Journal of Oceanology and Limnology   2023, Vol. 41 issue(6): 2444-2450     PDF       
http://dx.doi.org/10.1007/s00343-022-2279-7
Institute of Oceanology, Chinese Academy of Sciences
0

Article Information

LI Xiaolu, ZHANG Chi, TIAN Yongjun, LIN Longshan, LIU Shigang
Early life history traits of chub mackerel Scomber japonicus in the Oyashio water revealed by otolith microstructure
Journal of Oceanology and Limnology, 41(6): 2444-2450
http://dx.doi.org/10.1007/s00343-022-2279-7

Article History

Received Jul. 12, 2022
accepted in principle Oct. 12, 2022
accepted for publication Oct. 31, 2022
Early life history traits of chub mackerel Scomber japonicus in the Oyashio water revealed by otolith microstructure
Xiaolu LI1, Chi ZHANG1, Yongjun TIAN1,2, Longshan LIN3, Shigang LIU3     
1 Research Centre for Deep Sea and Polar Fisheries, and Key Laboratory of Mariculture, Ministy of Education, Ocean University of China, Qingdao 266100, China;
2 Frontiers Science Center for Deep Ocean Multispheres and Earth System(FDOMES), Ocean University of China, Qingdao 266100, China;
3 Third Institute of Oceanography, Ministry of Natural Resources, Xiamen 361000, China
Abstract: Information on survival and growth during the early life stage is essential to understand the mechanism of interannual variations in fish recruitment. Chub mackerel Scomber japonicus is a commercially important pelagic fish widely distributed in the northwestern Pacific. Its catch showed large fluctuations with changes in distribution and migration under climate change and strong fishing. We determined the hatch dates and growth rates of young-of-the-year of chub mackerel through otolith microstructure using samples collected in the Oyashio water in autumn 2018. Results show that the ages of young chub mackerel ranged between 120 and 180 d, and the estimated hatch date lasted from mid-January to late May with a peak from mid-March to mid-April. Average otolith daily increment width during the early life stages (from hatching to 25 d) showed an increasing trend. Chub mackerel grows slowly in the first 10 d, and then grows faster during the 10th to 25th d. Three groups with dissimilar growth histories and migration routes were identified using unsupervised random forest clustering analysis, but all eventually converge on the same nursery ground. The faster growth of young-of-the-year chub mackerel leads to better recruitment due to the hypothesis of growth-dependent mortality. Most chub mackerels hatched in March and April, the spawning period is longer and earlier, which could lead to strong year classes. These findings on population composition and life history traits of young-of-the-year of chub mackerel provide valuable information on its recruitment processes during the period of stock recovery.
Keywords: Scomber japonicus    otolith microstructure    growth history    unsupervised random forest clustering    recruitment    
1 INTRODUCTION

Chub mackerel Scomber japonicus is a commercially important small pelagic fish widely distributed in the northwest Pacific Ocean (Watanabe and Yatsu, 2006), and they can be divided into two stocks: Pacific stock and Tsushima Current stock (Yu et al., 2018). The Pacific stock of chub mackerel mainly spawns from April to June around the Izu Island and Boso Peninsula (Fig. 1). Larvae and early juveniles are transported offshore from April to September by the Kuroshio and Kuroshio extension to a large region stretching from the Kuroshio-Oyashio transition region to the Oyashio region. Then juveniles begin to migrate to the coastal waters of northern Japan and first reach sexual maturity at 250-mm fork length, then become the targets of fisheries (Takahashi et al., 2014; Kamimura et al., 2015; Go et al., 2020). As the most productive area for capture fisheries (FAO, 2016), the northwest Pacific supports the main stock of chub mackerel (Shiraishi et al., 2009). However, since the highest record in 1978 (4 742× 103 t), the total biomass of the Pacific stock dramatically decreased and was lowest in 2001 (151×103 t). The recruitment of chub mackerel also showed a trend similar to the total biomass, it fluctuated between 175 million (1998) and 14 206 million (1979) for age-0 fish (Kawai et al., 2002). However, the factors responsible for the dramatic fluctuations in the biomass of chub mackerel in the northwest Pacific since the 1970s have not been suitably elucidated (Wang et al., 2021). Despite strong fishing, recruitment and biomass have increased in recent years (Kamimura et al., 2015). With the large fluctuations in biomass, the biological characteristics of chub mackerel have also changed correspondingly. Kanamori et al. (2019) found that the spawning period extended and the geographic location of the spawning ground moved northward after the 2000s. Such phenological and distributional shifts in reproduction could influence subsequent early life history and recruitment success since larvae growth and survival are determinant factors (Kamimura et al., 2015). Therefore, information on the processes associated with survival during the early stage of chub mackerel is essential for understanding mechanisms of interannual variation in recruitment.

