Vol.:(0123456789)1 3 Marine Biology (2021) 168:138 https://doi.org/10.1007/s00227-021-03946-4 ORIGINAL PAPER The population genetic structure of the urchin Centrostephanus rodgersii in New Zealand with links to Australia L. J. Thomas1  · L. Liggins2,3  · S. C. Banks4  · L. B. Beheregaray5  · M. Liddy1 · G. A. McCulloch6  · J. M. Waters6  · L. Carter7  · M. Byrne8  · R. A. Cumming1  · M. D. Lamare1 Received: 16 December 2020 / Accepted: 20 May 2021 © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2021 Abstract The diadematid sea urchin Centrostephanus rodgersii occurs in Australia and New Zealand and has undergone recent south- ward range extension in Australia as a result of regional warming. Clarifying the population genetic structure of this species across its New Zealand range would allow a better understanding of recent and future mechanisms driving range changes in the species. Here, we use microsatellite DNA data to assess connectivity and genetic structure in 385 individuals from 14 locations across the Australian and New Zealand ranges of the species. We detected substantial genetic differentiation among C. rodgersii populations from Australia and New Zealand. However, the population from Port Stephens (located north of Newcastle), Australia, strongly clustered with New Zealand samples. This suggests that the New Zealand populations recently originated from this area, likely via larval transport in the Tasman Front flow that arises in this region. The weak population genetic structure and relatively low genetic diversity detected in New Zealand (global Fst = 0.0021) relative to Australia (global Fst = 0.0339) is consistent with the former population’s inferred history of recent climate-driven expan- sion. Population-level inbreeding is low in most populations, but were higher in New Zealand (global Fis = 0.0833) than in Australia (global Fis = 0.0202), suggesting that self-recruitment is playing an increasingly important role in the New Zealand region. Our results suggest that C. rodgersii is likely to spread southwards as ocean temperatures increase; therefore, it is crucial that researchers develop a clearer understanding of how New Zealand ecosystems will be reshaped by this species (and others) under climate change. Introduction Recent anthropogenic climate change has driven shifts in ocean currents, causing range shifts, novel introductions, and extinctions in marine communities (Poloczanska et al. 2013; IPCC 2019; Pinsky et al. 2020). Most marine species have indirect life cycles and responses in their free-living larval stages and can play a key role in climate-induced population responses and distributional shifts. The biological effects of recent marine climate change have been particularly prominent in temperate waters of southern Australia, with numerous species shifting poleward in response to warm- ing (Ling et al. 2009; Pitt et al. 2010; Wernberg et al. 2011; Monaco et al. 2021). Currently, there are few studies of cli- mate-driven range shifts in New Zealand marine ecosystems (although see Brooks 2020; Middleton et al. 2021). One of the best-documented examples of range extension due to the interaction of ocean warming and larval ecol- ogy is seen in the warm-temperate diadematid sea urchin Centrostephanus rodgersii (Agassiz 1863). This relatively Responsible Editor: T. Reusch. Reviewers: undisclosedexperts * L. J. Thomas leighton.thomas@otago.ac.nz 1 Department of Marine Science, University of Otago, Dunedin, New Zealand 2 School of Natural and Computational Sciences, Massey University, Albany Campus, Auckland 0745, New Zealand 3 Auckland Museum, Tāmaki Paenga Hira, Auckland 1010, New Zealand 4 Research Institute for the Environment and Livelihoods, Charles Darwin University, Darwin, NT 0810, Australia 5 Molecular Ecology Lab, College of Science and Engineering, Flinders University, Adelaide, Australia 6 Department of Zoology, University of Otago, P.O. Box 56, Dunedin 9054, New Zealand 7 University of Victoria, Wellington, New Zealand 8 School of Medical Sciences, University of Sydney, Sydney, Australia http://orcid.org/0000-0002-7106-0946 http://orcid.org/0000-0003-1143-2346 http://orcid.org/0000-0003-2415-0057 http://orcid.org/0000-0003-0944-3003 http://orcid.org/0000-0003-1462-7106 http://orcid.org/0000-0002-1514-7916 http://orcid.org/0000-0002-7502-9452 http://orcid.org/0000-0002-8902-9808 http://orcid.org/0000-0001-6999-6617 http://orcid.org/0000-0001-8656-7240 http://crossmark.crossref.org/dialog/?doi=10.1007/s00227-021-03946-4&domain=pdf Marine Biology (2021) 168:138 1 3 138 Page 2 of 11 large sea urchin is found along the coasts of eastern Australia and northern Aotearoa New Zealand, as well as Lord Howe Island, Norfolk Island, and the Kermadec Islands (Byrne and Andrew 2020). The species has been found to spawn in the Austral winter from June to August (Byrne and Andrew 2020), meaning their larvae develop in the coldest months of the year. C. rodgersii has a long-lived larval stage, reach- ing competency from 40 days to > 3 months depending on temperature (Hugget et al. 2005; Soars et al. 2009; Hardy et al. 2014; Mos et al. 2020), and is capable of long-distance passive dispersal (Banks et al. 2007b). Evidence suggests the geographic distribution of this species is limited by sea temperatures, with larvae having a minimum thermal limit of 12 °C (Ling and Johnson 2009). Over the past 50 years, warming winter sea temperatures have facilitated range expansions of C. rodgersii across the Tasman Sea and the Bass Strait (Ling and Johnson 2009). These increased sea temperatures are influenced by the East Australian Current (EAC; Fig. 1), which has extended at least 350 km further south over the past 60 years, bringing with it an increase in temperature and salinity (Hill et al. 2011). In south-eastern Australia (and particularly Tasmania), thermally facilitated southward range extension of C. rodgersii (and reduced pre- dation pressure) have led to kelp over-grazing by this species (Johnson et al. 2011). This range expansion has driven a regime shift from dense kelp beds to rocky barrens, reduc- ing biodiversity and fisheries yields (Ling et al. 2009; Ling and Johnson 2009). Centrostephanus rodgersii also occurs along coastal north-eastern New Zealand, as well as offshore islands including Rangitāhua (the Kermadec Archipelago, Fig. 1), Manawatāwhi (Three Kings Island, 34°9′S, 172°8′E) and Te Puia o Whakaari (White Island, 37°31′S, 177°11′E) within New Zealand’s Exclusive Economic Zone (Pecorino et al. 2013, Byrne and Andrew 2020, Byrne and Gall 2017). In contrast to the Australian populations, New Zealand populations of C. rodgersii have been less well studied. Reproduction and growth parameters reported in New Zealand populations are similar to those reported in Australian populations (Agassiz 1863). In contrast, New Zealand population densities appear to be lower (Pecorino et al. 2013; Edgar et al. 2017), and the species is not asso- ciated with extensive urchin barrens. Currently, C. rodg- ersii