Dispersal and recruitment are fundamental processes for population recovery following disturbances in sessile species. While both processes are well understood for many terrestrial species, they still remain poorly resolved for some macroalgal species. Here we experimentally investigated the effective dispersal and recruit survival of a mesophotic Mediterranean fucoid, Cystoseira zosteroides. In three isolated populations, four sets of settlement collectors were placed at increasing distances (from 0 to 10 m) and different orientations (North, South, East and West). We observed that effective dispersal was restricted to populations’ vicinity, with an average of 6.43 m and not further than 13.33 m, following a Weibull distribution. During their first year of life, survival was up to 50%, but it was lower underneath the adult canopy, suggesting a negative density-dependence. To put our results in a broader context we compared the effective dispersal of other fucoid and kelp species reported in the literature, which confirmed the low dispersal ability of brown algae, in particular for fucoids, with an effective dispersal of few meters. Given the importance of recruitment for the persistence and recovery of populations after disturbances, these results underline the vulnerability of C. zosteroides and other fucoid species to escalating threats.
Citation: Capdevila P, Linares C, Aspillaga E, Riera JL, Hereu B (2018) Effective dispersal and density-dependence in mesophotic macroalgal forests: Insights from the Mediterranean species Cystoseira zosteroides. PLoS ONE 13(1): e0191346. https://doi.org/10.1371/journal.pone.0191346
Editor: Bayden D. Russell, The University of Hong Kong, HONG KONG
Received: August 1, 2017; Accepted: January 3, 2018; Published: January 12, 2018
Copyright: © 2018 Capdevila et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Relevant data are available in the paper and its Supporting Information files. The dispersal data of Cystoseria zosteroides and the comparative dispersal data are available on Figshare (DOI: 10.6084/m9.figshare.5743221).
Funding: Funding was provided by the Spanish MINECO (CGL2012-32194) and the European Union’s Horizon 2020 research and innovation program under grant agreement No 689518 (MERCES). Support for this work was provided by an FI-DRG grant from the Secretaria d’Universitats i Recerca from the Economy and Knowledge Department of the Generalitat de Catalunya to Pol Capdevila and by a Ramon y Cajal research contract (RyC-2011-08134) to Cristina Linares. The authors are part of the Marine Biodiversity Conservation research group (2014SGR1297, www.medrecover.org) from Generalitat de Catalunya. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
The increase of anthropogenic stressors (e.g. coastal development, overexploitation, pollution) has driven the loss of key habitat-forming organisms, such as terrestrial plants, corals, and algae, rendering their populations more fragmented, isolated and vulnerable to further sources of disturbance [1–4]. In this context, dispersal and recruitment play a crucial role in maintaining the resilience and ensuring the long-term population stability of habitat-forming species [1–3]. At the local scale, dispersal and successful recruitment determine population dynamics and structure, also driving their recovery after disturbances; while at larger scales these processes ensure population connectivity, gene flow and the colonization of new locations [4–6]. Thus, understanding the dispersal of these species may be crucial to predicting their ability to respond to local and global stressors, as well as to assess the scale at which management strategies may be effective [7–9].
Nonetheless, the wealth of dispersal ecology studies in terrestrial plants [10,11] contrasts with the relative poor mechanistic understanding about the dispersal and settlement of other habitat-forming species like canopy-forming macroalgae [12–15]. Large brown algae of the orders Laminariales (kelps) and Fucales (fucoids) play a key ecological role as habitat-forming species in temperate marine ecosystems worldwide [16–18]. Their dispersal is likely a very passive process, as it occurs through the release of non-motile (or with limited mobility) propagules (spores or zygotes) in the water column [19,20]. Consequently, macroalgae are located at the lowest end of the range of dispersal distances reported for the marine realm [8,21,22]. Still, their propagule stages are microscopic and elusive to study, so their dispersal scales remain poorly understood for many species.
In the Mediterranean Sea, Cystoseira spp. are late successional species which conform important forest-like assemblages from the intertidal to sublittoral zone (in some cases deeper than 50 m), providing food and shelter for many associated organisms, enhancing local biodiversity [18,23]. During the last decades, a widespread decline of these assemblages has been documented in many regions [24,25]. The multiple anthropogenic stressors to which they are exposed, as well as their slow population dynamics (slow growth rates, scarce recruitment, longevity; e.g. ) and their limited population connectivity , have been argued to be the main causes of their decline [18,26,28]. Nevertheless, little is known about their dispersal abilities and population dynamics, especially for deep-water macroalgal species.
Cystoseira species are often assumed to present low dispersal distances , given that their zygotes develop closely attached to the thallus of adult stands, although quantitative data regarding this topic are still scant (but see [16,29,30]). Genetic tools have brought the opportunity to unravel large-scale connectivity patterns, but there are very few genetic studies dealing with Cystoseira species (but see [27,31]). In contrast to genetic studies, classical experiments have provided important insights in short- to medium-distance dispersal for other algae and terrestrial plant species [13,32]. Thus, when no genetic markers are available, experimental studies can provide useful information about the effective dispersal of species .
