Productivity and environmental and economic evaluation of integrated
duck-rice-fish systems on smallholdings in
the Mekong Delta of Vietnam
 

Bui Xuan Men and R Brian Ogle*
 

     Department of Animal Husbandry, Faculty of Agriculture, Cantho University,
* Department of Animal Nutrition and Management, Swedish University of Agricultural Sciences, Sweden

 

Abstract

Four experiments were conducted on smallholdings in the central part of the Mekong Delta in Vietnam to assess the ability of growing ducklings and fish to control insect and golden snail pests and weeds, and to improve soil fertility in high yielding rice fields. Experiment 1 was carried out in spring-summer 2002 and included three treatments allocated to three 559 m² plots of three families near a direct seeded rice field: (Ctrl) complete application of chemical fertilizers and pesticides, without ducks; (DR) no pesticides, fertilizer application as in (Ctrl) with ducklings, and (DRF) ducklings and fish, without pesticide applications. The study consisted of 90 Cherry Valley crossbred ducklings with 15 birds per plot.

Three weeks after the introduction of the ducks and fish most of the common species of weed, insect and golden snail pests affecting rice had been largely eliminated. Rice yields were 4,573, 4,712 and 4,848 kg per ha for the Ctrl, DR, and DRF treatments, respectively.

Experiment 2 was conducted on the summer-autumn crop of the year and included the same treatments as in Experiment 1, except that the fish from Experiment 1 remained and the ducklings used were local crossbred ducks. Three weeks after introduction of the ducklings and remaining fish the main pests had been eliminated. Rice yields were 3,904, 4,056 and 3,915 kg per ha for the Ctrl, DR and DRF treatments, respectively.

Experiment 3 was carried out in the autumn-winter season during the flooding season, with only ducks and fish raised in the duck-rice-fish areas of Experiment 1 and 2. Two months after introducing the ducks and 8 months after introducing fish in the DRF areas total live weights were 1,900 kg and 438 kg per ha for the ducks and the fish, respectively. Most of the insect and snail pests and weeds were eliminated after at the end of Experiment 3, and the nutrient concentrations of N, P and K in the soil were increased.

Experiment 4 was as conducted in winter-spring 2002-2003 and included the same treatments as in Experiment 2. Three weeks after introduction of the ducklings the important weed and insect pest and golden snails had been eliminated. The rice yields were 5906, 6056 and 6,493 kg per ha for the Ctrl, DR and DRF treatments, respectively.

The net economic benefits in these experiments were highest for the DRF treatment. It was concluded that completely replacing pesticides by growing ducks and fish and reducing nitrogen fertilizer application by 20 % increased rice yields by up to 10 % and increased total net economic benefits of the duck-rice or duck-rice-fish integrations by 55 to 144 % per year.  

Key words: Ducks, fish, rice, insects, weeds, fertilizer, pesticide, integrated system
 

Introduction

Rice is the main crop produced in the Mekong Delta. High yielding varieties have been introduced, planted in two or three crops annually, a system that has become dependent on the regular use of chemical fertilizers, herbicides and insecticides, and which has enabled farmers to abandon more traditional control strategies. However, an over-reliance on the use of chemicals has led to problems of resistance and environmental damage. Overuse of chemicals and overexploitation of the land has impoverished the soils and resulted in pollution of the environment in the Delta, damaging the health of humans, aquatic creatures and domestic animals. Income of farmers from the rice monoculture is still very low and there is an increasing dependence of farmers on this unsustainable system. Integrated duck-rice-fish cultivation in Vietnam and other Asian countries is seen as an efficient system for the control of pests and weeds in the growing rice fields, and preliminary trials with ducks and rice carried out in 1998 and 2000 (Men et al 2000) partly confirmed this. However, fish were not included in these trials, and so experiments were carried to evaluate the effects of duck-rice-fish integration. 
 