populations appear to be restricted to north of New Fig. 1 Map of collection sites of Centrostephanus rodgersii along mainland Australia and Tasmania, the Kermadec Islands, and New Zealand. Approximate locations of oceanic currents are indicated with blue arrows ( modified from Hill et  al. 2011). PSNB Port Ste- phens Bay, SHBP Sydney Harbour, TASC Tasmania (Safety Cove), TAFB Tasmania (Fortescue Bay), JBMB Jervis Bay, CH Cape Howe, TAEC Tasmania (East Cape), PKN Poor Knights Island, WIS White Island, ALD Alderman Islands, MOK Mokohinau Island, MER Mer- cury Island, NKER northern Kermadec Islands, SKER southern Ker- madec Islands Marine Biology (2021) 168:138 1 3 Page 3 of 11 138 Zealand’s East Cape (37.7°S), where winter sea temper- atures are above 15 °C, consistent with experimentally assessed larval temperature thresholds (Pecorino et al. 2013). While there have been no reports of recent range shifts in New Zealand C. rodgersii, it seems likely that this urchin will increase in abundance and expand its distribu- tion southward as sea temperatures continue to rise. Previous studies examining the genetic structure of C. rodgersii populations using microsatellite markers included only a single New Zealand population (Mokoh- inau Islands) (Banks et al. 2007a, b, 2010). These studies suggested that genetic structure of this species in south- eastern Australia is primarily constrained by the influence of regional oceanographic circulation on larval reten- tion. Specifically, while these studies detected little or no large-scale genetic differentiation, they revealed fine-scale autocorrelation indicative of self-recruitment linked to sea surface temperature (SST) variability and coastal topology (Banks et al. 2007b), especially in southern New South Wales (NSW). Banks et al. (2010) further characterised the genetic structure of Australian C. rodgersii popula- tions to better understand the process of range extension and connectivity in the region, with analyses supporting recent southward extension to Tasmania, and high ongoing connectivity between source (mainland) and founder (Tas- manian) populations. Banks et al. (2007b) detected little trans-Tasman genetic differentiation in C. rodgersii, sug- gesting the New Zealand populations arrived only recently. However, their sampling of New Zealand populations was insufficient to understand larval-mediated connectivity across the Tasman sea and within New Zealand, or an ongoing demographic response to regional climate change. Here, we examine the population genetic structure of C. rodgersii in New Zealand. We combine the previously published data of Banks et al. (2007b) with new genotypic data from seven populations across the New Zealand range of C. rodgersii. Using these datasets, we: (1) assess the extent to which New Zealand populations are self-recruit- ing versus reliant on larvae from Australia; and (2) assess population genetic structure and genetic diversity across the range of C. rodgersii in New Zealand. Where the New Zealand populations rely on the Australian populations for larval delivery, we expected that the New Zealand populations would have little genetic differentiation with Australia, few private alleles, and little genetic diversity in New Zealand. In the case of no contemporary larval delivery, we expected genetic differentiation among Aus- tralia and New Zealand, with signals of reduced genetic diversity in New Zealand. However, under a hypothesis of self-recruitment within New Zealand, and an absence of larval delivery, we expected that there would be genetic differentiation from Australia, but with equivalent levels of genetic diversity. Materials and methods Centrostephanus rodgersii were collected from rocky subtidal reefs (5–30 m) throughout the species’ New Zea- land range (with appropriate consent and permissions, see “Acknowledgements”). The number of individuals sam- pled per location was relative to the local abundance to avoid interfering with the species’ local demography. DNA was extracted from 193 C. rodgersii specimens from seven New Zealand locations (Fig. 1; Table 1) using a chelex protocol (Casquet et al. 2012). Specimens were genotyped using six microsatellite loci [Supplementary Table 1, as described in Banks et al. (2007a)]. Amplified fragments were analysed on an ABI 3730xl using GeneS- can 500 ROX. Fragment lengths were scored manually using Geneious R6 (Kearse et al. 2012; https:// www. genei ous. com) to avoid binning error. For quality control, and to ensure combinability of Australian and New Zealand data, a subset of Australian specimens used in a previous study of C. rodgersii (Banks et al. 2010) were re-genotyped in our laboratory and compared to the findings of Banks et al. (2010). These analyses yielded a combined dataset of 385 specimens from 14 locations across the Australian and New Zealand ranges of the species. The loci were tested for linkage disequilibrium and Hardy–Weinberg equilib- rium (HWE) using the R package pegas (Paradis 2010). We tested for presence of null alleles and scoring using errors using MICROCHECKER (Oosterhout et al. 2004). To estimate population genetic and demographic param- eters of C. rodgersii in New Zealand, we used several population genetic indices and analyses, using package specific default settings, unless specified. Observed het- erozygosity (Ho), expected (He) heterozygosity, inbreeding coefficient (Fis), and mean allelic richness (AR) for all loci in all populations and Fis values for each loci were calcu- lated using the R package hierfstat (Goudet 2005). Con- fidence intervals for population Fis with 1000 bootstrap replicates were calculated using the boot.ppfis function in heirfstat. We assessed genetic differentiation among sampled locations by estimating pairwise Fst values in Arlequin ver 3.5.2.2 (Excoffier and Lischer 2010). The significance of these values was tested with 10,000 permutations. Hier- archical population genetic structure (analysis of molecu- lar variance, AMOVA) was also assessed in Arlequin. As pairwise Fst values may not be appropriate for highly vari- able microsatellite data (see Jost 2008), we also calculated pairwise JOST’S D values. We calculated these values and 95% confidence intervals using 1,000 bootstrap replicates using the R package diveRsity (Keenan et al. 2013). Emer- gent genetic structuring across our 14 sampled locations was assessed using the Bayesian clustering algorithm in https://www.geneious.com https://www.geneious.com Marine Biology (2021) 168:138 1 3 138 Page 4 of 11 STRU CTU RE 2.3.4 (Pritchard et al. 2000) as well as a k-means clustering algorithm using the R package ade- genet (Jombart 2008). We initially ran the STRU CTU RE analysis across the entire dataset, using the admixture model with allele frequencies correlated. Ten replicate analyses were conducted for each value of K, from K = 1 to K = 15, with each run using a different starting seed. Each run consisted of 200,000 burn-in iterations, followed by 2,000,000 