Effects of silver carp and small indigenous species on pond ecology and carp polycultures in Bangladesh

Aquaculture 261 (2006) 1065 – 1076 www.elsevier.com/locate/aqua-online

Effects of silver carp and small indigenous species on pond ecology and carp polycultures in Bangladesh A. Kadir a , R.S. Kundu a , A. Milstein b,⁎, M.A. Wahab a a

Department of Fisheries Management, Bangladesh Agricultural University, Mymensingh-2202, Bangladesh b Fish and Aquaculture Research Station Dor, M.P. Hof HaCarmel, 30820 Israel Received 18 July 2006; received in revised form 7 September 2006; accepted 7 September 2006

Abstract A sustainable semi-intensive pond aquaculture technology including major carp species as cash-crop and small indigenous fish species (SIS) as food for the farmers' families is being optimized in Bangladesh. The inclusion of silver carp (Hypophthalmichthys molitrix), a cheap large species affordable by poor farmers, is now being considered. As part of a study on the effects of this filter feeder on polycultures including the large carps rohu (Labeo rohita), catla (Catla catla) and common carp (Cyprinus carpio) and the SIS punti (Puntius sophore) and mola (Amblypharyngodon mola), an experiment was carried out to test the effects of silver carp and of each SIS species on the growth, survival and yield of the large and small fish and on pond ecology. The ecology of the ponds was dominated by changes in time, strongly related to the development of a surface plankton scum at the beginning of the culture season and weather conditions. The surface scum increased decomposition processes and decreased algal development in the water body, promoted photosynthesis and ammonium release and reduced nitrification. Over those effects, the presence of silver carp in the ponds decreased algal biomass through grazing and promoted nitrification providing and resuspending particles in the water column. These effects were also produced by mola, but were evident only in the absence of silver carp. Punti stirring on the pond bottom increased nutrient flow from the sediments into the water column and promoted nitrification, but were also evident only in the absence of silver carp. The addition of 10 silver carp over the 99 large carps stocked in the 100 m2 fishponds did not affect punti and mola reproduction in the ponds, negatively affected rohu and catla growth and yield by about 20–25% but not their survival, did not affect common carp performance, reduced punti harvested biomass by 10%, reduced mola performance by about 50%, and silver carp's own biomass increased total yield and total income in about 20% each. The addition of 250 mola or punti to the large carp polycultures did not affect the performance of any of the large carps. The decreased income from selling the more expensive large carps was more than compensated by that obtained from silver carp, which increased total income by 13–24% as compared to the corresponding treatments without silver carp. This allows the option to the farmer of selling part of the silver carp to complete the cash income that would have been obtained from large carps only if silver carp would not be stocked, and consume the rest with the family. © 2006 Elsevier B.V. All rights reserved. Keywords: Catla; Food web; Mola; Polyculture; Punti; Rohu; Silver carp; SIS small indigenous species

1. Introduction ⁎ Corresponding author. Tel.: +972 4 6390651x23; fax: +972 4 6390652. E-mail address: [email protected] (A. Milstein). 0044-8486/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.aquaculture.2006.09.010

Freshwater aquaculture in Bangladesh has been increasing at a rate of 13% per year with a 44% contribution to the country total fish production of 2.102 million

1066

A. Kadir et al. / Aquaculture 261 (2006) 1065–1076

metric tones (DoF, 2005). Polyculture of Indian major carps and Chinese carps is the main aquaculture technology practiced in the country. As part of the pond management in this culture practice, fish poisons from organic and inorganic origin have been prescribed to the pond aquaculturists to eradicate small indigenous fish species (SIS, adults size up to 15 g) from the fish ponds (Wahab et al., 2003b). Polyculture technologies focused on large carps only have deprived the members of the farming households, especially the children and women, because large carps are grown for cash crop not for family consumption. Even the large carps seldom used for family consumption often are not distributed evenly in the family, rather the male senior members consume the largest share depriving the women and children. On the other hand, small fish, which are eaten whole and constitute a rich source of vitamins and micronutrients, if available for consumption are distributed evenly among all family members. The low intake of vitamin A and other micronutrients has caused an increase in night blindness, anemia and stunted growth, which can be mitigated by providing vitamin A rich small fishes like mola (A. mola), calcium rich punti (P. sophore) and other micronutrient rich fishes (Thilsted et al., 1997). A sustainable semi-intensive pond aquaculture technology under Bangladeshi conditions including major carp species as cash-crop and small indigenous fish species as food for the farmers' families is being developed. The first steps to optimize the cash-SIS technology were centered on the study of fish–environment relationships in several polyculture combinations, focusing on interference on the pond bottom by the bottom feeding fish (Wahab et al., 2001, 2002, 2003a; Milstein et al., 2002; Alim et al., 2004, 2005). At present the research concentrates on the intervention in the water column, through the addition of a fish with ecological and socio-economic potential advantages: silver carp (H. molitrix) is expected to have a strong impact on the pond ecology, because it is a very efficient filter feeder (Milstein et al., 1985a,b; Milstein, 1992), and also on the farmers' family nutrition, because it is a cheap fish that the family can afford to eat instead of selling. It is also easily accessible to the poorer section of the population because of its low market price. In the framework of the study of the introduction of silver carp on polycultures including the Indian Major Carps rohu L. rohita, catla C. catla and common carp C. carpio, and the SIS species punti and mola, the objective of the herein reported experiment was to test the effects of silver carp and of each SIS species on the growth, survival and yield of the large and small fish and on pond ecology.