Here we have experimentally determined the effective dispersal at small-scale of Cystoseira zosteroides C. Agradh (Fucales, Ocrophyta), a canopy-forming macroalga, which thrives in mesophotic Mediterranean rocky bottoms. Given previous studies on Cystoseira species, we expected that C. zosteroides would exhibit limited dispersal ability. Density-dependent survival of recruits was also expected because of the intraspecific competition with adults . Finally, in order to put our results in a broader context, we compared our data with similar experiments about brown macroalgae dispersal reported in the literature. The final aim of this study was to better understand the dispersal and recruitment dynamics of macroalgae, discussing how important these outcomes are for conservation actions and habitat restoration plans.
Material and methods
Cystoseria zosteroides is an endemic and one of the most representative species of Mediterranean deep-water Cystoseira assemblages, inhabiting in rocky bottoms exposed to unidirectional currents at depths ranging from 15 to 50 m, with light levels ranging from 1% to 0.3% of surface irradiance [23,34,35]. Their distribution spans all the western Mediterranean Sea with the exception of the Alboran Sean ; however, recent studies suggest that their distribution is increasingly fragmented, mainly due to human stressors . It is a long-lived species (about 50 years) and is very vulnerable to disturbances [23,26,37]. C. zosteroides has a perennial thallus with reservoir vesicles (tophules) at the top. Tophules are formed annually and constitute the origin of primary branches, which emerge during the productive season (early spring) until late summer or early autumn . C. zosteroides is considered monoecious with a diplontic and iteroparous life cycle. Gametes are enveloped on parenchymatic walls which form “small bags” named conceptacles, which tend to aggregate into groups of 5 to 8 conceptacles on the basis of primary deciduous branches named receptacles. Their reproduction starts between late March and early April, and lasts until late June-early July (pers. obs.). The liberation of male gametes results in the external fertilization of female gametes, which remain attached to the external part of the receptacles. The age of sexual maturity is reached at three years (P. Capdevila unpublished data). There is no planktonic stage, zygotes are non-motile, and this species does not present clonal reproduction.
Study site and experimental design
Experiments were conducted at three well-developed C. zosteroides populations on Montgrí, Illes Medes and Baix Ter Natural Park (Catalonia, Spain; Site 1: N 42° 2’ 53.272” E 3°12’ 50.943”; Site 2: N 42° 2’ 54.38” E 3° 12’ 49.546”; Site 3: N 42° 2’ 53.873” E 3° 12’ 53.928”; Fig 1a). These populations were selected because they showed densities (2–4 individuals/0.25m2) and size structures (dominated by mature individuals) similar to other mature populations studied at the same location and with a similar extension . All the three populations were found on isolated rocks surrounded by sand at a depth range of 20–28 m, separated between 50 and 100 meters from each other. For this reason, we assumed that the exchange of individuals among these populations and during the period of study was unlikely, given the limited dispersal of Cystoseira species . The sandy bottoms surrounding these populations provided the ideal conditions to deploy this dispersal experiment, because of the isolation of the populations studied (at the short-term) and thus, facilitating the identification of the propagule source.
(a). Map of the study area. Black dots represent the sites where the experiments were installed. Map was created using R software . (b). Scheme of the experimental design, with a population of C. zosteroides surrounded by sand and collectors placed at different distances (from 0 to 10 m). (c). Picture of a recruit collector after one year of being installed. The topographic base map (1:5.000), coastal limit and bathymetry are freely accessible through the Cartographic and Geologic Institute of Catalonia (www.icgc.cat) under Creative Commons Attribution License (CC BY 4.0).
The particular conditions of our studied populations enabled us to place four sets of recruitment collectors on the surrounding sandy bottom at a distance of 1, 2, 3, 4, 5, 7.5 and 10 m from the edge of the C. zosteroides populations, following a cross in N, E, S and W directions (Fig 1b). In addition, three additional collectors were located inside each population, which were considered as distance 0 (Fig 1b). This resulted in a total of 31 collectors placed at each site. Collectors were made by irregular calcareous tiles, with the same composition as the rocks of the area to simulate the natural recruitment substrate of this species (Fig 1c). Inside the populations, recruitment collectors were fixed to the bottom by epoxy putty; while outside the populations, we used a metal bar crossing a drilled hole in the center of the tile and nailed into the sand to immobilize them. To prevent collectors from being buried by sand, they were laid on a brick keeping them about 15 cm above the bottom (Fig 1c). The area of each tile was estimated by photography, comparing the number of pixels occupied by the tile with a reference surface of the known area, using the software Adobe Photoshop version CS5 (Adobe Systems, San Jose, California).
Collectors were placed at Site 1 in February 2014 and the experiment was replicated in February 2015 at Sites 2 and 3, before the reproductive period of C. zosteroides. In preliminary studies, we observed that the reproductive period of this species spans from the beginning of April to late June and recruits were visually distinguishable in July. For this reason, we first estimated recruitment abundance at all collectors in July and after 60, 120 and 365 days to improve the survival estimates, although for the density-dependence analyses we only considered the annual survival. The number of individuals was estimated in situ by visual census of the same observers. In Site 1 they were counted during 2014–2015 and in Sites 2 and 3 were counted during 2015–2016.
Permission to work in Medes Islands was granted by the Montgrí, Illes Medes and Baix Ter Natural Park Staff. This study did not involve the removal or harvesting of C. zosteroides, which is listed in the Bern Convention and in the Aspim Protocol.