iterations. We used CLUMPAK (Kopelman et al. 2015) to assess the most likely value of K (using the deltaK vs K method of Evanno et al. 2005), and to plot the results. We conducted additional STRU CTU RE runs (using the parameters outlined above) to assess evidence of genetic structuring within New Zealand populations. For the k-means clustering, the optimal number of clusters was identified using the Bayesian Information Criterion (BIC). To describe the genetic groups identified using the k-means algorithm, we conducted a discriminant analysis of principal components (DAPC; Jombart 2008, 2010). To avoid overfitting, only the principal components that retain 90% of the variance were retained for the DAPC. Contemporary migration and self-recruitment rates among the populations were estimated using BAYESASS v.3.0.1 (Wilson and Rannala 2003). Five independent Markov Chain Monte Carlo (MCMC) runs were performed, each with 10,000,000 reps and a burn-in of 1,000,000 steps, with sampling every 2000 steps. The reliability of migra- tion rate estimates was assessed by checking for consist- ency among runs. For all BAYESASS runs, we followed the recommendations of Faubet et al. (2007) namely five independent runs with different random number seeds, and we used Bayesian deviance (using the calculateDeviance.R script from Meirmans 2014) to select the best run. Conver- gence of parameter estimates was also assessed by manual examination of the trace files. Tests for recent population bottlenecks were performed using the program BOTTLENECK (Piry et al. 1999). A two-tailed Wilcoxon sign-rank test for heterozygosity excess (Luikart and Cornuet 1998) was applied to detect any recent bottlenecks. In non-bottlenecked populations, observed het- erozygosity (Ho) is consistent with Hardy–Weinberg equilib- rium heterozygosity (He), whereas in a population that has experienced a recent bottleneck the mutation-drift equilib- rium is disrupted and Ho will be significantly greater than He (Luikart and Cornuet 1998). He values were calculated using the stepwise mutation model (SMM; Kimura and Ohta 1978) and the infinite allele model (IAM; Kimura and Crow 1964). We also tested for “mode-shift”, to determine if allele Table 1 Location of Centrostephanus rodgersii sampling sites across New Zealand and Australia N sampling size, LAT latitude, LONG longitude, Ho observed heterozygosity, He expected heterozygosity. Fis inbreeding coefficient. PSNB Port Stephens Bay, SHBP Sydney Harbour, TASC Tasmania (Safety Cove), TAFB Tasmania (Fortescue Bay), JBMB Jervis Bay, CH Cape Howe, TAEC Tasmania (East Cape), PKN Poor Knights Island, WIS White Island, ALD Alderman Islands, MOK Mokohinau Island, MER Mercury Island, NKER northern Kermadec Islands, SKER southern Kermadec Islands Region Pop N LAT LONG Mean Ho Mean He Private alleles Mean rarefied allele count Mean within pop Fis (min–max) Australia PSNB 17 − 32.718 152.142 0.648 0.613 0 4.081 − 0.0230 (− 0.2020 to 0.1150) SHBP 21 − 34.594 150.904 0.473 0.55 1 3.403 0.1689(− 0.0004 to 0.3282) TASC 26 − 43.167 147.85 0.642 0.605 3 4.059 − 0.0519 (− 0.2034 to 0.0492) TAFB 27 − 43.135 148 0.537 0.595 10 3.888 0.1165 (0.0354 to 0.2517) JBMB 29 − 35.126 150.752 0.668 0.602 2 3.858 − 0.1027 (− 0.1543 to − 0.0358) CH 30 − 37.504 149.983 0.627 0.577 2 3.801 − 0.1036 (− 0.2199 to 0.0496) TAEC 42 − 39.473 147.35 0.502 0.56 3 3.729 0.1066 (0.0830 to 0.1264) Overall Fis = 0.0202 Overall Fst = 0.0339 New Zealand PKN 30 − 35.47 174.74 0.543 0.573 0 3.838 0.0904 (− 0.0813 to 0.2213) WIS 30 − 37.53 177.17 0.58 0.607 9 4.134 0.0714 (− 0.1690 to 0.1829) ALD 30 − 36.96 176.07 0.55 0.612 0 3.947 0.0944 (− 0.0878 to 0.247) MOK 32 − 35.93 175.12 0.637 0.602 1 3.859 − 0.0431 (− 0.2139 to 0.108) MER 6 − 36.61 175.85 0.383 0.447 0 2.745 0.1977 (0.0252 to 0.4604) Overall Fis = 0.0833 Overall Fst = 0.0021 Kermadecs NKER 55 − 29.281 − 177.9 0.4763 0.63 1 3.986 0.2198 (0.0604 to 0.3123) SKER 10 − 31.347 − 179 0.528 0.676 0 3.964 0.2128 (− 0.4634 to 0.5645) Overall Fis = 0.2311 Overall Fst = 0.0059 Marine Biology (2021) 168:138 1 3 Page 5 of 11 138 frequency distribution shifts have occurred in recently bot- tlenecked populations (Luikart and Cornuet 1998). Under HWE, the allele frequency distribution would exhibit an L-shaped distribution, recently bottlenecked populations would not exhibit this distribution. Results All the Australian samples genotyped were scored identi- cally to those previously genotyped by Banks et al. (2010), justifying the subsequent combination of the datasets from Australia and New Zealand C. rodgersii populations. Devi- ations from HWE were detected for some loci/population combinations (Supplementary Table 2). However, these deviations were not consistently detected for a single popula- tion of C. rodgersii or locus, so all loci and populations were retained for further analysis. Fis values for individual loci ranged from − 0.1540 to 0.2740 (Supplementary Table 1). However, none of these values was significantly different from 0. When testing for null alleles using all of the popula- tions, an excess of homozygotes was found for most of the loci (data not shown). However, the excess of homozygotes was not found when individual populations were tested, suggesting that these findings reflect incomplete mixing across sites, rather than null alleles. Scoring errors were not detected in the MICROCHECKER analysis. The mean number of private alleles per population was greater in Australia (three unique alleles per popula- tion) than in New Zealand (2.2 unique alleles per popula- tion) (Table 1). Population-level inbreeding was greater in New Zealand (global Fis = 0.0833) than Australia (global Fis = 0.0202). By contrast, the extent of population struc- turing is greater in Australia (global Fst = 0.0339) than in New Zealand (global Fst = 0.0021). Estimates of pairwise JOST’S D and Fst (Supplementary Table 3) show a similar pattern. Pairwise Fst values between New Zealand popu- lations ranged from 0.0004 (between White Island, WIS and Mercury Island, MER) to 0.1434 (between MER and southern Kermadec Islands, SKER). Pairwise JOST’S D values ranged from 0.0003 (between WIS and MER) to 0.0236 (between MER and SKER). Comparisons between New Zealand and Australian populations yielded values Fst ranging from 0.0011 (JOST’S D = 0.0329) (between Port Stephens, PSNB and Mokohinau Island, MOK) to 0.2213 (JOST’S D = 0.0428) (between Tasmania East Cape, TAEC and SKER). Fst values between Australian populations ranged from 0.0012 (JOST’S D = 0) (between Tasmania Safety Cove, TASC and Tasmania Fortescue Bay, TAFB) to 0.1345 (JOST’S D = 0.0281) (PSNB and Sydney Harbour, SHBP). Hierarchical AMOVA analysis revealed that most of the variation (90.5%) was found within individuals, with 5.9% of the variation attributed to variation across the two regions [Australia and New Zealand (including the Kerma- dec Islands)], and 3.7% among populations within regions (Supplementary Table 4). Bayesian clustering analyses indicated that the most likely