2. Methods The experiment was performed at the Fisheries Field Laboratory, Bangladesh Agricultural University, Mymensingh (BAU) in 18 ponds of 100 m2 area and 1.5 m depth each. These are experimental ponds that reflect on farm conditions, since most farm ponds are 100–800 m2 area. Before starting the experiment, ponds were drained to eradicate all predatory fishes, embankments and slopes were repaired, and agricultural lime (CaCO3) at 250 kg/ha (= 2.5 kg/pond) was applied. Ponds were filled up with pumped water and fertilized with urea and triple super phosphate (TSP), each at 100 kg/ha (= 1 kg/pond) to promote algal growth. The experiment lasted 138 days and had 6 treatments in a 2 × 3 factorial design. The 2-level factor was with or without silver carp addition. The 3-level factor was SIS: no small fish, mola addition, and punti addition. On 1-Jul-05 each 100 m2 pond was stocked with 33 rohu (44 g stocking weight), 33 catla (18 g) and 33 common carp (39 g). No other fish were added to the control treatment (Ctr–). Ten silver carp (37 g) and/or 250 small fish (mola of 2.7 g or punti of 2.1 g) were added to the appropriate treatments (treatments CtrS, MM–, MMS, PP– and PPS, where S and – indicate silver carp presence and absence respectively, MM indicates mola, and PP punti). Fish were weighed monthly to adjust feeding amounts. Harvesting was on 15-Nov-05. Fingerlings of the major carps were gathered from the local retailer, who collected them from rural nurseries that obtained the fertilized eggs from the nearby Government hatchery. Fingerlings of the small fish were collected from perennial ponds of the farmers, where farmers keep them together with major carps and the small fish naturally breed. Fertilizers and manure were applied at 10 days intervals, always at the same hour (10:00 AM). Fertilizers were urea and TSP (1 kg/100 m 2 pond each). Manure (6.5 kg/100 m2 pond, wet weight) was applied wet in the four corners of each pond. Liming was repeated on 30-Jul-05 at 1 kg/100 m2 pond, to reduce a surface plankton scum that developed in all ponds. Supplementary feed consisted of rice bran and soaked oil cake (2:1, both weighed dry, before oil cake was soaked in water), given 6 times a week at a daily rate of 3% of the large carps body weight. Environmental sampling was carried out at 10 day intervals, 1 or 2 days after fertilization, always at around the same hour (9:00 AM). Water samples were taken in the middle of the water column. The parameters measured were temperature, transparency (Secchi disk), pH (EC 10 Portable pH meter), dissolved oxygen (digital

Table 1 Results of ANOVA and Duncan mean multicomparisons of water quality parameters Temp. (°C) ANOVA models ⁎⁎⁎ Sign. r2 0.99 Sign ns ns ns ⁎⁎⁎

% SS 0 0 0 100

Secchi (cm)

Chl-a (μg/l)

DO (mg/l)

pH

PO4–P (mg/l)

NH4–N (mg/l)

NO2–N (mg/l)

NO3–N (mg/l)

⁎⁎⁎ 0.65

⁎⁎⁎ 0.68

⁎⁎⁎ 0.56

⁎⁎⁎ 0.82

⁎⁎⁎ 0.55

⁎⁎⁎ 0.52

⁎⁎⁎ 0.48

⁎⁎⁎ 0.42

⁎⁎⁎ 0.38

Sign ns ns ⁎⁎ ⁎⁎⁎

% SS 0 1 4 95

Sign ns ⁎ ns ⁎⁎⁎

% SS 0 3 0 97

Sign ns ns ⁎ ⁎⁎⁎

% SS 0 2 4 94

Sign ns ns ns ⁎⁎⁎

% SS 0 0 0 100

Sign ⁎ ⁎⁎⁎ ⁎⁎⁎ ⁎⁎⁎

% SS 3 8 17 72

Sign ⁎⁎ ns ns ⁎⁎⁎

% SS 5 0 0 95

Sign ns ns ns ⁎⁎⁎

% SS 0 0 0 100

Sign ns ns ⁎⁎⁎ ⁎⁎⁎

% SS 1 2 15 82

Sign ns ns ns ⁎⁎⁎

% SS 3 0 0 97

Mean multicomparisons by silver carp (n = 90) Present 29.5 a 115 a Absent 29.5 a 113 a

21 a 22 a

117 a 122 a

4.1 a 4.2 a

7.6 a 7.7 a

0.58 a_ 0.42 _b

0.35 a 0.39 a

0.018 a 0.016 a

0.07 a 0.06 a

Mean multicomparisons by SIS (n = 60) Mola 29.5 a 111 a Punti 29.4 a 115 a Absent 29.5 a 117 a

22 a_ 21 _b 22 a_

114 a 113 a 131 a

4.1 a 4.2 a 4.2 a

7.5 _b 7.8 a_ 7.7 a_

0.48 a 0.50 a 0.53 a

0.37 a 0.37 a 0.38 a

0.020 a 0.014 a 0.016 a

0.07 a 0.06 a 0.06 a

Mean multicomparisons by date (n = 18) 30-Jun 30.0___d____ 117 __c_ 16-Jul 29.0 ____e___ 112 __c_ 31-Jul 33.0 a_______ 162 a___ 15-Aug 31.7 _b______ 134 _b__ 30-Aug 30.1 ___d____ 116 __c_ 14-Sep 29.0 ____e___ 134 _b__ 29-Sep 27.0 ______g_ 81 ___d 15-Oct 30.9 __c_____ 104 __c_ 30-Oct 27.3 _____f__ 68 ___d 14-Nov 26.7 _______h 115 __c_

28 a____ 28 a____ 17 ___d_ 27 a____ 19 _bcd_ 21 _b___ 22 _b___ 15 ____e 18 __cd_ 20 _bc__

