Allometric growth and condition factor of Atlantic cod (Gadus morhua) fed to satiation: effects of temperature and body weight

Transcription

1 J. Appl. Ichthyol. 25 (2009), Ó 2009 The Authors Journal compilation Ó 2009 Blackwell Verlag, Berlin ISSN Received: March 21, 2008 Accepted: December 19, 2008 doi: /j x Allometric growth and condition factor of Atlantic cod (Gadus morhua) fed to satiation: effects of temperature and body weight By T. A rnason 1, B. Bjo rnsson 2 and A. Steinarsson 3 1 Marine Research Institute, Borgir v Nordurslo d, Akureyri, Iceland; 2 Marine Research Institute, Reykjavı k, Iceland; 3 Marine Research Institute, Grindavıḱ, Iceland Summary The effect of thermal environment on condition factor was examined for six different size-classes of Atlantic cod Gadus morhua fed to satiation. A weight length relationship for 8 to 1303 g fish reared at 4 20 C indicated an allometric growth (W = al b, a = , b = 3.257) of cod. Changes in relative condition factor (K rel ) with temperature were described with a second order polynomial. The most pronounced effect of temperature on body condition was found in the smallest size-classes, but the curves flattened with increased size. Temperature had size-dependent effects on the relative condition factor obtained from an overall weight length relationship for all fish in the experiment, i.e. K rel increased with weight at 4 C, but decreased with weight at 16 and 20 C. K rel remained high for most size-classes at 8 and 12 C. The slopes (b-values) of the weight length relationships decreased linearly with temperatures from 4 to 16 C. Introduction Atlantic cod Gadus morhua L. shows seasonal condition patterns in association with food availability, temperature, maturation and spawning. Body condition is the central determinant for swimming capabilities and thus for the ability to catch prey and escape from predators (Kjesbu et al., 1991; Lambert and Dutil, 1997; Lloret and Rätz, 2000; Dutil et al., 2003; Martínez et al., 2003). FultonÕs condition factor (K) has been used to describe a two-dimensional weight length relationship by converting it into a single statistic that gives an indication of the energy reserves of a fish. The disadvantage of the condition factor is the assumption of an isometric growth in fish. If this assumption is violated, problems of correlation between condition factor and length can arise (Lambert and Dutil, 1997; Mommesen, 1998). The functional regression value b = 3.0, from a weight length relationship of the type W = al b describes isometric growth where body form does not change (Froese, 2006). Allometric growth indicates that the weight of a fish does not change in proportion with the length. If the b-value from a weight length relationship is larger than three, the large specimens have increased in height or width more than in length. Conversely, if the b-value is smaller than three the large specimens become more elongated (Froese, 2006). The relative condition factor (K rel ) introduced by Le Cren (1951) compensates for changes in form or condition with length, and can be used to indicate whether an individual is in better (K rel 1) or worse (K rel 1) condition than an average individual of the same length. A disadvantage of K rel is that it can only be used to compare body conditions between fish within the sample and other samples with the same b-value (Cone, 1989). Moreover, the expression of K rel can be biased when the fish in the sample exhibit consistent changes in body form between size-classes (Bolger and Connolly, 1989; Pepin, 1995). The distinction between growth in length and growth in weight may be subtle, but in physiological terms, they involve different processes. An important part of growth in weight may be due to the accumulation and restoration of energy reserves, while growth in length refers to the synthesis of structural molecules (Couture et al., 1998). Food availability is the most important factor determining condition factor of wild cod, but temperature has also been shown to have an effect (Krohn et al., 1997; Lloret and Ra tz, 2000; Rätz and Lloret, 2003). However, the ever-changing thermal and trophic condition makes it difficult to elucidate the precise effect of each factor on the condition factor and growth pattern of wild cod. Thus, to better understand these effects, it may be useful to keep other variables such as food supply constant, i.e. by feeding groups of fish to satiation at different