Evaluating the importance of demographic connectivity in a marine metapopulation

Recently researchers have gone to great lengths to measure marine metapopulation connectivity via tagging, genetic, and trace-elemental fingerprinting studies. These empirical estimates of larval dispersal are key to assessing the significance of metapopulation connectivity within a demographic context, but the life-history data required to do this are rarely available. To evaluate the demographic consequences of connectivity we constructed seasonal, size-structured metapopulation matrix models for two species of mytilid mussel in San Diego County, California, USA. The self-recruitment and larval exchange terms were produced from a time series of realized connectivities derived from trace-elemental fingerprinting of larval shells during spring and fall from 2003 to 2008. Both species exhibited a strong seasonal pattern of southward movement of recruits in spring and northward movement in fall. Growth and mortality terms were estimated using mark-recapture data from representative sites for each species and subpopulation, and literature estimates of juvenile mortality. Fecundity terms were estimated using county-wide settlement data from 2006-2008; these data reveal peak reproduction and recruitment in fall for Mytilus californianus, and spring for M. galloprovincialis. Elasticity and life-stage simulation analyses were employed to identify the season- and subpopulation-specific vital rates and connectivity terms to which the metapopulation growth rate (lambda) was most sensitive. For both species, metapopulation growth was most sensitive to proportional changes in adult fecundity, survival and growth of juvenile stages, and population connectivity, in order of importance, but relatively insensitive to adult growth or survival. The metapopulation concept was deemed appropriate for both Mytilus species as exchange between the subpopulations was necessary for subpopulation persistence. However, highest metapopulation growth occurred in years when a greater proportion of recruits was retained within the predominant source subpopulation. Despite differences in habitat and planktonic duration, both species exhibited similar overall metapopulation dynamics with respect to key life stages and processes. However, different peak reproductive periods in an environment of seasonal current reversals led to different regional (subpopulation) contributions to metapopulation maintenance; this result emphasizes the importance of connectivity analysis for spatial management of coastal resources.     

Temporal patterns of larval settlement and survivorship of two broadcast-spawning acroporid corals

Acroporid corals are the main reef-building corals that provide three-dimensional habitats for other reef organisms, but are decreasing on many reefs worldwide due to natural and anthropogenic disturbances. In this study, temporal patterns of larval settlement and survivorship of two broadcast-spawning acroporid coral species, Acropora muricata and A. valida, were examined through laboratory rearing experiments to better understand the potential for larval dispersal of this important coral group. Many larvae were attached (but not metamorphosed) to settlement tiles on the first examination 3–4 days after spawning (AS). The first permanent larval settlement (i.e. metamorphosed and permanently settled juvenile polyps) occurred at 5–6 days AS, and most larval settlement (85–97% of total) occurred within 9–10 days AS. Larval survivorship decreased substantially to around 50% by the first week of the experiment and to approximately 10% by the second to third week. The rates of larval attachment, settlement, and the initial drop in survivorship of larvae suggest that effective dispersal of some acroporid species may largely be completed within the first few weeks AS.     

Spatial variability in habitat associations of pre- and post-settlement stages of coral reef fishes at Ishigaki Island,Japan

Successful settlement of pelagic fish larvae into benthic juvenile habitats may be enhanced by a shortened settlement period, since it limits larval exposure to predation in the new habitat. Because the spatial distribution of marine fish larvae immediately prior to settlement versus during settlement was unknown, field experiments were conducted at Ishigaki Island (Japan) using light trap sampling and underwater visual belt transect surveys to investigate the spatial distribution patterns of selected pre- and post-settlement fishes (Acanthuridae, Pomacentridae, Chaetodonidae and Lethrinidae) among four habitats (seagrass bed, coral rubble, branching coral and tabular coral). The results highlighted two patterns: patterns 1, pre- and post-settlement individuals showing a ubiquitous distribution among the four habitats (Acanthuridae) and pattern 2, pre-settlement individuals distributed in all habitats, but post-settlement individuals restricted to coral (most species of Pomacentridae and Chaetodontidae) or seagrass habitats (Lethrinidae). The first pattern minimizes the transition time between the larval pelagic stage and acquisition of a benthic reef habitat, the latter leading immediately to a juvenile lifestyle. In contrast, the second pattern is characterized by high settlement habitat selectivity by larvae and/or differential mortality immediately after settlement.     