75 ____e 101 __cde 63 ____e 83 ___de 119 _bcd_ 100 __cde 142 _b___ 282 a____ 100 __cde 126 _bc__

4.8 __c___ 7.3 a_____ 6.0 _b____ 3.2 ____e_ 2.0 _____f 2.0 _____f 2.2 _____f 5.3 __c___ 4.9 __c___ 4.0 ___d__

7.4 7.4 7.5 7.5 8.3 7.5 7.8 7.7 7.5 8.0

1.03 a___ 0.26 __cd 1.03 a___ 0.68 _b__ 0.32 __cd 0.20 ___d 0.70 _b__ 0.47 _bc_ 0.18 ___d 0.15 ___d

0.62 _b___ 0.16 ____e 0.27 __cde 0.18 ___de 0.15 ____e 0.43 __c__ 0.29 __cde 0.88 a____ 0.41 __c__ 0.34 __cd_

0.015 __cd 0.015 __cd 0.042 a___ 0.018 __c_ 0.005 ___d 0.005 ___d 0.031 _b__ 0.020 __c_ 0.013 __cd 0.006 ___d

0.03 ____ef 0.06 __cde_ 0.13 ab____ 0.08 __cd__ 0.14 a_____ 0.10 _bc___ 0.02 ____ef 0.03 ____ef 0.04 ___def 0.01 _____f

____e ____e ___de ___de a____ ___de __c__ __cd_ ___de _b___

A. Kadir et al. / Aquaculture 261 (2006) 1065–1076

Var. source Silver carp SIS Silver⁎SIS Date

Alkalinity (mg/l)

r2 = coefficient of determination. Significance levels: ⁎ = 0.05, ⁎⁎ = 0.01, ⁎⁎⁎ = 0.001, ns = not significant. % SS = percentage of total sums of squares. Mean multicomparisons: same letters in each column indicate no significant differences at the 0.05 level. aNbN… .

1067

1068

A. Kadir et al. / Aquaculture 261 (2006) 1065–1076

Jenway portable DO-meter Model-58), phosphate and nitrogen compounds (P–PO4, N–NH4, N–NO2, N– NO3, HACH Kit DR/2010), total alkalinity and chlorophyll-a (standard procedures, APHA, 1992). Ecological processes that account for the main variability of the measured variables were identified through factor analysis (Kim and Mueller, 1978; Milstein, 1993), run from the correlation matrix among water quality variables. The purpose of factor analysis is to explain the relationships among a set of variables in terms of a limited number of new variables, which are assumed to be responsible for the covariation among the observed variables. The first factor extracted from that matrix is the linear combination of the original variables that accounts for as much of the variation contained in the samples as possible. The second factor is the second linear function of the original variables that accounts for most of the remaining variability, and so on. The factors are independent of one another, have no units and are standardized variables (normal distribution, mean = 0, variance = 1). The coefficients of the linear functions defining the factors are used to interpret their meaning, using the sign and relative size of the coefficients as an indication of the weight to be placed upon each variable. The effect of silver carp and SIS presence or absence on the performance of each fish species, on water quality variables and on the factors resulting from the factor analysis were tested with ANOVA. Differences between treatment levels were tested with the Duncan multicomparison test of means. A significance level of P b 0.05 was used. Survival (percentage) data were normalized using the arcsine of the square root transformation. Feed conversion ratios were transformed to ranks (nonparametric technique appropriate for ratios) before performing further analyses. The analyses were run using the SAS statistical package. 3. Results 3.1. Water quality At the beginning of the experiment the ponds were covered with a plankton scum mainly composed by Euglenophytes, Cyanophytes, Bacillariophytes and Chlorophytes, which varied in color from light green to deep reddish-brown. The scum disappeared by midAugust. The water samples collected for chemical analyses were taken from bellow the scum. ANOVA results of the water column parameters are presented in Table 1. The model applied was highly significant for all variables, and accounted for all the variability of temperature, 82% of the variability of DO,

65–70% of alkalinity and Secchi, around 50% of chlorophyll, pH, phosphate and ammonium, and around 40% of nitrite and nitrate variability. Time was the most important source of variability of all variables, responsible

Table 2 Results of factor analysis, ANOVA and Duncan mean multicomparisons Factors:

F1

F2

F3

Secchi DO pH Chlorophyll NH4 NO2 NO3 PO4 Variance explained (%)

0.45 0.26 −0.64 −0.60 −0.11 0.57 0.26 0.60 22

−0.41 0.43 −0.26 0.55 0.73 0.39 −0.25 0.25 19

−0.56 −0.34 0.22 0.07 −0.10 0.40 0.75 0.14 16

Interpretation

Algal biomass vs decomposition (scum)

Ammonia and photosynthesis

Nitrification

⁎⁎⁎ 0.70 Sign

% SS

⁎⁎⁎ 0.72 Sign

% SS

⁎⁎⁎ 0.57 Sign

% SS

⁎⁎ ⁎⁎ ⁎⁎⁎ ⁎⁎⁎

2 2 6 90

ns ns ns ⁎⁎⁎

0 0 0 100

⁎ ns ⁎ ⁎⁎⁎

2 0 3 95

ANOVA models Significance r2 Variance source Silver carp SIS Silver ⁎ SIS Date

Mean multicomparisons by silver carp (n = 90) Present a_ a Absent _b a

a_ _b

Mean multicomparisons by SIS (n = 60) Mola a_ a Punti _b a Absent _b a

a a a

Mean multicomparisons by date (n = 18) 30-Jun ab__ _bc____ 16-Jul _b__ ___de__ 31-Jul a___ _b_____ 15-Aug _b__ _____f_ 30-Aug ___d ______g 14-Sep __c_ ____ef_ 29-Sep __c_ __cd___ 15-Oct ___d a______ 30-Oct __c_ _bc____ 14-Nov ___d ___de__

___d ___d a___ _bc_ a___ _b__ _b__ _b__ _bc_ __c_

Factor coefficients in bold were used for interpretation. r2 = coefficient of determination. Significance levels: ⁎ = 0.05, ⁎⁎ = 0.01, ⁎⁎ = 0.001, ns = not significant. % SS = percentage of the total sums of squares. Mean multicomparisons: same letters in each column indicate no significant differences at the 0.05 level. aNbN… Total N = 180.