Dispersal kernel fitting
Dispersal kernels represent the statistical distributions of dispersal distances in a population , or the probability density functions describing the distribution of the post-dispersal locations from a source point. To estimate the dispersal ability of C. zosteroides, we fitted seven commonly used dispersal models [39,40] to the field data: (i) power exponential, (ii) negative exponential, (iii) Gaussian, (iv) inverse Gaussian, (v) log-normal, (vi) Weibull and (vii) 2Dt (Table 1). The functions selected here are dispersal location kernels , projected in a two-dimensional space, so they consider the increasing area of the circle when increasing distance from source [39,40]. Then, we fitted the empirical models expressed in term of counts rather than densities [41,42]. We multiplied these functions by another fitted parameter Q, which represents the number of propagules dispersed, and A, the total area sampled by the recruit collectors at distance r. Then, the recruit counts (c) were estimated as: (1)
In all cases, a is a scale parameter, b is a shape parameter and r is the radial distance from the recruit source.
The respective parameters of the dispersal kernels in Table 1 and Q were fitted to our data by a maximum-likelihood approach using the function “mle2” in the package “bbmle”  of the R software . To evaluate goodness-of-fit of the models and select the best-fitted models to our data we used the Akaike Information Criterion (AIC ): (2) where Li, the maximum likelihood for the candidate model i, is determined by adjusting the Vi free parameters in such a way as to maximize the probability that the candidate model has generated the observed data. To select the best candidate model, we also computed the differences in AIC, with respect to the best candidate model (with the minimum AIC value) : (3)
Finally, to estimate the probability that the selected model was the best one given our data and the set of candidate models, we estimated the Akaike weights wi(AIC) : (4) where K is the number of fitted models.
Dispersal studies usually assume that dispersal is isotropic (i.e. is the same in all directions ), so they usually don’t split the data according to the direction (but see ). Here, given that C. zosteroides usually inhabits deep rocky bottoms exposed to unidirectional currents [23,35,37], to account for a potential anisotropy derived from dominant currents [12,48] we fitted individual models, splitting the data by direction and site. Once we obtained the models, we selected the most frequent best-fitting function across sites and directions . This enabled us to find the most representative dispersal kernel for our studied populations at a fine scale. In addition, dispersal distances estimated from different models depend much on the selected models and on the dispersal system . For this reason, to accurately predict and compare the dispersal distances between sites and directions we selected the best-fitting function as a general description of our system. Finally, at each direction and site, we estimated the mean and the tail of the dispersal distance as the median and the 95th percentile distances (m) respectively, formulated as the integral of the best-fitted dispersal kernel .
Recruit survival was estimated as the difference in the abundance between censuses, considering the starting point when recruits were visible to the unaided eye (July), which almost coincides with the end of the reproductive season. To test for differences in recruit survival between sites and time (fixed factor), we applied generalized linear mixed models (GLMM), with a binomial error distribution and a logit link function, using the ID of each tile as a random factor. We applied a Type II Wald χ2 test over the fitted model to determine the effect of site and time on recruit survival. GLMMs were used to deal with the non-independence between observations (i.e. repeated measures) and a binomial distribution was assumed to deal with the dichotomous response variable (survive or not survive).
We tested for differences in yearly recruit survival inside and outside the C. zosteroides canopy. To compensate for the unbalanced design of the number of recruitment plates inside and outside the canopy we only used the dispersed individuals at the 0 m and 1 m distance. We fitted a generalized linear model (GLM), with a binomial error distribution and a logit link function. To test the null hypothesis of no effect of the scaling parameter in both models (sites and inside/outside canopy respectively) we used a Wald χ2 test. All analyses were performed using the package “lme4”  of the R software . The normality of residuals and overall model performance was visually inspected by observing the residual distributions and quantile-quantile plots (Figures A and B in S1 File).
In order to put our results in a broader context, we compared our data with other experimental studies targeting brown macroalgae. Dispersal estimates for macroalgal species were obtained from two published reviews [8,21] and our own literature survey. A total of 42 studies were found but only direct or experimental estimates measuring the settlement or recruitment at different distances from a propagule source were used, leading a total 17 studies. We did not consider genetic studies in order to avoid potential biases due to different methodologies and approaches. Due to the scarcity of data for several macroalgae orders, comparisons were performed only between the orders Laminariales and Fucales, which were the ones with the highest number of studies (7 and 10, respectively).
A total of 727, 791 and 494 recruits were found at Sites 1, 2 and 3 respectively at the first census. Although some recruit collectors yielded no recruits, when considering all together, recruits were found at all distances and directions. For the Site 3 the directions North and East contained very low recruit densities, so we did not fit the individual distributions to avoid misleading dispersal estimates. At all sites, 95% of recruits fell within 7.5 m from the algal stand edge. The zygote abundance at 0 m was very variable.
The shape of the fitted dispersal kernels for all C. zosteroides populations and directions suggests a leptokurtic distribution, with many propagules deposited close to the source and a rapid decline in recruit densities with increasing distances (Fig 2). Of the 7 functions used to model the dispersal with distance, the Weibull function generally provided the best fit for all four cardinal directions and the three sites (Tables A-C in S1 File). The median and 95th quantile dispersal distances calculated from the Weibull dispersal kernels were very similar among sites and directions (Table 2). The 50th quantile of the dispersion values ranged from 4.65 to 8.31 m, with a mean value of 6.43 m, while the 95th quantile values ranged from 8.61 to 16.39 m, with a mean value of 13.33 m (Table 2).
