Source-sink Metapopulation Research Articles

Combining otolith chemistry and genetics to infer the population structure of yellow snapperLutjanus argentiventris

AbstractDeveloping a metapopulation framework contributes to the understanding of spatial processes and structures in populations, providing basic information for conservation biology. However, the extent to which the metapopulation structure differs geographically and across life history stages is unexplored for most fishes. Here, we compared the population structure and connectivity patterns of juvenile yellow snapperLutjanus argentiventrisin mangroves of the Eastern Tropical Pacific Ocean. In both the Gulf of California and Galápagos, these fish experience an array of environmental and oceanographic conditions that produce unique otolith microchemistry signatures and genetic population structures. In the Galápagos Archipelago, we used otolith microchemistry and microsatellite DNA, while for the Gulf of California, we only used otolith microchemistry. We observed relatively small differences in the otolith microchemistry at the ecoregion spatial scale (10–100 km2) for the Galápagos and the Gulf of California (classification accuracy of 57% and 64%, respectively). At the finer mangrove spatial scale (~1 km2), we found larger differences in the otolith microchemistry among the Gulf of California mangroves (classification accuracy ~90%) than in Galápagos (classification accuracy ~24%), potentially related to a stronger environmental gradient in the former. Based on otoliths, 75% of juveniles were migrants among mangroves in the Galápagos, in contrast to only around 10% of juveniles being migrants among the Gulf of California mangroves, suggesting a metapopulation structure characterized by high rates of self‐recruitment for snappers in the Gulf of California. For Galápagos, both otolith microchemistry and genetics supported the presence of a source–sink metapopulation structure for the yellow snapper, with high levels of connectivity in mangroves within the Eastern and Western ecoregions (FST = 0.003,p < 0.05). By combining tools that assess connectivity at multiple timescales (microchemistry and genetics) at different life stages and ecosystems, we provide a more complete elucidation of metapopulation structure forL. argentiventris. The approach employed here can be applied to other tropical fishes with broad distribution ranges and ontogenetic changes across complex habitats such as mangroves, seagrasses, and coral reefs.

Investigating local population dynamics of bottlenose dolphins in the northern Bahamas and the impact of hurricanes on survival

Little Bahama Bank in the northern Bahamas supports several populations of bottlenose dolphins (Tursiops truncatus). We provide the first estimates of birth rate and age-class-specific apparent survival rates for the local South Abaco population using data from a long-term (1997–2014) photo-identification (photo-ID) study and use the estimated life history parameters in a population viability analysis (PVA) to predict future population trends. Hurricane events are predicted to become more intense due to climate change but knowledge of how hurricanes may impact cetacean populations is limited. Little Bahama Bank is subject to hurricane activity, so we also investigate the potential impact of hurricanes on calf, juvenile and adult survival. Photo-ID data confirmed the existence of a core adult population with relatively high site fidelity in South Abaco, but also evidence of transient animals. Estimated annual birth rate was 0.278 (95% CI: 0.241–0.337). We found strong support for a decline in apparent survival for all age-classes. Estimated survival declined by 9% in adults (0.941 in 1998, to 0.855 in 2013), 5% in juveniles (0.820 in 2000, to 0.767 in 2013) and 36% in calves (0.970 in 1997, to 0.606 in 2013). Evidence that survival was influenced by repeated hurricane activity leading to increased mortality and/or emigration was stronger for calves and juveniles than for adults. PVA simulations of an assumed isolated South Abaco population showed that declines would lead to extinction within decades, even under the most optimistic scenario. Future work should focus on establishing if South Abaco is part of natural source–sink metapopulation dynamics on Little Bahama Bank by assessing trends in abundance in local populations and establishing how they interact; this will be important for assessing their conservation status in a potentially increasingly changing environment.