A. Kadir et al. / Aquaculture 261 (2006) 1065–1076

1069

Fig. 1. FACTOR1 and FACTOR3 by treatment showing silver ⁎SIS significant interactions.

for over 70% of the explained variability of pH and nitrite and over 95% for the other variables. The letters in the ‘mean multicomparison by date’ section of Table 1 that

indicate significant differences between dates are placed showing the time patterns of each variable. Silver carp accounted for a small (5%) but significant part of the

Table 3 Two-way-ANOVA and Duncan mean multicomparisons of rohu parameters at harvesting Rohu ANOVA

Number (/pond)

Weight (g)

Biomass (kg/pond)

Survival# (%)

Growth (g/day)

Yield (kg/pond)

Signif. r2 Source Silver carp SIS Silver ⁎ SIS

ns 0.25 Sign ns ns ns

⁎⁎ 0.65 Sign ⁎⁎⁎ ns ns

⁎⁎ 0.69 Sign ⁎⁎⁎ ns ns

ns 0.25 Sign ns ns ns

⁎⁎ 0.70 Sign ⁎⁎⁎ ns ns

⁎⁎ 0.70 Sign ⁎⁎⁎ ns ns

%SS . . .

%SS 81 10 9

%SS 82 9 9

%SS . . .

%SS 84 10 6

Mean multicomparisons by silver carp Present 32.7 227 _b Absent 32.4 273 a_

7.4 _b 8.9 a_

99 98

1.33 _b 1.67 a_

6.0 _b 7.4 a_

Mean multicomparisons by SIS No SIS 32.7 Mola 32.7 Punti 32.3

8.0 8.5 7.9

99 99 98

1.50 1.58 1.43

6.6 7.1 6.4

#

248 261 241

%SS 84 8 8

Statistical tests based on transformed data. Values of means given untransformed. r2 = coefficient of determination. Sign = significance levels: ⁎ = 0.05, ⁎⁎⁎ = 0.01, ⁎⁎⁎ = 0.001, ns = not significant. % SS = % of the total sum of squares. Mean multicomparisons: same letters in each column indicate no significant differences at the 0.05 level.

1070

A. Kadir et al. / Aquaculture 261 (2006) 1065–1076

Table 4 Two-way-ANOVA and Duncan mean multicomparisons of catla parameters at harvesting Catla ANOVA

Number (/pond)

Weight (g)

Biomass (kg/pond)

Survival# (%)

Growth (g/day)

Yield (kg/pond)

Signif. r2 Source Silver carp SIS Silver ⁎ SIS

ns 0.15 Sign ns ns ns

⁎ 0.59 Sign ⁎⁎ ns ns

⁎ 0.59 Sign ⁎⁎ ns ns

ns 0.12 Sign ns ns ns

⁎ 0.59 Sign ⁎⁎ ns ns

⁎ 0.58 Sign ⁎⁎ ns ns

% SS . . .

% SS 82 13 5

% SS 84 11 5

% SS . . .

% SS 90 7 4

Mean multicomparisons by silver carp Present 32.4 111 _b Absent 32.3 144 a_

3.6 _b 4.7 a_

98 98

0.67 _b 0.92 a_

3.0 _b 4.1 a_

Mean multicomparisons by SIS No SIS 32.2 Mola 32.5 Punti 32.5

4.4 4.0 3.9

97 98 98

0.85 0.78 0.76

3.8 3.5 3.4

137 124 122

% SS 90 7 3

#

Statistical tests based on transformed data. Values of means given untransformed. r2 = coefficient of determination. Sign = significance levels: ⁎ = 0.05, ⁎⁎ = 0.01, ⁎⁎⁎ = 0.001, ns = not significant. % SS = % of the total sum of squares. Mean multicomparisons: same letters in each column indicate no significant differences at the 0.05 level.

explained variability only of phosphate, with higher values when this fish was present. SIS accounted for a small (3–8%) but significant part of the explained variability of Secchi and pH, the former with lower values when punti was present, and the latter with lower values when mola was present. Significant silver carp ⁎ SIS interactions were found for alkalinity, chlorophyll, pH and nitrite. Factor analysis results performed on the water quality parameters are presented in Table 2, together with the

ANOVA of the extracted factors. Three factors accounted for 57% of the overall data variability. The first factor (FACTOR1) shows two groups of variables (one with positive and one with negative high coefficients) that were positively correlated within the group and negatively correlated with the other group. The factor reflects an opposition between algal biomass in the water column (chlorophyll negatively correlated to Secchi disk) and decomposition (that liberates phosphate and nitrite into

Table 5 Two-way-ANOVA and Duncan mean multicomparisons of common carp parameters at harvesting Common carp ANOVA

Number (/pond)

Weight (g)

Biomass (kg/pond)

Survival# (%)

Growth (g/day)

Yield (kg/pond)

Signif. r2 Source Silver carp SIS Silver ⁎ SIS

ns 0.20 Sign ns ns ns

ns 0.51 Sign ns ⁎ ns

ns 0.51 Sign ns ⁎ ns

ns 0.20 Sign ns ns ns

ns 0.50 Sign ns ⁎ ns

ns 0.50 Sign ns ⁎ ns

% SS . . .