number of populations (K) across the entire dataset was two (Figure S1). The seven New Zealand sites (including the two Kermadec Islands) and a single Australian site (PSNB) represented one population cluster, with the remaining six Australian sites forming the second cluster (Fig. 2a). No population structuring was evident when only the seven New Zealand populations were analysed (Fig. 2b). The optimal number of clusters with the lowest BIC using the k-means clustering algorithm was 10. The DAPC similarly indicates a genetic distinction between Australian and New Zealand C. rodgersii, but with the PSNB sample clustering with New Fig. 2 Structure analyses (K = 2) for Centrostephanus rodgersii col- lected from (a) 14 sites across Australia and New Zealand, and (b) only the 7 New Zealand sites (including the Kermadec Islands). Each bar represents an individual sample, with the proportion of colour on each bar indicating the posterior probability of assignment of the individual to each of the two clusters Marine Biology (2021) 168:138 1 3 138 Page 6 of 11 Fig. 3 Discriminant analysis of principal components for 385 Cen- trostephanus rodgersii individuals collected from 14 sites across Australia and New Zealand. Each small circle represents an indi- vidual. Individuals are grouped by k-means clustering, represented as inertia ellipses. Individuals from the Australian populations were mainly assigned to clusters 5, 6, 7, 9 and 10, apart from individuals from PSNB, which were mainly assigned to clusters 1, 2, 3, 4, and 8 with the New Zealand and Kermadec Islands individuals. This indi- cates that there is genetic differentiation among the Australian and New Zealand populations, and suggests the New Zealand popula- tions came from PSNB. The X- and Y-axes indicate the 1st (explain- ing 40.6% of genetic variation) and 2nd (explaining 15.2% of genetic variation) linear discriminant axes. For detailed information on how individuals from each population are assigned to each cluster, see Table S5 Fig. 4 Discriminant analysis of principal components for 193 Centrostephanus rodger- sii individuals collected from seven sites across New Zealand and the Kermadec Islands. Each small circle represents an individual. Individuals are grouped by k-means clustering, represented as inertia ellipses. Individuals from the New Zea- land and Kermadec populations were assigned relatively evenly across all clusters, suggest- ing there is very little genetic structure across this region. The X- and Y-axes indicate the 1st (explaining 35.4% of genetic variation) and 2nd (explain- ing 25.4% of genetic variation) linear discriminant axes. For detailed information on how individuals from each popula- tion are assigned to each cluster, see Table S6 Marine Biology (2021) 168:138 1 3 Page 7 of 11 138 Zealand populations (Figs. 3, 4). The percentage contribu- tion of the individuals assigned to each of the clusters esti- mated by the STRU CTU RE and DAPC analysis are shown in Supplementary Tables 5 and 6. With the exception of PSNB, most of the Australian individuals are assigned to clusters 5, 6, 7, 9, and 10 of the DAPC analysis. By contrast, most of the New Zealand individuals, along with individuals from PSNB, are assigned primarily to clusters 1, 2, 3, 4, and 8 (Supplementary Table 5). Estimates of self-recruitment were high for all populations, being highest in NKER population (0.9285 ± 0.0170) (Supplementary Table 7). Across all runs, the highest estimates of trans-Tasman migration parameters were found for PSNB to NKER (0.1767 ± 0.0309). Heterozygosity excess (an indication of a genetic bottle- neck) was found in all of the Australian populations under the Stepwise Mutation Model (SMM) (Supplementary Table 8). Heterozygosity excess was also found in the New Zealand populations from MER and MOK. Tests of mode- shift likewise detected a bottleneck in the MER population. Discussion Marine climate change is driving species distributional shifts in many regions of the globe (Perry et al. 2005; Monaco et al. 2021), and the urchin C. rodgersii presents one of the most dramatic examples of such climate-driven ecosystem change (Ling et al. 2009). Consistent with the expectations of a climate-driven range shift (Banks et al. 2010), our study detected very weak population structure among New Zea- land populations of this recently expanded species. In addi- tion, while our study detected substantial genetic differentia- tion between New Zealand and most Australian C. rodgersii populations, the finding that a single Australian population (PSNB) clustered tightly with New Zealand samples sug- gests that this region has been the source for larvae arriving in New Zealand via the Tasman Front flow. This potential source for trans-Tasman connectivity is also supported by oceanographic circulation patterns, where the Tasman Front develops as a flow from the EAC along the NSW coast at around 34°S in the region north of Newcastle (Ridgway and Dunn 2003) and meets north-eastern New Zealand as the East Auckland Current (Stanton 1979). These results thus appear to highlight the crucial interplay between physical and biological marine processes, and specifically the impor- tance of key oceanographic features in mediating marine biological dispersal and connectivity (Wolanski and Hamner 1988; Gaylord and Gaines 2000; Banks 2007b, 2010). A previous study found little genetic differentiation between Australian populations and a single New Zealand population (from MOK; Banks et al. 2007b), suggesting C. rodgersii larvae are readily transported within the Tasman Front. Indeed, Banks et al. (2007b) found pairwise Fst values between the MOK population and Australian populations north of about 34°S were 0. By contrast, some popula- tions south of PSNB had notably higher pairwise Fst values (0–0.37), indicating greater isolation from the New Zealand population. These patterns are consistent with our observa- tions, suggesting a strong connectivity among New Zealand with Australian populations from regions linked via the Tas- man Front. Centrostephanus rodgersii has increased its range and abundance in southern Australia markedly over the past 60 years (Ling et  al. 2009), mirroring southward range expansions detected in a host of additional temperate Aus- tralian marine taxa (Pitt et al. 2010; Last et al. 2011; Wern- berg et al. 2011). Oceanographic data suggest that strength- ening of the southerly EAC-extension along the Australian coastline is responsible for these changes in Australian C. rodgersii populations. In contrast, the offshoot of the EAC that crosses the Tasman Sea to New Zealand has weakened over recent decades (observed post-1980s; Sloyan and O’Kane 2015). Such weakening would presumably reduce transport of larvae from Australia to New Zealand, result- ing in recent reductions of gene flow. As a consequence, established populations of C. rodgersii in New Zealand have the potential to become more genetically isolated and more reliant on local larval supply as this