Comparing Genetic and Field-Based Estimates of Population Connectivity in Marbled Salamanders, Ambystoma opacum

Estimating connectivity is key for maintaining population viability for pond-breeding amphibians, especially in areas where habitat alterations occur. Here, we used genetic data (microsatellites) to estimate connectivity of marbled salamanders, Ambystoma opacum, among three focal ponds and compared it to field data (capture-mark-recapture estimates) of movement among the same ponds. In addition, we derived least-cost dispersal paths from genetic data and compared them to field connectivity estimates. We found that genetic and field estimates of dispersal were generally congruent, but field-based paths were more complex than genetic-based paths. While both methods complement each other in identifying important source-sink metapopulation dynamics to inform efficient conservation management plans, field data provide a more biologically accurate understanding of the spatial movement of individual marbled salamanders.

Dispersal and Good Habitat Quality Promote Neutral Genetic Diversity in Metapopulations.

Dispersal is a fundamental and crucial ecological process for a metapopulation to survive in heterogeneous or changing habitats. In this paper, we investigate the effect of the habitat quality and the dispersal on the neutral genetics diversity of a metapopulation. We model the metapopulation dynamics on heterogeneous habitats using a deterministic system of ordinary differential equations. We decompose the metapopulation into several neutral genetic fractions seeing as they could be located in different habitats. By using a mathematical model which describes their temporal dynamics inside the metapopulation, we provide the analytical results of their transient dynamics, as well as their asymptotic proportion in the different habitats. The diversity indices show how the genetic diversity at a global metapopulation scale is preserved by the correlation of two factors: the dispersal of the population, as well as the existence of adequate and sufficiently large habitats. The diversity indices show how the genetic diversity at a global metapopulation scale is preserved by the correlation of two factors: the dispersal of the population as well as the existence of adequate and sufficiently large habitats. Moreover, they ensure genetic diversity at the local habitat scale. In a source-sink metapopulation, we demonstrate that the diversity of the sink can be rescued if the condition of the sink is not too deteriorated and the migration from the source is larger than the migration from the sink. Furthermore, our study provides an analytical insight into the dynamics of the solutions of the systems of ordinary differential equations.

Metapopulation dynamics of the migratory fish Prochilodus lineatus (Characiformes: Prochilodontidae) in a lotic remnant of the Grande River, Southeastern Brazil

ABSTRACT River impoundments for electricity generation lead to environmental changes which severely affect fish migration and species richness. However, little is known about their effect on the genetic structure and population dynamics downstream from the reservoir. Here, we analyzed a set of ten microsatellite loci of Prochilodus lineatus, an important South American migratory fish. Specimens (n = 150) were sampled from five sites in a remnant lotic system that includes sections of the Grande, Pardo and Mogi Guaçu rivers, southeastern Brazil. The data showed that all microsatellites were polymorphic with the allele number per locus ranging from 5 to 32, and genetic diversity (H e ) varied from 0.74 to 0.80. Indices of genetic differentiation and Bayesian analysis showed a significant genetic structure and three genetic clusters inhabiting this river system. An asymmetric gene flow suggests source-sink metapopulation dynamics from tributaries (genetic source) to the main river (genetic sink). A genetic cluster that was not detected in the upper Mogi and Pardo rivers tributaries may indicate there is a “trapped gene pool” downstream from the Porto Colômbia dam. Thus, here we provide new insights into the genetic structure and population dynamics of a migratory fish species in a highly dammed river basin.

Number of Source Patches Required for Population Persistence in a Source-Sink Metapopulation with Explicit Movement.

We consider a simple metapopulation model with explicit movement of individuals between patches, in which each patch is either a source or a sink. We prove that similarly to the case of patch occupancy metapopulations with implicit movement, there exists a threshold number of source patches such that the population potentially becomes extinct below the threshold and established above the threshold. In the case where the matrix describing the movement of populations between spatial locations is irreducible, the result is global; further, assuming a complete mobility graph with equal movement rates, we use the principle of equitable partitions to obtain an explicit expression for the threshold. Brief numerical considerations follow.

Small-scale spatial variation in population- and individual-level reproductive parameters of the blue-legged hermit crab Clibanarius tricolor.