temperatures in controlled laboratory experiments. Several studies have investigated the effect of temperature on specific growth rate of cod at various sizes (e.g. Bjo rnsson and Steinarsson, 2002; Bjo rnsson et al., 2001, 2007; Brander, 1995) but limited information is available on the effect of temperature on condition factor of immature cod fed to satiation. In a previous paper (Bjo rnsson et al., 2007), we focused on the effects of temperature and body weight on growth. The present study presents a further inspection of the same data-set to examine the effects of temperature on body condition and growth pattern of cod from g. Materials and methods The fish in the present study were used in growth experiments carried out by Bjo rnsson et al. (2007). Hence, the feed, tanks, temperature, etc. were the same in both studies. A weight length relationship was obtained from six size-classes (experiments A F) with final mean weights of 8, 16, 70, 192, 409 and 1303 g; duration of the experiments was 27, 39, 54, 69, 61 and 154 days, respectively. The fish were reared in groups of 100 fish per tank except 75 to 88 specimens for the largest sizeclass. The two smallest size-classes were reared at six temperatures (0, 4, 8, 12, 16 and 20 C), the 70 g fish at five temperatures (4, 8, 12, 16 and 20 C) and the three largest sizeclasses at four temperatures (4, 8, 12 and 16 C), two tanks per temperature (Appendix A). The feeding frequencies were adjusted so that maximum growth could be obtained U.S. Copyright Clearance Centre Code Statement: /2009/ $15.00/0

2 402 T. Árnason, B. Björnsson and A. Steinarsson (Folkvord and Ottera, 1993; Rosenlund et al., 2004). All fish were hand-fed to satiation, 3 5 times during working hours (8:00 16:00 hours) with commercial dry feed from Danafeed Ltd. and Fo durblandan Ltd. In experiments A B, automatic beltfeeders with an equal amount of food as fed in the day were used additionally during the remainder of the 24-h period to ensure that the fish were fed to satiation. The feed in experiments B F contained 53, 15 and 11.5% of protein, fat and carbohydrates, respectively, whereas the respective ratios of macronutrients in experiment A were 62, 13 and 7%. The experimental fish were obtained from a large group of fish, reared in the laboratory at intermediate temperatures (8 10 C) and ample feeding. The total weight (W) (g) of each fish was obtained at the start and termination of each experiment. Before the fish were divided among tanks, total length (L) (cm) of fish was measured, except in experiment A where the initial length was not obtained. At the end of the experiments, 20 fish from experiment A and B, and 10 fish from experiments C E were sampled for length measurements in each tank. In experiment F, length, gutted weight, liver weight and weight of gonads were measured for all fish in that experiment. The formula W = al b was used to establish the relationship between W and L for fish in experiments A F, and also to establish weight length relationships for each temperature. Relative condition factor (K rel =W al b ) was obtained from the overall weight length relationship. Second order polynomial regression between K rel and L were performed to test the dependency of the condition indices on fish size. The regression was significant (P < 0.001) but weak (r 2 = 0.01). Fish reared at 4.1 C in experiment B (9.2 g) were omitted from the weight length relationships due to low feed intake, and fish reared at 19.7 C in experiment C (W = 41 g) because of high mortality. The fish in experiment A and B reared at 0 C lost weight during the experiments, and were therefore omitted from the data. Relative condition factor was used to describe changes in body condition with temperature with a second order polynomial for fish in experiments A F. For experiment F, it was possible to exclude the effects of temperature on gut content and weight of gonads by calculating a new condition factor (K rel.gl ) based on gutted weight plus liver weight (W gl ): K rel. gl =W gl al b. K rel was used to describe changes in body condition with weight on a ln scale for fish reared at five different temperatures (4 20 C). Linear regression was applied to the data for fish reared at 4, 16 and 20 C, and second order polynomial regression for fish reared at 8 and 12 C. Statistical analyses were performed to evaluate the correlations (Figs 2 and 3), where F-values indicated statistical significance (Tables 1 and 2). For