Fig.1 Schematic illustration of the current systems and main spawning ground of chub mackerel on the Pacific side of Japan The survey stations in 2018 are marked by black spots.

Otolith microstructure analysis is an essential tool in various fields of fish biology studies (Campana and Thorrold, 2001), which makes it possible to track early life histories, such as hatch date and early growth rate, and to identify stock composition with a different growth rate from a mixed stock (Brophy and Danilowicz, 2002; Clausen et al., 2007). Takahashi et al. (2014) standardized the aging method of juvenile chub mackerel and confirm the daily pattern for increment information. The technical progress led to significant advances in studies of the early recruitment process. Using otolith daily increments analysis, Higuchi et al. (2019) recovered the temperature history of chub mackerel during the early life stage and revealed that the chub mackerel population could be divided into several groups, in which individuals with high initial growth proactively enter cooler water temperature areas, accessing a highly nutritious diet resulting in further rapid growth. Taga et al. (2019) implied that faster growth of larvae led to better recruitment due to growth-dependent mortality. These studies deepened our understanding of the early life history of chub mackerel, but the data of Higuchi was not uniform for one year and the proportion that each migration group occupies in the nursery ground is not understood yet. The differing growth histories could result in different survival rates and affect the recruitment success consequently. Besides, little ecological information is available for chub mackerel, especially for young-of-the-year (YOY) in the nursery ground in the Oyashio water.

In this study, based on YOY chub mackerel collected in Oyashio waters, we tracked the early life history of chub mackerel by otolith microstructure analysis. According to otolith increment during the early life stage, potential groups with different growth patterns were identified using random forest clusters. Among the identified groups, early growth characteristics and hatch date distribution were further compared. The information on population composition could benefit the estimation of chub mackerel survival among groups and contribute to a better understanding of the recent population dynamics.

2 MATERIAL AND METHOD 2.1 Sample collection

Young-of-the-year chub mackerel were collected from nekton light-purse seine surveys conducted in the high seas area of the Northwest Pacific Ocean in the autumn of September 2018, in an area ranging from 146°E to 160°E and 35°N to 45°N (Fig. 1). The survey vessels were light-purse seine vessels, 50.5 m in length, 9.8 m in width, 4.6 m in depth, all with a gross tonnage of 830 t. The vessels had a main engine power of 800 kW and an auxiliary engine power of 698 kW. Each vessel is equipped with 110 fish-collecting lights with a power of 4 kW, a maximum mesh size of 4.5 cm, and a minimum mesh size of 3.5 cm. After arriving at the survey station, the fish were collected with lights on for 2 h. Individuals were randomly taken from each survey sample, preserved by freezing (-20 ℃), and thawed at room temperature just before analysis. The fork length (FL) of each juvenile was measured to accuracy of 1 mm (Table 1).

Table 1 Specification of the surveys of young-of-the-year of chub mackerel Scomber japonicus in the western North Pacific
2.2 Otolith measurement

Sagittal otoliths were dissected out, cleaned with pure water and air-dried. One of the otoliths was selected for microstructure analysis and embedded in epoxy resin. Then otoliths were sliced by a low-speed cut (BUEHLER IsoMet with a diamond-edged wafering blade, USA) with a thickness of 0.5 mm along the long axis. Both sides of otolith slices were ground with sandpaper (600–4 000 grit) on the frontal plane (BUEHLER MetaServ 250, USA), and polished with alumina polishing suspension (0.3 μm) until the primordium and daily increments can be clearly identified (Takahashi et al., 2014).