Spatial variation in population dynamics in a colonial ascidian (Botryllus schlosseri)

Recent interest in the dynamics of marine invertebrate populations has focused largely on taxa with an open population structure. However, in many colonial taxa with limited larval dispersal, settlers may be locally derived. Consequently, dynamics may vary among sites that are separated by relatively short distances. This study explored spatial variation in temporal dynamics of colonial ascidians (Botryllus schlosseri Pallas) inhabiting five sites distributed along a ≈ 17-km temperature and phytoplankton gradient in the Damariscotta River estuary, Maine, USA. Settlement and population densities and sexual reproductive status were assayed throughout the summer seasons of 1996 and 1997. Sexual reproduction and larval settlement commenced earlier in the summer in up-river populations, which subsequently underwent a seasonal population explosion that was much smalier in down-river populations. Two peaks in settlement density up-river (in early July and early September) suggest that colonies there may have completed two sexual generations, in contrast to a single generation at down-river sites. Similar spatial variation is expected among populations of other taxa with limited larval dispersal when they are distributed across environmental gradients. Published online: 18 September 2002     

Larval period and its influence on post-larval life history: comparison of lecithotrophy and facultative planktotrophy in the aeolid nudibranch Phestilla sibogae

The variable duration of the pelagic phase of metamorphically competent larvae of benthic marine invertebrates is set by an interaction between environmental factors and larval traits that together influence the chance that a larva will encounter and respond to a suitable settlement site. In the Hawaiian aeolid nudibranch Phestilla sibogae Bergh, an extended competent larval phase resulted in a cascade of negative effects on larval and post-larval life-history traits. When raised as fed (i.e., facultatively planktotrophic) larvae, an extended larval period resulted in lower larval survival, slightly lower metamorphic success, and delayed reproduction. When raised as unfed (i.e., lecithotrophic) larvae, an extended larval period resulted in lower larval and post-larval weights, survival, metamorphic success, and reproductive output, and also resulted in a longer juvenile period and delayed reproduction. The chance nature of locating a settlement site generally spreads these negative effects over all larvae of a cohort, and so balances the relative fitness of the genetic lineages within a population.     

Integrating larval connectivity into the marine conservation decision-making process across spatial scales

Larval dispersal connectivity is typically integrated into spatial conservation decisions at regional or national scales, but implementing agencies struggle with translating these methods to local scales. We used larval dispersal connectivity at regional (hundreds of kilometers) and local (tens of kilometers) scales to aid in design of networks of no-take reserves in Southeast Sulawesi, Indonesia. We used Marxan with Connectivity informed by biophysical larval dispersal models and remotely sensed coral reef habitat data to design marine reserve networks for 4 commercially important reef species across the region. We complemented regional spatial prioritization with decision trees that combined network-based connectivity metrics and habitat quality to design reserve boundaries locally. Decision trees were used in consensus-based workshops with stakeholders to qualitatively assess site desirability, and Marxan was used to identify areas for subsequent network expansion. Priority areas for protection and expected benefits differed among species, with little overlap in reserve network solutions. Because reef quality varied considerably across reefs, we suggest reef degradation must inform the interpretation of larval dispersal patterns and the conservation benefits achievable from protecting reefs. Our methods can be readily applied by conservation practitioners, in this region and elsewhere, to integrate connectivity data across multiple spatial scales.     

Evolutionary ecology of settlement size in planktotrophic neritimorph gastropods

Body size during larval development is one of the most important attributes of aquatic animals. The optimal size for changing form or habitat may vary according to ecological traits of species, while phylogenetic constraints also play a significant role. The major goal of this study is to reveal the patterns in the settlement size of planktotrophic larvae in an archaic gastropod superorder Neritimorpha. We take advantage of the fact that size at various ontogenetic stages of neritimorphs can be rigorously estimated by measuring features of the adult opercula. This unique feature of neritimorphs has allowed us to generate the largest data set so far on larval settlement sizes within a group of marine invertebrates that recruit into very different post-metamorphic habitats. Eighty-eight species that represent most extant genera from rocky shores, seagrass beds, mangroves, estuaries, streams, submarine caves, deep-sea vents or seeps showed negligible intraspecific variation and considerable interspecific differences in settlement size, particularly between genera or families. Settlement size is determined primarily by phylogenetic constraints, while parallel evolution toward smaller sizes was shown to occur exclusively in four independent clades (two living and two extinct) of amphidromous snails with a marine larval period followed by a limnic adult phase. The smaller settlement size may possibly reduce the risk of being wafted away from the estuaries of their natal streams through less time achieving metamorphic competence, while ability to make occasional long-distance trips is retained by the presence of a sufficiently long delay period. This delay period also seems to obscure the possible correlation between settlement size and geographic distribution range of neritimorph species, both fully marine and amphidromous.     

