Báo cáo nghiên cứu khoa học " Better Breeds of Common Carp (Cyprinus carpio L.) for Small-scale Fish Farmers - Milestone 4 " pptx

ANOVA table testing for differences in daily growth rate in relation to the effects of common carp strain HP3 and LOC and type of feed input.. Daily growth rate of HP3 and LOC common car

Collaboration for Agriculture & Rural Development

002/04VIE Project

Better Breeds of Common Carp (Cyprinus carpio L.)

for Small-scale Fish Farmers

Milestone 4: On-farm Trials and Uptake of Improved Breeds of

Contents

Contents 2

List of Figures 3

List of Table 4

1 Introduction 6

2 Materials and methods 7

2.1 Experimental Design 7

2.2 Selection of common carp strains 8

2.3 Common carp rearing and experimental procedures 9

2.4 Data collection and analysis 11

3 Results 14

3.1 Fingerling growth rate 14

3.2 Growth and survival of HP3 and LOC strains and effects of feed input 14

3.3 Growth rate and survival of HP3, H3B and LOC strains 17

3.4 Growth rate and survival for HP3, VNW and LOC Strains 19

3.5 Growth rate and survival of HP3, H3B, VNW and LOC strains in low feed input farms controlled for variation among farmer ponds 21

3.6 Common carp biomass production in farms stocked with three common carp strain (HP3, H3B, LOC) 23

3.7 Common carp biomass production comparing farms stocked with two common carp strains (HP3 and LOC) .25

3.8 Demand of the improved common carp seed 27

4 Conclusion and Recommendations 29

Appendices 33

Reference 34

List of Figures

Fig 2.1 Rice field used for farm trials in Yen Bai province 9

Fig 2.2 Farmer’s pond in Thai Nguyen province 9

Fig 2.3 Measuring common carp .11

Fig 2.4 Ethnic people harvesting common cap in rice field in Yen Bai province 12

Fig 2.5 Harvesting common carp in pond in Thai Nguyen 13

Fig 2.6 A farmer is happy with common carp growth 13

Fig 3.1 Daily growth rate of HP3 and LOC common carp strains in each of 18 farms 15

Fig 3.2 Example of relative growth of three strains of common carp in rice field in Yen Bai province after six months of culture 18

Fig 3.3 Daily growth rate of HP3, H3B and LOC common carp strain in each of 11 farms in Yen Bai and Thai Nguyen provinces 18

Fig 3.4 Daily growth rate of HP3, VNW and LOC common carp strain for a period of 10 months in each farm in Yen Bai and Thai Nguyen provinces .20

Fig 3.5 Daily growth rate of HP3, H3B, VNW and LOC common carp strains in 3 farms 22

Fig 3.6 Common carp from HP3 strain after seven months of cultured in a high input feed farm in Thai Nguyen province 23

Fig 3.7 Proportion of genetically improved common carp fry and fingerlings cultured in Vinh Phuc, Thai Nguyen and Yen Bai provinces during 2004-2006 .29

List of Table

Table 2.1 Farms, pond type, and number of fish stocked of each strain (some farms were

remove from the data set because fish escaped during flooding) .10 Table 2.2 Data analyses for common carp culture farm trails 13 Table 3.1 Mean (± SD) of body length and weight of fingerling carp after 60 days of

nursing 14 Table 3.2 ANOVA table testing for differences in daily growth rate in relation to the

effects of common carp strain (HP3 and LOC) and type of feed input .15 Table 3.3 Daily growth rate of HP3 and LOC common carp strain for period of 10 months

in 18 farms in two type of feeding regimes in Yen Bai and Thai Nguyen provinces Superscripts indicate significant differences among groups based on Tukey’s test 15 Table 3.4 ANOVA table testing for differences in survival rate in relation to the effects of

common carp strain (HP3 and LOC) and type of feed input .16 Table 3.5 Survival rate of HP3 and LOC common carp strains grown for a period of 10

months in 18 farms with two types of feeding rates in Yen Bai and Thai Nguyen provinces Superscripts indicate significant differences among groups based on Tukey’s test 16 Table 3.6 ANOVA table testing for differences in daily growth rate in relation to the

effects of common carp strain (HP3, H3B and LOC) 17 Table 3.7 Daily growth rate of HP3, H3B and LOC common carp strains grown for a

period of 10 months in 11 farms in Yen Bai and Thai Nguyen provinces

Superscripts indicate significant differences among groups based on Tukey’s test 17 Table 3.8 ANOVA table testing for differences in survival rate in relation to the effects of

common carp strains (HP3, H3B and LOC) .19 Table 3.9 ANOVA table testing for differences in daily growth rate in relation to the