Materials and methods

The study consisted of three experiments on ducks-rice-fish integration in different crops of rice through the year in which one crop of duck-fish integration in the flooding season was introduced between two crops of rice. The treatments were: Rice area completely controlled by insecticides, herbicides, and chemical fertilizers (Ctrl); Ducks-rice integration with fertilizers applied as in (Ctrl) without fish and pesticides (DR);  Rice area combined with ponds with ducklings and fish, and fertilized as in (Ctrl) but without insecticides or herbicides (DRF). These treatments were repeated in the same rice areas in three crops of rice. The trial on duck-fish integration included only ducks and fish raised in the DRF treatment, without rice during flooding season.

Experimental site

The experiments were carried out on small farmer holdings in Phuocthoi village in Omon district, Cantho province. The main agricultural product is rice, with some supplementary income from animals and fish (Men et al 1996, 1998, 2001). 

Experimental design

Experiment 1

Three treatments were allocated to three 559m² plots in each area (block) of a 15m x 125m rice field on 3 smallholdings in Omon district, Cantho Province in the spring-summer crop: Ctrl (control treatment) - conventional application of pesticides and fertilizers, without ducks and fish; DR - ducks to control insect and golden snail pests and weeds (no pesticides), fertilizer application as in Ctrl; DRF - ducks and fish to control insect and golden snail pests and weeds (no pesticides), fertilizer application as in Ctrl and DR treatments. 

Experiment 2 

The three treatments (Ctrl, DR and DRF) in Experiment 1 were repeated in the same areas on the summer-autumn crop (after the spring-summer crop) with 15 local crossbred ducks per plot. 

Experiment 3 

Ducks were kept in the integrated duck-rice-fish area (DRF treatment) together with fish, without rice in the flooding season. The density of ducks was 60 ducks per 559 m² plot.  

Experiment 4

The three treatments (Ctrl, DR and DRF) in Experiment 2 were repeated for the third crop of rice in the same areas in the winter-spring crop (after the flooding time). 

Management of the rice plots and ducklings

In three experiments the plots were separated by clear nylon sheets tightly attached to one meter high bamboo frames buried in the mud 0.3 m under the soil surface and completely surrounding each plot in order to minimize insect and chemical movements between plots. The ducks on treatments DR and DRF were driven onto the direct-seeded rice plots at 9 (Cherry Valley crossbred ducks in Experiment 1) and 7 (local crossbred ducks in Experiment 2 and 4) days of age and 0.172 and 0.105 kg mean live weight in Experiment 1, and 2 and 4, respectively. The ducks were allowed free access to the plots at all times from 24 to 50 days after sowing for all these crops. The ducklings were given 158 g / bird (Experiment 1) and 124 g / bird (Experiment 2 and 4) of supplementary grower feed three times per day, that contained 15 % crude protein and was formulated to meet recommended nutrient requirements. The densities were 15 ducklings per 559m² plot in all experiments. The scavenging area of 37m² per bird in all experiments was in the normal range in the Mekong Delta of 20 - 40m² per bird, and based on trials done in the same village in 1998 and 2000. When the plants started flowering the ducks were removed to prevent damage to the developing grains of rice. Between flowering and harvesting, all rice plots were sprayed with pesticides (if any) according to the farmers’ normal practice. During the experimental period the pesticides and chemical fertilizers applied and application rates were as shown in Tables 1 and 2.

 

Data collection and analysis

The numbers of insect and snail pests, and weeds on each plot were determined by placing a counting frame at random at 4 different sites in each of the rice plots. Important insect and weed pests in the frame were identified and counted. This was done on the day before the ducks were driven onto the rice plots, and then at 46 days of age of the rice plants. At harvesting, rice yields, duck and fish gains, input costs and net economic benefits were calculated. Feed materials and rice soil samples were analysed. All data collected were analysed by ANOVA, using the General Linear Models of Minitab, Version 12 (Minitab 1998).

 

Table 1. Pesticide and herbicide applications (Experiments 1, 2 and 4)
	Brand name	Against
1	Pesticides  Hopkill   Endosol   Hopsan   Tilt Super	Leaf rolling worms Hoppers Golden snails Rice bugs Fungi
2	Herbicides   Whip’S   Cantosin   Nominee	Weeds Weeds Weeds
3	Fertilizers   UREA   P2O5   K2O   Bioted

 

 

   

Results

Effect of treatment on insect pest populations 

Experiment 1 

The population densities of  the major insect pests and golden snails  decreased within three weeks of the introduction of scavenging ducklings, the extent of the decreases being similar to that seen for the control treatment (Table 3). Total Hoppers (Brown and Green) were eliminated by ducks and chemicals after three weeks. Observations of leaf and stem damage caused by the insects showed that most green leaf parts were changed from a green colour to white as a result of leaf-rolling worm attack.

