Climate change is precipitating large changes to a range of ecological conditions, leading to directional shifts in species’ distributions, richness, abundance, demography, and phenology (García Molinos et al., 2015; Poloczanska et al., 2013, 2016). These collective impacts of climate change are likely to alter population connectivity, a vital component of metapopulation and fishery ecology (Kool et al., 2013; Wasserman et al., 2012). Due to their reliance on seasonal resource availability in habitats that can be far apart, highly migratory species appear to be particularly susceptible to these climate change-induced consequences, which may make adaptation challenging (Robinson et al., 2009). Previous studies exploring the potential effects of climate change on population connectivity mainly focused on the pelagic larval duration, highlighting the importance of ocean fronts, currents and circulations (Bharti et al., 2022; Carson et al., 2010; Munday et al., 2009; Wolanski et al., 2021; Woodson et al., 2012). Nevertheless, evaluating those effects is more challenging for highly migratory fish, as explicit assessment of species migration potential and understanding connectivity patterns at a large spatial scale are necessary.
In marine environments, instead of reacting to changes in long-term mean conditions, responses to extreme climatic events may be more overarching (Frolicher and Laufkotter, 2018; Wernberg et al., 2012). The biggest source and control of yearly fluctuations in the climate is the El Niño Southern Oscillation (ENSO). (Cai et al., 2014; Santoso et al., 2017; Wang et al., 1999). Though characterized by recurring (2 to 7-year) oscillations between a warming and a cooling phase in tropical Pacific sea surface temperatures (SST), it is a “global pattern of anomalies” (Cane, 1986) that has strong impacts on global and regional marine ecosystems. The China Seas have shown multiple physical and biochemical responses to El Niño events, including increased SST (Ma et al., 2019), lower sea levels (Wang et al., 2018), abnormal currents (Li, 2016), weakened monsoons (Zhou et al., 2007), and southward-shifted rain bands (Zhang et al., 2017). The abundance and distribution of marine organisms were significantly impacted by this environmental variation, for instance, phytoplankton showed significant decreased correlated with elevated SST (Liu et al., 2019) while some small pelagic fish increased substantially in biomass (Ma et al., 2019). By impacting physiological processes or disrupting biotic interactions (e.g. predator–prey interactions), ENSO events are likely to cause great changes in population connectivity of migratory fish in China Seas.
Japanese Spanish mackerel, Scomberomorus niphonius, is widely distributed in the temperate waters of northwestern North Pacific, supporting an economically valuable commercial fishery in China, Japan and Korea (Qiu and Ye, 1996; Horikawa et al., 2001; Shoji and Tanaka, 2005). S. niphonius undertakes long, seasonal migrations and has been observed to move into coastal waters of China Seas during warmer seasons to breed and spawn, and move back to deeper waters in the cooler seasons (Fig. 1). The traditional acknowledgement claimed that there are two populations along the coast of China: (1) the Yellow Sea and Bohai Sea population; and (2) the East China Sea population (Horikawa et al., 2001). However, most recent studies applying advanced technologies like mitochondrial DNA analysis (Shui et al., 2009), otolith phenotypic analysis (Zhang et al., 2016) and otolith chemistry analysis (Pan et al., 2020b) provided evidence for the existence of a metapopulation and large-scale connectivity between the Yellow Sea and the East China Sea. Since the 1990s, the distribution of S. niphonius showed a northward expansion with increasing water temperature (Fujiwara et al., 2013; Yang et al., 2022). Furthermore, the spawning grounds of S. niphonius would move northward when SST was high (Wan et al., 2020). Considering that the 2015–2016 super El Niño has spread its impacts to the ecosystem of China Seas (Yin et al., 2021), the climate-induced range shifts and migration alteration are probably influence the population connectivity of S. niphonius.
Despite this breadth of background, the complex life histories of marine migratory fish make the investigations of their responses to extreme climate events difficult and rare. Otolith biogeochemistry offers a potentially efficient and cost-effective means to evaluate the impacts of ENSO events on population connectivity (Walther 2019; Reis-Santos et al. 2022; Pan et al., 2020a). As they continuously accrete, trace and minor metals from the local environment are permanently incorporated into the crystalline matrix of the otolith. Otolith increments are unlikely to be subject to resorption due to the metabolic inertness (Powles et al., 2006). Although otolith microchemistry can represent a combination of local ambient chemistry and individual physiology, the resulting elemental composition can produce a distinctive chronological ‘signature’ that can be used as a natural tag (Campana and Thorrold, 2001; Elsdon et al., 2008) to distinguish location and infer ontogenetic change. This approach has been successfully applied to identify the population connectivity and natal origin in the highly migratory species, such as Chilean jack mackerel (Trachurus murphyi), bigeye tuna (Thunnus obesus) and yellowfin tuna (Thunnus albacares) (Ashford et al., 2011; Rooker et al., 2016). In regions where there is a paucity of long-term data, fish otoliths have been increasingly used to evaluate the climatic effects on fish populations (Lee and Punt, 2018; Reis-Santos et al., 2021). However, studies using otolith biogeochemistry to evaluate the impacts of extreme climate events on population connectivity are still limited.
In this study, we assessed the population connectivity of S. niphonius collected from the main spawning grounds throughout its distribution in the Yellow Sea and East China Sea. By analyzing the otolith biogeochemistry signatures of samples from 2016 to 2018, during which 2015–2016 was a strong ENSO year, we aimed to evaluate the temporal variations in population connectivity and figure out the relationship with the ENSO events. We first compared the chemistry of the whole life to examine empirically how otolith chemistry varies spatially and temporally. Then we focused on the larval and adult stages, using random forest classification and clustering to identify the natal origins and to see if the source-sink mechanism differed greatly for the ENSO years. Our study will help identify potential changes in population connectivity of highly migratory species in response to extreme climate events and highlight the further application of otolith biogeochemistry to explore the climate-induced impacts on fish population.