% SS . . .

% SS . . .

% SS . . .

% SS . . .

Mean multicomparisons by silver carp Present 32.8 199 Absent 32.9 187

6.5 6.1

99 100

1.17 1.07

5.2 4.8

Mean multicomparisons by SIS No SIS 32.8 Mola 32.8 punti 32.8

6.9 a_ 6.3 ab 5.7 _b

99 99 99

1.25 a_ 1.12 ab 0.99 _b

5.6 a_ 5.0 ab 4.5 _b

#

210 a_ 192 ab 175 _b

% SS . . .

Statistical tests based on transformed data. Values of means given untransformed. r2 = coefficient of determination. Sign = significance levels: ⁎ = 0.05, ⁎⁎ = 0.01, ⁎⁎⁎⁎ = 0.001, ns = not significant. % SS = % of the total sum of squares. Mean multicomparisons: same letters in each column indicate no significant differences at the 0.05 level.

A. Kadir et al. / Aquaculture 261 (2006) 1065–1076

the water column reducing pH). The ANOVA model accounted for 90% of the variability of this factor, most of it due to sampling date. High decomposition and low algal biomass in the water column occurred until mid-August, when the thick surface scum was present in the ponds. After the scum collapsed the phytoplankton population in the water column developed and the decomposition of scum material decreased. Over the date effect there were significant silver carp, SIS and interaction effects, with higher FACTOR1 values (lower algal biomass and higher decomposition) in ponds with silver carp and in ponds where the SIS was mola. The interaction effect indicates that while in the presence of silver carp there was a decreased algal biomass independently of the SIS, in the absence of silver carp this effect was produced by mola while the opposite was produced by punti (Fig. 1). The second factor (FACTOR2) shows ammonium, chlorophyll and oxygen negatively correlated to water transparency (Secchi disk). This combination reflects photosynthesis, which increases chlorophyll and oxygen while decreasing transparency (Secchi disk) due to algal growth, coupled with ammonium concentration. The ANOVA indicates that this factor was related to time and not to the fish composition in the ponds. The time pattern of this factor (multicomparison by date section in Table 2) indicates a high photosynthesis and ammonium levels at the beginning of the culture season and by midOctober. The former was related to the scum that was present at that time. The latter was related to weather conditions. The mid-October sampling was performed after a rather long (10 days) period of continuous sunshine in the absence of rain, when air and water temperatures strongly increased enhancing both, photosynthesis and decomposition increase. Together with this, the lack of rain

Table 6 ANOVA and Duncan mean multicomparisons of silver carp parameters at harvesting

1071

Table 7 ANOVA and Duncan mean multicomparisons of punti parameters at harvesting Punti (adult) ANOVA

Number (/pond)

Weight (g)

Biomass (kg/pond)

Growth (g/day)

Significance r2

ns 0.06

ns 0.02

ns 0.08

ns 0.01

1.33 1.41

0.05 0.05

Mean multicomparisons by silver carp Present 146 9.2 Absent 150 9.4 Punti fry

Total punti

ANOVA

Number Weight Biomass Biomass (/pond) (g) (kg/pond) (kg/pond)

Significance r2

ns 0.04

ns 0.00

ns 0.04

Mean multicomparisons by silver carp Present 87 1.26 0.12 Absent 117 1.25 0.17

Yield (kg/pond)

⁎ 0.79

⁎⁎ 0.88

1.44 _b 1.58 a_

0.94 _b 1.04 a_

r2 = coefficient of determination. Significance levels: ⁎ = 0.05, ⁎⁎ = 0.01, ⁎⁎⁎ = 0.001, ns= not significant. Mean multicomparisons: same letters in each column indicate no significant differences at the 0.05 level.

during those 10 days led to low pond water depth (seepage and evaporation without water refill by rain), which also allowed high water temperature from surface to bottom.

Table 8 ANOVA and Duncan mean multicomparisons of mola parameters at harvesting Mola (adult) ANOVA

Number (/pond)

Weight (g)

Biomass (kg/pond)

Growth (g/day)

Significance r2

⁎ 0.77

ns 0.31

⁎⁎ 0.88

ns 29

0.39 _b 0.72 a_

0.03 0.02

Mean multicomparisons by silver carp Present 56 _b 7.0 Absent 137 a_ 5.5

Silver carp ANOVA Significance r2

Number Weight Biomass Survival# Growth Yield (/pond) (g) (kg/pond) (%) (g/day) (kg/pond) ns 0.09

ns 0.09

Mean multicomparisons by SIS No SIS 10 751 7.5 Mola 10 734 3.4 Punti 10 739 3.4 #

100 100 100

ns 0.05

ns 0.05

5.19 5.10 5.13

7.1 7.0 7.0

Statistical tests based on transformed data. Values of means given untransformed. r2 = coefficient of determination. ns = not significant.

Mola fry

Total mola

ANOVA

Number Weight Biomass (/pond) (g) (kg/pond)

Biomass Yield (kg/pond) (kg/pond)

Significance r2

ns 0.01

⁎⁎⁎ 0.99

⁎⁎⁎ 0.96

0.52 _b 0.84 a_

− 0.15 _b 0.17 a_

ns 0.01

ns 0.00

Mean multicomparisons by silver carp Present 101 1.21 0.124 Absent 110 1.23 0.126

r2 = coefficient of determination. Significance levels: ⁎ = 0.05, ⁎⁎ = 0.01, ⁎⁎⁎ = 0.001, ns = not significant. Mean multicomparisons: same letters in each column indicate no significant differences at the 0.05 level.

