Effects of grazing by two species of sea urchins (Strongylocentrotus franciscanus and Lytechinus anamesus) on recruitment and survival of two species of kelp (Macrocystis pyrifera and Pterygophora californica)

We studied the effects of grazing by two species of sea urchins on two species of kelp (Macrocystis pyrifera and Pterygophora californica) in the San Onofre kelp bed in southern California from 1978 through 1981. Both red sea urchins, Strongylocentrotus franciscanus, and white sea urchins, Lytechinus anamesus, were abundant and lived in aggregations. The purple sea urchin (S. purpuratus) was rare at the study site and was not studied. The aggregations of red urchins were either relatively small and stationary (for over 3 yr) or relatively large and motile (advancing at about 2 m mo^–1). Both stationary and moving aggregations were observed at the same time, and within 100 m of one another. Stationary aggregations of red urchins probably subsisted mainly on drift kelp and had no effect on kelp recruitment or on adult kelp abundance. In contrast, red sea urchins in large, motile aggregations or fronts ate almost all the macroalgae in their path. The condition of their gonalds indicated that red urchins in fronts were starved relative to red urchins in the small, stationary aggregations. Large, motile aggregations developed after 2 yr of declining kelp abundance (probably due largely to storms). We propose that a scarcity of drift algae for food results in a change in the behavior pattern of the red urchins and thus leads to the formation of large, motile aggregations. The aggregations of white urchins, which occurred along the offshore margin of the kelp bed, were large, but relatively stationary. The white urchins rarely ate adult kelps, but grazed extensively on early developmental stages of kelps and evidently prevented seaward expansion of the bed. The spatial distribution of both types of red urchin aggregations appeared to be unrelated to predation pressure from fishes or lobsters.Please address all requests for reprints to the senior author at his present address.     

Feeding aggregations of sea stars (<Emphasis Type="Italic">Asterias</Emphasis> spp. and <Emphasis Type="Italic">Henricia sanguinolenta</Emphasis>) associated with sea urchin (<Emphasis Type="Italic">Strongylocentrotus droebachiensis</Emphasis>) grazing fronts in Nova Scotia

Migrating feeding aggregations (or fronts) of sea urchins can dramatically alter subtidal seascapes by destructively grazing macrophytes. While direct effects of urchin fronts on macrophytes (particularly kelps) are well documented, indirect effects on associated fauna are largely unknown. Secondary aggregations of predators and scavengers form around fronts of Strongylocentrotus droebachiensis in Nova Scotia. We recorded mean densities of the sea stars Asterias spp. (mainly A. rubens) and Henricia sanguinolenta of up to 11.6 and 1.7 individuals 0.25 m⁻² along an urchin front over 1 year. For Asterias, mean density at the front was 7 and 15 times greater than in the kelp bed and adjacent barrens, respectively. There was strong concordance between locations of peak density of urchins and sea stars (Asterias r = 0.98; H. sanguinolenta r = 0.97) along transects across the kelp–barrens interface, indicating that sea star aggregations migrated along with the urchin front at rates of up to 2.5 m per month. Size–frequency distributions suggest that Asterias at the front were drawn from both the barrens (smaller individuals) and the kelp bed (larger individuals). These sea stars fed intensively on mussels on kelp holdfasts and in adjacent patches. Urchin grazing may precipitate aggregations of sea stars and other predators or scavengers by incidentally consuming or damaging mussels and other small invertebrates, and thereby releasing a strong odor cue. Consumption of protective holdfasts and turf algae by urchins could facilitate feeding by these consumers, which may obtain a substantial energy subsidy during destructive grazing events.     