effects of common carp strain (HP3, VNW and LOC) 19 Table 3.10 Daily growth rate of HP3, VNW and LOC common carp strains grown for a

period of 10 months in 5 farms in Yen Bai and Thai Nguyen provinces

20 Table 3.11 ANOVA table testing for differences in survival rate in relation to the effects

of common carp strains (HP3, VNW and LOC) 21 Table 3.12 ANOVA table testing for differences in daily growth rate in relation to the

effects of common carp strain (HP3, H3B, VNW and LOC) 21 Table 3.13 Daily growth rate of HP3, H3B, VNW and LOC common carp strains for

period of 10 months in 3 farms in Yen Bai and Thai Nguyen provinces

Superscripts indicate significant differences among groups based on Tukey’s test 22 Table 3.14 ANOVA table testing for differences in survival rate in relation to the effects

of common carp strains (HP3, H3B, VNW and LOC) and culture types (low and high feed input) .23 Table 3.15 Biomass of common carp strains in 11 farms with 3 strains (HP3, H3B and

LOC) and cultured for 300 days in Yen Bai and Thai Nguyen provinces 24 Table 3.16 ANOVA analyses of biomass of HP3, H3B, and LOC common carp strains.25 Table 3.17 ANOVA analysis of biomass of HP3 and LOC common carp strains .25 Table 3.18 Biomass of common carp strains in 18 farms with 2 strains (HP3 and LOC)

cultured for 300 days Superscripts indicate significant differences among groups based on Tukey’s test 26 Table 3.19 Biomass of HP3 and LOC common carp strains grown for a period of 10

months in 18 farms with two types of feed input in Yen Bai and Thai Nguyen provinces .27 Table 3.20 Number of hatcheries and fry nursing farmers in 3 provinces .27 Table 3.21 Number of common carp fry produced in five hatcheries in Thai Nguyen, Yen

Bai and Vinh Phuc provinces in 2006 28 Table 3.22 Proportion of fry produced from genetically improved common carp lines in

five hatcheries in Thai Nguyen, Yen Bai, and Vinh Phuc provinces in 2006 .28

1 Introduction

In Vietnam aquaculture plays a very important role in economic development and food security for small scale farmers and contributes 35% of Vietnamese protein consumption Aquaculture production is increasing at rate of 10% per year and contributes significantly

to the country’s export income (MOFI, 2007)

Common carp is one of the most popular freshwater aquaculture species in Vietnam and is cultured in pond, cages, reservoir and rice field for household consumption and income generation A recent survey of 133 carp farmers indicated that pond and rice field

production are the preferred form of culture systems (98%) with the pond culture the most common (Austin et al., 2007a) Most farmers culture common carp with up to 8 other fish species, both indigenous (silver carp, black carp) and exotic (silver grass carp, bighead, rohu, mrigal, pirapitinga, Tilapia) In polyculture ponds, common carp is the predominate species making up 30.1% of the biomass based on a recent survey (Austin et al., 2007a)

The level of culture intensity for common carp varies from small scale extensive farming, with fish deriving all their nutrition from natural pond productivity, through semi-

intensive farming using fertilization from organic material such as bran, agriculture products and household wastes to high intensive culture system with high stocking

by-densities and the use of manufactured fish foods Semi-intensive culture systems are the most popular in Vietnam using ponds or a combination of ponds and rice field cultivation (Austin et al., 2007a)

There are many factors that affect production and yield of farmed fish species, including seed (fry and fingerlings) quality, feed type and rate, fertilisation and pond management, including water exchange Of these, seed quality, which is directly related to the genetic quality of the broodstock used to produce the fry and fingerlings, has been identified as being of major concern to researchers (Thai et al., 2006; 2007), but which is not generally understood by farmers to be a potential issue affecting farm productivity (Austin et al., 2007a)

In Vietnam there are many different local varieties of common carp that have been used

by farmers but they usually have small size and low growth rate (Tran, 1983) Over recent times the Research Institute for Aquaculture No.1 (RIA-I) has bred genetically improved common carp strains to enhance the productivity of small scale fish farms that utilise this species This program has used crossbreeding and mass and family selection

methodologies to produce genetically improved strains and is considered to have achieved

an average increase of 5% in growth rate per generation over a number of generations (Thien and Thang, 1992) However, all the selective breeding and associated growth trials have been conducted in research ponds, often without the availability of unselected lines

as control populations for comparative studies

As a consequence, on-farm growth trials of different strains were undertaken to allow for a more effective analysis of carp growth under environments directly relevant to small scale carp farmers and as a strategy to encourage uptake by farmers of genetically improved strains This report presents, firstly, a report on the statistical analysis of growth and production of different common carp strains in small scale farmer ponds in Yen Bai and Thai Nguyen provinces The farmers participating in this research project included those using both pond and rice field culture from mostly highland environments All farmers had previously participated in the socio-economic survey and one of two workshops on fish breeding and genetic improvement conducted as part of this project Secondly, a survey of the uptake of improved breeds by producers is presented

