Fine-scale reconstruction of pelagic fish migration by iso-logging of eye lens

Understanding the evolution and maintenance of partial migration, in which a population includes both migratory and resident individuals, is an important challenge in ecology. However, studies of fish partial migration have exclusively focused on terrestrial and coastal ecosystems, and the difficulty of tracking highly migratory fish has hampered studies in the open ocean. Here, we performed isotope-based migration tracking (iso-logging) of 33 adult skipjack tuna (Katsuwonus pelamis) in the western Pacific by measuring carbon and nitrogen stable isotope ratios (δ13C and δ15N) in the eye lens. We compared distribution maps of δ13C and δ15N (isoscapes) with the δ13C and δ15N of individual eye lens and estimated the most likely migration routes by using a state–space model. Our results showed that i) skipjack in the western Temperate North Pacific included migrants from the western Tropical Pacific, and ii) most individuals caught in the western Tropical Pacific resided in the tropics or subtropics for their entire lives. Thus, this study is the first to observe partial migration in a highly migratory pelagic marine fish. Iso-logging offers a simpler and more cost-effective method of migration tracking than conventional bio-logging, and could provide a breakthrough for research on marine behavioral ecology.

Methods
Stable isotope data for skipjack eye lens sections (Dataset 1) were obtained through isotope analysis of 33 skipjacks caught in the western Pacific. The data include the date of capture, latitude and longitude of the capture location, and the fork length (FL) at the time of catch. The eye lens from each fish was extracted, dried under mild vacuum conditions, and sectioned serially while measuring the lens diameter. Using an equation that relates the dried lens diameter to the fork length of skipjack, the estimated fork length (Estimated FL) at the time each lens section was formed was calculated. Each lens section was analyzed with an isotope mass spectrometer (EA-IRMS) to determine the stable carbon and nitrogen isotope ratios, including the analytical error and the carbon-to-nitrogen molar ratio (C/N). The corrected δ¹³C and δ¹⁵N values were used to compare the carbon and nitrogen isotope ratios in the eye lens sections with the isoscapes. These corrections accounted for isotope ratio offsets between the eye lens and muscle tissue and adjusted for increases in isotope ratios associated with somatic growth in skipjack tuna. Dataset 2 represents the fork length and eye lens diameter of skipjack tuna (n = 88).