Lines indicate the predictions from Weibull model fit to the data. Panels contain the different studied sites, and colors indicate the different orientations.
All the parameters showed were significantly different to 0, see Table A in S1 File. In addition, the median and 95th dispersal distances (in meters) are presented as measures of the average dispersal distance and its tail, respectively.
First year of survival and density-dependence
Survival since the first census decreased substantially with time, reaching values lower than 50% after a year (Fig 3a). Significant differences were found in recruit survival among the three populations studied (χ2 = 17.73, df = 2, P < 0.01) and between the consecutive years (χ2 = 753.52, df = 2, P < 0.01), with a significant interaction between the aforementioned factors (χ2 = 7.82, df = 4, P = 0.02).
(a) Recruit survival (mean ± SE) vs census time. Grey scale indicates the different populations studied. (b) Yearly recruit survival inside and outside the Cystoseira zosteroides canopy. Boxes represent the interquartile range, the horizontal line represents the median, vertical line represents the upper and lower extreme values.
In agreement with our expectations, we found that recruit survival was significantly lower inside than outside the C. zosteroides canopy (χ2 = 107.35, df = 1, P < 0.01; Fig 3b), indicating negative density-dependence on post-settlement survival. There were significant differences in the survival between sites (χ2 = 8.37, df = 2, P < 0.05) and the interaction between factors was significant (χ2 = 6.12, df = 2, P < 0.05).
Cystoseira zosteroides was the fucoid species with the highest mean dispersal estimates (Table 3). The maximum reported mean dispersal distance was for Pterygophora californica, dispersing as far as 500 m (Table 3). However, another kelp species displayed the lowest dispersal distance reported, Undaria pinnatifida dispersing only 17.5 cm. On the other hand, the maximum mean dispersal reported for a fucoid species apart from C. zosteroides was Sargassum horneri (6.3 m), while C. compressa and C. amantacea showed the lowest, only 20 cm (Table 3).
Few studies have examined effective dispersal of macroalgal species at small spatial scales [21,32] and very few have fitted empirical dispersal kernels . Our results suggest that C. zosteroides effective recruitment is limited to the vicinity of their populations, corroborating the limited connectivity of Cystoseira populations [27,31]. The recruit abundance declined with distance from the source populations, with a sharp decrease beyond the first meters, following a Weibull distribution. This dispersal kernel is commonly used in terrestrial ecology [39,51], although is usually outperformed by other fat-tailed kernels such as the log-normal or the 2Dt . Still, comparing it with other fucoid species, C. zosteroides shows the highest mean dispersal reported.
Estimates of the median and the tail of the dispersal were very similar among sites and directions. Currents and waves can extend the dispersal of macroalgae propagules far from population sources , so given that C. zosteroides usually inhabits in rocky bottoms exposed to unidirectional currents [35,37], we expected higher dispersal distances in some directions, showing anisotropy in the movement of propagules. Nevertheless, the influence of major currents is difficult to detect with experimental approaches, which usually underestimate large-scale dispersal . Genetic  or experimental studies including larger scales and in situ physical measures  are more likely to be able to detect directionality on the propagules dispersal. Furthermore, dispersal patterns were quite variable among the three studied sites, supporting the notion that dispersal variability tends to be large at very small spatial scales .
Cystosiera zosteroides recruit survival was low, in line with other long-lived macroalgal [52–54] and sessile species, both marine [55,56] and terrestrial . C. zosteroides recruits displayed a high mortality (>50%), which contrasts with the high survival of adults (more than 90% ). It is worth noting that here we missed a substantial proportion of this mortality, which happens during settlement phases when individuals are not visually distinguishable . Besides, recruit survival was site-dependent, supporting the high variability at small spatial scales observed in C. zosteroides population dynamics studied elsewhere [35,38]. These results suggest that a deeper comprehension of the environmental factors driving recruit survival is still needed for this species.
Although recruitment was high inside C. zosteroides populations, recruit survival was lower than outside the adult canopy [23,33,38]. These results show that adult canopy exerts a negative influence on the recruit survival, yet the underlying mechanisms behind this negative density-dependence cannot be revealed with the data presented here. However, given the depth at which C. zosteroides dwell, light inhibition by conspecifics has been suggested to be a major restricting factor . Indeed, laboratory studies show that light limitation can diminish growth and survival rates of early macroalgal stages [58,59]. On the other hand, propagules arriving in high densities at natural gaps can experience higher mortalities than those distributed sparsely over larger areas [60,61]. This suggests that given the higher densities of recruits inside the populations, early competition could lower the survival of recruits. Although with the present data we cannot confirm this hypothesis, our results support that C. zosteroides do have a strong ability to compensate disturbance pulses through the arrival of recruits to cleared zones [26,33]. Further, the negative density-dependence does explain the recruitment limitation observed in natural C. zosteroides populations [23,33] and it is likely in many other subtidal macroalgae .