Endemism and Regional Color and Genetic Differences in Five Putatively Cosmopolitan Reef Fishes

Abstract:  Endemism is thought to be relatively rare in marine systems due to the lack of allopatric barriers and the potential for long-distance colonization via pelagic larval dispersal. Although many species of coral reef fishes exhibit regionally restricted color variants that are suggestive of regional endemism, such variation is typically ascribed to intraspecific variation. We examined the genetic structure in 5 putatively monospecific fishes from the Indo-West Pacific ( Amphiprion melanopus , Chrysiptera talboti , and Pomacentrus moluccensis [Pomacentridae] and Cirrhilabrus punctatus , and Labroides dimidiatus [Labridae]) that express regional color variation unique to this area. Mitochondrial-control-region sequence analysis showed shallow to deep genetic divergence in all 5 species (sequence divergence 2–17%), with clades concordant with regional color variation. These results were partially supported by nuclear RAG2 data. An analysis of molecular variation (AMOVA) mirrored the phylogenetic results; Φ_ST values ranged from 0.91 to 0.7, indicating high levels of geographic partitioning of genetic variation. Concordance of genetics and phenotype demonstrate the genetic uniqueness of southwestern Pacific color variants, indicating that these populations are at a minimum distinct evolutionarily significant units and perhaps distinct regionally endemic species. Our results indicate that the alpha biodiversity of the southwestern Pacific is likely underestimated even in well-studied groups, such as reef fishes, and that regional endemism may be more common in tropical marine systems than previously thought.     

Experimental investigation into the effects of suspended sediment on fertilisation, larval survival and settlement in a scleractinian coral

Laboratory and field experiments were used to determine whether high (≃100 mg l⁻¹), low (≃50 mg l⁻¹) and control (≃0 mg l⁻¹) levels of suspended sediment affected fertilisation, larval survival, and larval settlement in the scleractinian coral Acropora digitifera (Dana, 1846). Both high- and low-sediment treatments significantly decreased fertilisation, but post-fertilisation embryonic development was not inhibited by suspended sediments. Larval survival and larval settlement were significantly reduced in high- and low-sediment treatments. No difference was found between high- and low-sediment treatments in any of the three post-spawning processes investigated, suggesting that they are susceptible to sediment concentrations which are not exceptionally high even under natural conditions (>50 mg l⁻¹). The introduction of an additional stress in the form of high levels of suspended sediments coupled with naturally high variability in recruitment may have a considerable effect on the successful supply and settlement of coral larvae to a reef. Given that many coral communities are open reproductive systems, the consequences of disturbance events are not likely to be restricted to the impact area. Recruitment to a population may be reduced significantly in the presence of high levels of suspended sediments because of effects on larval survival and settlement. Recruitment of larvae to adjacent populations may also be affected due to a decreased fertilisation success and potential increases in mortality of larvae passing through the affected site. Received: 13 August 1998 / Accepted: 22 July 1999     

Larvae of a colonial ascidian use a non-contact mode of substratum selection on a coral reef

The rate at which larvae successfully recruit into communities of marine benthic invertebrates is partially dependent upon how well larvae avoid benthic predators and settle on appropriate substrata. Therefore, to be able to predict recruitment success, information is needed on how larvae search for settlement sites, whether larvae preferentially settle on certain substrata, and the extent to which there are adequate cues for larvae to find these substrata. This article describes how larvae of the colonial ascidian Diplosoma similis find settlement sites on a coral reef. Direct field observations of larval settlement were made on a fringing reef in Kaneohe Bay, Oahu, Hawaii, between September 1985 and April 1986. A comparison of the substrata that larvae contacted prior to settlement relative to the percentage cover of these substrata on the study reef suggests that larvae are using a non-contact mode of substratum identification to locate suitable settlement sites. This mode of substratum identification allowed 74% of larvae to evade predation by benthic organisms who would otherwise have eaten larvae if they had been contacted. Of those larvae that evaded predation, 88% subsequently settled on the same two substrata upon which most adults are found (dead coral or the green alga Dictyosphaeria cavernosa). This pattern of settlement was probably a result of active selection, since the two substrata cover only 14.4% of the reef's surface and currents had little effect on the direction in which larvae swam. An important contributing factor to the high success rate of larval settlement on suitable substrata was the lack of any temporal decay in substratum preference. It is concluded that for Diplosoma similis larval supply is a sufficient predictor of larval settlement rate. However, for marine invertebrates whose larvae are passively dispersed and exhibit a greater temporal decay in substratum preference, larval settlement should generally have a greater dependency on spatial variation in the abundance of benthic predators and suitable substrata.     

