Northward shift of a biogeographical barrier on China’s coast

Understanding the formation and maintenance of biogeographical breaks is fundamental for developing analyses related to biodiversity and conservation. Biogeographical patterns along China's coast are changing dramatically in the face of climate change and alterations in land‐use. In this paper, we sought to clarify the mechanisms responsible for the formation and maintenance of a biogeographical barrier on China's coast.


| INTRODUC TI ON
Biogeographical breaks, defined as regions where multiple species reach their ranges limit simultaneously, can lead to changes in species composition and species richness, and thus are important for the formation and maintenance of biogeographical patterns (Costello et al., 2017;Spalding et al., 2007). Great uncertainty remains about the roles of climate change and human land-use activities in the formation and maintenance of biogeographical barriers (Ayre & Rosser, 2021;Waters, 2008).
Along China's coast, the Yangtze (Changjiang) River Estuary (30°-31°N) has been regarded as a break that separates the North Pacific Temperate Biotic Region (Cold Temperate Northwest Pacific Province) and the Indo-West Pacific Warm-water Biotic Region (the Warm Temperate Northwest Pacific Province; Liu, 2013;Spalding et al., 2007). The formation of this biogeographical barrier has been attributed to complex interactions among historic events, Yangtze River Diluted Water, coastal circulation patterns, climatic factors and substrate types Dong et al., 2012;He et al., 2010;Ni et al., 2017;Qu et al., 2021;Wang et al., 2015).
In the last decades, the biogeographical barrier for some intertidal macrobenthos has moved northward to ~33°-34°N and has formed a new biogeographical barrier, the Subei Biogeographical Barrier (SBB).
Mechanisms promoting the northward shift of the biogeographical barrier are mainly associated with oceanographic features, seascape transformation and climatic factors Wang et al., 2020). These abiotic changes can consequently affect larval dispersal, larval settlement and post-settlement survival of the marine macrobenthos (Cowen & Sponaugle, 2009;D'Aloia et al., 2015;O'Connor et al., 2007;Sassa et al., 2006;Wilson et al., 2018). The northward shift of the biogeographical barrier and the formation of the SBB reveal the changes of species distribution limits and biogeographical patterning in the context of climate change and human land-use activities.

| Biogeographical and phylogeographical references search
To investigate the spatiotemporal changes of community structures and population genetic patterns of intertidal macrobenthos, literature related to biogeographical and phylogeographical patterns of intertidal macrobenthos along the coast of Jiangsu Province and adjacent areas has been searched and downloaded from Web of Science and Google scholar using the following general query: TS = "Jiangsu biogeography*" or "China* biogeography*" or "Jiangsu population connectivity*" or "China* population connectivity*" or "Jiangsu phylogeography*" or "China* phylogeography*" or "Yangtze (Changjiang) River barrier*" or "Yangtze (Changjiang) River freshwater" or "Yangtze (Changjiang) River Diluted Water" or "China* artificial structure*." (TS denotes a search for "Topic," * is a regular expression used to match all words including that string of characters).

| Species life history references collection
To clarify the environmental factors controlling the shift of the biogeographical break, literature was search using the following general query: TS = ("species name" or "genus name") AND TS= ("China* life history*" or "China* life character*" or "China* larva* dispersal" or "China* larva* settle*" or "China* larva* post-settle*" or "China* postsettle* establish*" or "China* population establish*". (TS denotes a search for "Topic," * is a regular expression used to match all words including that string of characters).

| Coastal biogeographical pattern and biogeographical barriers
Coastal biogeographical patterns are characterized by varying areas, ranging from regional scales to the global ocean (Costello et al., 2017;Liu, 2013;Spalding et al., 2007). The classification of these patterns is helpful in many ways, notably for development of marine reserves and assessment of biodiversity (Costello et al., 2017;Spalding et al., 2007). Coastal biogeographical barriers are ubiquitous across the globe. For instance, on the southeast coast of Australia, a biogeographical barrier called the Southeast Australian Biogeographic Barrier (SEABB) occurs, which plays an important role in shaping species distribution and phylogeographic patterns (Ayre et al., 2009;Ayre & Rosser, 2021;Waters, 2008). Similar biogeographical barriers occur in the southeast Pacific on the northern coast of Peru (Barahona et al., 2019), the west coast of the United States (Burton, 1998;Palumbi, 1996;Taylor & Hellberg, 2006), the east coast of North America (Pappalardo et al., 2015), the Atlantic-Mediterranean coast (El Ayari et al., 2019), and the coast of China (Dong et al., 2012;Liu, 2013). These barriers are not hard and fast, however, but shift when environmental changes occur. Thus, under the influences of climate change and human activities that modify shorelines, many species have managed to cross these biogeographical barriers and enter regions in which they formerly were rare, if not absent altogether (Adams et al., 2014;Bulleri & Airoldi, 2005;Dong et al., 2016;Miller et al., 2013;Wang et al., 2020).