further details of the experimental conditions, refer to Bjo rnsson et al., (2007). Results The b-value of the overall weight length relationship for fish in experiments A F was significantly higher than 3.0 (t-test, P < 0.001), i.e. the experimental fish exhibited positive allometric growth in the pooled sample (Fig. 1). Furthermore, separate weight length relationships for 8 to 192 g fish and 409 to 1303 g fish reared at 4 16 C showed that both groups exhibited positive allometric growth, with b = and 3.177, respectively. The b-values for both size-ranges were significantly higher than 3.0 (t-test, P < 0.001), and the confidence limits of a and b for the two size-ranges did not overlap with each other. There were significant relationships between K rel and temperature in experiments A and B, but no significant relations were obvious in experiments C E (Fig. 2, Table 1). In experiment F, 12% of the males and 2% of the females had ripening gonads (gonadosomatic index >5%) at 4 C, but no mature fish were found at other temperatures. Further, the fish at 4 C had slightly more gut contents than fish at other temperatures. K rel was therefore highest at 4 C, but there was no relation between temperature and condition based on gutted weight plus liver weight (K rel.gl ) (Fig. 2f, Table 1). On a ln scale, K rel increased linearly with weight at 4 C, but was highest for fish of intermediate weight ( g) at 8 C. Most size-classes showed high condition factors at 12 C, but K rel decreased linearly with ln weight at 16 and 20 C (Fig. 3). A statistically significant relation was found between ln weight and K rel at all temperatures except at 12 C (Table 2). Separate Exp. Temperature range ( C) Condition factor Number of fish Regression statistics r 2 n d.f. F-values P-values A 4 20 K rel B 8 20 K rel C 4 16 K rel D 4 16 K rel E 4 16 K rel F 4 16 K rel F 4 16 K rel.gl Table 1 Relationships between condition factors (K rel and K rel.gl ) and temperature in experiments A (8 g), B (16 g), C (70 g), D (192 g), E (409 g) and F (1303 g). Data fitted with a second order polynomial. K rel.gl = relative condition factor calculated from gutted weight plus liver weight d.f., degrees of freedom. T ( C) Regression coefficients Regression statistics a b c (polynomial) r 2 n d.f. F-values P-values ) ) ) ) Table 2 Relationship between relative condition factor (K rel ) and weight on a ln scale for fish reared at five different temperatures (see Fig. 3). Linear regression applied to data from 4, 16 and 20 C, and second order polynomial regression from 8 and 12 C

3 Allometric growth and condition factor of Atlantic cod 403 at 20 C (experiments A and B) showed negative allometric growth (b = 2.763). Fig. 1. Double-logarithmic plot of weight length data for fish reared at 4 20 C in experiments A F. Overall regression line is W = L 3.257, with n = 1053, r 2 = 0.997, 95% CL of a = , 95% CL of b = weight length relationships for fish reared at 4, 8, 12, and 16 C showed that the b-values decreased linearly with increasing temperature (T) from 4 16 C in accordance with the following equation: b = T (Fig. 4, Table 3). Fish reared Discussion We observed allometric growth for fish ranging between 10 and 48 cm. The allometry was more pronounced for cm fish (8 192 g) in experiments A D (a = , b = 3.295), compared to cm fish ( g) in experiments E and F (a = , b = 3.177). Most fish species change their shape as they grow and b-values may be different for larval, immature and mature fish. In most cases, growth in length appears to dominate in early life stages while growth in weight becomes relatively more important as fish reach adulthood. Allometric growth is therefore commonly observed when juveniles are included in weight length relationships (Le Cren, 1951; Ricker, 1979; Froese, 2006). Daan (1974) presented weight length data for Atlantic cod in the North Sea, including specimens from cm. The inflection point in the relationship between K and length was found to be at 22 cm, i.e. growth of fish longer than 22 cm was isometric. The strong allometric growth in the present study relative to nearly isometric growth for wild cod stocks (Daan, 1974; Lloret and Ra tz, 2000; Rätz and Lloret, 2003) may be due to difference in food availability. The experimental fish were at all times provided with excess feed whereas wild cod have commonly been found to be food-limited (Bjo rnsson, 1999; Bjo rnsson et al., 2001). The present study suggests that large changes