Early recognition of sediment during settlement of<Emphasis Type="Italic"> Eupolymnia nebulosa</Emphasis> (Polychaeta: Terebellidae) larvae

Analysis of the displacements and activity of larval stages of the terebellid polychaete Eupolymnia nebulosa were carried out on different sediment types collected in the Bay of Banyuls-sur-Mer (western Mediterranean). Results indicated sediment recognition by the larvae. Time devoted to exploration and settlement trials were significantly longer when sediments of similar granulometry had been previously reworked by adults. There were increased responses of settling larvae if the tentacles of adults of the same species had previously been in contact with the tested sediment. The early settlement activity was stronger at the sediment boundaries and increased in the coarse sediment. Our results suggest that larvae are able to recognize biologically bioturbed sediments and modify their behaviour accordingly. There is a larval settlement before the competent stage. Possible consequences of early settlement and the limitation of dispersal in Eupolymnia nebulosa larvae are discussed together with the results of automated measurements of early larval swimming.Communicated by S.A. Poulet, Roscoff     

Critical swimming speeds of late-stage coral reef fish larvae: variation within species,among species and between locations

The swimming abilities of larval fishes are important for their survival, potentially affecting their ability to avoid predators, obtain food and control dispersal patterns. Near settlement swimming abilities may also influence spatial and temporal patterns of recruitment. We examined Critical speed (U-crit) swimming ability in late stage larvae of 89 species of coral reef fishes from the Great Barrier Reef and the Caribbean. Coefficients of variation in U-crit calculated at the individual level were high (28.4%), and this was not explained by differences in size or condition factor of these same larvae. Among species U-crit ranged from 5.5 cm s⁻¹ to 100.8 cm s⁻¹ (mean=37.3 cm s⁻¹), with 95% of species able to swim faster than the average current speed around Lizard Island, suggesting that most species should be capable of influencing their spatial and temporal patterns of settlement. Inter-specific differences in swimming ability (at both the family and species levels) were significantly correlated with size and larval morphology. Correlations were found between swimming performance and propulsive area, fineness ratio and aspect ratio, and these morphological parameters may prove useful for predicting swimming ability in other taxa. Overall, the swimming speeds of larvae from the same families at the two locations were relatively similar, although the Lutjanidae and Acanthuridae from the Caribbean were significantly slower than those from the great barrier reef. Differences in swimming speed and body form among late stage larvae suggests that they will respond differently to factors influencing survival and transport during their pelagic phase, as well as habitat use following settlement.     

Settlement-competency periods of larvae of three species of scleractinian corals

The pattern of settlement over time of three broadcast spawning coral species (Cyphastrea serailia, Acanthastrea lordhowensis, and Goniastrea australensis) from the Solitary Islands (30°00′S; 153°20′E) was studied in 1995 and 1996 in order to determine the maximum length of time these larvae could remain in the water column and still retain the ability to settle and metamorphose. Larvae were maintained in aquaria and the number which had settled on biologically-conditioned tile pairs was monitored every 5 to 10 d. While the majority of larvae settled quickly after becoming competent, some larvae survived and settled for extended periods after spawning. Competency periods ranged from 26 d for C. serailia to 56 d for G. australensis and 78 d for A. lordhowensis. These data greatly extend the known competency periods for larvae of broadcast-spawning corals and indicate the potential for transport of broadcast-spawned coral larvae over large distances. Medium to long-distance larval dispersal of the species studied provides a mechanism for their widespread distribution in subtropical regions, on reefs which are often widely spaced and relatively isolated. Received: 27 May 1997 / Accepted: 27 November 1997     

Reproduction and recruitment of scleractinian corals in a high-latitude coral community, Amakusa, southwestern Japan