| Habitat continuity
Habitat continuity is an important driver shaping the biogeographic boundaries of species (Buonomo et al., 2017;Fenberg & Rivadeneira, 2019;Fraser et al., 2012;Platts et al., 2019). For example, along the eastern Pacific rocky shore, habitat continuity is a top predictor of biogeographic structure and the richness gradient of gastropods. In the extratropical regions rocky shore habitat continuity is low, species turnover is relatively low, emphasizing the importance of habitat continuity on biogeographical patterns and processes (Fenberg & Rivadeneira, 2019).
The muddy tidal flat habitat is a dominant coastline feature on the north region of the Yangtze River Estuary up to the northern shoreline of Jiangsu Province (~35°N). The absence of exposed natural rocky shore habitat has impeded mass settlement of rocky intertidal species in the Yangtze River Delta Wang et al., 2018Wang et al., , 2020Zhao et al., 2017). Discontinuity of habitat type, as in the absence of hard substrate in muddy tidal flats, plays an important role in the genetic differentiation among populations and for the formation of biogeographic barriers (Brown, 1995;Connell & Irving, 2008;Riginos & Nachman, 2001). With the construction of hard artificial structures in last decades along the coastline in Jiangsu Province (30.75°-35.3°N), the natural muddy shore seascape has been changed dramatically; many regions now are characterized by seawalls, breakwaters and other artificial hard shores.
These artificial hard surfaces serve as stepping stones that promote larval settlement by rocky shore species, and thereby affect the biogeographical pattern .

| Historical events
Pleistocene glacial-interglacial cycles have led to changes of marginal sea level and these have influenced vicariance of marine species (Marko et al., 2010). With sea level dropping during the last glacial maximum (LGM), the East China Sea (ECS) was reduced to the Okinawa Trough and the South China Sea (SCS) became a semienclosed gulf. These glacial refuges were important for species persistence during the LGM (Shen et al., 2011). With the rising sea levels that followed the LGM, a northward migration from these refuges occurred, which has potential impacts on current species distributions and biogeographical patterns (Cheng & Sha, 2017;Dong et al., 2012;Ni et al., 2017).

| Oceanographic circulation
Oceanographic circulation is another important factor shaping current coastal biogeographical and phylogeographical patterns (Blanchette et al., 2008;Gosnell et al., 2014;Spalding et al., 2012;Wang et al., 2016). Oceanographic discontinuity can impede dispersal of marine larvae and prompt the formation of biogeographical breaks.
The coastline of China comprises four marginal seas, including Bohai Sea (BS), the Yellow Sea (YS), the ECS and the SCS (Figure 1), and is influenced by three major oceanographic systems: the Kuroshio Current and its branches, the Coastal Currents and Diluted Water (Liu, 2013;Williams et al., 2019). In YS and ECS, the water circulation is mainly wind-driven, leading to the strength and direction of oceanographic circulations highly variable seasonally . These coastal current systems influenced the present population genetic patterns and biogeographical histories of marine species, and delimitated the biogeographical boundary for marine species (Ni et al., 2014;Wang et al., 2016). The Kuroshio Current and its branch currents connect the SCS, ECS, YS and the Sea of Japan (Lie & Cho, 2016;Sassa et al., 2006). The Yangtze River Diluted Water has significant effects on the physical, chemical and biological environments around the Yangtze River Estuary (Lin et al., 2002), and has been proposed as a physical barrier critical for determining the coastal biogeographical pattern (Dong et al., 2012;Wang et al., 2015).

| Climate factors
Climate factors, particularly temperature, are often assumed to be the primary abiotic drivers that limit species distributions (Harley et al., 2006;Spence & Tingley, 2020). Ocean temperature can also lead to a direct and predictable influence on population connectivity, as the duration of the larval period and survival have strong correlations with ambient temperature (O'Connor et al., 2007). After settlement, ambient temperature can also determine population dynamics through influencing organisms' reproduction, development, growth and survival (O'Connor et al., 2007;Punzón et al., 2021;Wang et al., 2018).