Management of the few regulated ornamental fisheries relies on inadequate information about the life history of the target species. Herein, we investigated the reproductive biology of the most heavily traded marine invertebrate in the western Atlantic; the blue-legged hermit crab Clibanarius tricolor. We report on density, individual-level, and population-level reproductive parameters in 14 populations spanning the Florida Keys. In C. tricolor, abundance, population-level, and individual-level reproductive parameters exhibited substantial small-scale spatial variation in the Florida Keys. For instance, the proportion of brooding females varied between 10–94% across localities. In females, average (±SD) fecundity varied between 184 (±54) and 614 (±301) embryos crab-1 across populations. Fecundity usually increases with female body size in hermit crabs. However, we found no effect of female body size on fecundity in three of the populations. Altogether, our observations suggest that C. tricolor may fit a source-sink metapopulation dynamic in the Florida Keys with low reproductive intensity and absence of a parental body size—fecundity relationship resulting in net reproductive loses at some localities. We argue in favor of additional studies describing population dynamics and other aspects of the natural history of C. tricolor (e.g., development type, larval duration) to reveal ‘source’ populations, capable of exporting larvae to nearby populations. Our observations imply that future studies aimed at assessing standing stocks or describing other aspects of the life history of this hermit crab need to focus on multiple localities simultaneously. This and future studies on the reproductive biology of this species will form the baseline for models aimed at assessing the stock condition and sustainability of this heavily harvested crustacean.

Spatial metrics effect of forest fragmentation on forest bird abundance and site occupancy probability: the influence of patch size and isolation

The persistence of species taxa within fragmented habitats is dependent on the source–sink metapopulation processes, and forest patch size and isolation are key factors. Unveiling species–patch area and/or species–patch isolation relationships may help provide crucial information for species and landscape management. In this study, relationship between forest patch size and isolation with abundance and occupancy probability of forest-dependent birds was investigated. This study was based within a coastal landscape that faces deleterious human activities such as clearing for agriculture. The study aimed to answer the question of whether the size and extent of isolation of forest patches influence abundance and/or occupancy probability of forest-specialist and generalist birds. Two bird species, namely Tiny Greenbul Phyllastrephus debilis subsp. rabai and Yellow-bellied Greenbul Chlorocichla flaviventris, were used as models. Birds were surveyed using distance sampling methods, and spatial metrics were measured from satellite imagery. Focal forest size and distance between forest patches were the most influential metrics whereby abundance and occupancy probabilities increased with increasing patch size, but were negatively influenced by increasing gaps between patches. These findings provide evidence of the existence of patch size/ isolation–occupancy relationships characterised by higher occupancy rate of large patches and distance-dependent dispersal, which decreased with increasing gaps between patches. Controlling deleterious human activities that reduce forest size should be a priority for the long-term conservation of forest-dependent birds.

Interspecific differences in larval production and dispersal in non-migratory galaxiids: implications for metapopulation structure

Metapopulation structure can strongly influence how species respond to extinction threats. Isolated populations are highly vulnerable to extirpation, whereas interconnected populations are generally more resilient to pressures. In fish with sedentary adults, population connectivity can depend on the movements of early life stages. The present study examined how previously identified interspecific differences in life history affect larval production and dispersal in a species-complex of non-migratory galaxiid fish. Consistent with a priori predictions, ‘fast’ life-history species (high size-relative fecundity, early maturation, small egg size) showed high abundance of potential recruits, whereas the larvae of ‘slow’ life-history species (low size-relative fecundity, late maturation, large egg size) were, on average, between six and nine times less abundant. The species with intermediate traits showed moderate larval abundance compared with other species. The small, poorly swimming larvae of a ‘fast’ life-history species dispersed over large spatial scales (up to over 12km), whereas the larger, better-swimming larvae of a ‘slow’ life-history species showed minimal dispersal (<300m). These findings suggest that the ‘fast’ life-history species are likely to follow a classic source–sink metapopulation structure, whereas ‘slow’ life-history species are more likely to form isolated population structures. The implications for how these species respond to pressures from invasive salmonids, a principal cause of their decline, are discussed.

