease with which the fish could breed in controlled conditions prompted the departmental fish farms throughout the country to produce the seed of common carp in large numbers and to stock it in the reservoirs. However, these stocking attempts lacked any ecological considerations.

Figure 11 Catch of siver carp in relation to total fish ctch in Gobindsagar reservoir

The Bangkok strain of common carp has been stocked in a large number of reservoirs in the plains while the European strain was introduced in the reservoirs of temperate zones and at high altitudes. Their performance in reservoirs is erratic despite heavy stocking. Common carp is not a suitable fish for stocking in Indian reservoirs, especially the larger ones, for a number of reasons. Being a torpid fish, its chances of survival in a predator-dominated reservoir are very poor. It is not frequently caught in passive fishing gear such as gillnets because of its slow movement and tendency to dwell at the bottom of the reservoir. It is not surprising that despite regular stocking for 13 years (involving 537000 fingerlings), not a single common carp was ever caught from Nagarjunasagar. Obviously, the stocked fish failed to survive among the marauding predators. With a few exceptions such as Krishnarajasagar, this has been the case for common carp stocked in all the deep reservoirs. A more significant reason for its disqualification is its propensity to compete with some important indigenous carp such as Cirrhinus mrigala, C. cirrhosa and C. reba with which common carp shares its food niche. In Girna and Krishnarajasagar, cases where the presence of common carp has resulted in the decline of Cirrhinus sp. were reported.

Following its introduction in some of the upland lakes of Kumaon Himalayas, the Dal Lake in Kashmir, Gobindsagar, and the reservoirs of the Northeast, the mirror carp threatened the survival of a number of native fish species.

The exotic carp, being considered for introduction in the country are the bighead carp (A. nobilis), already introduced unofficially, the mud carp (Cirrhinus molitorella), and the snail carp (Mylopharyngodon piceus). After a thorough investigation of the ecological implications of their introduction, Natarajan (1988) concluded that their introduction was an irrational step, as all of them infringe on the food niche of the economic carp species of India and the alien species were likely to cause the extinction of their native counterparts.

India has not been surpassed by many of the South East Asian countries in adopting technology for increasing fish and shrimp production, notwithstanding the fact that intensive aquaculture has only recently been undertaken in the country. The production of fish and shrimp from aquaculture has increased from about 0.2 million tonnes in 1980 to over 1.35 million tonnes in 1992 mainly because of the national effort to introduce scientific composite fish farming through a network of Fish Farmers' Development Agencies (FFDA). The establishment of a large number of fish seed hatcheries for making available the requisite quantity of quality fish seed to the farmers has also acted as a catalyst. In 1992, a total area of about 0.74millionha was used for aquaculture (see Table 3.9). India is one of the top world exporters of shrimp. There has been a phenomenal growth in export of marine products from the country during the last 30 years. The quantum of exported marine products increased from about 15700t during 1961-62 to 208 600t during 1992-93 while their value increased from Rs. 39.2 million to Rs.19085 million ($US615 million) during the same period. Nearly 60% of the quantum and about 80% of the value of marine products exported is contributed by shrimp (Table 3.10)

The formation of Fish Farmers' Development Agencies (FFDAs) is a major national scheme launched by the Ministry of Agriculture to popularize fish farming. Introduced during 1973-74 as district level organizations, the project finances the designated agencies to provide a package of financial, technical and extension support to the identified beneficiaries in order to enable them to undertake scientific composite fish farming in the village tanks and ponds. The impact of the programme was so great that, on the basis of the evaluation of the FFDA programme in 1978-79, it was extended to cover more districts in almost all the states. At present FFDAs are functioning in different parts of the country. Introduction of improved fish farming technology through FFDAs has helped to increase the average productivity of the ponds from a low 50 kg/ha/yr prior to the introduction of the FFDA scheme during 1973-74 to over 2100 kg/ha during 1992-93. To date FFDAs have applied scientific fish farming to a water area of over 0.4 millionha and trained over 0.46 million fish farmers.

The increasing demand for quality fish seed owing to the popularization of scientific fish farming in freshwater tanks and ponds has resulted in corresponding growth of fish seed production. Fish seed production which was 1048 million fry in 1979-80 increased to 14429 million fry by 1993 because of the extraordinary success of the Chinese type of eco-hatcheries and the introduction of indoor hatchery facilities in commercial fish seed farms (Table 3.11). The government has constructed 47 commercial fish seed farms and hatcheries under both the National Programme for Fish Seed Development and the Inland Fisheries Project with World Bank assistance. Moreover, a number of private entrepreneurs with over 210 carp hatcheries, particularly in West Bengal and Andhra Pradesh, have ventured into fish seed production providing these states with a surplus of fish seed.

Brackishwater aquaculture, mainly shrimp farming, involves three categories of farmers/entrepreneurs, i.e small and medium farmers, middle-level entrepreneurs and big entrepreneurs. The involvement of the private sector in shrimp farming based on middle-level technology, utilizing the resource and technological inputs available locally or elsewhere, is of very recent origin. Along India's east and west coasts, particularly those of Andhra Pradesh and Tamil Nadu, a large number of commercial integrated shrimp farming units with foreign collaboration are emerging and adopting scientific culture systems with facilities for production of shrimp seed, shrimp feed and processing. The total area which is expected to be under semi-intensive shrimp farming in these units by the year 2000 is likely to be over 10000 ha. The liberalized industrial policy and the revised customs/excise tariff announced by the government have acted as incentives for the aquaculturists and encouraged their increased involvement in aquaculture activity in the country.

An analysis of the overall performance of the brackishwater aquaculture programmes in the country reveals that progress has been hampered owing to a number of constraints such as:

• inadequacy of indigenous technology for commercial scale shrimp seed production, shrimp culture and production of cheap and acceptable supplementary pellet feed for shrimp;
• selection of sites which were later found to be unsuitable for shrimp farming;
• absence of a suitable land-leasing policy for brackishwater shrimp/fish farming;
• inadequacy of technical staff for the implementation of brackishwater aquaculture programmes;
• overdependence on the natural collection of shrimp seed for culture operations;
• lack of local availability of suitable equipment and machinery for shrimp farming, shrimp hatcheries, feed formulation and preparation, etc.;
• lack of suitable larval feed and technology for artificial propagation of commercial shrimp species.

The use of an additional 10000ha for shrimp farming has been proposed. A proposal has also been made to increase the total number of brackishwater Fish Farmers' Development Agencies from 37 to 45 and increase the level of average productivity per unit from about 600 kg/ha/yr to 3-4 t/ha/yr. The other major policies and strategies for development of shrimp aquaculture in particular within the year 2000 are as follows. Since 50 to 75% of the brackishwater land will be reserved for the weaker sectors and fishers, extensive shrimp farming will continue to play a major role in brackishwater shrimp farming by providing an appropriate package of technology and inputs. Private entrepreneurship in commercial shrimp seed and feed production and semi-intensive and intensive shrimp farming will be encouraged by providing financial and technology inputs.