Limited dispersal and low early survival seem to be common traits among brown habitat-forming macroalgae species [16,21]. Compared to other macroalgal species, C. zosteroides displays a dispersal range similar to most fucoids, but lower than most kelps. The decline of Cystoseira species across the Mediterranean has been argued to be a consequence of the paucity of their populations and their limited dispersal abilities . Supporting these studies, we show that C. zosteroides effective dispersal is very limited, also in line with the high genetic differentiation observed among other Cystoseira species [27,31]. Sporadic dispersal events through drifting thalli or dislocated fertile algae have been suggested to enhance connectivity among distant Cystoseira populations . Yet, these seem very unlikely in our case, given the lack of air bladders for C. zosteroides and the low survival of recruits, suggesting that these rare events may not be enough to recover entire extensions of their populations at short-time scales, as seen in other macroalgae species . For our studied species, which inhabits in relatively stable habitats  and present slow population dynamics , short-dispersal could be a mechanism to maintain local populations over more competitive species. Although this local retention of propagules intensifies intraspecific competition, through negative density-dependence, it acts as a reservoir of recruits to compensate disturbance pulses [26,33,38]. Nevertheless, the low recruit survival and the low dispersal ability of C. zosteroides highlights their poor ability to colonize new habitats and respond to large disturbances affecting broad extensions [23,38]. This, coupled with the late age at maturation, the slow somatic growth and the patchiness of C. zosteroides populations could arguably explain local extinctions [25,26,63].
The patchy distribution of C. zosteroides populations across the Mediterranean and the uncertainty of their distribution in several regions, coupled with the lack of genetic markers, prevented us from studying the connectivity of this species at a larger temporal and spatial scales. Furthermore, their deep distribution challenges more exhaustive studies that could help us to clear up the underlying dispersal mechanisms. Nevertheless, our approach enabled us to have an estimate of their short-scale dispersal, as well as their survival, during their macroscopic recruit stage. The limited dispersal and recovery ability of other fucoid species [64,65] strengthens the patterns observed here. For this reason, our study constitutes the first step in understanding the dispersal mechanisms of this species, but future studies should consider incorporating broader spatial and temporal scales.
Overall, our results illustrate that the effective dispersal of the deep-water macroalga C. zosteroides, similarly to other fucoid species, is limited to few meters. Although C. zosteroides is the fucoid species with the highest propagule dispersal ability, their effective dispersal is still much lower than for other marine species [21,22]. These results indicate that their loss in fragmented areas could be hardly reversible, given that the limited dispersal and the paucity of their populations would prevent rapid recolonization and would extend the recovery time. Indeed, the recovery of C. zosteroides populations can take decades . This contrasts with many ephemeral kelp forests, which can be eliminated within a year but are able to recover as quickly as they have disappeared [66,67]. The differences in the dispersal patterns, as well as contrasting population dynamics , may explain divergences in the recovery dynamics of macroalgae species, suggesting that dispersal scales must be taken into account when designing management plans for these and other marine species . In the case of our studied C. zosteroides populations, they may be favoured by increasing their protection against human stressors (e.g. pollution, physical disturbances), but their recovery may need restoration actions, such as adult transplantation  or seeding techniques . This could be the case of other fucoid species with limited dispersal and low recovery abilities. Given the global decline of macroalgal forests [24,70–72], ensuring the connectivity of their populations should be a conservation priority in temperate seas, but we highlight here that this requires a clear comprehension of the dispersal mechanisms and the life history of targeted species.
We thank I. Montero-Serra, M. Pagès, and A. Medrano for their assistance in the field during the study years and for their comments on early versions of the manuscript. The authors are part of the Marine Biodiversity Conservation research group (www.medrecover.org) from Generalitat de Catalunya.
- 1. Lloret F, Escudero A, Iriondo JM, Martínez-Vilalta J, Valladares F. Extreme climatic events and vegetation: The role of stabilizing processes. Glob Chang Biol. 2012;18: 797–805.
- 2. Wernberg T, Thomsen MS, Tuya F, Kendrick GA, Staehr PA, Toohey BD. Decreasing resilience of kelp beds along a latitudinal temperature gradient: Potential implications for a warmer future. Ecol Lett. 2010;13: 685–694. pmid:20412279
- 3. Hughes TP, Baird AH, Dinsdale EA, Moltschaniwskyj NA, Pratchett MS, Tanner JE, et al. Patterns of recruitment and abundance of corals along the Great Barrier Reef. Nature. 1999;397: 59–63.
- 4. Nathan R, Muller-Landau HC. Spatial patterns of seed dispersal, their determinants and consequences for recruitment. Trends Ecol Evol. 2000;15: 278–285. pmid:10856948
- 5. Levin SA, Muller-Landau HC, Nathan R, Chave J. The ecology and evolution of seed dispersal: A theoretical perspective. Annu Rev Ecol Evol Syst. 2003;34: 575–604.
- 6. Coulson SJ, Bullock JM, Stevenson MJ, Pywell RF. Colonization of grassland dispersal versus by sown species : Colonization in responses to management microsite limitation in responses to management. J Appl Ecol. 2011;38: 204–216.
- 7. Carr MH, Neigel JE, Estes JA, Andelman S, Warner RR, Largier JL. Comparing marine and terrestrial ecosystems: Implications for the design of coastal marine reserves. Ecol Appl. 2003;13: 90–107.
- 8. Shanks AL, Grantham BA, Carr MH. Propagule dispersal distance and the size and spacing of marine reserves. Ecol Appl. 2003;13: 159–169.
