On habitat changes of Cololabis saira based on 

carbon and nitrogen stable isotopes technology

Abstract: Cololabis saira is a highly migratory pelagic coldwater fish. It is a key species in the ecosystem and one of the important fishing targets in the Pacific Ocean. According to Cololabis saira samples collected in open seas of the Northwest Pacific Ocean near 40°28′N48°56′N, 150°08′E164°20′E by the Cololabis saury fishing boat “Lu Peng Yuanyu 027” from July to November 2018, 127 muscle tissues were taken for stable isotope determination, and δ13C (carbon stable isotope ratio) and δ15N (nitrogen stable isotope ratio) of the muscle were analyzed, and the growth process of δ13C and δ15N was discussed. The range of δ13C in muscle of Cololabis saira was between -23.58% and -20.69 ‰, the average was (-21.71±0.69) ‰, the range of δ15N in muscle of Cololabis saira was between 6.49% and 11.11 ‰, the average was (8.51±1.00)‰. Compared with previous results, it was found that δ13C and δ15N values of Cololabis saira in this study were lower than those of previous studies, which might be because the latitude and offshore distance of the sea area where Cololabis saira were located were quite different, leading to differences in the baseline values of δ13C and δ15N, and thus the differences in δ13C and δ15N of Cololabis saira muscles at different sampling points. Analysis of variance (ANOVA) showed that there was no significant difference between male and female saury in δ13C and δ15N (P>0.05). There were significant differences among different body lengths in δ13C and δ15N (P<0.05). The range of body length of Cololabis saira was between 244 mm and 307 mm, the average value of body length was (281±13) mm. In combination with the relationship between body length and age of Cololabis saira , it could be found that the age range of Cololabis saira samples in this study was between 385 d and 473 d, so it could be considered that samples in this study were mainly  1 year old. The changes of δ13C and δ15N between different body length groups were relatively similar, and showed a trend of increasing first and then stabilizing. In 240260 mm, 260280 mm and 280300 mm body length groups, δ13C and δ15N increased with the increase of body length. These results indicated that δ13C increased gradually during individual growth, and the trophic level also increased. The δ13C of marine phytoplankton decrease from equatorial to two poles (low latitude to high latitude). Therefore, 1yearold saury may migrate from north to south near the Oyashio waters along with individual growth and habitat changes. In the range of body length from 280300 mm to >300 mm, δ13C and δ15N almost remained unchanged, and a plateau appeared. For adjacent body length groups, the same δ13C value meant that the primary productivity level (δ13C value of baseline phytoplankton) was similar, and the same δ15N value meant that the trophic level was similar. Therefore, it could be inferred that the body length of Cololabis saira was 280307 mm and they inhabited the same sea area without longdistance migration. To sum up, the results indicated that 1yearold saury inhabited the Oyashio waters, and then migrated south to the KuroshioOyashio mixed waters. Based on the collinearity test results of the prediction variables, a model of fifteen different variable combinations was established for four prediction variables, and the model with the minimum AIC value was selected as the best fitting model. The best fitting model of δ13C was BL+Lat+DSB+chla, and these variables had significant effect on the δ13C of muscle of Cololabis saira (P<0.05), deviation explanation rate of the model was 56.8%. The best fitting model of δ15N was Lat+DSB, and these variables had significant effect on the δ15N of muscle of Cololabis saira (P < 0.05), deviation explanation rate of the model was 45.8%. Based on the generalized additive model (GAM), the relationship between δ13C, δ15N and latitude, body length, distance from shore, and chlorophyll was established respectively based on the generalized additive model (GAM). The variation trend of δ13C and δ15N values with explanatory variables was significant. Based on the results of GAM models, δ13C was negatively correlated with latitude, it increased slightly in the range between 40° and 42°N, but decreased significantly in the range between 42° and 48°N. This was mainly because isotopic fractionation was one of the factors affecting the variation of δ13C, and fractionation existed in high latitude sea area, resulting in low δ13C value. δ13C was positively correlated with body length. It could be considered that Cololabis saira at this stage migrated and inhabited in different sea areas during their growth. To sum up, we inferred that 1yearold saury in the Oyashio waters migrated southward from high latitude to low latitude with the growth of individual. δ13C first decreased and then increased with the increase of the distance from the shore, a minimum occured at the nearshore distance of 800 km. This might be because the nutrients brought by the Oyashio extension current and other sea currents improved the primary productivity of the sea area, resulting in a higher δ13C level of baseline organisms, which had a positive effect on the δ13C value of Cololabis saira. The variation trend of δ13C with chlorophyll was complex. The δ13C had a positive correlation among 00.5 mg·m3 and among 1.52.5 mg·m3, and a negative correlation among 0.51.5 mg·m3. These results indicate that the δ13C value of baseline organisms is not only related to chlorophyll content, but also may be caused by low sea surface temperature, low inorganic nutrient content or “iron” restriction; δ15N is negatively correlated with latitude; δ15N decreases first, then slowly increases, and then continues to decrease with the increase of the nearshore distance.The variation of δ15N in latitude and offshore distance indicates that 1yearold saury migrates highly and inhibits different sea areas.

Keywords: Cololabis saury; habitat; δ13C；δ15N; GAM