I. INTRODUCTION

Sustainable aquaculture, innovation of modern technologies, enhancing livelihoods, and global food security are the long-term goals of aquaculture development (Rai et al., 2008). Nepal is a landlocked country with an abundance of freshwater water bodies having a high possibility of aquaculture. Fish is considered auspicious and symbolizes a sign of fertility, power, and prosperity in Nepal (Gurung et al., 2003). In Nepal, fish culture is the prevailing type of aquaculture and is cultured in different systems. Aquaculture is at its blooming phase with an annual growth rate of about 8-9% in Nepal (Gurung, 2016). Aquaculture and fisheries contribute about 4.29% in agriculture domestic production (AGDP) and nearly

1.34% in GDP (DoFD, 2017). National production of fish was 77,000 mt of which about 72% contributed by aquaculture and 28% from capture fisheries (Kunwar and Adhikari, 2016).

Stocking the complementary species of fishes in a pond can increase the maximum standing crop by allowing a wider range of available foods and ecological niches (Da silva et al., 2006). Polyculture is also known as multi-trophic aquaculture, co-culture, or integrated aquaculture (Bunting, 2008). One of the most widely practiced pond aquaculture systems in Central Asia is carp polyculture (Woynarovich, 2010). Literally, polyculture fits the principles of sustainable aquaculture. With advanced ecological stability and optimizing the use of available resources, this system reduces the environmental impact of the activity and increases producer profitability (McKinnon et al., 2002). In Nepal, pond fish culture contributes about 89.1% of total production from aquaculture which is mainly prevailed by the carp polyculture in earthen ponds (CFPCC, 2018).

Combination of mainly IMC (Indian Major Carps) and CMC (Chinese Major Carps) along with Common carp (Cyprinus carpio) is the most commonly used concept in carp polyculture Polyculture of carp species contributes about 70% of total aquaculture production in four countries (India, Myanmar, Nepal and Pakistan) of south Asia (FAO, 2016). Among different experiments, the addition of Nile tilapia (Oreochromis niloticus) and Sahar (Tor putitora) has been successfully proven to increase the total production and gross margin in pond aquaculture (Shrestha et al., 2011).

Nile tilapia, in comparison to other species, has many aquaculture attributes such as excellent growth rates, low dietary protein requirement, and its prolific breeding nature. Tilapia can tolerate wide ranges of environmental conditions, less susceptibility to disease, and responsive to handling and captivity.

Integrated fish farming mainly focuses on production, integrated management, and comprehensive use of aquaculture, agriculture, and livestock, with an emphasis on aquaculture. The major features of this system include recycling of by-products in which the waste of one system becomes the input of other systems and efficient utilization of space. According to Latif (1993), integrating duck farming with aquaculture is an economically viable and productive system for both farmers and commercial entrepreneurs.

The duck droppings act as excellent organic fertilizer for the fish pond which accounts for 60% of the total input cost in fish culture (Shrestha& Pandit, 2012). Duck manure is considered one of the effective nutrients for enhancing the growth of natural food (Latif, 1993). Some fish species like common carp (Cyprinus carpio) intake duck dropping directly as their feed (Biswas, 2015).

Integrating duck with fish farming has many benefits like utilization of duck droppings by the fish as natural food, space utilization, and droppings can be used as manure. If the ducks are raised in ponds 2-3% of protein in duck feeds will be reduced. Ducks act as natural aerators by their swimming and dabbling activities. Integrating duck with fish culture ensures the farmers high profit with less investment (Majhi, 2018).

II. MATERIALS AND METHODS

a) Study Site

The duck cum fish integrated farming was practiced in the Aquaculture farm of Fisheries Program,

Agriculture and Forestry University located in Rampur, Chitwan. The study was conducted for 120 days from March 15 to mid-July. The area of the pond was $575m^2$ with a duck shed on the dyke previously constructed by the farm.

b) Materials and Methodology

Pond was prepared by draining, thorough cleaning, and removal of the existing fish and aquatic vegetation. Dry liming was done at the rate of 200 kg/ha with agriculture lime ( $CaC{O}_3$ ). After 7 days of liming, the pond was fertilized with fresh cow dung at the rate of 3000 kg/ha. The ponds were filled with fresh water after the organic fertilization. Water depth was maintained at 1 meter deep. Inorganic fertilizers such as Urea and DAP were applied at the rate of 4.7 g/m ${}^2$ /week and 3.5 g/m ${}^2$ /week respectively.