 

 

Experiment 2

 

Population densities of the major insect pests were lower than in Experiment 1, (Table 3) and significant numbers of only two species were recorded: Leaf rolling worms, and Hoppers (Brown and Green). Numbers tended to be lower on the plots with ducks and fish, and only ducks compared to the control treatment (P<0.05), and with the exception of Leaf rolling worms decreased to very low levels three weeks after the start of the experiment. In particular, the number of golden snails in the Ctrl treatment was higher than in the Ctrl treatment in Experiment 1 as a result of the lower effect of chemical applications than in the treatments with ducks or ducks and fish.

 

Experiment 4

 

Population densities of the major insect pests were lower than in Experiment 1 and 2, and significant numbers of only two species were recorded: Leaf rolling worms and Hoppers (Brown, and Green) in Ctrl and DR treatments. In particularly the number of golden snails tended to be lower on the plots with ducks and fish compared to the ducks only and control treatments (P<0.05), and with the exception of Leaf rolling worms and Golden snails decreased to very low levels three weeks after the start of the experiment.

 

 

Effect of treatment on the development of weeds

Experiment 1 

Eight weed species were identified in the rice plots, but densities of only four of these were higher than two plants per m². Chat weed (Fimbristylis miliacea), Chao weed (Cyperus difformis), and to a lesser extent, Lacran (Cyperus iria L.) and Longcong (Echinochloa crus-gali) populations were high initially and within three weeks were reduced (Table 5).  Especially Echinochloa species were difficult to control by the ducks, as in favourable conditions for rice plants it usually grew faster and flowered earlier than the rice plants and is similar in appearance. However, it was easily identified and eliminated by hand during its flowering period. 

 

Experiment 2 

 

Densities of Chao weed were highest initially, followed by Chat and Longcong weed, and numbers of all three species were reduced to insignificant levels 21 days after the start of the experiment (Table 6). The ducks tended to be more efficient in reducing numbers of these weed species than the herbicides used. 

Experiment 4 

After Experiment 3, buds and seeds of all weed species were eliminated by ducks and fish on the rice land in the DRF treatment.

 

Soil fertility 

The nitrogen, phosphorus and potassium concentrations in the soil were increased  in the DRF treatment compared to the Ctrl treatment after two crops of duck-rice-fish integration plus one crop of duck-fish combined in the duck-rice-fish plot without rice (Table 5). The organic matter was higher in both DRF and DR treatments compared to the Ctrl treatment.

 
Rice yields and economic analysis 

Experiment 1

 

The grain yield of the duck-rice-fish treatment (DRF) was higher (4,847 kg/ha), although there was no significant difference in yield compared to the DR and Ctrl treatments (P>0.05). However, due to the high costs of pesticide and herbicide applications of the Ctrl treatment the net economic benefits from rice were lower (Table 6).

 

Experiment 2

 

Replacing conventional pesticides and herbicides by scavenging ducks not only reduced toxic chemicals purchased but did not reduce rice yields (Table 6). Average net benefits from rice were higher for the DR and DRF treatments compared with the control.

 

Experiment 4

 

Replacing conventional pesticides and herbicides by scavenging ducks and fish in 3 crops (two of duck-rice-fish integration plus one duck-fish combination during the flooding season) reduced toxic chemical and nitrogen used and increased yield by 10 %. Average net profits from rice in 3 crops of rice were 7.46, 8.93 and 9.73 million VND/ha for the Ctrl, DR and DRF treatments, respectively.

 

Live weight gains of ducks and fish 

The growth rate of the Cherry Valley crossbred ducks in Experiment 1 was higher than that of the local crossbred ducks in Experiments 2, 3 and 4 (Table 7).