The role of behavioral responses to predators in modifying urchins' (Strongylocentrotus droebachiensis) destructive grazing and seasonal foraging patterns

We documented spatial and temporal patterns of urchins (Strongylocentrotus droebachiensis) and periwinkles (Littorina littorea) in three habitats: a persistent Laminaria longicruris and L. digitata bed; an urchin dominated barrens, and the edge of the kelp bed that formed a boundary between the two. Urchins were rare in the kelp and, when present, always large and well hidden, a pattern we interpret as a response to crab and lobster predation. Urchins were abundant in the barrens, and, in the summer when predaceous fish were active during the day, foraged only at nigh. We observed the formation of a dense urchin feeding front along the kelp bed edge, and these urchins remained exposed and feeding even during the summer. Laboratory experiments demonstrated that aggregations are an effective defense against some predators, and that the presence of crabs increases the tendency of large urchins to aggregate. We hypothesize that healthy Laminaria spp. beds persist because kelp bed associated predators keep urchins at low densities and in hiding. A reduction in predation pressure permits urchin densities to increase to the point where they form aggregations, which provide better defense than hiding. These aggregations then graze destructively on Laminaria spp., forming barrens. These barrens seem to be a new, stable configuration of the system.     

Severe storm disturbances and reversal of community structure in a southern California kelp forest

Regular observations made over a period of 5 yr in four permanent transects provided data on plant, sea urchin, and fish densities which indicate that two unusually severe winter storms in 1980 (Storm I) and 1983 (Storm II) had different effects on a southern California kelp-forest community. Storm I removed all canopies of the giant kelp Macrocystis pyrifera, but spared most understory kelps, mainly Pterygophora californica. Hence, the previously large accumulation of detached drift kelp, mostly M. pyrifera, disappeared. Denied their preferred diet of drift kelp, the sea urchins Strongylocentrotus franciscanus and S. purpuratus then emerged from shelters to find alternative food. Without effective predators, they consumed most living plants, including the surviving understory kelps. This weakened the important detritus-based food chain, as indicated indirectly by declining abundances of algal turf and fish (Embiotocidae) that eat small animals living in turf. In 1983, Storm II reversed the process by eliminating exposed urchins, while clearing rock surfaces for widespread kelp settlement and growth. By summer 1984, the kelp grew to maturity to form extensive canopies despite elevated water temperatures during summer and fall of 1983. Thus, severe storms may have vastly different effects on community structure, depending on the state of the community before the disturbance.     

Experimental evaluation of aggregation behavior in the sea urchin Strongylocentrotus droebachiensis

Defensive aggregation by Strongylocentrotus droebachiensis has been invoked as the fulcrum for the transformation of macroalgal beds into coralline barrens in the northwest Atlantic. We critically examined some of the mechanisms contributing to aggregation behavior by experimentally manipulating sea urchins, purported predators and food during autumn 1983 and spring 1984 both in the laboratory and field. We utilized several approaches to examine a range of sea urchin responses to the presence of food, tethered predators, caged predators, crushed con-specifics in the field and predators in laboratory tanks. Some of the field cages had the property of allowing free passage to sea urchins while retaining lobsters; this allowed distinctions to be made between artifacts caused by cage walls or topographic barriers and unrestricted behavioral responses of sea urchins. The results falsified the hypothesis that aggregations of S. droebachiensis are elicited by predators. Except in the presence of algae, sea urchins always avoided decapod predators (but not sea stars) and fled from the vicinity of injured conspecifics. However, avoidance behavior was subordinate to feeding responses, demonstrated by mass migrations of sea urchins into cages (with lobsters) to feed on algae. We reaffirmed by observation and manipulation previous studies which showed that sea urchins aggregate only in the presence of food. Two types of sea urchin groupings were delineated: (1) surficial two dimensional associations, often caused by topographic or other features which inhibited dispersal and (2) cohesive three-dimensional aggregations induced by food.     