The polished otolith slices were then photographed with microphotographic equipment (OLYMPUS BX53 and DP74, Japan) at 200× magnification (Fig. 2). Total of 136 YOY samples (131–243 mm) were aged and used for the otolith microstructure analysis. Using ImageJ software, the number of daily increments was counted twice, with a 20-d interval for cross-validation. Only otolith slices for which the counting error was < 10% were included in the subsequent analysis. Because secondary growth primordium is produced on otolith at around 25 d and to focus on the early growth history of YOY chub mackerels (Fig. 2), otolith increments from hatching to age 25 d were measured. Increment width was measured along the axis from the nucleus to the posterior margin of the otolith. Daily-age of each chub mackerel was calculated by adding 3 to the average number of otolith daily increments because the first daily ring was usually deposited 3 d after hatch (Takahashi et al., 2014). Hatching dates were back-calculated based on the sampling dates and daily ages.

Fig.2 Otolith microstructure on the frontal plane from a 152-day individual
2.3 Clustering analysis

Unsupervised random forest (URF) clustering analysis was applied to detect potential groups of chub mackerel with different migration routes for the main spawning season (March–May, see Section 3). Random forests bootstrap the data from the samples from the training set with randomly selected feature subsets that were evaluated at each node of the decision tree, while the final decision is made by decision fusion of all trees by majority voting (Kandaswamy et al., 2011). It is an unsupervised classification approach that, by generating an ensemble of individual tree predictors, leads to a measure of natural dissimilarity between the observations (Breiman, 2001; Seligson et al., 2005). Classifier ensembles promote an optimal trade-off between diversity and accuracy. URF has high classification accuracy, good robustness to noise and outliers, and strong generalization ability. It contains multiple decision trees trained by the Bagging ensemble learning technology; the final classification result is voted by the output result decided by a single decision tree. With the R package "Random Forest" (Liaw and Wiener, 2002; Shi and Horvath, 2006), larval increments characterizing the juvenile growth rate that is important for recruitment were selected as clustering factors and 125 individuals were clustered for analysis. Because of the existence of growth-dependent mortality and the part of the lifespan proved to be important for recruitment (Higuchi et al., 2019; Taga et al., 2019), only increments during 10–25 d were used. Before clustering, we determined the number of clusters using the R package "cluster". Among the identified clusters, growth and hatch date differences were compared by repeated-measurement ANOVA because of the autocorrelations resulting from repeated measurements for each individual.

3 RESULT 3.1 Hatch date

For the 136 fish samples, the estimated hatch dates ranged from January 14 to May 22, 2018. Hatch dates were concentrated to March and April, accounting for approximately 85.3% of all the individuals (Fig. 3).

Fig.3 Hatch date distribution in 2018 for the young-of-the-year of chub mackerel in the Oyashio waters
3.2 Otolith daily increment width through early life stages

Average otolith daily increment width during the early life stages showed an increasing trend (Fig. 4). For all samples, it increased from 1.5 μm to the maxima of 21.1 μm. Chub mackerel otoliths grew slowly in the first 10 d, after that they grew faster during the 10th to 20th d. The among-individuals difference in growth became greater with increasing age.

Fig.4 Changes in average otolith daily increment width during the early life stage for the whole samples Error bars indicate the range of otolith daily increment width. Error bars: 95% confidence intervals.
3.3 Group identification

Three groups of chub mackerel with a significant difference in growth (P < 0.001) were identified by URF clustering, with 40, 38, and 47 individuals in each, respectively (Fig. 5). No significant difference in the first 10 d's average otolith daily increment width was detected among the three groups, but the growth rate was significantly different after 10 d. Group 1 (G1) showed a faster growth trend, Group 3 (G3) showed a relatively slower growth trend and Group 2 (G2) was the slowest one.

Fig.5 Changes in mean otolith daily increment width in the early life stage for young-of-the-year chub mackerel in the three groups G1 (n=40), G2 (n=38), and G3 (n=47). n is the number of samples tested. Error bars: 95% confidence intervals.

No significant difference in hatch dates was detected among the three groups (Fig. 6). Hatch dates of G1 mackerels ranged from March 1 to May 22, and the peak was in early April (32.5%). G2 finished hatching by early May (March 1–May 4) and most hatched in late March (31.6%) (Fig. 6). Hatch dates of G3 mackerels ranged from March 2 to May 7, and the peak was in April (53.2%).