Population connectivity shifts at high frequency within an open-coast marine protected area network.

A complete understanding of population connectivity via larval dispersal is of great value to the effective design and management of marine protected areas (MPA). However empirical estimates of larval dispersal distance, self-recruitment, and within season variability of population connectivity patterns and their influence on metapopulation structure remain rare. We used high-resolution otolith microchemistry data from the temperate reef fish Hypsypops rubicundus to explore biweekly, seasonal, and annual connectivity patterns in an open-coast MPA network. The three MPAs, spanning 46 km along the southern California coastline were connected by larval dispersal, but the magnitude and direction of connections reversed between 2008 and 2009. Self-recruitment, i.e. spawning, dispersal, and settlement to the same location, was observed at two locations, one of which is a MPA. Self-recruitment to this MPA ranged from 50–84%; within the entire 60 km study region, self-recruitment accounted for 45% of all individuals settling to study reefs. On biweekly time scales we observed directional variability in alongshore current data and larval dispersal trajectories; if viewed in isolation these data suggest the system behaves as a source-sink metapopulation. However aggregate biweekly data over two years reveal a reef network in which H. rubicundus behaves more like a well-mixed metapopulation. As one of the few empirical studies of population connectivity within a temperate open coast reef network, this work can inform the MPA design process, implementation of ecosystem based management plans, and facilitate conservation decisions.

Individual, population, and ecosystem effects of hypoxia on a dominant benthic bivalve in Chesapeake Bay

Hypoxia is an environmental stressor that affects abundance, biomass, diversity, and ecosystem function of benthic assemblages worldwide, yet its collective impact at individual, population, and ecosystem levels has rarely been investigated. We examined the effects of hypoxia on the biomass‐dominant clam, Macoma balthica, in the York and Rappahannock Rivers (Chesapeake Bay, USA). We (1) surveyed the M. balthica populations in both rivers in 2003 and 2004, (2) determined the effects of low dissolved oxygen (DO) on M. balthica fecundity in a laboratory experiment, and (3) employed a predator‐exclusion field experiment to establish the effects of hypoxia and prey density on predation upon M. balthica. The resultant data were used to parameterize a matrix model, which was analyzed to define potential effects of hypoxia at the population level. In both rivers, hypoxia decreased individual clam growth and caused local extinction of populations. Hypoxia reduced egg production of M. balthica by 40% and increased protein investment per egg. In the predator‐exclusion field experiment, hypoxia magnified predation rates threefold and altered the functional response of predators to M. balthica from a stabilizing type III functional response to a destabilizing type II functional response. In a density‐independent matrix model, hypoxia resulted in coupled source–sink metapopulation dynamics, with hypoxic areas acting as black‐hole sinks. Increases in the spatial and temporal extent of hypoxia caused the populations to decline toward extinction. In a second model that incorporated density dependence, under mild hypoxic conditions trophic transfer from M. balthica to predators increased, but at increased spatial or temporal extent of hypoxia trophic transfer decreased. The major decline in trophic transfer to predators under severe hypoxia resulted from diversion of M. balthica biomass into the microbial loop. Our model predicted that there are multiple stable states for M. balthica populations (high and very low densities), such that the saddle point (threshold at which the population switches from one state to the other) increased and resilience decreased with the spatial extent of hypoxia. We underscore how effects of a stressor at the individual level can combine to have substantial population and ecosystem‐level effects.