- 9. Durrant HMS, Burridge CP, Kelaher BP, Barrett NS, Edgar GJ, Coleman MA. Implications of macroalgal isolation by distance for networks of marine protected areas. Conserv Biol. 2014;28: 438–445. pmid:24373031
- 10. Bullock JM, Shea K, Skarpaas O. Measuring plant dispersal: An introduction to field methods and experimental design. Plant Ecol. 2006;186: 217–234.
- 11. Clark JS, Beckage B, Camill P, Cleveland B, HilleRisLambers J, Lichter J, et al. Interpreting recruitment limitation in forests. Am J Bot. 1999;86: 1–16. pmid:21680341
- 12. Gaylord B, Reed DC, Raimondi PT, Washburn L, McLean SR. A physically based model of macroalgal spore dispersal in the wave and current-dominated nearshore. Ecology. 2002;83: 1239–1251.
- 13. Gaylord B, Reed DC, Raimondi PT, Washburn L. Macroalgal spore dispersal in coastal environments: Mechanistic insights revealed by theory and experiment. Ecol Monogr. 2006;76: 481–502.
- 14. Taylor D, Delaux S, Stevens C, Nokes R, Schiel D. Settlement rates of macroalgal algal propagules: Cross-species comparisons in a turbulent environment. Limnol Oceanogr. 2010;55: 66–76.
- 15. Stevens CL, Taylor DI, Delaux S, Smith MJ, Schiel DR. Characterisation of wave-influenced macroalgal propagule settlement. J Mar Syst. Elsevier B.V.; 2008;74: 96–107.
- 16. Schiel DR, Foster MS. The population biology of large brown seaweeds: Ecological consequences of multiphase life histories in dynamic coastal environments. Annu Rev Ecol Evol Syst. 2006;37: 343–372.
- 17. Cheminée A, Sala E, Pastor J, Bodilis P, Thiriet P, Mangialajo L, et al. Nursery value of Cystoseira forests for Mediterranean rocky reef fishes. J Exp Mar Bio Ecol. Elsevier B.V.; 2013;442: 70–79.
- 18. Mineur F, Arenas F, Assis J, Davies AJ, Engelen AH, Fernandes F, et al. European seaweeds under pressure: Consequences for communities and ecosystem functioning. J Sea Res. 2014;98: 91–108.
- 19. Norton TA. Dispersal by macroalgae. Br Phycol J. 1992;27: 293–301.
- 20. Chapman ARO. Functional ecology of fucoid algae: twenty-three years of progress. Phycologia. 1995;34: 1–32.
- 21. Kinlan BP, Gaines SD. Propagule dispersal in marine and terrestrial environments : A community perspective. Ecology. 2003;84: 2007–2020.
- 22. Kinlan BP, Gaines SD, Lester SE. Propagule dispersal and the scales of marine community process. Divers Distrib. 2005;11: 139–148.
- 23. Ballesteros E, Garrabou J, Hereu B, Zabala M, Cebrian E, Sala E. Deep-water stands of Cystoseira zosteroides C. Agardh (Fucales, Ochrophyta) in the Northwestern Mediterranean: Insights into assemblage structure and population dynamics. Estuar Coast Shelf Sci. Elsevier Ltd; 2009;82: 477–484.
- 24. Airoldi L, Beck MW. Loss, status and trends for coastal marine habitats of Europe. Oceanogr Mar Biol. 2007;45: 345–405.
- 25. Thibaut T, Pinedo S, Torras X, Ballesteros E. Long-term decline of the populations of Fucales (Cystoseira spp. and Sargassum spp.) in the Albères coast (France, North-western Mediterranean). Mar Pollut Bull. 2005;50: 1472–1489. pmid:16026805
- 26. Capdevila P, Hereu B, Riera JL, Linares C. Unravelling the natural dynamics and resilience patterns of underwater Mediterranean forests: insights from the demography of the brown alga Cystoseira zosteroides. J Ecol. 2016;104: 1799–1808.
- 27. Buonomo R, Assis J, Fernandes F, Engelen A, Airoldi L, Serrão E. Habitat continuity and stepping-stone oceanographic distances explain population genetic connectivity of the brown alga Cystoseira amentacea. Mol Ecol. 2016; pmid:27997043
- 28. Gianni F, Bartolini F, Airoldi L, Ballesteros E, Francour P, Guidetti P, et al. Conservation and restoration of marine forests in the Mediterranean Sea and the potential role of Marine Protected Areas. Adv Oceanogr Limnol. 2013;4: 83–101.
- 29. Mangialajo L, Chiantore M, Susini ML, Meinesz A, Cattaneo-Vietti R, Thibaut T. Zonation patterns and interspecific relationships of fucoids in microtidal environments. J Exp Mar Bio Ecol. 2012;412: 72–80.
- 30. Kendrick GA, Walker DI. Dispersal of propagules of Sargassum spp. (Sargassaceae: Phaeophyta): Observations of local patterns of dispersal and consequences for recruitment and population structure. J Exp Mar Bio Ecol. 1995;192: 273–288.
- 31. Susini M-L, Thibaut T, Meinesz A, Forcioli D. A preliminary study of genetic diversity in Cystoseira amentacea (C. Agardh) Bory var. stricta Montagne (Fucales, Phaeophyceae) using random amplified polymorphic DNA. Phycologia. 2007;46: 605–611.