| Subei biogeographical barrier (SBB)
The biogeographical break for coastal species at the Yangtze River Estuary has been weakened since the 1950s (Xu et al., 2020).
At the species level, some southern species have expanded their leading edges northward to ~33°-34°N Xu & Zhang, 2011). For example, the warm water sea urchin Schizaster lacunosus has extended its northern distribution range limit to ~34°N (personal communication, Yong Xu) in the YS during the period from 1958 to 2016 (Xu et al., 2020). Some intertidal macrobenthos, like Nerita yoldii  and Crassostrea sikamea , have moved northward to ~33°-34°N (Figure 3).
Based on molecular markers, a clear phylogeographical break appeared around 33°-34°N for widely distributed coastal macrobenthos ( Figure 3). In addition, with mitochondrial cytochrome oxidase subunit I (mtDNA COI) and nuclear ribosomal internal transcribed spacer (nrDNA ITS), Cheng and Sha (2017) found that the southern genetic lineage and northern genetic lineage of the Japanese mantis shrimp Oratosquilla oratoria were sympatric in their distribution ranges, with the southern Yellow sea as an overlapped zone (Cheng & Sha, 2017).
Community structure similarity of intertidal macrobenthos further confirmed the existence of the SBB. Cluster analyses based on Bray-Curtis similarity have showed that the intertidal communities can be clustered into three groups along the coast of Jiangsu

| MECHANIS MS AC TING TO THE NORTHWARD S HIF T OF B IOG EOG R APHIC AL B RE AK
Biogeographical distributions of marine species are changing rapidly over the world in the context of global change and human activities like shoreline construction (Amarasekare & Simon, 2020; Pinsky et al., 2020), and the rearrangements of biogeographical patterns can be largely attributed to these environmental changes (Blanchette et al., 2008;Gaylord & Gaines, 2000;Gosnell et al., 2014;Pappalardo et al., 2015). The weakened YREBB and emerging SBB provide model systems for clarifying the impacts of environmental changes on the formation and maintenance of biogeographical breaks ( Figure 5).

| Larval dispersal potential and oceanographic features
With high connectivity in the marine ecosystem, pelagic larval dispersal is the key process of population connectivity for species with biphasic life cycle (Cowen & Sponaugle, 2009;D'Aloia et al., 2015).
Life history characters, such as reproductive mode, reproductive period, larval type and pelagic larval duration (PLD) can affect larvae dispersal potential (Table 1) PLD is usually regarded as the most important trait influencing larvae dispersal distance and thus shaping population connectivity (Liggins et al., 2016;Wang et al., 2020). A previous study using the results of an operational hydrodynamic model has shown that along China's coast, species with a relatively long PLD (>17 days) can disperse across the Yangtze River Estuary, and have high connectivity between southern and northern populations . A cold-water species Chthamalus challengeri widely distributes along China's coast without a clear genetic break around the Yangtze River Estuary (Cheang et al., 2012;Liu et al., 2015), and the lack of genetic break of C. challengeri might attributed to its long PLD (Cheang et al., 2012).
Increasing evidence has shown that oceanographic features such as upwelling systems, fronts, moving convergences, eddies and counter currents provide conditions for larvae transport and can enhance population connectivity (Hidas et al., 2007)

| Larval settlement and seascape transformation
The characteristics of the sites that can be recognized and selected for settlement can influence post-settlement mortality and thereby determine biogeographical patterns (Bell et al., 2015;Harrington et al., 2004). Persistence of range expansion varies greatly among species, and most of the variation between species can be explained by differences in habitat availability and habitat specificity (Harrington et al., 2004;Platts et al., 2019).
The availability of suitable substrate for planktonic larval settlement can affect the success of metamorphosis from larvae into benthic juveniles (Cowen & Sponaugle, 2009). Construction of artificial hard structures is a vital factor contributing to the transformation of the biogeographic pattern around the Yangtze River Delta that has taken place in the last several decades. Over the past 70 years, about 3000 km 2 of tidal mudflats have been reclaimed using artificial hard structures along Jiangsu's coastline (Zhang et al., 2013). These artificial hard structures, including seawalls, breakwaters, ripraps and other structures, provide appropriate substrates for rocky intertidal species to settle upon and thereby facilitate their northward distribution shift (Bishop et al., 2017;Strain et al., 2019;Strain, Cumbo, et al., 2020;Wang et al., 2020).
Various abiotic and biotic cues on the substrate, such as surface texture/roughness and complexity (Coombes et al., 2015;Loke & Todd, 2016;Sedano et al., 2020) and chemical cues produced by conspecifics and microbial biofilms (Coombes et al., 2015;Ding et al., 2018;Pawlik, 1992), can influence the rate and patterns of larval supply and settlement Hunt & Scheibling, 1997), and thereby the structuring of macrobenthic communities (Loke & Todd, 2016;Sedano et al., 2020;. Furthermore, the composition of the intertidal community on the artificial structures is also related to the environmental tol-  Zhejiang Y a n g tz e R iv e r°N°E 6 months . The rapid occupation of these new habitats promotes the northward distribution range extension of intertidal species.