Florida Gulf Bay Scallop (Argopecten Irradians Concentricus) Population Genetic Structure: Form, Variation, and Influential Factors

Knowledge of changes in the interdependence of a species' populations for continued existence (connectivity) and of the forces driving connectivity patterns is critical for management and conservation of the species. Population genetics can be used to investigate population connectivity. Combining population genetics with climatic, environmental, and biological (external) factors that can influence population structure and connectivity can lead to a greater understanding of the forces influencing population dynamics. We deciphered the population genetic structure and connectivity patterns of bay scallops (Argopecten irradians concentricus Lamarck) from Florida Gulf of Mexico waters using allozyme-locus and mitochondrial DNA population genetics data from samples representing four consecutive generations. We evaluated bay scallop population genetic relationships within the context of habitat, hydrodynamic, and environmental variation, and concomitant spatial patterns in the abundance of adult and recently recruited scallops to infer the influence of these factors on bay scallop population structure. Florida Gulf bay scallops form a hierarchical, mixed-model, source-sink metapopulation with intergenerational variation in connectivity that seems to be influenced by the factors we considered. An El Niño event that occurred during 1 y caused climatic and environmental change that greatly reduced subpopulation connectivity, allowing us to understand more completely the potential relative importance of the other external factors on change in metapopulation connectivity over time. Our study illustrates the value of using multiple genetic markers, sampling for multiple years, and integrating data from multiple external factors for understanding the population genetic structure of species for adaptive management.

Meso-scale hydrodynamic and reproductive asynchrony affects the source–sink metapopulation structure of the coastal polychaete Pectinaria koreni

The polychaete Pectinaria koreni exhibits a complex life cycle characterized by non-overlapping generations and widespread larval dispersal. To explore how “local” metapopulation genetic structure varies spatially and temporally during population turnover, we combined observations on demography, larval dispersal through hydrodynamic modelling and population genetics of successive age cohorts in the Baie de Seine (eastern English Channel, France). Mature adults (March), newly settled (July) and later-stage juveniles (September) were sampled in 2003 on the edge and in the main demes of the metapopulation. Demes displayed an asynchronous dynamics due to variations in habitat quality affecting reproductive timing (e.g. three distinct spawning events observed) and in local larval supply linked to temporal fluctuations of hydrodynamism. Two-source populations were identified among dense areas with the greatest larval retention and self-recruitment rates: one with a single recruitment event, stable temporal genetic variation and a strong spatial genetic re-homogenization during turnover, and the other with two recruitment events and significant allele frequency changes over time. Sink habitats displayed single recruitment event and experienced strong inter-generational (juveniles vs adults) genetic changes due to genetic drift associated with strong winter mortalities. Altogether, results suggested that adult spawning asynchrony and sweepstakes reproductive success, together with genetic drift, played a greater role than hydrodynamics itself in determining effective recruitment rates at some sites and generating genetic patchiness.

Experimental demonstration of accelerated extinction in source-sink metapopulations.

Population extinction is a fundamental ecological process which may be aggravated by the exchange of organisms between productive (source) and unproductive (sink) habitat patches. The extent to which such source-sink exchange affects extinction rates is unknown. We conducted an experiment in which metapopulation effects could be distinguished from source-sink effects in laboratory populations of Daphnia magna. Time-to-extinction in this experiment was maximized at intermediate levels of habitat fragmentation, which is consistent with a minority of theoretical models. These results provided a baseline for comparison with experimental treatments designed to detect effects of concentrating resources in source patches. These treatments showed that source-sink configurations increased population variability (the coefficient of variation in abundance) and extinction hazard compared with homogeneous environments. These results suggest that where environments are spatially heterogeneous, accurate assessments of extinction risk will require understanding the exchange of organisms among population sources and sinks. Such heterogeneity may be the norm rather than the exception because of both the intrinsic heterogeneity naturally exhibited by ecosystems and increasing habitat fragmentation by human activity.

New approaches to source–sink metapopulations decoupling demography and dispersal

Source–sink systems are metapopulations of patches with different, and possibly temporally varying, habitat qualities, which are commonly used in ecology to study the fate of spatially extended populations.We propose new techniques that disentangle the respective contributions of demography and dispersal to the dynamics and fate of a single species in a source–sink system. Our approach is valid for a general class of stochastic, individual-based, stepping-stone models, with density-independent demography and dispersal, provided that the metapopulation is finite or else enjoys some transitivity property.We provide (1) a simple criterion of persistence, by studying the motion of a single random disperser until it returns to its initial position; (2) a joint characterization of the long-term growth rate and of the asymptotic occupancy frequencies of the ancestral lineage of a random survivor, by using large deviations theory. Both techniques yield formulas decoupling demography and dispersal, and can be adapted to the case of periodic or random environments, where habitat qualities are autocorrelated in space and possibly in time.In this last case, we display examples of coupled time-averaged sinks allowing survival, as was previously known in the absence of demographic stochasticity for fully mixing (Jansen and Yoshimura, 1998) or partially mixing (Evans et al., 2012; Schreiber, 2010) metapopulations.