Temporal variation in community interfaces: kelp-bed boundary dynamics adjacent to persistent urchin barrens

We examined, over 2 years, factors affecting the temporal stability of the lower limit of kelp beds (Alaria esculenta) at five subtidal sites in the Mingan Islands, northern Gulf of St. Lawrence. The position of the lower limit of the beds varied markedly among sites and over time and was largely controlled by the green sea urchin, Strongylocentrotus droebachiensis, which formed dense (up to 500 individuals m^–2) feeding fronts at the lower edge of the beds. These aggregations advanced over the kelp most rapidly during the summer (at rates as high as 2.5 m month^–1), and there appeared to be a threshold urchin biomass of ~5 kg m^–2 below which the fronts could not substantially reduce the limit of the beds. The fronts consisted mainly of large individuals, whereas smaller urchins predominated in the barrens zone below the kelp beds. At one site, we recorded large seasonal shifts in overall urchin densities, with large increases and decreases during the summer and winter, respectively. An urchin exclusion experiment indicated that algal recruitment in the barrens was two orders of magnitude greater in the absence than in the presence of urchins. The kelp Agarum cribrosum greatly restricted urchin movements, and the greater temporal stability of the kelp bed at one site appeared related to the gradual replacement of Alaria esculenta in the lower kelp bed by a large stand of Agarum cribrosum. We propose that perturbations by abiotic factors (e.g., ice scouring and water motion) trigger important but localized changes in urchin densities that, in turn, largely determine the limits of kelp bed distribution in this region of the Atlantic where urchin barrens are a persistent community state.Communicated by R.J. Thompson, St. Johns     

Changes in sea urchin populations after the destruction of kelp beds

In St. Margaret's Bay, Nova Scotia, Canada, there are large areas in which sea urchins (Strongylocentrotus droebachiensis) have eliminated beds of kelp (Laminaria spp.). Sites were identified where destruction of kelp beds had taken place 1, 2, 3, 3.5 and 4 years ago. With increase of time since kelp disappearance, the sea urchins showed decreased growth rate, reduced gonad size, but an increase in numbers resulting from high recruitment rates in the first two years after kelp bed destruction. These sea urchin populations, by their browsing, effectively prevent the regeneration of kelp. There is, as yet, no evidence of the sea urchin populations being starved out to allow the kelp to return.     

Interactions between sea urchins (Strongylocentrotus droebachiensis) and their predators in field and laboratory experiments

Field observations and manipulative experiments in a nearshore cobble bed (2 to 3 m below mean low water) at Eagle Head, Nova Scotia, Canada, between 1984 and 1986, showed that small juveniles ofStrongylocentrotus droebachiensis (3 to 6 mm diam) sheltering beneath cobbles had a refuge from predators such as rock crabs, small lobsters, and fish. Sea urchins gradually outgrew these refuges and small adults (25 to 30 mm) required larger rocks as shelter from predators, particularly large cancrid crabs. Small juveniles were usually solitary and well dispersed beneath cobbles, whereas small adults tended to aggregate on the undersides and in the interstices of boulders. These aggregations may develop passively as sea urchins accumulate in suitablysized refuges. Chemotaxis experiments indicate that juvenileS. droebachiensis are repelled by waterborne stimuli from conspecifics. In a factorial experiment, effects of the presence of potential predators (rock crabs and lobsters) and/or food (kelp) on the behaviour of large juvenile (10 to 15 mm) and small adult sea urchins were examined in flowing seawater tanks. Both size classes formed exposed feeding aggregations when kelp was provided as food, irrespective of the presence or absence of predators. In the absence of kelp, each size class responded differently to the presence of a predator: juveniles became more cryptic, whereas adults aggregated on the tank sides. Increased movement to the sides of a tank in the presence of a predator may reflect a flight response, since chemotaxis experiments indicated thatS. droebachiensis is repelled by waterborne chemical stimuli from predators. Observational and experimental data suggest that predation at the late juvenile and early adult stages may influence population structure, distribution and abundance ofS. droebachiensis.     