- 32. Dudgeon SR, Kubler JE, Wright WA, Vadas RLS, Petraitis PS. Natural variability in zygote dispersal of Ascophyllum nodosum at small spatial scales. Funct Ecol. 2001;15: 595–604.
- 33. Capdevila P, Linares C, Aspillaga E, Navarro L, Kersting DK, Hereu B. Recruitment patterns in the mediterranean deep-water alga Cystoseira zosteroides. Mar Biol. 2015;162: 1165–1174.
- 34. Giaccone G, Bruni A. Le Cistoseire e la vegetazione sommersa del Mediterraneo. Univ. degli Studi; 1973.
- 35. Hereu B, Mangialajo L, Ballesteros E, Thibaut T. On the occurrence, structure and distribution of deep-water Cystoseira (Phaeophyceae) populations in the Port-Cros National Park (north-western Mediterranean). Eur J Phycol. 2008;43: 263–273.
- 36. Giaccone G. Écologie et chorologie des Cystoseira de Méditerranée. Rapp Commun Int Mer Méditerranéenne. 1973;22: 49–50.
- 37. Ballesteros E. Structure and dynamics s f the community of Cystoseira zosteroides (Turner) C. Agardh (Fucales, Phaeophyceae) in the northwestern Mediterranean. Sci Mar. 1990;54: 217–229.
- 38. Navarro L, Ballesteros E, Linares C, Hereu B. Spatial and temporal variability of deep-water algal assemblages in the Northwestern Mediterranean: The effects of an exceptional storm. Estuar Coast Shelf Sci. Elsevier Ltd; 2011;95: 52–58.
- 39. Clobert J, Baguette M, Benton TG, Bullock JM. Dispersal ecology and evolution. Oxford, UK: Oxford University Press; 2012.
- 40. Tufto J. Stochastic Dispersal Processes in Plant Populations. Theor Popul Biol. 1997;26: 16–26.
- 41. Bullock JM, Clarke RT. Long distance seed dispersal by wind: measuring and modelling the tail of the curve. Oecol. 2000;124: 506–521. pmid:28308389
- 42. Skarpaas O, Stabbetorp OE, Rønning I, Svennungsen TO. How far can a hawk ‘ s beard fly ? Measuring and modelling the dispersal of Crepis praemorsa. J Ecol. 2004;92: 747–757.
- 43. Bolker B, R Development Core Team. bbmle: Tools for General Maximum Likelihood Estimation. R package version 1.0.18. http://CRAN.R-project.org/package=bbmle. 2016. pp. 1–15.
- 44. R Development Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.; 2014.
- 45. Akaike H. A New Look at the Statistical Model Identification. IEEE Trans Automat Contr. 1974;19: 716–723.
- 46. Burnham K, Anderson DR. Model selection and multi-model inference. 2nd ed. New York: Springer-Verlag, Inc.; 2002. https://doi.org/10.1007/b97636
- 47. Bullock JM, Mallada González L, Tamme R, Götzenberger L, White SM, Pärtel M, et al. A synthesis of empirical plant dispersal kernels. J Ecol. 2017;105: 6–19.
- 48. Coleman MA, Roughan M, Macdonald HS, Connell SD, Gillanders BM, Kelaher BP, et al. Variation in the strength of continental boundary currents determines continent-wide connectivity in kelp. J Ecol. 2011;99: 1026–1032.
- 49. Hirsch BT, Visser MD, Kays R, Jansen PA. Quantifying seed dispersal kernels from truncated seed-tracking data. Methods Ecol Evol. 2012;3: 595–602.
- 50. Bates D, Mächler M, Bolker B, Walker S. Fitting Linear Mixed-Effects Models using lme4. J Stat Softw. 2014;67: 1–48.
- 51. Clark JS, Silman M, Kern R, Macklin E, HilleRisLambers J. Seed dispersal near and far: patterns across temperate and tropical forests. Ecology. 1999;80: 1475–1494.
- 52. Vadas RL, Johnson S, Norton TA. Recruitment and mortality of early post-settlement stages of benthic algae. Br Phycol J. 1992;27: 331–351.
- 53. Wright JT, Steinberg PD. Effect of variable recruitment and post-recruitment herbivory on local abundance of a marine alga. Ecology. 2001;82: 2200–2215.
- 54. Dudgeon S, Petraitis PS. First year demography of the foundation species, Ascophyllum nodosum, and its community implications. Oikos. 2005;109: 405–415.
- 55. Hughes TP, Tanner JE. Recruitment failure, life histories, and long-term decline of caribbean corals. Ecology. 2000;81: 2250–2263.
- 56. Doropoulos C, Roff G, Bozec YM, Zupan M, Werminghausen J, Mumby PJ. Characterizing the ecological trade-offs throughout the early ontogeny of coral recruitment. Ecol Monogr. 2016;86: 20–44.
- 57. Forbis TA, Doak DF. Seedling establishment and life history trade-offs in alpine plants. Am J Bot. 2004;91: 1147–1153. pmid:21653470
- 58. Cervin G, Åberg P, Jenkins SR. Small-scale disturbance in a stable canopy dominated community: implications for macroalgalrecruitment and growth. Mar Ecol Prog Ser. 2005;305: 31–40.
- 59. Irving AD, Balata D, Colosio F, Ferrando GA, Airoldi L. Light, sediment, temperature, and the early life-history of the habitat-forming alga Cystoseira barbata. Mar Biol. 2009;156: 1223–1231.