| Post-settlement establishment and global warming
A growing amount of evidence has suggested that post-settlement establishment plays a more important role in the maintenance of biogeographic breaks in the oceans than larval dispersal (Keith et al., 2015). Invertebrates usually exhibit high rates of post-settlement mortality, which are mainly due to the delay of metamorphosis, competition, predation, physiological stress (Hunt & Scheibling, 1997) and genetic inviability driven by genotype-environment interactions (Barahona et al., 2019;Plough et al., 2016). Importantly, the capacity of species to tolerate physiologically marginal conditions influences their ability to establish a viable population after successful dispersal (Keith et al., 2015) and the persistence of their present distribution ranges Han et al., 2019).
Along China's coast, the most thermally vulnerable locations occur in the Yangtze River Delta, based on analyses using thermal safety margin as a proxy (Deutsch et al., 2008;Dong et al., 2017;Ma et al., 2021). Mapping of operative temperatures between 2010 and 2020, as calculated using heat budget models (Wethey et al., 2011), shows that the region between 32°N and 33°N is one of the hottest areas in summer along China's coast, and this area can potentially act as a thermal barrier for intertidal species. Wang et al., 2020). (b) Cluster analysis for community structure of rocky shore species based on Bray-Curtis similarity of presence/absence data collected from 2013 to 2015 . (c) Cluster analysis using data collected from 2013 to 2017   Hirano and Inaba (1980) Monodonta labio M-L M Pelagic development July to September (Japan) Lijima and Furota (1996) Echinolittorina radiata Note: E, exposed shores; M, moderately exposed shores; S, sheltered; R, rock pools; L, low-intertidal; M, mid-intertidal; H, high-intertidal; Sp: Splash zone; I/BR, internal fertilization/eggs brooded; I/ EM, internal fertilization/egg mass; E/EM, external fertilization/egg mass; E/BS, external fertilization/gametes broadcast spawn; PP, planktonic planktotrophic; PL, planktonic lecithotrophic, CJ, crawling juveniles; BL, benthic lecithotrophic.

F I G U R E 4 Sampling sites and cluster patterns of communities of intertidal macrobenthos. (a) Sampling sites included in previous studies
The low temperatures in winter can also impede the northward shift of southern species and contribute to the formation of the SBB. Northern distributional boundaries of larvae of two southern species, N. yoldii and C. sikamea, could be found at 34.5°N, which is beyond the current northern distributional limit of adults at ~33°N and ~34°N, respectively. This can be explained by low temperatures during winter that reach the species' thermal tolerance limits, and thus define their current distributional boundaries . In the process of colonization by N. yoldii to the north of the Yangtze River Estuary, some specific haplotypes have been filtered out , and a clear phylogeographical break may consequently have formed (Kawecki, 2008).
The thermal plasticity of intertidal macrobenthos varies greatly at different spatiotemporal scales in the face of global warming and increasing occurrences of heat wave (Li, Tan, et al., 2021;Zhang et al., 2021). Structural complexity of the habitat could provide microhabitat for intertidal species as "thermal refugia," and micro-scale physiological thermal tolerance can play an important role in determining whether species could survive in the face of extreme events (Li, Tan, et al., 2021;Lima et al., 2016). The accumulation of microscale physiological variations could enhance the macro-scale thermal resilience of intertidal species Strain, Cumbo, et al., 2020), and thereby influence species distributional shifts affected by climate change (Bates et al., 2018;Li, Tan, et al., 2021;Liao et al., 2021;Schils & Wilson, 2006). At different time-scales, heat hardening can work synchronously with seasonal acclimatization to increase resistance of raze clam Sinonovacula constricta to high temperatures .

| PER S PEC TIVE S
The biogeographical pattern of intertidal species has been experi- However, integrative studies about biogeographical pattern and regional oceanography systems are still relatively rare, and demand more attention in future research.
3. Species' response to global climate change has been verified to be largely determined by their physiological performance (Han et al., 2019;Liao et al., 2021). In order to accurately assess the impact of global climate change on the survival and distribution of intertidal macrobenthos, physiological tolerance of intertidal species at multiple spatiotemporal scales must be considered.

ACK N OWLED G EM ENTS
We sincerely appreciate the criticisms of this manuscript given by Professor George Somero. This work was supported by National Natural Science Foundation of China (42025604, 41776135, 41976142, and 42106141), and the Fundamental Research Funds for the Central Universities.

CO N FLI C T O F I NTE R E S T
The authors have no conflict of interest.

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1111/ddi.13463.

DATA AVA I L A B I L I T Y S TAT E M E N T
All used data were downloaded from literature, and some have been restructured.