Fig.6 Frequency distribution of hatch dates of chub mackerel in 2018 by group (G1, G2, and G3) n is the number of samples tested.
4 DISCUSSION

This study provided information on early growth history and hatching date for YOY chub mackerel in the feeding ground of Oyashio water. Our results showed that chub mackerel hatched from January to May with a peak in March and April. The spawning of the Pacific stock of chub mackerel mainly occurs from April to June in the coastal waters around the Izu Islands, the main spawning ground, and off southern Japan (Watanabe, 1970; Usami, 1973; Murayama et al., 1995; Watanabe and Yatsu, 2006). Long-term survey data (40 years) on egg density showed that the spawning period of chub mackerel in the Izu Islands lasting from March to June, was extended in response to increasing SST, but not advanced (Kanamori et al., 2019). We collected some individuals hatched in January and February. They probably came from spawning grounds farther south, where spawning started earlier (Yukami et al., 2009). Most chub mackerels hatched in March and April, the spawning period is longer and earlier than that of Kamimura et al. (2015) from 2002 to 2011. An earlier spawning period could lead to strong year classes (Kamimura et al., 2015). Therefore, our result could lend higher reliability in interpreting spawning events in 2018. The detected earlier spawning period is consistent with the strong recruitment of the 1970s (Takahashi et al., 2014). This finding provides further evidence for stock recovery and depicts a strong year-class.

For the main spawning cohorts, three groups with different growth histories were successfully identified by URF. Higuchi et al. (2019) separated YOY individuals collected in the Kuroshio-Oyashio transition area for 10 years survey into 6 groups based on otolith increments, and then further classify the six groups as three groups with similar size (i.e., fast growth with high δ18O and slow growth with low δ18O, as higher δ18O mean values indicate cooler experienced temperatures, given constant δ18O of in situ seawater) combined with stable oxygen isotope analysis. The otolith increments trend and proportion for the three groups identified in this study are consistent with the result of Higuchi et al. (2019). The result could be helpful to understand the speculation of the migration route of chub mackerel. Based on the period studies (Higuchi et al., 2019; Nakamura et al., 2020), we speculate that the group with high growth (G1) proactively enters cooler water in the Kuroshio-Oyashio transition area experiencing lower water temperature by about 2 ℃, the rest groups would migrate eastwards along the Kuroshio extension and then move northwards into the transition area (Fig. 7). The three different migratory groups all eventually converge on the same nursery ground in the Oyashio waters.

Fig.7 Sea surface water temperature (T) off Japan in March (a), April (b), and May (c), 2018 Red curves are presumed migratory routes (data source: https://www.jma.go.jp/jma/menu/menureport.html; Higuchi et al., 2019).

Various factors could influence the early growth of chub mackerel in different ways, especially temperature and prey density (Sassa and Tsukamoto, 2010; Kanamori et al., 2019; Sogawa et al., 2019; Taga et al., 2019). Unfortunately, there is no more uniform conclusion on this research. Since no significant difference in hatch dates was detected among the three groups, we speculate that the growth difference could be caused by spatial environmental heterogeneity. As Guo et al. (2022) have indicated, the temperature did not have a direct significant effect on growth rates, but coastal temperature changes in the Kuroshio axis can lead to changes in the zooplankton community, which in turn can affect growth rate and replenishment of chub mackerel. Sogawa et al. (2019) pointed out that the environmental variation within the spawning grounds was small, but because of the water mass, the distribution of eggs and larvae showed obvious patchiness, and there was a clear transport process. These further match growth to environmental factors. The daily increments of our samples in the first 10 d are similar to the result of Taga et al. (2019). We add the missing link in the middle because the individuals in his study were 1-year-old adult or juvenile fish and the individuals in our study are YOY chub mackerel. Besides, the change in otolith daily increment width (Fig. 4) is similar to the high recruitment per spawning year class (2016) in the study of Taga et al. (2019), we further confirmed that faster growth of YOY chub mackerel leads to better recruitment due to the hypothesis of growth-dependent mortality. According to the result of Taga et al. (2019), the surviving individuals all had otolith daily incremental growth of more than 10 μm after 10 d and eventually became supplemental individuals. Based on the average otolith daily increment width (Fig. 4), although there are some differences in the daily incremental growth in the latter day, the otolith daily incremental growth of our study was more than 10 μm after 10 d. Therefore, we can infer that this is the threshold for early survival, which gives us more insight into better understanding replenishment and survival.