- 60. Kendrick GA. Effects of propagule settlement density and adult canopy on survival of recruits of Sargassum spp. (Sargassaceae: Phaeophyta). Mar Ecol Prog Ser. 1994;103: 129–140.
- 61. Steen H, Scrosati R. Intraspecific competition in Fucus serratus and F. evanescens (Phaeophyceae: Fucales) germlings: Effects of settlement density, nutrient concentration, and temperature. Mar Biol. 2004;144: 61–70.
- 62. Paine RT, Buhle ER, Levin SA, Kareiva P. Short‐range dispersal maintains a volatile marine metapopulation: the brown alga Postelsia palmaeformis. Ecology. 2017;0: 1–14. pmid:28328145
- 63. Thibaut T, Blanfuné A, Boudouresque CF, Verlaque M. Decline and local extinction of Fucales in the French Riviera: The harbinger of future extinctions? Mediterr Mar Sci. 2015;16: 206–224.
- 64. Underwood AJ. Physical disturbances and their direct effect on an indirect effect: Responses of an intertidal assemblage to a severe storm. J Exp Mar Bio Ecol. 1999;232: 125–140.
- 65. Jenkins SR, Norton TA, Hawkins SJ. Long term effects of Ascophyllum nodosum canopy removal on mid shore community structure. J Mar Biol Assoc UK. 2004;84: 327–329.
- 66. Tegner M, Dayton PK, Edwards P, Riser K. Large scale, low frequency oceanographic effects on kelp forest succession: A tale of two cohorts. Mar Ecol Prog Ser. 1997;146: 117–134.
- 67. Steneck R, Graham MH, Bourque BJ, Corbett D, Erlandson JM, Estes JA, et al. Kelp Forest Ecosystems : Biodiversity, Stability, Resilience and Future. Environ Conserv. 2002;29: 439–459.
- 68. Campbell AH, Marzinelli EM, Vergés A, Coleman MA, Steinberg PD. Towards restoration of missing underwater forests. PLoS One. 2014;9: e84106. pmid:24416198
- 69. Hernández-Carmona G, García O, Robledo D, Foster M. Restoration techniques for Macrocystis pyrifera (Phaeophyceae) populations at the southern limit of their distribution in México. Bot Mar. 2000;43: 273–285.
- 70. Krumhansl KA, Okamoto DK, Rassweiler A, Novak M, Bolton JJ, Cavanaugh KC, et al. Global patterns of kelp forest change over the past half-century. Proc Natl Acad Sci U S A. 2016;113: 1–6. pmid:27849580
- 71. Wernberg T, Bennett S, Babcock RC, De Bettignies T, Cure K, Depczynski M, et al. Climate-driven regime shift of a temperate marine ecosystem. Science. 2016;353: 169–172. pmid:27387951
- 72. Thibaut T, Pinedo S, Torras X, Ballesteros E. Long-term decline of the populations of Fucales (Cystoseira spp. and Sargassum spp.) in the Albères coast (France, North-western Mediterranean). Mar Pollut Bull. 2005;50: 1472–1489. pmid:16026805
- 73. Pearson GA, Brawley SH. Reproductive ecology of Fucus distichus (Phaeophyceae): An intertidal alga with successful external fertilization. Mar Ecol Prog Ser. 1996;143: 211–223.
- 74. Fredriksen S, Sjøtun K, Lein TE, Rueness J. Spore dispersal in Laminaria hyperborea (Laminariales, Phaeophyceae). Sarsia. 1995;80: 47–53.
- 75. Reed DC, Laur DR, Ebeling AW. Variation in algal dispersal and recruitment: the importance of episodic events. Ecol Monogr. 1988;58: 321–335.
- 76. Dayton PK. Dispersion, dispersal, and persistence of the annual intertidal alga, Postelsia palmaeformis Ruprecht. 1973;54: 433–438.
- 77. Deysher L, Norton TA. Dispersal and colonization in Sargassum muticum (Yendo) Fensholt. J Exp Mar Bio Ecol. 1981;56: 179–195.
- 78. Engelen AH, Olsen JL, Breeman AM, Stam WT. Genetic differentiation in Sargassum polyceratium (Fucales: Phaeophyceae) around the island of Curaçao (Netherlands Antilles). Mar Biol. 2001;139: 267–277.
- 79. Kendrick GA, Walker DI. Dispersal distances for propagules of Sargassum spinuligerum (Sargassaceae, Phaeophyta) measured directly by vital staining and venturi suction sampling. Mar Ecol Prog Ser. 1991;79: 133–138.
- 80. Schiel DR, Thompson GA. Demography and population biology of the invasive kelp Undaria pinnatifida on shallow reefs in southern New Zealand. J Exp Mar Bio Ecol. 2012;434–435: 25–33.
- 81. Parada GM, Tellier F, Martínez EA. Spore dispersal in the intertidal kelp Lessonia spicata: Macrochallenges for the harvested Lessonia species complex at microscales of space and time. Bot Mar. 2016;59: 283–289.
- 82. Kang Y-H, Choo H-S, Sin J-A, Lee C. Numerical modeling of propagule dispersal for Sargassum bed restoration in Gamak Bay, Korea. J Appl Phycol. 2016;28: 1859–1874.