The pond was stocked with Common carp (25%), Bighead carp (5%), Grass carp (15%), Rohu (25%), Mrigal (10%), and Nile tilapia (20%). The number and amount of fingerlings stocked is tabulated below:

Feed containing 20% CP was given twice a day at 10 am and 3 pm at the rate of 2% of total body weight. Farm-made feed was fed to the fish for the reduction of the cost of production. The mixture of locally available rice bran and mustard oil cake in a 1:1 ratio was made in a dough form each day. The vitamin and mineral mixture was added at the rate of 1kg per 100 kg feed. For Grass carp, different types of vegetation like Colocassia, banana leaves, Para grass, and Napier were fed by chopping them into small pieces. Sampling was done every 2 weeks to check the growth performance and to estimate the amount of feed required. After 3 months of stocking the fish, partial harvesting was initiated. Complete harvesting was done by draining the pond water completely.

c) Rearing of Duck

Prior to stocking, the duck shed was cleaned with water thoroughly. Total 14 ducklings with an average weight of $161\pm 69.8$ g were stocked. Mainly, rice husk was fed by mixing homogeneously with water in a feeding tray. Feed was given 4 times daily at 10 am, 12 pm, 3pm, and 5 pm. Sampling of duck was done every month to observe the growth rate. Each duckling was weighed individually on a weighing machine separately. After 4 months of rearing, ducks were harvested.

d) Analytical Method

i. Fish Growth Measurements

Growth and production was calculated using the following formulae:

Extrapolated

e) Water quality analysis

Water quality parameters were monitored and recorded daily during the entire culture period. Physical parameters like dissolved oxygen (DO), temperature with DO meter (Lutran, DO 5519), and pH by using pH meter (Lutran, pH 222) were recorded daily.

f) Economic analysis

Simple gross margin analysis was done after the complete harvesting of the fish. Gross margin analysis was based on the farm prices for the harvested fish. The rate of fish per kg was estimated as NRs.300 per/day for all species of fish.

g) Statistical analysis

Statistical analysis of data was performed by using MS-Excel. Mean and standard deviation was calculated and differences were compared. Means were given with standard deviation (Mean± SD).

III. RESULTS AND DISCUSSION

a) Water quality analysis

Water quality parameters like temperature, dissolved oxygen, and pH were recorded daily during the culture period of 117 days. The mean and range of water quality parameters recorded are presented in Table 2 and Figures 1-3. The range of all the water quality parameters is similar as reported by Jena et al., (2002) and Jha et al., (2018).

b) Fish growth and production Table 3: Growth and production parameters

Table 3 shows the growth and production parameters of all fish species during the culture period of 117 days. The daily weight gain and total weight gain of Rohu was found to be 1.5 g/fish/day and 2.05kg/100m2 respectively. Similarly, the survival rate was 61.4%, and extrapolated GFY and NFY were 1.0 t/ha/yr and 0.64 t/ha/yr respectively. The daily weight gain and total weight gain of Mrigal was found to be 2.0 g/fish/day and 240.3 g/100m2 respectively. Similarly, the survival rate was 14.5% and extrapolated GFY and NFY 0.1 t/ha/yr and 0.07 t/ha/yr respectively. The daily weight gain and total weight gain of Common carp were found to be 1.7 g/fish/day and 3742.0 g/100m2 respectively. Similarly, the survival rate was 80.7% and extrapolated GFY and NFY 1.3 t/ha/yr and 1.17 t/ha/yr respectively. The daily weight gain and total weight gain of Bighead carp were found to be 3.0 g/fish/day and 1736.0 g/100m2 respectively. Similarly, the survival rate was 93.3%, and extrapolated GFY and NFY was 0.6 t/ha/yr and 0.54 t/ha/yr respectively. The daily weight gain and total weight gain of Grass carp were found to be 0.6

c) Growth Trend of Fish

i. Rohu (Labeorohita)