weakening continues. Other sea urchin species in the region also appear to be strongly influenced by the EAC and the Tasman Front. The ranges of Tripneustes kermadecensis and Heliocidaris tuberculata extend from eastern Australia to the Kermadec Islands, and also locations within north-eastern New Zea- land (Bronstein et al. 2017, 2019; Byrne and Gall 2017). Both species have planktotrophic larvae that are thought to be advected from Australia to the Kermadec Islands and New Zealand via the Tasman Front (Bronstein et al. 2019). Transport of larvae from Australia and New Zealand is also reflected in the genetic structure found in other coastal marine taxa around the Tasman Sea. For example, the gastro- pod mollusc Nerita melanotragus, a species with a pelagic larvae duration of 5–6 months, shows little or no mtDNA differentiation among eastern Australian and New Zealand populations (Waters et al., 2005; Spencer et al. 2007). Despite the high potential for long-distance dispersal in C. rodgersii, our analyses suggested that self-recruit- ment may be important in maintaining some populations (see Burgess et al. 2014). The northern Kermadec Island population (Raoul Island) has the highest levels of self- recruitment (0.9285 ± 0.0170), perhaps a reflection of its geographic isolation. Notably, these islands sit outside of the dominant flow of the Tasman Front and origin of the East Auckland Current in a region of weak and fluctuat- ing eddy systems (Sutton et al. 2012). Self-recruitment at Raoul Island (NKER) has similarly been surmised as important in maintaining populations of crown-of-thorns Marine Biology (2021) 168:138 1 3 138 Page 8 of 11 sea stars (Liggins et al. 2014), despite this species having a planktonic larval dispersal. By contrast, Reisser et al. (2014) noted an absence of genetic structure between populations of Nerita melanotragus, perhaps a reflection of this species’ relatively extensive larval dispersal capac- ity. Our results here similarly suggest that migration of C. rodgersii larvae from New Zealand to the northern Ker- madec population may be substantial (0.1767 ± 0.0309). Although not surprising based on previous larval dispersal modelling [estimated dispersal time for planktonic larvae was 20–50 days; (Sutton et al. 2012)], population densities of C. rodgersii at the Kermadec Islands are much higher than around mainland New Zealand (Edgar and Stuart- Smith 2017). In contrast, most of the inferred migrants arriving in New Zealand came from PSNB, supporting the results of the DAPC and STRU CTU RE analyses. Although the expansion of the Australian C. rodgersii populations over the past 60 years is well documented (Ling et al. 2009), previous analyses of Tasmanian populations imply that ongoing gene flow may mitigate the effects of genetic bottlenecks at the front of southward expanding populations (Banks et al. 2010). By contrast, higher levels of population-level inbreeding and lower levels of genetic structuring in New Zealand suggest higher levels of self- recruitment, and the likely influence of complex oceano- graphic conditions off north-eastern New Zealand as key drivers of genetic structure (see Ross et al. 2009; Gardner et al. 2010). These features include the persistent mesoscale East Cape Eddy and North Cape Eddy systems that are capable of retaining long-lived oceanic larvae, with reten- tion of nearshore also likely enhanced through wind driven upwelling and ageostrophic secondary circulation (Chiswell and Roemmich 1998). Species such as C. rodgersii with long-lived dispersive phases (1–3 months) that are temperature-limited will likely undergo ongoing range extensions under global warming (Banks et al. 2010). Such climate-driven shifts are increas- ingly evident for a broad range of coastal marine species in Australasia (Last et al. 2011; Wernberg et al. 2011; Monaco et al. 2021; Middleton et al. 2021) and globally (Perry et al. 2005; Poloczanska et al. 2013; Pinsky et al. 2020). Winter sea temperatures and warming has been an important control of C. rodgersii in Australia, with populations establishing in regions where at least one-in-three winters are warmer than 12 °C (the larval thermal threshold; Ling et al. 2009). In New Zealand, the larval thermal threshold appears to be around 15 °C, corresponding with the present distribution of the species (Pecorino et al. 2013). SSTs on average are increasing in the region and marine heatwaves are becoming more frequent (IPCC 2019; Misra et al. 2021). Law et al. (2018) modelled a predicted increase in SSTs of between 1 and 3 °C under RCP8.5 by 2100 across the current New Zealand range of C. rodgersii. If populations of C. rodgersii in New Zealand respond to warming in the same way as Australian populations, we expect increases in population sizes and an expansion southwards. Based on present-day New Zealand sea surface tem- peratures in the winter, a 3 °C of ocean warming would result in all of coastal North Island having winter sea tem- peratures 15 °C or warmer, equating to 400 km expansion south in the species’ potential range. Changes in C. rodg- ersii distributions are driven by favourable currents (such as the EAC driving the southward expansion in Australia), and any changes in the New Zealand population would also be influenced by regional currents. Importantly, the south- ern-most reported distribution of New Zealand populations near East Cape not only coincides with the 15 °C winter isotherm, but is also within the south-flowing coastal East Cape Current that could facilitate the transport of larvae south and potentially population expansion. Given the well-documented biodiversity and socio-economic impacts that C. rodgersii can have (Johnson et al. 2011; Pecl et al. 2017), if there is such a southern expansion of the species’ range, it is essential to develop a clearer understanding of how New Zealand ecosystems may be reshaped by this species and others under climate change. Supplementary Information The online version contains supplemen- tary material available at https:// doi. org/ 10. 