2 Materials and methods

2.1 Experimental Design

The original objective of the on-farm trials was to compare a genetically improve carp strain and a local strain communally reared in six ponds owned by different farmers With additional support from the Research Institute of Aquaculture No 1 (RIA1), the

experimental objectives were scaled up to include 37 separate farmers and four strains The majority (34) of farmers used ponds as their culture systems with six farmers using

rice field systems, which reflects the proportion of these different farming systems in these provinces based on the socio-economic survey (Austin et al., 2007a) (Fig 2.1&2.2) Farms were classified according to their levels of feed inputs as either high or low to determine if this important management aspect influenced the relative performance of the different carp strains Farmers who did not provide food more than once per month were classified

as having “Low” input pond systems and those who fed at least once per week or more frequently were classified as having “High” input pond systems The experiment was conducted over a 12 months period from March 2006 to March 2007, which included spawning and fry rearing March – May, 2006 and fingerling grow out May 2006 to March

2007, which largely coincides with the normal carp farming culture cycle The trials were conducted in pond systems owned by 20 households in each of the Thai Nguyen and Yen Bai provinces Representatives of these households were interviewed for the socio-

economic survey and participated in one of the farmer workshops on fish breeding and selection held as part of this project

As described in more detail below the experimental design could not be achieved due to differential reproduction and survival of fry Thus different farms were stocked with different numbers of strains and in varying combinations Further, fish could not be

harvested from several farms due to flooding and other management problems Appendix

1 lists the farms that participated in the project, and details of their culture systems and fish that were stocked and if data collection at harvest was possible

2.2 Selection of common carp strains

Fours common carp strains were used for growth trials and included one strain (HP3) recently produced through hybridization between the three blood Hungarian strain and a recently imported pure line of Hungarian carp, the three blood Hungarian strain (H3B), an unselected Vietnamese strain (VNW) and a locally available strain (LOC) produced from broodstock available from the Yen Bai provincial hatchery Genetic analysis of this strain (Thai et al., 2006: 2007) indicates it represented a mixture of Indonesian, Hungarian and Vietnamese strains with the latter strain predominating

Fig 2.1 Rice field used for farm trials in Yen Bai province

Fig 2.2 Farmer’s pond in Thai Nguyen province

2.3 Common carp rearing and experimental procedures

Common carp breeding was undertaken at the National Broodstock Centre Hai Duong Broodstock of each experimental strain were induced to breed on the same day, using standard practices for gamete stripping and fertilization (Thai and Ngo, 2004) Between 10 -12 families of fish were obtained from each strain After fertilization eggs produced from

different families of the one strain were pooled and raised in 200l upwelling incubators After 4-5 days when larvae had reached 8-12 mm they were transferred to four ponds and stocked at a rate of 100 larvae/m2 and grown for a period of two months Every effort was made to keep the conditions under which the fry and fingerlings of each strain were raised

as similar as possible, especially in relation to stocking density and feeding regime When the fingerlings had reached of 3-5g, they were tagged by using Coded Wire Tag (CWT) The strains of common carp lines were marked by placing the CWTs on different

locations on the body The tagged fish were stocked into 40 ponds or pond-rice field systems over a two day period Communally stocked fish were in equal proportions with the exception of three ponds The details of the fish stocked in farms is given Table 2.1

Table 2.1 Farms, pond type, and number of fish stocked of each strain (some farms were remove from the data set because fish escaped during flooding)

Farm Provinces Culture Area Feed

systems (m2) input HP3 H3B VNW LOC Total

Lien Yen Bai Ricefield 1500 LF 113 113 113 113 450

Strain and No of fish stocked

HF: High feeding rate LF: Low feeding rate

Each farm pond or rice field was stocked at a rate of 0.3 fish/m2 and was classified as either “Low Feed” or “High Feed” based on the level of food inputs into the ponds Rice field systems fed less than once per week were classified as low input systems and more than once per week as high input

All experimental ponds and rice fish fields were managed by farmers Fish in ponds and rice fields were fed by available foods from farmers’ households such as rice bran, corn and cassava A log book was provided to each farmer to keep a record of food inputs into their experimental pond and other relevant information and this information was used to classified ponds into high and low feed input The farmers were visited on a monthly basis

by project staff to assist in record keeping (Fig 2.3)

Fig 2.3 Measuring common carp

2.4 Data collection and analysis

Inequalities in fry weight among strains after the three month nursing period was tested by weighing a sample of 30 individuals to the nearest 0.1 g and analysed using a one –way ANOVA Prior to this analysis the FMax test was used to determine if variances amongst strains were homogenous