1072

A. Kadir et al. / Aquaculture 261 (2006) 1065–1076

Table 9 ANOVA and Duncan mean multicomparisons of total fish harvesting biomass and yield and feed conversion ratio (FCR) ANOVA

Large Total Total Large carp yield biomass yield carp (kg/pond) (kg/pond) (kg/pond) FCR#

Significance ⁎⁎⁎ r2 0.81 Source Sign. % SS Silver carp ⁎⁎⁎ 90 SIS ns 8 Silver ⁎ ns 2 SIS

⁎⁎⁎ 0.80 Sign. % SS ⁎⁎⁎ 96 ns 0 ns 4

⁎⁎ 0.78 Sign. % SS ⁎⁎⁎ 96 ns 0 ns 4

All fish FCR#

ns 0.26 Sign. % SS ns . ns . ns .

ns 0.33 Sign. % SS ns . ns . ns .

Mean multicomparisons by silver carp Present 21.2 a_ 25.6 a_ 21.9 a_ Absent 16.4 _b 20.5 _b 17.1 _b

2.38 2.40

2.31 2.28

Mean multicomparisons by SIS No SIS 19.6 23.1 Mola 19.1 23.2 Punti 17.8 22.8

2.35 2.41 2.41

2.35 2.33 2.21

19.6 19.7 19.3

#

Statistical tests based on rank-transformed FCR data. Values of means given untransformed. r2 = coefficient of determination. Sign. = Significance level: ⁎ = 0.05, ⁎⁎ = 0.01, ⁎⁎⁎ = 0.001, ns = not significant. % SS = percentage of the total sum of squares. Mean multicomparisons: same letters in each column indicate no significant differences at the 0.05 level.

thus the higher the amount of particles (lower Secchi) the higher the amount of nitrifiers and their produce. The main variability source of this factor was time, with low nitrification level at the beginning of the experiment when the surface scum was present and bacteria of the second nitrification phase (leading to nitrate) were still not well developed, to pick after the scum collapsed, decreasing slowly thereafter. Over the time effect there was a small but significant effect of silver carp and of silver ⁎ SIS interaction. The presence of more fish in the pond (added silver carp and/or SIS to the basic 99 Indian Major Carps, Fig. 1 Ctr vs. other treatments) promoted nitrification increasing suspended particles in the water column, which are substrate for nitrifying bacteria. This effect was through the direct contribution of particles (faeces, mucus, scales, etc.) and indirectly through their activity that helped in maintaining particles suspended in the water column preventing their quick precipitation onto the bottom. The addition of silver carp produced a stronger nitrification increase than the addition of punti or mola in relation to the control treatment without silver carp and SIS. When silver carp and SIS were stocked together, the nitrification increase produced by silver carp was reduced by SIS grazing on particles (and the nitrifying bacteria on them), mainly by mola. 3.2. Fish performance

The excess presence of particles in the water column (dead algae derived from the scum and the large phytoplankton population in mid-October) might have increased the decomposition in the water column and bottom, which released higher amounts of ammonium into the water. The third factor (FACTOR3) shows a negative correlation between nitrate and nitrite on one hand and Secchi on the other, which indicates nitrification. Nitrifiers are attached on particles in the water column,

Large fish ANOVA results are presented in Tables 3–6. Survival of the four large carp species was at least 97%. Common carp performance and rohu and catla survival were not affected by the presence of silver carp and/or SIS. In contrast, the ANOVA models for weight and biomass at harvest, growth rate and yield of rohu and catla accounted for about 70% and 60% respectively of the variability of each fish species. Over 80% of the explained

Fig. 2. Average total yield in each treatment including each species in the polyculture. Yield scale in a per ha basis.

A. Kadir et al. / Aquaculture 261 (2006) 1065–1076

variability of each parameter were accounted for by silver carp. In the presence of this filter feeder fish, those four parameters were 17–20% lower for rohu and 23–27% lower for catla than in the absence of silver carp. SIS had no significant effects on the performance of the four large carp species. Punti ANOVA results are presented in Table 7. Punti reproduced in all ponds. All punti adult and fry per-

1073

formance parameters were not significantly affected by the presence of silver carp. However, total (adult + fry) harvested biomass and yield were significantly affected by this fish, which reduced them in 10% in relation to treatments without silver carp. Mola ANOVA results are presented in Table 8. Mola (adults) harvesting number and biomass as well as total (adult + fry) harvested biomass and yield were

Fig. 3. Conceptual representation of the pond ecosystem functioning. a) without silver carp, b) with silver carp. Size of fish, community circles and arrows represent importance of effects. Empty arrows = inorganic nutrients flow. Plain arrows = food flow, dotted arrows = sedimentation of excreta and other organic particles.