in temperature (4 16 C) have little effect on condition factor of cod of different sizes ( g, cm) constantly fed to satiation. Furthermore, the curves describing changes in K rel with temperature flattened as the fish grew larger, thus indicating diminishing effect of temperature on condition with increased size. Rätz and Lloret (2003) analysed the (a) (d) (b) (e) Fig. 2. Relationship between relative condition factor and temperature in experiments A (8 g) (a), B (16 g) (b), C (70 g) (c), D (192 g) (d), E (409 g) (e) and F (1303 g) (f) and K rel.gl and temperature in experiment f. Data fitted with a second order polynomial. Each data point = average condition factor for each sample (tank). K rel ( ), K rel.gl ( ). K rel.gl = relative condition factor calculated from gutted weight plus liver weight (c) (f)

4 404 T. Árnason, B. Björnsson and A. Steinarsson Table 3 b-values (standard error in parentheses) from weight length relationships of fish reared at 4, 8, 12 and 16 C (see Fig. 4). Range in L between shortest and longest fish Temperature ( C) b r 2 n Range in L (cm) (0.011) (0.012) (0.013) (0.013) Fig. 3. Relative condition factor as a function of body weight on an ln scale for five different temperatures average habitat temperature effect on FultonÕs K of cm cod in various stocks in the north Atlantic. The comparison between the cod stocks showed a positive Fig. 4. Linear regression of b values from weight length relationships on temperature (T). Regression line is b = T, with n=4,r 2 = 0.984, P = for significance of parameter T relationship between average annual bottom temperature and K. Our results indicate that these effects of temperature on K of wild cod may be more indirect than direct as a result of changes in food availability with temperature. However, the ratio between growth in length and growth in weight in juvenile cod (8 16 g) was influenced by the thermal environment, i.e. juveniles seemed to enter a new growth stanza characterized by growth in weight rather than length at high temperatures (16 20 C). Condition factors at 20 C were obtained from three size-classes (experiments A C). Mortalities for the largest size-class in experiment C ranged between 58 65%, but between 4 5% and 24 26% for the fish in experiment A and B, respectively. Therefore, it is likely that high mortalities affected the condition factors of fish reared at 20 C in experiments B and C. Although the growth rate in experiment A at 20 C was considerably lower than for fish reared at 8 16 C, the fish at 20 C exhibited the highest K rel in the experiment. Furthermore, fish in experiments A and B reared at 16 C showed higher body condition than fish at 12 C, despite lower growth rate at 16 C (Bjo rnsson et al., 2007). A growth study carried out by Otterlei et al. (1999), including cod in larval and early juvenile stages showed similar results, where weight at length increased with temperatures from 4 to 14 C. Body condition of fish reared at 4 C may have been slightly overestimated since the fish reared at 4 C in experiments C E must have had higher gut content than fish at higher temperature. Digestion rate in cod reared at low temperatures is considerably lower than at intermediate and high temperatures (Tyler, 1970; Knutsen and Salvanes, 1999), and therefore the fish in experiment F reared at 4 C had relatively heavier gut than fish at intermediate and high temperatures. Starvation times in experiments C E were similar as in experiment F (16 21 h), but starvation times in experiment A and B were only a few hours. The body condition of fish reared at extreme temperatures (0, 4 and 20 C) in experiments A and B may have been underestimated because the experiments were short (27 and 39 days) and the adjustment time not long enough (1 day) (Bjo rnsson et al., 2007). Thus, at extreme temperatures the fish may stop feeding for several days and will lose weight until normal feeding has resumed. Body conditions of some of these fish may not have fully recovered by the end of these relatively short experiments. The lack of adjustment to extreme temperatures had clearly much effect on food intake of fish reared at 0 C. However, observed daily feeding rates (data not shown) suggested that short adjustment time had little effect on the final condition factor of fish reared at 4 20 C. Only fish reared at 0 C and fish in experiment B reared at 4 C showed reduced feeding rates in the beginning of the experiment, and were therefore omitted.