1007/ s00227- 021- 03946-4. Acknowledgements We thank Tania King for lab assistance, Felix Vaux for assisting with figure production and Robert Smith for discus- sions on physical oceanography and various members of MEGMAR (Molecular Ecology Group for Marine Research at Flinders University) for assistance with the sampling in Australia. Expeditions to the Ker- madec Islands were made possible by the Sir Peter Blake Trust, the Commanding Officer and ship’s company of HMNZS Canterbury, the RV Braveheart crew (Stoney Creek Shipping Company), and the RV Tangaroa with the support of the Auckland Museum Institute, Auck- land Museum, the Pew Charitable Trusts, and the School of Natural and Computational Sciences (SNCS) Massey University, the Marine Fund- ing Advisory Research Group, NIWA (Project COBR1705), Te Papa Tongarewa, and Ministry for the Environment. We thank J. David Agu- irre, Tom Trnski, Adam N. H. Smith, David Acuña-Marrero, Emma Betty, Mat Betty, Crispin Middleton, Phil Ross, Sam McCormack, and Severine Hannam for collecting help. We wish to thank and acknowl- edge mana whenua (the traditional owners), the Māori iwi of Ngāti Kuri and Ngātiwai, for their support. We thank Te Komite Rakahau ki Kāi Tahu (the Ngāi Tahu Research Consultation Committee) for reviewing our proposed research and providing advice on establishing partnerships with iwi and exchanging new knowledge generated by the research. We thank the two anonymous reviewers for their careful and detailed suggestions on earlier drafts which have greatly improved this manuscript. Authors contributions Conceptualisation: MDL, LL, SCB, LBB, JMW, MB, and LC. Methodology: MDL, JMW, LL, SCB, LBB, GAMcC, and LJT. Field sampling: MDL, LL, and ML. Formal analysis and investi- gation: all the authors. Writing—original draft preparation: MDL, LL, SCB, LBB, JMW, MB, LC, GAMcC, and LJT. Writing—review and editing: MDL, LL, SCB, LBB, JMW, MB, GAMcC, and LJT. Funding https://doi.org/10.1007/s00227-021-03946-4 Marine Biology (2021) 168:138 1 3 Page 9 of 11 138 acquisition: MDL, LL, SCB, LBB, JMW, MB, and LC. Resources: MDL, LL, and JMW. Funding Research was supported by a University of Otago Research Grant to M. Lamare. L. Liggins was supported by a Rutherford Founda- tion New Zealand Postdoctoral Fellowship. Availability of data and materials All voucher specimens are stored at the Department of Zoology, University of Otago and tissue collections held at Massey University are findable via the Genomics Observatories Metadatabase (GEOME). All genotype files and sample metadata are deposited in appropriate repositories (genotype files are available at https:// doi. org/ 10. 6084/ m9. figsh are. 14450 163. v1; sample metadata are available at GEOME [GUID https:// n2t. net/ ark:/ 21547/ DpW2] as part of the Ira Moana Project, https:// sites. massey. ac. nz/ iramo ana/). These samples and derived data have a Biocultural (BC) Notice attached. The BC Notice is a visible notification that there are accompanying cultural rights and responsibilities that need further attention for any future sharing and use of this material or data. The BC Notice recognises the rights of Indigenous peoples to permission the use of informa- tion, collections, data and digital sequence information generated from the biodiversity or genetic resources associated with traditional lands, waters, and territories. The BC Notice may indicate that BC (Biocul- tural) Labels are in development and their implementation is being negotiated. For more information about the BC Notices, visit https:// local conte xts. org/ notice/ bc- notice/. Code availability N/A. Declarations Conflict of interests The authors have no conflicts of interest to de- clare. Ethical approval Authorisation from the New Zealand Department of Conservation was provided for collections within the Kermadec Islands Marine Reserve (Authorisation Number: 47976 MAR) and Poor Knights Islands Marine Reserve (Authorisation Number: 40106 MAR), and the Ministry for Primary Industries for all other collections (MPI Special Permit 632, 644-2). References Agassiz A (1863) Synopsis of the echinoids collected by Dr. W. Stimp- son on the North Pacifis explering expedition, under the command of captains ringgold and rodgers. In: Proceedings of the academy of natural sciences of Philadelphia, pp 352–361 Banks SC, Piggott MP, Williamson JE, Beheregaray LB (2007a) Microsatellite DNA markers for analysis of population structure in the sea urchin Centrostephanus rodgersii. Mol Ecol Notes 7:321–323. https:// doi. org/ 10. 1111/j. 1471- 8286. 2006. 01594.x Banks SC, Piggott MP, Williamson JE, Bové U, Holbrook NJ, Behere- garay LB (2007b) Oceanic variability and coastal topography shape genetic structure in a long-dispersing sea urchin. Ecology 88:3055–3064. https:// doi. org/ 10. 1890/ 07- 0091.1 Banks SC, Ling SD, Johnson CR, Piggott MP, Williamson JE, Behere- garay LB (2010) Genetic structure of a recent climate change- driven range extension. Mol Ecol 19:2011–2024. https:// doi. org/ 10. 1111/j. 1365- 294X. 2010. 04627.x Bronstein O, Kroh A, Tautscher B, Liggins L, Haring E (2017) Cryptic speciation in pan-tropical sea urchins: a case study of an edge-of-range population of Tripneustes from the Ker- madec Islands. Sci Rep 7:5948. https:// doi. org/ 10. 1038/ s41598- 017- 06183-2 Bronstein O, Kroh A, Miskelly AD, Smith SDA, Dworjanyn SA, Mos B, Byrne M (2019) Implications of range overlap in the com- mercially important pan-tropical sea urchin genus Tripneustes (Echinoidea: Toxopneustidae). Mar Biol 166:34. https:// doi. org/ 10. 1007/ s00227- 019- 3478-4 Brooks A (2020) Using ecological niche modelling to predict climate change responses of ten key fishery species in Aotearoa New Zea- land. Dissertation, Victoria University of Wellington, Wellington Burgess SC, Nickols KJ, Griesemer CD, Barnett LA, Dedrick AG, Sat- terthwaite EV, Yamane L, Morgan SG, White JW, Botsford LW (2014) Beyond connectivity: how empirical methods can quantify population persistence to improve marine protected-area design. Ecol App 24:257–270 Byrne M, Andrew N (2020) Centrostephanus rodgersii and Cen- trostephanus tenuispinus. Dev Aquacult Fish Sci 43:379–396. https:// doi. org/ 10. 1016/ B978-0- 12- 819570- 3. 00022-6 Byrne M, Gall M (2017) A tale of two urchins Heliocidaris erythro- gramma and H. tuberculata. Australian Echinoderms: Biology Ecology and Evolution, CSIRO Publishing Melbourne and ABRS, Canberra Casquet J, Thebaud C, Gillespie RG (2012) Chelex without boiling, a rapid and easy technique to obtain stable amplifiable DNA from small amounts of ethanol-stored spiders. Mol Ecol Resour 12:136–141. https:// doi. org/ 10. 1111/j. 1755- 0998. 2011. 03073.x Chiswell SM, Roemmich D (1998) The East Cape Current and two eddies: a mechanism for larval retention? N Z J Mar Freshw Res 32:385–397. https:// doi. org/ 10. 1080/ 00288 330. 1998. 95168 33 Edgar GJ, Stuart-Smith RD, Thomson RJ, Freeman DJ (2017) Consist- ent multi-level trophic effects of marine reserve protection across northern New Zealand. PLoS ONE 12:e0177216. https:// doi. org/ 10. 1371/ journ al. pone. 01772 16 Evanno G, Regnaut S, Goudet J (2005) Detecting the number of clus- ters of individuals using the software structure: a simulation study. Mol Ecol 14:2611–2620. https:// doi. org/ 10. 1111/j. 1365- 294X. 2005. 02553.x Excoffier L, Lischer HEL (2010) Arlequin suite ver 3.5: a new series of programs to perform population genetics analyses under Linux and Windows. Mol Ecol Resour 10:564–567. https:// doi. org/ 10. 1111/j. 1755- 0998. 2010. 02847.x Faubet P, Waples RS, Gaggiotti OE (2007) Evaluating the performance of a multilocus Bayesian method for the estimation of migration rates. Mol Ecol 16:1149–1166. https:// doi. org/ 10. 1111/j. 1365- 294X. 2007. 