 

 

The live weight gains of the four fish species after 8 months in the integrated duck-rice-fish systems on the rice fields of smallholdings are shown in Table 8.

 

 

The net income per hectare of the duck-rice-fish treatment was highest (Table 9).

 

 

 

Discussion

The eight species of insect pests monitored in our study had been identified previously as being those mainly responsible for the damage to growing rice in the Mekong Delta (Pham Van Kim 1984). However, only a few species of insect pest were found in sufficiently high numbers to cause major damage, and these were, in order of importance (translation of local name, with Latin names in italics (Reissig 1985): Hoppers (Brown plant hopper, Nilapavata lugens, Green leaf hopper, Nephotettix cincticeps) and Leaf-rolling worms (Snaphalocrosis medinalis and Parnara guttata). Green measure worms (Naganra aenescens Moore), Stem borers (Chilo suppressalis) and Locusts (locusta migratoria manilensis) were also identified, but not in sufficiently high numbers to cause any real damage. The major insect pest species were quickly reduced in number or eliminated in the treatments with ducks, or ducks and fish, or insecticides. Numbers of these insect pests were lower than reported in 1998 in the same village (Men et al 2001). The golden snail epidemic was still a major problem to the rice producers in the village in 2002. Due to rapid infestation and reproductive characteristics the pests can dominate and damage the young rice plants in a short time. Use of toxic chemicals to eliminate the snails has been shown to have very little effect in the control treatment compared to the duck-rice-fish treatment. Also, the pesticides killed many valuable creatures in the rice fields, such as fish, but the snails are able to escape as they are buried in the mud. After two duck-rice-fish crops plus one crop of duck-fish integration without rice in the flooding season, most golden snails were eliminated by the ducks and fish with an exception of those that were too big for the ducks and fish to consume (Table 3). However, with only small numbers of the big snails remaining, producers could easily detect and collect them by hand. Raising ducks and fish in the rice field in the flooding season for over two months without rice, allowed the ducks and fish to consume all weeds, including roots (observations of the contents of the fish stomachs) and pests. The integrated duck-fish treatment at flooding time not only eliminated the pests but also cut off the life cycles of the pests. The results in the duck-rice-fish treatment (Tables 3 and 4) show that after flooding time no pesticides were needed for controlling insect or golden snail pests, the only exception being fungicide used to control Sheath Blight disease caused by Rhizoctonia solani fungi (Vo Tong Xuan and Hoang Van Thon  2000).

 

Weeds were also a major problem in the high yielding rice fields, and if not cleaned out by chemicals (Ctrl) or by ducks (DR) or the ducks and fish (DRF), soon dominated the young rice plants. If the weeds become established, and are not eliminated by either hand-weeding or chemical treatment, they continue producing seeds, which spread widely and will affect the following crop. In three crops of rice in Experiments 1, 2 and 4, two kinds of weed: Chao weed (Cyperus difformis), and Chat weed (Fimbristylis miliacea), in particular, were major problems when they appeared and grew alongside the young rice plants. If these weeds are not eliminated they compete strongly for nutrients and water, and reduce grain yields by up to 50% (Nyarko 1991). However, in our studies they were mostly eliminated 3 weeks after the ducks were introduced into the rice plots and so their effects were probably minimal. The numbers of these weeds were lower than those on the rice land reported by Men et al (2001). Especially, Echinochloa species of weed usually grew faster than the rice plants. They mixed with and looked like rice plants in the direct-seeded rice fields and the ducks could not eliminate all of them. However, these weeds were easily detected and removed by hand, which prevented them from producing seed in the following crop. These weeds were not seen in the duck-rice-fish treatment of Experiment 4.

 

Our experiments were carried out on three crops through the year (spring-summer, summer-autumn and winter-spring crop), and yields on all treatments were typical. In the Mekong Delta on this soil type three crops of rice can be grown per year with varying yields. The winter-spring crop has the highest yield (around 6.5 tonnes / ha), while the spring-summer crop has the lowest (around 5 tonnes / ha) (unpublished data, Norsk Hydro Fertiliser Project 1996). In particular, in the duck-rice-fish treatment the amount of N application was reduced by 20% in the third crop of duck-rice-fish integration, but the yield still increased 10% compared to the control treatment without ducks and fish. Ducks and fish integrated with rice cultivation in three crops improved and increased levels of the nutrients N, P and K in the soil and thus increased the rice yields.