5 CONCLUSION

Our study provided valuable information on the early growth history and composition of the chub mackerel in the Oyashio waters. Results present good evidence of the current-recovering condition of the chub mackerel stock. The findings contribute to a better description of the recruitment process during a strong recruitment period. We systematically described the recruitment concept of chub mackerel and gained more insight into the origin of the groups. However, the exact migratory route and the water temperature environment they experienced are not yet known. Our subsequent studies will be supplemented with otolith stable oxygen isotopes. In addition, subsequent studies will aim to examine growth studies over a wider range of years and larger sample sizes to prove the effectiveness of our methodology. Moreover, Future research could focus on systematic research and study of spawning grounds, estimation of overall complement, mortality, and main spawning ground area.

6 DATA AVAILABILITY STATEMENT

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

7 ACKNOWLEDGMENT

We acknowledge the crew members with the Zhongtai Oceanic Fishery Co. for their help during the survey.

References
Breiman L. 2001. Random forests. Machine Learning, 45: 5-32. DOI:10.1023/A:1010933404324
Brophy D, Danilowicz B S. 2002. Tracing populations of Atlantic herring (Clupea harengus L.) in the Irish and Celtic Seas using otolith microstructure. ICES Journal of Marine Science, 59(6): 1305-1313. DOI:10.1006/jmsc.2002.1304
Campana S E, Thorrold S R. 2001. Otoliths, increments, and elements: keys to a comprehensive understanding of fish populations?. Canadian Journal of Fisheries and Aquatic Sciences, 58(1): 30-38. DOI:10.1139/f00-177
Clausen L A W, Bekkevold D, Hatfield E M C, et al. 2007. Application and validation of otolith microstructure as a stock identification method in mixed Atlantic herring (Clupea harengus) stocks in the North Sea and western Baltic. ICES Journal of Marine Science, 64(2): 377-385. DOI:10.1093/icesjms/fsl036
FAO. 2016. The State of World Fisheries and Aquaculture 2016. New York: FAO. p. 200.
Go S, Lee K, Jung S. 2020. A Temperature-Dependent Growth Equation for Larval Chub Mackerel (Scomber japonicus). Ocean Science Journal, 55: 157-164. DOI:10.1007/s12601-020-0004-z
Guo C Y, Ito S I, Kamimura Y, et al. 2022. Evaluating the influence of environmental factors on the early life history growth of chub mackerel (Scomber japonicus) using a growth and migration model. Progress in Oceanography, 206: 102821. DOI:10.1016/j.pocean.2022.102821
Higuchi T, Ito S I, Ishimura T, et al. 2019. Otolith oxygen isotope analysis and temperature history in early life stages of the chub mackerel Scomber japonicus in the Kuroshio-Oyashio transition region. Deep Sea Research Part Ⅱ: Topical Studies in Oceanography, 169-170: 104660. DOI:10.1016/j.dsr2.2019.104660
Kamimura Y, Takahashi M, Yamashita N, et al. 2015. Larval and juvenile growth of chub mackerel Scomber japonicus in relation to recruitment in the western north Pacific. Fisheries Science, 81(3): 505-513. DOI:10.1007/s12562-015-0869-4
Kanamori Y, Takasuka A, Nishijima S, et al. 2019. Climate change shifts the spawning ground northward and extends the spawning period of chub mackerel in the western north Pacific. Marine Ecology Progress Series, 624: 155-166. DOI:10.3354/meps13037
Kandaswamy K K, Chou K C, Martinetz T, et al. 2011. AFP-Pred: a random forest approach for predicting antifreeze proteins from sequence-derived properties. Journal of Theoretical Biology, 270(1): 56-62. DOI:10.1016/j.jtbi.2010.10.037
Kawai H, Yatsu A, Watanabe C, et al. 2002. Recovery policy for chub mackerel stock using recruitment-per-spawning. Fisheries Science, 68(5): 963-971. DOI:10.1046/j.1444-2906.2002.00520.x