Brownsville Common Yellowthroats Maintain Genetic Diversity in a Fragmented Landscape

Habitat fragmentation is particularly severe in the Lower Rio Grande Valley (LRGV) of Texas, where greater than 90% of the native wetland habitat has been replaced by agriculture and urban development. The Brownsville Common Yellowthroat (Geothlypis trichas insperata), a rare subspecies of the Common Yellowthroat restricted to this area, is a good model for studying the effects of this extensive fragmentation on avian genetics and ecology. In this study, three hypotheses addressing genetics and habitat fragmentation were tested: (1) the density of Brownsville Yellowthroats is greatest along the Rio Grande rather than away from the river; (2) individuals found on or near the Rio Grande are more closely related than individuals farther away; and (3) a source-sink metapopulation model is best fit to describe the population dynamics of this subspecies. We used 15 microsatellites to examine the genetic diversity of 128 individuals from Cameron and Hidalgo Counties (Texas) sampled during 2008–2009. The densities of yellowthroats at sites near the Rio Grande were not significantly different from densities at sites farther from the river. Genetic analysis indicated that individuals were as related to each other as would be expected in distant relatives and indicated the presence of a single, admixed population. The results suggest that the Brownsville Common Yellowthroat is able to move freely among isolated habitat patches and that wetlands along the Rio Grande do not act as an exclusive corridor for this species.

Population structure of sexually reproducing carp gudgeons: does a metapopulation offer refuge from sexual parasitism?

Australian carp gudgeons (Hypseleotris spp.) of the Murray–Darling basin are a species complex including sexually reproducing taxa and unisexual hybrid lineages that reproduce via hybridogenesis. Unisexual fish require gametes of a sexual species to propagate themselves and can be regarded as ‘sexual parasites’ capable of driving closed populations to extinction. Metapopulation dynamics have been proposed as a mechanism that could facilitate coexistence between a sexual parasite and its ‘host’. This study evaluates whether patterns of spatial genetic variation are compatible with metapopulation dynamics for a sexually reproducing member of the carp gudgeon complex (Hypseleotris sp. HA), in the Granite Creeks system of central Victoria. Genetic differentiation of fish among all study sites was accommodated by a model of migration-drift equilibrium using decomposed pairwise regression analysis. Given that the population was divided into discrete patches in the form of refugial waterholes during the time of this study, we infer that spatially constrained source–sink metapopulation dynamics may be responsible for producing this pattern. It is therefore possible that metapopulation dynamics contribute to coexistence in the Granite Creeks carp gudgeon hybridogenetic system, and further analysis is required to determine the relative importance of environmental versus demographic factors towards patch extinction.

Monitoring a Puma (Puma concolor) Population in a Fragmented Landscape in Southeast Brazil

ABSTRACT The northeast area of São Paulo state has been intensively deforested, resulting in a highly fragmented landscape composed of a few large patches and several small patches of natural vegetation surrounded by sugarcane, eucalyptus, and citrus plantations. In this scenario, we investigated the puma (Puma concolor) population size, sex ratio, and relatedness in two of the last, and largest, natural refuges in the area using a noninvasive method during 2004–2008. By collecting and individualizing fecal samples by microsatellites, we identified 17 individuals, 13 females (76.4%) and 4 males (23.6%) in these areas. Five females were sampled in distinct years and over an extended time and probably represented resident adults. By investigating the relatedness among individual pumas inhabiting the area, we found that only three animals were not related to each other. We also found evidence that young females might establish an adjacent or overlapping territory to their mothers (phylopatry). Moreover, we registered 11 road‐killed individuals nearby the study area, ten males and one female, and six human–puma conflicts. The study area may act as a source of individuals that disperse across the matrix to occupy new home ranges, maintaining some degree of gene flow in a source–sink metapopulation structure. Finally, we recommend that puma management should be conducted at the landscape level to provide effective puma conservation in northeastern São Paulo state.Abstract in Portuguese is available in the online version of this article.