Pond trials were conducted for 300 – 330 days Fish were harvesting by draining and by netting Data from were collected as fish wet weight to the nearest 0.1 g and length (snout – caudal) measured to the nearest mm (Fig 2.4, 2.5 & 2.6) Fish were allocated to strain based on the identification of the position of the CWT using a detector scanner (North West Marine technology, Shaw Island, WA, American) To compare growth between strains, fish weights were converted to daily growth rate (DGR) after first subtracting the mean fingerling weight for that strain

The appropriate statistical procedure for the experimental design for analyzing for

differences in growth rate is a three-way ANOVA with nesting The first level of the analysis allows partitioning of variance due to Feed Input as either low (LF) or high (HF), the second level is Farmer, which is nested within level 1 and the third level is Fish Strain Due to the unequal numbers fry available for each of the four carp strains and their

stocking in different combinations into farms and the unavailability of data from a number farms, several kinds of ANOVA were conducted The number of farms with different combinations of stocked strains is given in Table 2.2 together with the type of ANOVA conducted

Fig 2.4 Ethnic people harvesting common cap in rice field in Yen Bai province

Fig 2.5 Harvesting common carp in pond in Thai Nguyen

Fig 2.6 A farmer is happy with common carp growth

Table 2.2 Data analyses for common carp culture farm trails

Type of analysis

Farm input Strains

Percentage of recovered fish and harvested fish biomass of each common carp strain were analyzed as a 2-way ANOVA The biomass of fish harvested in ponds with unequal

stocking rates was adjusted to the average stocking rate to allow for statistical comparison All analyses were conducted using the Excel and SPSS software packages

3 Results

3.1 Fingerling growth rate

The average weight of common carp fingerlings after 60 days of nursing was 3.18 g The average weight of each common carp strain is given in Table 3.1 However, the difference

in body weigh among strains was not significantly different (P>0.05)

Table 3.1 Mean (± SD) of body length and weight of fingerling carp after 60 days of nursing

3.2 Growth and survival of HP3 and LOC strains and effects of feed input

Growth rate for the HP3 and LOC strains were compared in 18 farms which included both low and high feed input systems Significant differences were observed for all factors and for the interaction between production systems and strain (Table 3.2) (P<0.01) The mean daily growth rate of HP3 strain (0.48g) was 60% higher than that of local strain (LOC) (0.30g) (Table 3.3) Daily growth rate of each fish farm is present in Fig 3.1

common carp strain (HP3 and LOC) and type of feed input

Source Type III Sum of Squares df Mean Square F Sig.

Fig 3.1 Daily growth rate of HP3 and LOC common carp strains in each of 18 farms

This last finding is particularly significant as it indicates that while there is a significant difference among strains, this difference is greatly diminished in production systems with low feed inputs Table 3.3 gives the average daily growth rate for the 2 strains in each

system and it can be seen there is only a 0.06g difference in average daily growth in low feed input systems compared with a 0.36 in high input system

The results from an analysis of survival rate of two strains are shown in Table 3.4 and Table 3.5 There is a significant difference in survival rate between low and high feed input farms but no significant difference between strains The survival rate of HP3 and LOC strains is 24.77% and 22.76% respectively and it is 62% higher in High feed input ponds compared to Low feed input ponds

Table 3.4 ANOVA table testing for differences in survival rate in relation to the effects of common carp strain (HP3 and LOC) and type of feed input

Source Type III Sum of Squares df Mean Square F Sig

Table 3.5 Survival rate of HP3 and LOC common carp strains grown for a period of 10 months

in 18 farms with two types of feeding rates in Yen Bai and Thai Nguyen provinces Superscripts indicate significant differences among groups based on Tukey’s test

Total 36 26.90a 3.58 12.95 45.67

Growth data from three strains, HP3, H3B and LOC were compared in 11 farms which included 10 low and 1 high feed input systems Significant differences were observed for strain and farmer as presented Table 3.6 Table 3.7 gives the average growth rate for the 3 strains The effect of the level of feed input could not be tested statistically due lack of replicates The differences between strains can be seen from Figure 3.2 and Table 3.7 The daily growth rate of common carp in each farm is presented in Fig 3.3 Each strain was significantly different on the basis of Tukey’s test with HP3>H3B>LOC

Table 3.6 ANOVA table testing for differences in daily growth rate in relation to the effects of common carp strain (HP3, H3B and LOC)

Source Type III Sum of Squares df Mean Square F Sig.

Table 3.7 Daily growth rate of HP3, H3B and LOC common carp strains grown for a period of

10 months in 11 farms in Yen Bai and Thai Nguyen provinces Superscripts indicate significant differences among groups based on Tukey’s test