5 Allometric growth and condition factor of Atlantic cod 405 K rel was found to increase with weight at 4 C and decrease with weight at 16 and 20 C. Most size-classes exhibited high condition factors at intermediate temperatures (8 and 12 C) (Fig. 3). Similar results have been found for specific growth rate, where most size-classes grow well at intermediate temperatures (8 12 C) but growth at 4 and 16 C is highly size-dependent (Bjo rnsson et al., 2007). The weak but statistically significant second order polynomial relationship between K rel and L (P < 0.001, r 2 = 0.01) indicates that K rel was slightly underestimated in the smallest and largest size classes, and slightly overestimated in medium sized fish. However, the correlation coefficient (r 2 ) indicates that only 1% of the variation in K rel can be explained by L, and therefore we conclude that K rel can be adequately used to compare body condition between size-classes in Fig. 3. The significant relationship between K rel and L is likely caused by the decrease in b-value with fish size. The decrease in b-value with temperature (Fig. 4) corresponds to the results in Fig. 3, i.e. small fish thrive well at high temperatures, whereas larger fish prefer low and intermediate temperatures. A weight length relationship for fish reared at 20 C showed negative allometric growth (b = 2.763), due to rapidly decreasing condition factors with increasing length as a result of increased difficulty in coping with extreme temperatures as the fish grew larger. In conclusion, this study determined the weight length relationship for six different size-classes of Atlantic cod reared at different temperatures and fed to satiation. The most pronounced effect of temperature on body condition was found in the smallest size-classes (8 16 g), and the curves showed that growth of juveniles reared at high temperatures (16 20 C) was characterized by growth in weight rather than growth in length. Condition factors of larger fish ( g) were not significantly affected by variation in temperature from 4 16 C (Fig. 2). Acknowledgements Mr Njáll Jo nsson, Mr Kristja n Sigurdsson and Mr Matthı as Oddgeirsson, Marine Research Institute, Grindavı k helped with the sampling and took good care of the fish in the experiments. Dr Gudmundur Tho rdarson, Marine Research Institute, and two anonymous reviewers were helpful in the development of this study. References Bjo rnsson, B., 1999: Is the growth rate of Icelandic cod (Gadus morhua L.) food-limited? A comparison between pen-reared cod and wild cod living under similar thermal conditions. Rit Fiskideildar 16, Bjo rnsson, B.; Steinarsson, A., 2002: The food-unlimited growth rate of Atlantic cod (Gadus morhua). Can. J. Fish. Aquat. Sci. 59, Bjo rnsson, B.; Steinarsson, A.; Oddgeirsson, M., 2001: Optimal temperature for growth and feed conversion of immature cod (Gadus morhua L.). ICES J. Mar. Sci. 58, Bjo rnsson, B.; Steinarsson, A.; Árnason, T., 2007: Growth model for Atlantic cod (Gadus morhua): effects of temperature and body weight on growth rate. Aquaculture 271, Bolger, T.; Connolly, P. L., 1989: The selection of suitable indices for the measurement and analysis of fish condition. J. Fish Biol. 34, Brander, K. M., 1995: The effect of temperature on growth of Atlantic cod (Gadus morhua L.). ICES J. Mar. Sci. 52, Cone, R. S., 1989: The need to reconsider the use of condition indices in fishery science. Trans. Am. Fish. Soc. 118, Couture, P.; Dutil, J.-D.; Guderley, H., 1998: Biochemical correlates of growth and condition in juvenile Atlantic cod (Gadus morhua) from Newfoundland. Can. J. Fish. Aquat. Sci. 55, Daan, N., 1974: Growth of North Sea cod, Gadus morhua. Neth. J. Sea Res. 8, Dutil, J.