Population structure in the South American tern Sterna hirundinacea in the South Atlantic: two populations with distinct breeding phenologies

The South American tern Sterna hirundinacea is a migratory species for which dispersal, site fidelity and migratory routes are largely unknown. Here, we used five microsatellite loci and 799 bp partial mitochondrial DNA sequences (Cytochrome b and ND2) to investigate the genetic structure of South American terns from the South Atlantic Ocean (Brazilian and Patagonian colonies). Brazilian and Patagonian colonies have two distinct breeding phenologies (austral winter and austral summer, respectively) and are under the influence of different oceanographic features (e.g. Brazil and Falklands/Malvinas ocean currents, respectively), that may promote genetic isolation between populations. Results show that the Atlantic populations are not completely panmictic, nevertheless, contrary to our expectations, low levels of genetic structure were detected between Brazilian and Patagonian colonies. Such low differentiation (despite temporal isolation of the colonies) could be explained by demographic history of these populations coupled with ongoing levels of gene flow. Interestingly, estimations of gene flow through Maximum likelihood and Bayesian approaches has indicated asymmetrical long term and contemporary gene flow from Brazilian to Patagonian colonies, approaching a source–sink metapopulation dynamic. Genetic analysis of other South American tern populations (especially those from the Pacific coast and Falklands–Malvinas Islands) and other seabird species showing similar geographical distribution (e.g. royal tern Thalasseus maximus), are fundamental in gaining a better understanding of the main processes involved in the diversification of seabirds in the southern hemisphere.

Evidence of source–sink metapopulations in a vulnerable native galaxiid fish driven by introduced trout

Introduced predators with patchy distributions can create demographic sinks within native prey populations. Such invasions may give rise to source-sink metapopulations if there are still sources of native species colonists in the landscape. In New Zealand, introduced brown trout (Salmo trutta) and rainbow trout (Oncorhychus mykiss) are linked with declines in native non-diadromous galaxiids but co-occur with these galaxiids in some locations. We investigated whether trout create sinks in Galaxias vulgaris populations, and whether trout-free reaches could act as sources, allowing persistence in the sink habitat. We conducted quantitative seasonal monitoring of G. vulgaris population structure across two subcatchments of the Waimakariri River, South Island. Two trout-free and seven trout-invaded sites in the Porter River catchment and two trout-free and five trout-invaded sites in the Broken River catchment were monitored over two winters and the adjoining summer. Spatially continuous monitoring of young-of-the-year (YOY) galaxiid distributions and apparent survival across the Broken River catchment was also undertaken. Galaxias vulgaris YOY recruitment was high in trout-free reaches, indicating positive population growth. Galaxias vulgaris was absent from three trout-invaded sites, and the remaining invaded sites had significantly depleted juvenile recruitment. Information-theoretic model selection indicated that trout, rather than habitat, drove recruitment failure. Trout-invaded sites could be divided into "sinks" that retained no YOY galaxiids, indicating no local recruitment, and "pseudosinks," which had very few recruits. Absence of small G. vulgaris at sink sites suggested population maintenance through immigration of adults from sources, whereas pseudosink sites appear capable of self-recruitment at low carrying capacities. Trout-free reaches appear to act as sources in a river network but are susceptible to future invasions by trout. Thus, not only may invasive species cause source-sink metapopulations in native species, but also the potential of refugia for natives (sources) to become future sinks highlights the vulnerability of these metapopulations when invasive predators are the principal demographic driver.