Diel foraging patterns of the sea urchin Centrostephanus coronatus as a predator avoidance strategy

During the day, the diadematid sea urchin Centrostephanus coronatus occupies holes and crevices in shallow subtidal rocky substrata. Individuals emerge from these after sunset and forage on organisms attached to the surrounding rock surface. Each urchin travels <1 m from its shelter and returns to the same one before sunrise. The sheephead wrasse Pimelometopon pulchrum does not remove urchins from their shelters, but will attack and consume urchins placed in normal feeding locations during the daytime. The active periods of the sheephead and the urchin do not overlap; urchins begin foraging about 20 min after the diurnal sheephead retire in the evening and return to their shelters 1 to 2 h before sheephead resume feeding in the morning. We infer that the urchin's daytime crevice-dwelling and nocturnal foraging habits have evolved as a response to sheephead predation. Moreover, because shelters are limited in supply, shelter fidelity may have evolved to insure refuge from sheephead.     

The decline and recovery of four predatory fishes from the Southern California Bight

What to do about fisheries collapse and the decline of large fishes in marine ecosystems is a critical debate on a global scale. To address one aspect of this debate, a major fisheries management action, the removal of gill nets in 1994 from the nearshore arena in the Southern California Bight (34°26′30″N, 120°27′09″W to 33°32′03″N, 117°07′28″W) was analyzed. First, the impetus for the gill net ban was the crash of the commercial fishery for white seabass (Atractoscion nobilis; Sciaenidae) in the early 1980s. From 1982 to 1997 catch remained at a historically low level (47.8 ± 3.0 mt) when compared to landings from 1936–1981, but increased significantly from 1995–2004 (r = 0.89, P < 0.01) to within the 95% confidence limit of the historic California landings. After the white seabass fishery crashed in the early 1980s, landings of soupfin (Galeorhinus galeus; Triakidae) and leopard shark (Triakis semifasciata; Triakidae) also significantly declined (r = 0.95, P < 0.01 and r = 0.91, P < 0.01, respectively) until the gill net closure. After the closure both soupfin and leopard shark significantly increased in CPUE (r = 0.72, P = 0.02 and r = 0.87, P < 0.01, respectively). Finally, giant sea bass (Stereolepis gigas; Polyprionidae) the apex predatory fish in this ecosystem, which was protected from commercial and recreational fishing in 1981, were not observed in a quarterly scientific SCUBA monitoring program from 1974 to 2001 but reappeared in 2002–2004. In addition, CPUE of giant seabass increased significantly from 1995 to 2004 (r = 0.82, P < 0.01) in the gill net monitoring program. The trends in abundance of these fishes return were not correlated with sea surface temperature (SST), the Pacific Decadal Oscillation (PDO) or the El Niño/Southern Oscillation (ENSO). All four species increased significantly in either commercial catch, CPUE, or in the SCUBA monitoring program after the 1994 gill net closure, whereas they had declined significantly, crashed, or were absent prior to this action. This suggests that removing gill nets from coastal ecosystems has a positive impact on large marine fishes.     

Movement and feeding activity of red sea urchins (Strongylocentrotus franciscanus) adjacent to a kelp forest

Movement and feeding were studied in a population of red sea urchins, Strongylocentrotus franciscanus (Agassiz, 1863), found within and immediately seaward of a kelp forest offshore from Santa Cruz, California, USA. Mean sea urchin movements varied from 7.5 cm/day inside the kelp forest to over 50 cm/day at 15 and 100 m outside the kelp forest. The percentage of sea urchins feeding decreased from 66% inside the kelp forest to 16 and 15% at 15 and 100 m outside the kelp forest. These data indicate that movement by these sea urchins is a response to a low food supply.     