From the graph, it is clear that the growth of Rohu followed a normal trend from an average stocking weight of $42.1\pm 4.6$ g/fish to an average harvested weight of $212.2\pm 6.8$ g/fish. The daily weight gain of Rohu in the present study showed lower (1.5 g/fish/day) than as reported ( $2.5\pm 0.1$ g/fish/day) by Mandal et al., (2018) but higher than as reported by Jha et al. (2018). The higher weight gain might be due to the availability of natural food due to duck droppings incorporation. The survival rate of present work was 61.4% which is lower than as reported by Uddin et al., (2012), lower than that reported (91.0%) by Roy(2016), and also lower than as reported by Azim & Wahab (2003) which was 71%.

ii. Mrigal(Cirrhinusmrigala)

From the graph, it is clear that the growth of Mrigal followed a normal trend from an average stocking weight of $9.8\pm 0.4$ g/fish to an average harvesting weight of $242.8\pm 32.4$ g/fish. In the present work, the daily weight gain was 2.0 g/fish/day which was slightly higher than as reported ( $1.8\pm 0.1$ g/fish/day) by Mandal et al., (2018) and also higher than as reported ( $0.7\pm 0.1$ g/fish/day) by Jha et al., (2018). According to Uddin et al., (2012), the survival rate of Mrigal was $90.2\pm 2.20\mathrm{\%}$ which was 14.5% during the present work. Also, the survival rate of Mrigal as reported by Mandal et al., (2018) in previous work was $40.5\pm 4.1\mathrm{\%}$. and $66.0\pm 5.3\mathrm{\%}$ as reported by Jha et al., (2018).

iii. Common Carp (Cyprinus carpio)

From the graph, it is clear that the growth of Common carp followed a normal trend from an average stocking weight of $15.1\pm 3.7$ g/fish to an average harvesting weight of $208.6\pm 84.0$ g/fish. DWG of

Common carp during present work is 1.7 g/fish/day which is lower than $2.2\pm 0.1$ g/fish/day as reported by Mandal et al., (2018) and also lower than as reported ( $5.1\pm 1.7$ g/fish/day) by Jha et al., (2018) but higher than as reported ( $1.11\pm 0.07$ g/fish/day) by Bhandari (2016). The survival rate of Common carp is higher in present work (80.7%) than as reported ( $22.7\pm 3.8\mathrm{\%}$ ) by Jha et al., (2018) and ( $78\pm 7\mathrm{\%}$ ) by Bhandari(2016) but lower than as reported (84%) by Azim & Wahab(2003) in development of a duckweed fed carp polyculture system in Bangladesh.

iv. Bighead Carp(Aristichthys nobilis)

From the graph, it is clear that the growth of Bighead carp followed a normal trend from an average stocking weight of $57.2\pm 2.5$ g/fish to an average harvesting weight of $411.1\pm 42.7$ g/fish. Bighead Carp showed a daily weight gain of 3.0 g/fish/day in the present work which is higher than as reported ( $2.6\pm 0.8$ g/fish/day) by Mandal et al., (2018) and also higher than $2.1\pm 0.1$ g/fish/day as reported by Jha et al. (2018). The survival rate of bighead carp was relatively similar as reported by Mandal et al., (2018). It was remarkably higher than that as reported ( $45.4\pm 2.5\mathrm{\%}$ ) by Jha et al. (2018) in the production of periphyton to enhance yield in polyculture ponds with carps and small indigenous species.

v. Grass Carp(Ctenopharyngodon idella)

From the graph, it is clear that the growth of Grass carp followed a normal trend from an average stocking weight of $15.1\pm 3.7$ g/fish to average harvesting weight of $208.6\pm 84.0$ g/fish. As reported in Jha et al., (2018), the daily weight gain of Grass carp is $2.2\pm 0.2$ g/fish/day which is only 0.6 g/fish/day in the current work. Pandit et al., (2004) reported the daily weight gain of Grass carp to be $3.14\pm 0.15$ g/fish/day when stocked at $0.5fish/m^2$. The survival rate of Grass carp is 47.6% in the present work is similar to as reported ( $45.4\pm 2.6\mathrm{\%}$ ) by Jha et al., (2018) but reported higher by Bhandari (2016) in carp and tilapia culture.