1074

A. Kadir et al. / Aquaculture 261 (2006) 1065–1076

significantly affected by silver carp, which reduced the first three respectively by 59%, 46% and 38% and led to negative yields. Mola reproduced in the ponds, the numbers, weight and biomass of their fry at harvest not being affected by silver carp. Total biomass, yields and feed conversion ratios (FCR) results are presented in Table 9. While FCR was not affected by the presence of silver carp or SIS, the biomass and yield models were significant and accounted for about 80% of the variability, almost all due to silver carp. Large carp and total harvested biomass and yields were 20–22% higher when silver carp was also present, as expected. Fig. 2 presents average fish yields in each treatment by fish species. It can be seen that although rohu, catla and mola yields were lower in the presence of silver carp, addition of silver carp into the polyculture led to higher total yields. 4. Discussion 4.1. Ecological considerations The ecology of the ponds was dominated by changes in time, strongly related to the development of the surface scum at the beginning of the culture season, and to weather conditions. The surface scum provided organic matter for decomposition and reduced light for photosynthesis in the water column (FACTOR1), promoted photosynthesis and ammonium release (FACTOR2) and reduced nitrification (FACTOR3), without producing any noticeable damage to the fish. Over those effects the polyculture composition affected the pond ecology. Fig. 3 presents a conceptual representation of the pond ecosystem functioning, based on the significant interactions observed in this experiment. Since SIS effects were minor compared to silver carp effects, only the situations with and without silver carp are represented including both SIS species in each case. Size of fish, community circles and arrows represent importance of effects. The effects of both SIS are represented in Fig. 3, without showing the control without SIS. The optimum stocking density of the small fish to avoid competition with the large fish was determined in our previous studies (Wahab et al., 2003a; Alim et al., 2004), so that as expected the SIS had practically no effect on the other fish. The addition of a rather large amount of small fish, as well as the addition of a few large silver carp, provided and resuspended particles in the water column that promoted nitrification (FACTOR3). Mola, being a phytoplankton grazer (Miah and Siddique, 1992; Kohinoor, 2000) like silver carp, also decreased algal

biomass thus reducing the absorption of inorganic compounds liberated by decomposition (FACTOR1). However, this effect was evident in the absence of the highly efficient filter feeder silver carp. A rather different effect on pond ecology was produced by punti, which can graze on phytoplankton but mainly feeds on the bottom and on detritus (Kohinoor, 2000). Stirring of sediments by benthivorous fishes has two effects: (1) it increases diffusion rates across the sediment–water interface (Hohener and Gachter, 1994), and (2) it increases aerobic decomposition by aerating anaerobic sediments (Graneli, 1979; Beristain, 2005). These two effects enhance the ammonia and phosphorous flux from the sediments to the water column (Hargreaves, 1998), which support our decomposition part of the FACTOR1 interpretation. In our experiment bottom was stirred by the common carp in all the ponds, and by punti when present. However, the added punti effect was evident only in the absence of silver carp, since the increased phytoplankton biomass related to punti activity on the bottom was reduced by silver carp grazing, leading to the accumulation of decomposition products in the water column. In the absence of silver carp (Fig. 3a), common carp and punti searching for food stir the mud bottom, facilitating nutrient flow through the autotrophic food web and resuspending particles that provide substrate for nitrifiers in the water column. The abundant phytoplankton provided food mainly for the zooplankton, for the herbivorous water column feeder rohu (Das and Moitra, 1956; Jhingran and Pullin, 1985), and the phytoplankton grazer mola. Catla is a surface feeder mainly capturing zooplankton (Natarajan and Jhingran, 1961; Jhingran and Pullin, 1985; Rahman, 1989). Particles originating in all ecosystem components gradually settle onto the bottom, providing organic matter for the decomposing bacteria. In the presence of silver carp (Fig. 3b), nitrification was promoted and large amounts of phytoplankton and also zooplankton and suspended particles were grazed Table 10 Income obtained by selling the fish of each treatment, in Taka (1 U$ = 65 Tk.) Income

Ctr–

CtrS

MM–

MMS

PP–

PPS

Rohu Catla C. carp S. carp Punti Mola Total

392 169 184 0 0 0 745

246 132 210 391 0 0 978

448 157 176 0 0 33 814

240 120 178 384 0 20 941

387 164 148 0 55 0 754

231 107 164 386 50 0 938

A. Kadir et al. / Aquaculture 261 (2006) 1065–1076

1075

Fig. 4. Average income (in Taka, 1 U$ = 65 Tk.) in each treatment by species in the polyculture, extrapolated to a 1 ha pond.

by this very efficient filter feeder. These depleted resources did not affect the survival of the other large fish or the reproduction of the small fish, but affected the plankton eating fish performance. The harvested biomass of rohu was reduced in about 20%, that of catla in 25%, and that of mola in 50%. From the bottom feeding fish common carp was not affected at all and punti biomass was reduced in only 10%. In compensation, the silver carp biomass increased total fish biomass by 20%. 4.2. Nutritional and economic considerations The quantities of small fish harvested in the experiment may seem to be very small to the western consumers. However, these are very reasonable in Bangladesh, where a weekly consumption of 250 g of mola or punti can supply essential nutrients to a 5–6 members family. Although for the same fish size range silver carp is cheaper than the Indian Major Carps, its faster growth rate led to larger silver carp and total yield when this species was stocked, which was reflected in the income received when selling the fish. In rural markets in Bangladesh SIS and large fish species weighing 100– 250 g are sold at 35–45 Tk./kg, catla and rohu weighing 250–500 g at 45–65 Tk./kg, and silver carp weighing 500–1000 g at 50–60 Tk./kg (1 U$ = 65 Tk.). Table 10 and Fig. 4 present the income obtained applying those prices to the harvested biomass in each treatment. The decreased income from selling the more expensive Indian Major Carps was more than compensated by that obtained from silver carp, which increased total income by 13–24% as compared to the corresponding treatments without silver carp. This allows the option to the farmer of selling part of the silver carp to complete the