Liaw A, Wiener M. 2002. Classification and Regression by randomForest. R News, 23(23): 18-22.
Murayama T, Mitani I, Aoki I. 1995. Estimation of the spawning period of the Pacific mackerel Scomber japonicus based on the changes in gonad index and the ovarian histology. Bulletin of the Japanese Society of Fisheries Oceanography (Japan), 59(1): 11-17.
Nakamura M, Yoneda M, Ishimura T, et al. 2020. Temperature dependency equation for chub mackerel (Scomber japonicus) identified by a laboratory rearing experiment and microscale analysis. Marine and Freshwater Research, 71(10): 1384-1389. DOI:10.1071/MF19313
Sassa C, Tsukamoto Y. 2010. Distribution and growth of Scomber japonicus and S. australasicus larvae in the southern East China Sea in response to oceanographic conditions. Marine Ecology Progress Series, 419: 185-199. DOI:10.3354/meps08832
Seligson D B, Horvath S, Shi T, et al. 2005. Global histone modification patterns predict risk of prostate cancer recurrence. Nature, 435(7046): 1262-1266. DOI:10.1038/nature03672
Shi T, Horvath S. 2006. Unsupervised learning with random forest predictors. Journal of Computational and Graphical Statistics, 15(1): 118-138. DOI:10.1198/106186006X94072
Shiraishi T, Ketkar S D, Katoh Y, et al. 2009. Spawning frequency of the Tsushima current subpopulation of chub mackerel Scomber japonicus off Kyushu, Japan. Fisheries Science, 75(3): 649-655. DOI:10.1007/s12562-009-0094-0
Sogawa S, Hidaka K, Kamimura Y, et al. 2019. Environmental characteristics of spawning and nursery grounds of Japanese sardine and mackerels in the Kuroshio and Kuroshio Extension Area. Fisheries Oceanography, 28(4): 454-467. DOI:10.1111/fog.12423
Taga M, Kamimura Y, Yamashita Y. 2019. Effects of water temperature and prey density on recent growth of chub mackerel Scomber japonicus larvae and juveniles along the Pacific coast of Boso-Kashimanada. Fisheries Science, 85(6): 931-942. DOI:10.1007/s12562-019-01354-8
Takahashi M, Yoneda M, Kitano H, et al. 2014. Growth of juvenile chub mackerel Scomber japonicus in the western north Pacific Ocean: with application and validation of otolith daily increment formation. Fisheries Science, 80(2): 293-300. DOI:10.1007/s12562-013-0698-2
Usami S. 1973. Ecological studies of life pattern of the Japanese mackerel, Scomber japonicus Hottuyn. On the adult of the Pacific subpopulation. Bulletin of Tokai Regional Fisheries Research Laboratory, 76: 71-178.
Wang L M, Ma S Y, Liu Y, et al. 2021. Fluctuations in the abundance of chub mackerel in relation to climatic/oceanic regime shifts in the northwest Pacific Ocean since the 1970s. Journal of Marine Systems, 218: 103541. DOI:10.1016/j.jmarsys.2021.103541
Watanabe C, Yatsu A. 2006. Long-term changes in maturity at age of chub mackerel (Scomber japonicus) in relation to population declines in the waters off northeastern Japan. Fisheries Research, 78(2-3): 323-332. DOI:10.1016/j.fishres.2006.01.001
Watanabe T. 1970. Morphology and ecology of early stages of life in Japanese common mackerel, Scomber japonicus Houttuyn, with special reference to fluctuation of population. Bulletin of Tokai Regional Fisheries Research Laboratory, 62: 1-283.
Yukami R, Ohshimo S, Yoda M, et al. 2009. Estimation of the spawning grounds of chub mackerel Scomber japonicus and spotted mackerel Scomber australasicus in the East China Sea based on catch statistics and biometric data. Fisheries Science, 75(1): 167-174. DOI:10.1007/s12562-008-0015-7
Yu W, Guo A, Zhang Y, et al. 2018. Climate-induced habitat suitability variations of chub mackerel Scomber japonicus in the East China Sea. Fisheries Research, 207: 63-73. DOI:10.1016/j.fishres.2018.06.007