Seasonal variation in movement,aggregation and destructive grazing of the green sea urchin (<Emphasis Type="Italic">Strongylocentrotus droebachiensis</Emphasis>) in relation to wave action and sea temperature

Hydrodynamic forces are an important determinant of subtidal community structure, particularly when they limit the distribution and foraging ability of mobile consumers. We examined the effect of wave action on the rate of movement and destructive grazing of a kelp bed by the green sea urchin (Strongylocentrotus droebachiensis) under field conditions. We measured density and rate of advance at fixed intervals along ∼100 m of a grazing front over 1 year, and quantified individual movement rates in the barrens 5–10 m behind the urchin front using a time-lapse videography. Seasonal variation in the mean rate of advance of the front (range: 0–4 m month⁻¹) was explained by changes in urchin density at the front (120–360 individuals m⁻²), which in turn varied inversely with significant wave height (0.5–2 m). Water temperature (0.8–17.6°C) had no effect on the rate of advance or on urchin density (aggregation) at the front, except when temperature exceeded 17°C. Movement of individual urchins also was affected by wave action: we observed a significant decrease in speed and displacement of urchins with increasing significant wave height. Wave action had no effect on the proportion of urchins moving or the degree of linearity of their movements. We propose that the decrease in urchin density at the front associated with increased wave action, results from de-aggregation, which reduces the risk of dislodgement, combined with a reduction in urchin movement in barrens, which supplies new urchins to the front.     

Patterns of gamete incompatibility between the blue mussels<Emphasis Type="Italic"> Mytilus edulis</Emphasis> and<Emphasis Type="Italic"> M</Emphasis>.<Emphasis Type="Italic"> trossulus</Emphasis>

Previous research on gametic incompatibility in marine invertebrates suggests that for highly dispersive marine invertebrate species, barriers to fertilization among closely related taxa are often incomplete and sometimes asymmetric. The nature of these barriers can dramatically affect the patterns of gene flow and genetic differentiation between species, and thus speciation. Blue mussels, in the genus Mytilus, are genetically distinct in allopatry yet hybrids are present wherever any two species within the group co-occur. The present study sampled M. edulis (L.) and M. trossulus (Gould) in May and June 2001 from the East Bay section of Cobscook Bay, Maine, USA (latitude 44°56′30″N; longitude 67°07′50″W), where the two species are sympatric. Gamete incompatibility was investigated in a series of laboratory fertilizations carried out in July 2001. The proportion of fertilized eggs typically exceeded 80% at sperm concentrations of 10³–10⁴ ml^?1 among intraspecific matings (n=18), but was <30% even at sperm concentrations in excess of 10⁵–10⁶ ml^?1 for interspecific matings (n=13). Further analysis indicated that approximately 100- to 700-fold higher sperm concentrations were required to achieve 20% fertilization in interspecific matings relative to intraspecific matings, indicating strong barriers to interspecific fertilization. The proportion of fertilized eggs did not follow this general pattern in all matings, however. The eggs from two (out of five) M. edulis females were almost as receptive to M. trossulus sperm as they were to M. edulis sperm. In contrast, the eggs from all M. trossulus females (n=3) were unreceptive to M. edulis sperm, suggesting that fertilization barriers between these species may be asymmetric. Given the experimental design employed in this study, the results are also consistent with a strong maternal or egg effect on the level of interspecific gamete compatibility in M. edulis.     

Predators of juvenile sea urchins and the effect of habitat refuges

We evaluated the effects of potential predators from intertidal habitats on Strongylocentrotus purpuratus survival using laboratory experiments and assessed abundances of main predatory species along the Pacific coast of North America. The interactive effects of urchins’ and predators’ sizes in mediating predation were quantified. Habitat complexity (substrate pits, adult spine canopy) was manipulated to examine its effects on predation of most susceptible individuals (<14 mm). Pachygrapsus crassipes was identified as a major predator of urchins up to ≈30 mm. A positive effect of predator size on consumption of progressively larger urchins was detected, probably due to a mechanical limitation on crabs’ ability to consume large prey. Larger claws of males with respect to females of comparable sizes facilitated the handling of larger prey. Substrate refuges significantly reduced mortality on juvenile urchins. These results show that crab predation may be important in organizing intertidal communities, despite multiple ecological mechanisms promoting sea urchin survival.     