cash income that would have been obtained from Indian Major Carps only if silver carp would not be stocked, and in addition the family can consume the rest. Acknowledgements This research was supported under Grant No. TAMOU-03-C23-022 U.S.–Israel Cooperative Development Research Program, Economic Growth, U.S. Agency for International Development (USAID). Prof. R.I. Sarker, Director of BAURS, Bangladesh Agricultural University (BAU), Mymensingh, is acknowledged for his cooperation during this study. This work could not have been done without the assistance of the staff of the Fisheries Field Laboratory and Water Quality and Pond Dynamics Laboratory, BAU, Mymensingh Bangladesh. References Alim, M.A., Wahab, M.A., Milstein, A., 2004. Effects of adding different proportions of the small fish punti (Puntius sophore) and mola (Amblypharyngodon mola) to a polyculture of large carp. Aquaculture Research 35, 124–133. Alim, M.A., Wahab, M.A., Milstein, A., 2005. Effects of increasing by 20% the stocking density of large carps of the basic “cash” carp — small fish polyculture of Bangladesh. Aquaculture Research 36, 317–325. APHA, 1992. Standard Methods for the Examination of Water and Waste Water, 18th ed. American Public Health Association, Washington DC. 1268 pp. Beristain, B.T., 2005. Organic matter decomposition in simulated aquaculture ponds. PhD thesis, Fish Culture and Fisheries Group, Department of Animal Science, Wageningen University, The Netherlands. 138 pp. Das, S.M., Moitra, S.K., 1956. Studies on the food of some common fishes of Uttar Pradesh, India. 1. Surface-feeders, mid-feeders and bottom-feeders. Proceedings of the National Academy of Sciences (B) 25, 1–6.

1076

A. Kadir et al. / Aquaculture 261 (2006) 1065–1076

DoF (Department of Fisheries), 2005. Fish Fortnight Compendium, 2005. Department of Fisheries, Ministry of Fisheries and Livestock, Dhaka, Bangladesh, p. 154. Graneli, W., 1979. The influence of Chironomus plumosus larvae on the oxygen uptake of sediment. Archiv für Hydrobiologie 87, 385–403. Hargreaves, J.A., 1998. Nitrogen biochemistry of aquaculture ponds. Aquaculture 166, 181–212. Hohener, P., Gachter, R., 1994. Nitrogen cycling across the sediment– water interface in the eutrophic, artificially oxygenated lake. Aquatic Science 56, 115–132. Jhingran, V.G., Pullin, R.S.V., 1985. A Hatchery Manual for the Common, Chinese and Indian Major Carp. ICLARM Contribution, 252. ADB & ICLARM Publication. 191 pp. Kim, J.O., Mueller, C.W., 1978. Factor analysis. Statistical Methods and Practical Issues. Quantitative Applications in the Social Sciences, vol. 14. Sage University. 8 pp. Kohinoor, A.H.M., 2000. Development of culture technology of three small indigenous fish mola (Amblypharyngodon mola), punti (Puntius sophore) and chela (Chela cachius) with notes on some aspects of their biology. Ph.D. Thesis, Department of Fisheries Management, Bangladesh Agricultural University, Mymensingh, 263 pp. Miah, M.J.U., Siddique, W.H., 1992. Studies on the food and feeding habits of mola, Amblypharyngodon mola. Bangladesh Journal of Agricultural Sciences 19 (2), 165–170. Milstein, A., 1992. Ecological aspects of fish species interactions in polyculture ponds. Hydrobiologia 231, 177–186. Milstein, A., 1993. Factor and canonical correlation analyses: basic concepts, data requirements and recommended procedures. In: Prein, M., Hulata, G., Pauly, D. (Eds.), Multivariate Methods in Aquaculture Research: Case Studies of Tilapias in Experimental and Commercial Systems. ICLARM Studies and Reviews, vol. 20, pp. 24–31. Milstein, A., Hepher, B., Teltsch, B., 1985a. Interactions between fish species and the ecological conditions in mono- and polyculture

pond system. I. Phytoplankton. Aquaculture and Fisheries Management 16, 305–317. Milstein, A., Hepher, B., Teltsch, B., 1985b. Interactions between fish species and the ecological conditions in mono- and polyculture pond system II. Zooplankton. Aquaculture and Fisheries Management 16, 319–330. Milstein, A., Wahab, M.A., Rahman, M.M., 2002. Environmental effects of common carp Cyprinus carpio (L.) and mrigal Cirrhinus mrigala (Hamilton) as bottom feeders in major Indian carp polycultures. Aquaculture Research 33, 1103–1117. Natarajan, A.V., Jhingran, A.G., 1961. Index of preponderance — a method of grading the food elements in the stomach analysis of fishes. Indian Journal of Fisheries 8, 54–59. Rahman, A.K.A., 1989. Freshwater Fishes of Bangladesh. Zoological Society of Bangladesh, Dhaka. 364 pp. Thilsted, S.H., Roos, N., Hasan, N., 1997. The role of small indigenous fish species in food and nutrition security in Bangladesh. NAGA News letter, July–December (Supplement) 13 pp. Wahab, M.A., Rahman, M.M., Milstein, A., 2001. The effect of common carp Cyprinus carpio and mrigal Cirrhinus mrigala as bottom feeders in Indian carp polyculture. World Aquaculture 32 (4), 50–52 and 69. Wahab, M.A., Rahman, M.M., Milstein, A., 2002. The effect of common carp Cyprinus carpio (L.) and mrigal Cirrhinus mrigala (Hamilton) as bottom feeders in major Indian carp polycultures. Aquaculture Research 33, 547–557. Wahab, M.A., Alim, M.A., Milstein, A., 2003a. Effects of adding the small fish punti (Puntius sophore) and/or mola (Amblypharyngodon mola) to a polyculture of large carp. Aquaculture Research 34 (2), 149–164. Wahab, M.A., Thilsted, S.H., Hoq, E.M. (Eds.), 2003b. Small Indigenous Species of Fish in Bangladesh. Proceedings of BAUENRECA/DANIDA Workshop on 30–31 October, 2002, BAU, Mymensingh. Bangladesh Agricultural University, Mymenshingh, Bangladesh. 150 pp.