-D.; Lambert, Y.; Chabot, D., 2003: Winter and spring changes in condition factor and energy reserves of wild cod compared with changes observed during food-deprivation in the laboratory. ICES J. Mar. Sci. 60, Folkvord, A.; Otterå, H., 1993: Effects of initial size distribution, day length, and feeding frequency on growth, survival, and cannibalism in juvenile Atlantic cod (Gadus morhua L.). Aquaculture 114, Froese, R., 2006: Cube law, condition factor and weight length relationships: history, meta-analysis and recommendations. J. Appl. Ichthyol. 22, Kjesbu, O. S.; Klungsøyr, J.; Kryvi, H.; Witthames, P. R.; Walker, M. G., 1991: Fecundity, atresia, and egg size of captive Atlantic cod (Gadus morhua) in relation to proximate body composition. Can. J. Fish. Aquat. Sci. 48, Knutsen, I.; Salvanes, A. G. V., 1999: Temperature-dependent digestion handling time in juvenile cod and possible consequences for prey choice. Mar. Ecol. Prog. Ser. 181, Krohn, M.; Reidy, S.; Kerr, S., 1997: Bioenergetic analysis of the effects of temperature and prey availability on growth and condition of northern cod (Gadus morhua). Can. J. Fish. Aquat. Sci. 54(Suppl. 1), Lambert, Y.; Dutil, J.-D., 1997: Can simple condition indices be used to monitor and quantify seasonal changes in the energy reserves of Atlantic cod (Gadus morhua)? Can. J. Fish. Aquat. Sci. 54, Le Cren, E. D., 1951: The length weight relationship and seasonal cycle in gonad weight and condition in the perch (Perca fluviatilis). J. Anim. Ecol. 20, Lloret, J.; Rätz, H.-J., 2000: Condition of cod (Gadus morhua) off Greenland during Fish. Res. 48, Martínez, M.; Guderley, H.; Dutil, J.-D.; Winger, P. D.; He, P.; Walsh, S. D., 2003: Condition, prolonged swimming performance and muscle metabolic capacities of cod (Gadus morhua). J. Exp. Biol. 206, Mommesen, T. P., 1998: Growth and metabolism. In: The physiology of fishes. D. H. Evans (Ed.), CRC Press, New York, pp Otterlei, E.; Nyhammer, G.; Folkvord, A.; Stefansson, S.O., 1999: Temperature- and size-dependent growth of larval and early juvenile Atlantic cod (Gadus morhua): a comparative study of Norwegian costal cod and northeast Arctic cod. Can. J. Fish. Aquat. Sci. 56, Pepin, P., 1995: An analysis of the length weight relationship of larval fish: limitations of the general allometric model. Fish. Bull. 93, Rätz, H.-J.; Lloret, J., 2003: Variation in fish condition between Atlantic cod (Gadus morhua) stocks, the effect on their productivity and management implications. Fish. Res. 60, Ricker, W. E., 1979: Growth rates and models. In: Fish physiology. Vol. VIII. W. S. Hoar, D. J. Randall, J. R. Brett (Eds), Academic Press, London, pp Rosenlund, G.; Karlsen, Ø.; Tveit, K.; Mangor-Jensen, A.; Hemre, G.-I., 2004: Effect of feed composition and feeding frequency on growth feed utilization and nutrient retention in juvenile Atlantic cod, Gadus morhua L. Aquacult. Nutr. 10, Tyler, A. V., 1970: Rates of gastric emptying in young cod. Fish. Res. Board Can. 27, AuthorÕs address: T. Árnason, Marine Research Institute, Borgir v Nordurslód, 600 Akureyri, Iceland. tommi@hafro.is Appendix A Mean temperature (T), mean length (L), mean weight (W), mean relative condition factor (K rel ), sample size (n), mortality as % of initial number of fish (mt). Initial and final dates in the experiments: A, ; B, ; C, ; D, ; E, , F,

6 406 T. Árnason, B. Björnsson and A. Steinarsson Replicate 1 Replicate 2 Exp. T( C) L W K rel n mt (%) T( C) L W K rel n mt (%) A A A A A A B B B B B B C C C C C D D D D E E E E F F F F