Variation in fertilisation abilities between hemiclonal hybrid and sexual parental males of sympatric water frogs (<Emphasis Type="Italic">Rana lessonae</Emphasis>,<Emphasis Type="Italic"> R. esculenta</Emphasis>,<Emphasis Type="Italic"> R. ridibunda</Emphasis>)

In many species, males and females mate with multiple partners, which gives rise to sperm competition and multiple paternity. The experiments on water frogs presented here demonstrate that such sperm competition can affect the structure and dynamics of mixed-species communities. The hybrid frog Rana esculenta (LR) mates with one of its parental species, usually R. lessonae (LL), although in some areas R. ridibunda (RR), to regain the premeiotically eliminated parental genome ("hybridogenesis"). Mixed LL/LR-populations are stable although hybrid numbers should continuously increase at the expense of parental animals, because of differences in female fecundity and other factors. This would finally lead to the extinction of the sexual host, followed by that of the sexual parasite, unless the reproductive superiority of R. esculenta is reduced by other factors, such as lower hybrid male fertility. Eggs from LL- and LR-females were fertilised in vitro by single- and multi-male sperm suspensions of LL-, LR- and RR-males. In all experiments, the proportion of offspring sired by R. esculenta sperm was significantly lower than that sired by R. lessonae or R. ridibunda sperm. Gonad mass, sperm morphology, sperm swimming velocity, and sperm survival did not explain these differences in fertilisation success; nor did gamete recognition and compatibility. Sperm density was the only trait that paralleled fertilisation success, but it offers no explanation either, because densities were equalised for the in-vitro fertilisations. In natural LL/LR populations, the significantly smaller amount, poorer competitive ability and lower long-term survival of R. esculenta compared to R. lessonae sperm will reduce the initial reproductive superiority of hybrids and contribute to the stabilisation of mixed water-frog populations. Differences in fertilisation ability are also likely to be relevant for the structure and dynamics of several other systems with encounters between eggs and sperm from different genotypes, ecotypes, ploidy levels and/or species.     

Semi-parametric Bayesian density estimation using ranked set sample in the presence of ranking error

In this paper, we propose a Bayesian method to estimate the underlying density function of a study variable Y using a ranked set sample in which an auxiliary variable X is used to rank the sampling units. The amount of association between X and Y is not known, resulting in an unknown degree of ranking error. We assume that (X, Y) follows a Morgenstern family of distributions. The study variable Y is assumed to have a parametric distribution, with the distribution of the parameters having a Dirichlet process prior. A Markov chain Monte Carlo procedure is developed to obtain a Bayesian estimator of the desired density function as well as of the ranking error. A simulation study is used to evaluate the performance of the proposed method. An example from forestry is used to illustrate a real-life application of the proposed methodology.     

Ecology of a corallivorous gastropod,<Emphasis Type="Italic"> Coralliophila abbreviata</Emphasis>, on two scleractinian hosts. I: Population structure of snails and corals

Despite their potential importance in structuring reef communities, invertebrate corallivores and their population structures are poorly understood. We found distinct differences in the population structures (length-frequency distribution and sex ratio) of the corallivorous gastropod Coralliophila abbreviata residing on two coral-host taxa, Montastraea spp. and Acropora palmata, in the Florida Keys. In each of two survey years, around 50% of the Montastraea spp. colonies were infested, with a mean snail density of eight snails per infested colony (range 1–45), while around 20% of A. palmata colonies harbored three snails per infested colony (range 1–23). Variation in patterns of snail occurrence was also observed within a host taxon. A. palmata occurred in low- and high-density stands (0.4 and 1 colony m^–2, respectively, at the initial survey) at different sites. Hurricane Georges struck the area in September 1998. When resurveyed in 1999, density of colonies in low-density stands had decreased by 75% to 0.1 colonies m^–2. This decrease was accompanied by a doubling in the proportion of colonies infested with snails (from 19% to 46%) and an increase in snail density per infested colony (from 3.7±3.3 SD to 5.4±4.6 SD) as snails apparently concentrated on surviving A. palmata. In contrast, sites with high density A. palmata stands (thickets) retained colony densities of about ~1 colony m^–2 among years, while snail infestation increased only from 9% to 14% of colonies surveyed and snail density essentially remained unchanged (from 2.7±1.8 to 2.9±1.9 snails per infested colony). Snails collected from Montastraea spp. were shorter than those from A. palmata in low-density stands and were longest on A. palmata in thickets. On both host taxa, female snails were longer than males. The sex ratio of snails on Montastraea spp. hosts was even (1:1), while that of snails on A. palmata was skewed (70% males). Factors that could explain observed differences in size structure and sex ratio between Coralliophila populations on the two coral host taxa include: differential susceptibility to predators, influence of host tissue nutritional quality and/or secondary metabolite content, and genetic differences (cryptic species). The host-specific characteristics of C. abbreviata populations imply that the impact of gastropods on reef communities will vary with the coral species composition.     