Reproduction and recruitment in high-latitude coral populations in Japan have been little studied. A comprehensive study of the reproduction and early life history was conducted on nine common scleractinian coral species in Amakusa, southwestern Japan (32°N) from 2001 to 2003 including; (1) fecundity (the proportion of colonies with mature eggs), (2) timing and synchrony of spawning, (3) initial larval settlement pattern, (4) recruitment, (5) post-settlement mortality. The fecundity was high (76.7–100%) in six of seven species examined in 2002 and 2003. Annual spawning of the seven species occurred from mid July to August in 2001–2003, when seawater temperature was at the annual maximum. Spawning was highly synchronised among conspecific colonies and species in 2002 and 2003, with five species spawning five to nine nights after the full moon and another two spawning around the new moon. Temporal patterns of larval settlement of three spawning species during the first 10 days after spawning were similar to those of other spawning species from low latitudes. The number of scleractinian recruits on settlement plates, deployed from July to October (the major recruitment period at the study site), was low (2 recruits/m²) for the three consecutive years. Post-settlement mortality of 1–1.5 month old spat of five species ranged between 88 and 100% over 3–10 months in the field, similar to the values reported for both high and low latitude species (>94–99%). Among the key stages examined, the low recruitment rate may be the most important step in limiting successful reproduction and recruitment of these high-latitude scleractinian populations. The low recruitment rate may be attributable to (1) the reduced influx of larval supply from other coral populations, which are smaller and more isolated at high-latitudes and (2) the longer precompetent larval phase of broadcast-spawning corals which results in an increased chance of larvae being dispersed away from parent populations.     

Spatially stochastic settlement and the coexistence of benthic marine animals

For sessile organisms, dispersal and recruitment are typically spatially stochastic, but there is little understanding of how this variability scales up to influence processes such as competitive coexistence. Here we argue that coexistence of benthic marine animals is enhanced by stochastic differences between species in the spatial distribution of larval settlement. Differentiation of settlement distributions among competitors results in intraspecifically aggregated settlement, which can reduce overall interspecific competition and increase overall intraspecific competition. We test for the components of this mechanism using a pair of subtidal invertebrates, and we find that the mean interspecific effect of the dominant competitor is substantially reduced by natural settlement variability. Using a simulation parameterized with experimental data, we find that variable settlement could play an important role in long-term coexistence between these species. This mechanism may apply broadly to benthic marine communities, which can be highly diverse and typically exhibit large settlement fluctuation over a range of scales.     

A multi-event model to study stage-dependent dispersal in radio-collared hares: when hunting promotes costly transience

Behavioral ecologists have often assumed that dispersal is costly mainly because of unfamiliarity with traversed habitats during dispersal and energy costs of the movement per se; thus, dispersers that have successfully settled should experience survival rates comparable to those of philopatric individuals. In this paper, we tested that hypothesis using 152 radio-collared European hares in a harvested population. We developed a multi-event capture recapture model, combining telemetry data and recoveries and separately modeling the foray probability, the settlement probability, and the permanent dispersal probability. The parameterization introduced here raises the possibility of separately testing effects on survival and dispersal probabilities at each stage of dispersal (departure, transience, and settlement). In accordance with our expectations, we reveal that dispersers incur higher mortality risks during transience and the early settlement period than philopatric individuals or settled dispersers. We also found that dispersers suffer from higher risks of being shot. Those results illustrate that unfamiliarity with the habitat during transience makes dispersal costly and that settled dispersers may enjoy survival rates comparable to those of philopatric individuals. Surprisingly, we also found that individuals have a higher probability of foraying during the hunting season. We suggest that hunting and related disturbances increase dispersal costs both by increasing mortality risk during transience and (perhaps) by increasing movement rates. We emphasize the need to take human pressures into account as factors that may drive the demographics of movements in populations.     

Larval survivorship, competency periods and settlement of two brooding corals, Heliopora coerulea and Pocillopora damicornis

Larval dispersal and recruitment are important in determining adult coral distribution; however, few studies have been made of coral larval dispersal. This study examined the larval behavior, survivorship competency periods and settlement of two brooding corals, Heliopora coerulea and Pocillopora damicornis, in relation to different potential larval dispersal patterns. We also examined the lipid content of H. coerulea as a means of flotation and a source of energy. Planulae of H. coerulea were on average 3.7 mm in length, lacked zooxanthellae, and were mostly benthic, probably because of restricted movement and low lipid content (54% by dry weight). Planulae of P. damicornis were on average 1.0 mm in length, had zooxanthellae and swam actively. The competency period of H. coerulea was shorter (30 days) than that of P. damicornis (100 days). Forty percent of H. coerulea planulae crawled onto the substrata within 1 h of release, and 47% settled within 6 h. By contrast, fewer than 10% of P. damicornis planulae crawled onto the substrata within the first hour and 25% settled within 6 h of release. The planulae of H. coerulea may have a narrower dispersal range than those of P. damicornis, settling and recruiting near parent colonies. Thus, brooding corals exhibit variations in larval dispersal patterns, which are characterized by their position in the water column and competency periods.     