03218.x Gardner JPA, Bell JJ, Constable HB, Hannan D, Ritchie PA, Zuccarello GC (2010) Multi-species coastal marine connectivity: a literature review with recommendations for further research. New Zealand Aquatic Environment and Biodiversity Report No. 58. Victoria University, Wellington, New Zealand Gaylord B, Gaines SD (2000) Temperature or transport? Range limits of marine species mediated solely by flow. Am Nat 155:769–789 Goudet J (2005) hierfstat, a package for r to compute and test hierar- chical F-statistics. Mol Ecol Notes 5:184–186. https:// doi. org/ 10. 1111/j. 1471- 8286. 2004. 00828.x Hardy NA, Lamare M, Uthicke S, Wolfe K, Doo S, Dworjanyn S, Byrne M (2014) Thermal tolerance of early development in tropi- cal and temperate sea urchins: inferences for the tropicalization of eastern Australia. Mar Biol 161:395–409. https:// doi. org/ 10. 1007/ s00227- 013- 2344-z Hill KL, Rintoul SR, Ridgway KR, Oke PR (2011) Decadal changes in the South Pacific western boundary current system revealed in observations and ocean state estimates. J Geophys Res 116:C01009. https:// doi. org/ 10. 1029/ 2009J C0059 26 https://doi.org/10.6084/m9.figshare.14450163.v1 https://n2t.net/ark:/21547/DpW2 https://sites.massey.ac.nz/iramoana/ https://localcontexts.org/notice/bc-notice/ https://localcontexts.org/notice/bc-notice/ https://doi.org/10.1111/j.1471-8286.2006.01594.x https://doi.org/10.1890/07-0091.1 https://doi.org/10.1111/j.1365-294X.2010.04627.x https://doi.org/10.1111/j.1365-294X.2010.04627.x https://doi.org/10.1038/s41598-017-06183-2 https://doi.org/10.1038/s41598-017-06183-2 https://doi.org/10.1007/s00227-019-3478-4 https://doi.org/10.1007/s00227-019-3478-4 https://doi.org/10.1016/B978-0-12-819570-3.00022-6 https://doi.org/10.1111/j.1755-0998.2011.03073.x https://doi.org/10.1080/00288330.1998.9516833 https://doi.org/10.1371/journal.pone.0177216 https://doi.org/10.1371/journal.pone.0177216 https://doi.org/10.1111/j.1365-294X.2005.02553.x https://doi.org/10.1111/j.1365-294X.2005.02553.x https://doi.org/10.1111/j.1755-0998.2010.02847.x https://doi.org/10.1111/j.1755-0998.2010.02847.x https://doi.org/10.1111/j.1365-294X.2007.03218.x https://doi.org/10.1111/j.1365-294X.2007.03218.x https://doi.org/10.1111/j.1471-8286.2004.00828.x https://doi.org/10.1111/j.1471-8286.2004.00828.x https://doi.org/10.1007/s00227-013-2344-z https://doi.org/10.1007/s00227-013-2344-z https://doi.org/10.1029/2009JC005926 Marine Biology (2021) 168:138 1 3 138 Page 10 of 11 Huggett MJ, King CK, Williamson JE, Steinberg PD (2005) Larval development and metamorphosis of the Australian diadematid sea urchin Centrostephanus rodgersii. Invertebr Reprod Dev 47:197– 204. https:// doi. org/ 10. 1080/ 07924 259. 2005. 96521 60 IPCC (2019) IPCC Special Report on the Ocean and Cryosphere in a Changing Climate. In: Pörtner H-O, Roberts DC, Masson- Delmotte V, Zhai P, Tignor M, Poloczanska E, Mintenbeck K, Alegría A, Nicolai M, Okem A, Petzold J, Rama B, Weyer NM (eds) In press Johnson CR, Banks SC, Barrett NS, Cazassus F, Dunstan PK, Edgar GJ, Frusher SD, Gardner C, Haddon M, Helidoniotis F, Hill KL, Holbrook NJ, Hosie GW, Last PR, Ling SD, Melbourne-Thomas J, Miller K, Pecl GT, Richardson AJ, Ridgway KR, Rintoul SR, Ritz DA, Ross DJ, Sanderson JC, Shepherd SA, Slotwinski A, Swadling KM, Taw N (2011) Climate change cascades: Shifts in oceanography, species’ ranges and subtidal marine community dynamics in eastern Tasmania. J Exp Mar Biol Ecol 400:17–32. https:// doi. org/ 10. 1016/j. jembe. 2011. 02. 032 Jombart T (2008) Adegenet: a R package for the multivariate analysis of genetic markers. Bioinformatics 24:1403–1405. https:// doi. org/ 10. 1093/ bioin forma tics/ btn129 Jombart T, Devillard S, Balloux F (2010) Discriminant analysis of principal components: a new method for the analysis of geneti- cally structured populations. BMC Genet 11:94. https:// doi. org/ 10. 1186/ 1471- 2156- 11- 94 Jost L (2008) GST and its relatives do not measure differentiation. Mol Ecol 17:4015–4026. https:// doi. org/ 10. 1111/j. 1365- 294X. 2008. 03887.x Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, Buxton S, Cooper A, Markowitz S, Duran C, Thierer T (2012) Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bio- informatics 28:1647–1649. https:// doi. org/ 10. 1093/ bioin forma tics/ bts199 Keenan K, McGinnity P, Cross TF, Crozier WW, Prodöhl PA (2013) diveRsity: an R package for the estimation and exploration of pop- ulation genetics parameters and their associated errors. Methods Ecol Evol 4:782–788. https:// doi. org/ 10. 1111/ 2041- 210X. 12067 Kimura M, Crow JF (1964) The number of alleles that can be main- tained in a finite population. Genetics 49:725–738 Kimura M, Ohta T (1978) Stepwise mutation model and distribution of allelic frequencies in a finite population. Proc Natl Acad Sci 75:2868–2872. https:// doi. org/ 10. 1073/ pnas. 75.6. 2868 Kopelman NM, Mayzel J, Jakobsson M, Rosenberg NA, Mayrose I (2015) Clumpak: a program for identifying clustering modes and packaging population structure inferences across K. Mol Ecol Resour 15:1179–1191. https:// doi. org/ 10. 1111/ 1755- 0998. 12387 Last PR, White WT, Gledhill DC, Hobday AJ, Brown R, Edgar GJ, Pecl G (2011) Long-term shifts in abundance and distribution of a temperate fish fauna: a response to climate change and fishing practices. Glob Ecol Biogeogr 20:58–72 Law CS, Rickard GJ, Mikaloff-Fletcher SE, Pinkerton MH, Behrens E, Chiswell SM, Currie K (2018) Climate change projections for the surface ocean around New Zealand. N Z J Mar Freshw Res 52:309–335. https:// doi. org/ 10. 1080/ 00288 330. 2017. 13907 72 Liggins L, Gleeson L, Ciginos C (2014) Evaluating edge-of-range genetic patterns for tropical echinoderms, Acanthaster planci and Tripneustes gratilla, of the Kermadec Islands, southwest Pacific. Bull Mar Sci 90(1):379–397. https:// doi. org/ 10. 5343/ bms. 2013. 1015 Ling SD, Johnson CR (2009) Population dynamics of an ecologically important range-extender: kelp beds versus sea urchin barrens. Mar Ecol Prog Ser 374:113–125. https:// doi. org/ 10. 3354/ meps0 7729 Ling SD, Johnson CR, Ridgway K, Hobday AJ, Haddon M (2009) Climate-driven range extension of a sea urchin: inferring future trends by analysis of recent population dynamics. Glob Change Biol 15:719–731. https:// doi. org/ 10. 1111/j. 1365- 2486. 2008. 01734.x Luikart G, Cornuet J-M (1998) Empirical evaluation of a test for iden- tifying recently bottlenecked populations from allele frequency data. Conserv Biol 12:228–237 Meirmans PG (2014) Nonconvergence in Bayesian estimation of migration rates. Mol Ecol Resour 14:726–733. https:// doi. org/ 10. 1111/ 1755- 0998. 12216 Middleton I, Aguirre JD, Trnski T, Francis M, Duffy C, Liggins L (2021) Introduced alien, range extension or just visiting? Combin- ing citizen science observations and expert knowledge to classify range dynamics of marine fishes. Divers Distrib. https:// doi. org/ 10. 1111/ ddi. 13273 Misra R, Sérazin G, Meissner K, Gupta AS (2021) Projected changes to Australian marine heatwaves. Geophys Res Lett 48:e2020GL091323. https:// doi. org/ 10. 1029/ 2020G L0913 23 Monaco CJ, Booth DJ, Figueira WF, Gillanders BM, Schoeman DS, Bradshaw CJA, Nagelkerken I (2021) Natural and anthropogenic climate variability shape assemblages of range-extending coral- reef fishes. J Biogeogr 2021(00):1–13. https:// doi. org/ 10. 