Relationships Between Fish Size and Otolith Measurements for 33 Fish Species Caught by Bottom Trawl in Haizhou Bay, China
X. -X. Zan,C. Zhang,B. -D. Xu,C. -L. Zhang
DOI: https://doi.org/10.1111/jai.12751
2015-01-01
Journal of Applied Ichthyology
Abstract:Information from otoliths, together with data such as fish length and growth, can be a useful tool for population identification, aging, and feeding studies as well as stock assessments (Harvey et al., 2000; Hüssy et al., 2010; Jawad et al., 2011). Since Trout (1954) and Templemann and Squires (1956) studied the relationship between cod size and otolith measurements, relationships between fish size and otolith dimensions have been explored for many species (Jawad et al., 2012). For example, Steward et al. (2009) used the relationship between fish length and otolith dimensions to predict age in gray angelfish, and Agüera and Brophy (2011) used otolith shape analysis for the discrimination of Atlantic saury stocks. Haizhou Bay is an open bay on the western margin of the southern Yellow Sea and historically an important fishing ground. The bay still serves as a critical spawning and nursery habitat for many commercial fish species, such as Larimichthys polyactis, Setipinna taty and Engraulis japonicas (Editorial Board of Annals of Bays in China, 1993). The relationship between fish length and otolith dimensions can be described by using a simple linear regression (Harvey et al., 2000; Jawad and Al-Mamry, 2012), but these relationships between fish size and their otolith dimensions are lacking in the literature for species living in the Haizhou Bay. The objective of this study was to determine the relationships between fish length and otolith dimensions (otolith length, width, long radius and short radius) using simple linear regression for 33 species in Haizhou Bay. These results can provide a foundation for further research on these species, which may require knowledge of these relationships. Sampling occurred between 34°20′, 35°40′ N and 119°20′, 121°10′E in Haizhou Bay. Samples were collected by bottom trawl surveys in March, May, July, September and December 2011 at depths of 5–40 m. Trawling speed was 2–3 knots. Trawl net width was 25 m and cod end mesh size 17 mm. Fish specimens were identified to the species level following Meng et al. (1995). Scientific names were rechecked using the website FishBase (Froese and Pauly, 2014). Length of fish (L) was measured using a fish measuring board to the nearest mm. Sagittal otoliths were removed, cleaned, and stored dry in plastic vials with a code number. Otoliths were photographed on the convex outer side using a Nikon SMZ800 at magnifications 1, 2, 3, and 4×. Measurements (otolith length, width, long radius and short radius) were made on the otolith images by a micrometer with an accuracy of 0.01 mm. Each sagitta was weighed with a Mettler Toledo MS105DU analytical balance to an accuracy of 0.0001 g. Otolith length (OL) is the horizontal distance between the anterior and the posterior tips of the otolith (Harvey et al., 2000). Otolith width (OW) is the vertical distance from the dorsal to the ventral otolith edge (Battaglia et al., 2010). Long otolith radius (LR) was measured along the longest distance from the center of the otolith to the tip of the ventral edge (Hüssy et al., 2010). Short otolith radius (SR) is the shortest distance from the center of otolith to the tip of the ventral edge (Fig. 1). where L is fish length, β0 is the intercept, β 1 is the slope and O is the otolith dimensional measurement. anova tests were used to determine the statistical significance of the linear regressions (Norman and Streiner, 2008). In total, 2703 specimens of 33 fish species belonging to 18 families were analyzed. Sample size, ranges, and the means of L, OL, OW, LR and SR for each fish species are listed in Table 1. Sample sizes for species ranged in numbers between 16 and 429. The fitted regression models of fish size against otolith measurements for each species are presented in Table 2. anova tests revealed statistically significant linear relationships for 129 of the regressions at a significance level of P < 0.05 (Table 2); only the L-OW and L-SR regressions for Conger myriaster and the L-SR regression for Sebastiscus marmoratus were not statistically significant (P > 0.05, Table 2). In the 132 linear regression models, the coefficients of determination (R2) were larger than 0.8 in 90 cases, between 0.7 and 0.8 in 19 cases, and lower than 0.7 in 23 cases (Table 2). The simple linear regression model fitted the data poorly for Conger myriaster with considerably low R2 values (R2 < 0.1; Table 2). The relationships between fish size and otolith measurements of 33 fish species could provide useful information for the back-calculation of fish size from otolith measurements, and for further research on marine trophodynamics. Fish size–otolith measurement relationships for Callionymus beniteguri, Pseudoblennius cottoides, Rhinogobius giurinus, Tridentiger barbatus, Zebrias zebra, Pampus cinereus, and Saurida elongata were examined in Haizhou Bay for the first time (Table 2). Otoliths are a powerful tool for stomach content analysis because of their stable structures and species-specific features (Polito et al., 2011). The relationships between fish size and otolith measurements provide useful information for the back-calculation of size from otolith measurements of prey species. However, the relationships between fish size and otolith measurements in Haizhou Bay have rarely been studied. This research fills an information gap for the region that will be useful in understanding predator-prey relationships in the bay. Previous studies focused mainly on the relationship between otolith dimensions OL and OW and fish size (Harvey et al., 2000; Battaglia et al., 2010; Jawad et al., 2011; Jawad and Al-Mamry, 2012). This study provides additional information by considering four otolith measurements (OL, OW, LR, and SR). In many cases the tip or the dorsal or ventral edges of an otolith might be damaged, making it difficult to measure OL or OW accurately (Battaglia et al., 2010; Jawad et al., 2011); this could influence the reliability of subsequent calculations. Presenting the four models (L-OL, L-OW, L-LR, and L-SR) for each species helps mitigate this potential problem. It is appropriate to use the functions presented in this paper within the ranges of fish size given in Table 1. Further studies on fish size–otolith measurement relationships with a larger sample size, wider range of fish lengths, and different growth phases would help to support the analyses presented in this study. This research was funded by the Public Science and Technology Research Funds Projects of the Ocean (No. 201305030), the Specialized Research Fund for the Doctoral Program of Higher Education (No. 20120132130001) and the Fundamental Research Funds for the Central Universities (No. 201262004). We are grateful to Dr. Samuel Truesdell, Michigan State University and Jocelyn Runnebaum, University of Maine for their help in both science and English language editing. Special thanks are given to Jing Wang and Xiao Liu, Ocean University of China for their assistance in image analysis, and Dr. Yan Jiao, Virginia Tech for valuable comments on the manuscript. We also would like to thank the anonymous reviewers for their helpful and constructive comments that greatly improved the initial manuscript.