Seasonality,lunar periodicity of settlement and microhabitat association of juvenile humpback red snapper <Emphasis Type="Italic">Lutjanus gibbus</Emphasis> (Lutjanidae) in an Okinawan coral reef

Snappers (Lutjanidae) are one of the important fisheries resources in tropical and subtropical waters. However, there have been few studies clarifying the ecology of newly settled juvenile snappers. The aims of the present study were to clarify the seasonality and lunar periodicity of the larval settlement, and the microhabitat association of newly settled juveniles for Lutjanus gibbus in an Okinawan coral reef. Fifteen coral patches were chosen and underwater visual surveys were conducted during a 2-year period. The larval settlement of the species mainly occurred between May and October during the 2-year survey period. In terms of lunar periodicity of the settlement, the newly settled juveniles were mainly found between the last quarter moon and new moon during the main settlement season. Back calculations of settlement dates based on otolith microstructure showed that most juveniles settled during the last quarter moon and new moon. Stepwise multiple regression analyses revealed that the newly settled juveniles were mainly found on coral patches with a large number of holes on the top of the coral patch and this tendency was mainly consistent during the study period. The volume of holes at the base of coral patches had also a positive effect on the abundance of newly settled juveniles in some cases. The results of the present study suggest that larval settlement has a strong seasonal and lunar periodicity, and newly settled juveniles select coral patches with available refuge spaces.     

Predicting regional spread of non-native species using oceanographic models: validation and identification of gaps

Predicting spread is a central goal of invasion ecology. Within marine systems, researchers have increasingly made use of oceanographic circulation models to estimate currents and track species dispersal. However, the accuracy of these models for predicting biological patterns, particularly for non-native species, has generally not been validated. Particularly, we wished to examine the ability of models to predict physical and biological processes, which jointly determine the spread of marine larval organisms. We conducted two empirical studies—a recruitment study and a drift card study—along the coast of New England, USA, focusing on two invaders of concern—the European green crab (Carcinus maenas) and the Asian shore crab (Hemigrapsus sanguineus), to explicitly evaluate the ability of oceanographic models to predict patterns of spread. We used data from the large-scale drift card study to validate our ability to capture dispersal patterns driven purely by physical processes. Next, we conducted a recruitment study to evaluate our ability to reproduce patterns of biological dispersal. We were generally capable of reproducing drift cards patterns—suggesting that the physical mechanics in the model were predictive. However, predicted biological patterns were inconsistent—we were able to predict dispersal patterns for H. sanguineus but not for C. maenas. Our results highlight the importance of validating models and suggest that more work is necessary before we can reliably use oceanographic models to predict biological spread of intertidal organisms.     

How do dispersal costs and habitat selection influence realized population connectivity?

Despite the importance of dispersal for population connectivity, dispersal is often costly to the individual. A major impediment to understanding connectivity has been a lack of data combining the movement of individuals and their survival to reproduction in the new habitat (realized connectivity). Although mortality often occurs during dispersal (an immediate cost), in many organisms costs are paid after dispersal (deferred costs). It is unclear how such deferred costs influence the mismatch between dispersal and realized connectivity. Through a series of experiments in the field and laboratory, we estimated both direct and indirect deferred costs in a marine bryozoan (Bugula neritina). We then used the empirical data to parameterize a theoretical model in order to formalize predictions about how dispersal costs influence realized connectivity. Individuals were more likely to colonize poor-quality habitat after prolonged dispersal durations. Individuals that colonized poor-quality habitat performed poorly after colonization because of some property of the habitat (an indirect deferred cost) rather than from prolonged dispersal per se (a direct deferred cost). Our theoretical model predicted that indirect deferred costs could result in nonlinear mismatches between spatial patterns of potential and realized connectivity. The deferred costs of dispersal are likely to be crucial for determining how well patterns of dispersal reflect realized connectivity. Ignoring these deferred costs could lead to inaccurate predictions of spatial population dynamics.