The influences of industrial gross domestic product,urbanization rate,environmental investment,and coal consumption on industrial air pollutant emission in China

With the rapid economic development and urbanization, there would be severe air pollution. The path analysis method was employed to investigate the specific effects of Industrial Gross Domestic Product (IGDP), Urbanization Rate (UR), Environmental Investment (EI), and Coal Consumption (CC) on Industrial SO₂ Emissions (ISE) and Industrial Dust Emissions (IDE). The results shown that (1) IGDP is the main reason for the increase in ISE and IDE, with the direct positive influence more than 20% and a total contribution of more than 35%, and the combined effects of UR can reduce air pollutants emission considerably, especially in ISE. (2) The effects of EI and CC on ISE could be ignored, given that their contributions comprised <?10%. Meanwhile, the contribution of EI to IDE was only 8.92%. For one reason, EI is underfunded which the percentage of Gross Domestic Product was <?2%.     

Marine reserves reduce risk of climate-driven phase shift by reinstating size- and habitat-specific trophic interactions

Spatial closures in the marine environment are widely accepted as effective conservation and fisheries management tools. Given increasing human-derived stressors acting on marine ecosystems, the need for such effective action is urgently clear. Here we explore mechanisms underlying the utility of marine reserves to reinstate trophic dynamics and to increase resilience of kelp beds against climate-driven phase shift to sea urchin barrens on the rapidly warming Tasmanian east coast. Tethering and tagging experiments were used to examine size- and shelter-specific survival of the range-extending sea urchin Centrostephanus rodgersii (Diadematidae) translocated to reefs inside and outside no-take Tasmanian marine reserves. Results show that survival rates of C. rodgersii exposed on flat reef substratum by tethering were approximately seven times (small urchins 10.1 times; large urchins 6.1 times) lower on protected reef within marine reserve boundaries (high abundance of large predatory-capable lobsters) compared to fished reef (large predatory lobsters absent). When able to seek crevice shelter, tag-resighting models estimated that mortality rates of C. rodgersii were lower overall but remained 3.3 times (small urchins 2.1 times; large urchins 6.4 times) higher in the presence of large lobsters inside marine reserves, with higher survival of small urchins owing to greater access to crevices relative to large urchins. Indeed, shelter was 6.3 times and 3.1 times more important to survival of small and large urchins, respectively, on reserved relative to fished reef. Experimental results corroborate with surveys throughout the range extension region, showing greater occurrence of overgrazing on high-relief rocky habitats where shelter for C. rodgersii is readily available. This shows that ecosystem impacts mediated by range extension of such habitat-modifying organisms will be heterogeneous in space, and that marine systems with a more natural complement of large and thus functional predators, as achievable within no-take reserves, will minimize local risk of phase shifts by reinstating size and habitat-specific predator-prey dynamics eroded by fishing. Importantly, our findings also highlight the crucial need to account for the influence of size dynamics and habitat complexity on rates of key predator-prey interactions when managing expectations of ecosystem-level responses within marine reserve boundaries.