vi. Nile Tilapia(Oreochromis niloticus)

From the graph, it is clear that the growth of Tilapia followed a normal trend from an average stocking weight of $59.4\pm 13.4$ g/fish to an average harvesting weight of $132.2\pm 15.9$ g/fish. According to Guerrero III et al.,(1988), the survival rate of Nile tilapia was 100% in commercial diet and 93% in chicken manure but it is 81.5% in the present work where duck droppings were used. Also, the survival rate was reported lower as compared to the present work by Bhandari(2016) which was $69\pm 5\mathrm{\%}$. The DWG of Nile tilapia was estimated to be 0.8-1.0 g/fish/day by Shrestha & Jaiswal (2011) is higher compared to the present work (0.6 g/fish/day). But the DWG in the present work is higher as compared to the result reported by Pandit et al., (2004) which was $0.40\pm 0.02$ in polyculture of grass carp and Nile tilapia with Napier grass as the sole nutrient input in the subtropical climate of Nepal where tilapia was stocked at 0.5 fish/m ${}^2$.

d) Combined Fish Production

Table 4 presents the production and yield parameters of fish during the culture period of 120 days. Total fish yield of 53.21 kg of fish was gained from the pond of $575m^2$. Extrapolated NFY and GFY were calculated to be 4.0 t/ha/yr and 2.9 t/ha/yr respectively. The overall survival rate was 66%. The apparent food conversion ratio was found to be 1.4.

The overall survival rate of the present work was estimated to be 66% which is lower than that as reported by Bhandari (2016) in the value of Nile tilapia and Sahar in carp polyculture pond in improving pond productivity. Similarly, Mandal et al., (2018) reported a survival rate of $75.2\pm 5.8\mathrm{\%}$, and Jena et al., (2002) reported an $88.0\pm 0.2\mathrm{\%}$ survival rate. Bhandari (2016) reported the FCR to be $2.62\pm 0.17$ which in present work is 1.4. Similarly, Mandal et al., (2018) reported the FCR value of $1.5\pm 0.2$ which is nearly equal to that of the present work.

e) Duck growth and production

Figure 5 indicates the average weight of duck during the rearing period. The growth of duck showed a

Table 5 indicates the growth of duck during the rearing period. The rearing period of duck was about 120 days. The mean stock weight and mean harvest weight was $161\pm 69.8$ g/duck and $1114.4\pm 296.4$ g/duck respectively. Similarly, daily weight gain was 7.9 g/duck/day.

During the rearing period of a duck the mean weight harvest was $1114.1\pm 296.4$ g/duck in the present work which is similar as reported(1304g) by Kumar et al.,(2012) for the same work period. According to Latif et al.,(1993) the final mean weight ranged from 1200-1800g for a period of 4-6 months.

f) Gross margin Analysis

i. Gross margin analysis of duck cum fish integration The total variable cost involved in the fish production was NRs.18,286.54. Similarly, the production cost was NRs. 150.7 per kg and benefit: cost ratio was 1.69. All the variables and costs are tabulated below in Table 6.

g) Cost analysis of duck farming

Table 7 indicates the economic analysis of duck farming integrated with the fish culture. The ducklings were reared for 120 days. The production cost was estimated as NRs.465.99 per kg and the benefit: cost ratio was 1.24.

IV. CONCLUSION

Integrated fish farming is a sustainable and effective tool for improving the livelihood of rural people. It offers the effective and efficient utilization of the locally available resources and diversification of the income of the small-scale farmers. This research concluded that integrated duck-fish farming can resolve the issues of sustainability effectively.

ACKNOWLEDGMENT

We would like to thanks the Department of aquaculture and Fisheries, Agriculture and Forestry

University for providing research facilities to carry out this work. We would also like to thanks Mr. Rahul Ranjan, Prof. Dr. Madav Kumar Shrestha, Prof. Dr. Sunila Rai, Prof. Dr. Dillip Kumar Jha, and Asst. Prof. Dr. Narayan Prasad Pandit for their invaluable suggestion, persistent co-operation, continuous supervision, and guidance throughout the period of this research.