 

Farmers producing rice in the rural areas of the Delta, can easily apply the duck-rice-fish systems in their rice fields because nowadays 90% of high yielding varieties of rice are applied in practice. In addition, the rice producers have more experience and tradition in raising ducks and aquatic species in their fields. Also, they can easily use the enormous labor resources in the rural areas to develop integrated duck-rice fish systems in their limited rice land areas to increase net income for their families. With available capital with low interest levels that the farmers can borrow from the Agricultural Bank and promoted by extension policies with improved dissemination of knowledge, producers can adopt sustainable agricultural strategies to gradually improve the quality of their lives. 

 

Conclusions

• The results of these studies demonstrate the inherent sustainability of the traditional system of using growing ducks and fish as a biological control strategy in rice production in the Mekong Delta.
• Integrating rice, duck and fish production results in substantial benefits, both with respect to improving the local environment by decreasing agricultural chemicals and increasing the incomes of resource-poor farmers.
• It was confirmed that in integrated duck-rice-fish systems, crossbred ducks and local fish, with their strong instinctive scavenging nature, could find a high proportion of their food by scavenging, and almost eliminate damage from insect and snail pests and weeds.
• Including ducks and fish therefore in an integrated ducklings-growing rice or ducklings-growing rice-fish systems, without any use of pesticides or herbicides, and reducing the recommended amount of chemical N fertilizer by 20%, increased nutrients in soil, rice yields and net profits from rice by around 18to 30 % compared to conventional chemicals-based rice systems.
• If the net profits from duck and fish sales are included, total net incomes from the integrated systems were between 55 and 144 % higher than the conventional rice system.

• In addition to the economic benefits of integrating ducks with rice production, the reduction or elimination of agricultural chemicals will result in substantial environmental benefits, especially with respect to pesticides and herbicides, overuse of which causes serious health problems for humans and pollutes waterways in the Mekong Delta. 
   

Acknowledgements 

The authors would like to thank the Swedish International Development Authority (SIDA), Department for Research Co-operation (SAREC), and to the International Foundation for Science (IFS) and the Food and Agricultural Organization of the United Nation (FAO) for financial support and would also like to thank research assistants Ngoc and Phong, and the farmers in Phuocthoi village, Cantho Province, in particular Nam Bong, Tu Hau, Chin Vien and Tu Tuyet for their cooperation.
 

References 

Bui Xuan Men, Ogle R B and Lindberg J E 2002 Studies on Intgrated Duck-Rice Systems in  the Mekong Delta of Vietnam. Journal of Sustainable Agriculture. Food Products Press, 10 Alice Street, Binghamton, NY 13904-1580. Vol. 20, Number 1. pp 27-40.

 

Bui Xuan Men,  Ogle R B and Lindberg J E  2002  Effect of Diet and Management System on Growing Duck Performance in the Mekong Delta of Vietnam. Journal of Sustainable Agriculture. Food Products Press, 10 Alice Street, Binghamton, NY 13904-1580. Vol. 20, Number 3. pp 21-31.

 

Minitab Reference Manual. 1998 Release 12 for Windows. Minitab Inc. USA.

 

Norsk Hydro fertilizer trials project 1996  Soil Science Department, Cantho University (unpublished data).

 

Nyarko K M and De Datta S K 1991  A handbook for weed control in rice. International Rice Research Institute. P.O. Box 933, 1099 Manila, Philippines.

 

Pham Van Kim 1984 Insect pests and rice production in the Mekong Delta. Dongthap Publishing House, Vietnam. pp 5-80.

 

Reissig W H 1985  Illustrated Guide to Integrated Pest Management in Rice in Tropical Asia. International Rice Research Institute. Manila, Philippines. 

 

Vo Tong Xuan and  Hoang Van Thon 2000  Manual of Rice Producers. Angiang Information and Culture Service Press. 162 pp.