1111/ jbi. 14058 Mos B, Byrne M, Dworjanyn SA (2020) Effects of low and high pH on sea urchin settlement, implications for the use of alkali to counter the impacts of acidification. Aquaculture 528:735618. https:// doi. org/ 10. 1016/j. aquac ulture. 2020. 735618 Oosterhout CV, Hutchinson WF, Wills DPM, Shipley P (2004) Micro- checker: software for identifying and correcting genotyping errors in microsatellite data. Mol Ecol Notes 4:535–538. https:// doi. org/ 10. 1111/j. 1471- 8286. 2004. 00684.x Paradis E (2010) pegas: an R package for population genetics with an integrated–modular approach. Bioinformatics 26:419–420. https:// doi. org/ 10. 1093/ bioin forma tics/ btp696 Pecl GT, Araújo MB, Bell JD, Blanchard J, Bonebrake TC, Chen I-C, Clark TD, Colwell RK, Danielsen F, Evengård B, Falconi L, Ferrier S, Frusher S, Garcia RA, Griffis RB, Hobday AJ, Janion-Scheepers C, Jarzyna MA, Jennings S, Lenoir J, Linnet- ved HI, Martin VY, McCormack PC, McDonald J, Mitchell NJ, Mustonen T, Pandolfi JM, Pettorelli N, Popova E, Robinson SA, Scheffers BR, Shaw JD, Sorte CJB, Strugnell JM, Sunday JM, Tuanmu M-N, Vergés A, Villanueva C, Wernberg T, Wapstra E, Williams SE (2017) Biodiversity redistribution under climate change: impacts on ecosystems and human well-being. Science 355:eeai9214. https:// doi. org/ 10. 1126/ scien ce. aai92 14 Pecorino D, Lamare MD, Barker MF, Byrne M (2013) How does embryonic and larval thermal tolerance contribute to the distribu- tion of the sea urchin Centrostephanus rodgersii (Diadematidae) in New Zealand? J Exp Mar Biol Ecol 445:120–128. https:// doi. org/ 10. 1016/j. jembe. 2013. 04. 013 Perry AL, Low PJ, Ellis JR, Reynolds JD (2005) Climate change and distribution shifts in marine fishes. Science 308:1912–1915 Pinsky ML, Selden RL, Kitchel ZJ (2020) Climate-driven shifts in marine species ranges: scaling from organisms to communities. Annu Rev Mar Sci 12:153–179 Piry S, Luikart G, Cornuet J-M (1999) BOTTLENECK: a computer program for detecting recent reductions in the effective size using allele frequency data. J Heredity 90:502–503. https:// doi. org/ 10. 1093/ jhered/ 90.4. 502 Pitt NR, Poloczanska ES, Hobday AJ (2010) Climate-driven range changes in Tasmanian intertidal fauna. Mar Freshw Res 61:963–970 Poloczanska ES, Brown CJ, Sydeman WJ, Kiessling W, Schoeman DS, Moore PJ, Brander K, Bruno JF, Buckley LB, Burrows MT, Duarte CM (2013) Global imprint of climate change on marine life. Nat Clim Chang 3:919–925 https://doi.org/10.1080/07924259.2005.9652160 https://doi.org/10.1016/j.jembe.2011.02.032 https://doi.org/10.1093/bioinformatics/btn129 https://doi.org/10.1093/bioinformatics/btn129 https://doi.org/10.1186/1471-2156-11-94 https://doi.org/10.1186/1471-2156-11-94 https://doi.org/10.1111/j.1365-294X.2008.03887.x https://doi.org/10.1111/j.1365-294X.2008.03887.x https://doi.org/10.1093/bioinformatics/bts199 https://doi.org/10.1093/bioinformatics/bts199 https://doi.org/10.1111/2041-210X.12067 https://doi.org/10.1073/pnas.75.6.2868 https://doi.org/10.1111/1755-0998.12387 https://doi.org/10.1080/00288330.2017.1390772 https://doi.org/10.5343/bms.2013.1015 https://doi.org/10.5343/bms.2013.1015 https://doi.org/10.3354/meps07729 https://doi.org/10.3354/meps07729 https://doi.org/10.1111/j.1365-2486.2008.01734.x https://doi.org/10.1111/j.1365-2486.2008.01734.x https://doi.org/10.1111/1755-0998.12216 https://doi.org/10.1111/1755-0998.12216 https://doi.org/10.1111/ddi.13273 https://doi.org/10.1111/ddi.13273 https://doi.org/10.1029/2020GL091323 https://doi.org/10.1111/jbi.14058 https://doi.org/10.1111/jbi.14058 https://doi.org/10.1016/j.aquaculture.2020.735618 https://doi.org/10.1016/j.aquaculture.2020.735618 https://doi.org/10.1111/j.1471-8286.2004.00684.x https://doi.org/10.1111/j.1471-8286.2004.00684.x https://doi.org/10.1093/bioinformatics/btp696 https://doi.org/10.1093/bioinformatics/btp696 https://doi.org/10.1126/science.aai9214 https://doi.org/10.1016/j.jembe.2013.04.013 https://doi.org/10.1016/j.jembe.2013.04.013 https://doi.org/10.1093/jhered/90.4.502 https://doi.org/10.1093/jhered/90.4.502 Marine Biology (2021) 168:138 1 3 Page 11 of 11 138 Pritchard JK, Stephens M, Donnelly P (2000) Inference of population structure using multilocus genotype data. Genetics 155:945–959 Reisser CM, Bell JJ, Gardner JP (2014) Correlation between pelagic larval duration and realised dispersal: long-distance genetic connectivity between northern New Zealand and the Kermadec Islands archipelago. Mar Biol 161:297–312 Ridgway KR, Dunn JR (2003) Mesoscale structure of the mean East Australian Current System and its relationship with topography. Prog Oceanogr 56:189–222. https:// doi. org/ 10. 1016/ S0079- 6611(03) 00004-1 Ross PM, Hogg ID, Pilditch C, Lundquist CJ (2009) Phylogeography of New Zealand’s coastal benthos. N Z J Mar Freshwater Res 43:1009–1027 Sloyan BM, O’Kane TJ (2015) Drivers of decadal variability in the Tasman Sea. J Geophys Res 120:3193–3210. https:// doi. org/ 10. 1002/ 2014J C0105 50 Soars NA, Prowse TAA, Byrne M (2009) Overview of phenotypic plasticity in echinoid larvae, “Echinopluteus transversus” type vs. typical echinoplutei. Mar Ecol Prog Ser 383:113–125. https:// doi. org/ 10. 3354/ meps0 7848 Spencer HG, Waters JM, Eichhorst TE (2007) Taxonomy and nomen- clature of black nerites (Gastropoda: Neritimorpha: Nerita) from the South Pacific. Invertebr Syst 21:229–237. https:// doi. org/ 10. 1071/ IS060 38 Stanton BR (1979) The Tasman front. N Z J Mar Freshw Res 13:201– 214. https:// doi. org/ 10. 1080/ 00288 330. 1979. 95157 95 Sutton P, Chiswell S, Gorman R, Kennan S, Rickard G (2012) Physi- cal marine environment of the Kermadec Islands region. New Zealand Department of Conservation Waters JM, King TM, O’loughlin PM, Spencer HG (2005) Phylogeo- graphical disjunction in abundant high-dispersal littoral gastro- pods. Mol Ecol 14:2789–2802. https:// doi. org/ 10. 1111/j. 1365- 294X. 2005. 02635.x Wernberg T, Russell BD, Thomsen MS, Gurgel CFD, Bradshaw CJA, Poloczanska ES, Connell SD (2011) Seaweed communities in retreat from ocean warming. Curr Biol 21:1828–1832 Wolanski E, Hamner W (1988) Topographically controlled fronts in the ocean and their biological influence. Science 241:177–181 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. https://doi.org/10.1016/S0079-6611(03)00004-1 https://doi.org/10.1016/S0079-6611(03)00004-1 https://doi.org/10.1002/2014JC010550 https://doi.org/10.1002/2014JC010550 https://doi.org/10.3354/meps07848 https://doi.org/10.3354/meps07848 https://doi.org/10.1071/IS06038 https://doi.org/10.1071/IS06038 https://doi.org/10.1080/00288330.1979.9515795 https://doi.org/10.1111/j.1365-294X.2005.02635.x https://doi.org/10.1111/j.1365-294X.2005.02635.x The population genetic structure of the urchin Centrostephanus rodgersii in New Zealand with links to Australia Abstract Introduction Materials and methods Results Discussion Acknowledgements References