Effect of marine protected areas on distinct fish life-history stages

Marine Environmental Research xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Marine Environmental Research journal homepage: www.elsevier.com/locate/marenvrev

Effect of marine protected areas on distinct fish life-history stages Fabiana C. Félix-Hackradt1, Carlos W. Hackradt1,∗, Jorge Treviño-Otón, Ángel Pérez-Ruzafa, José A. García-Charton Department of Ecology and Hydrology, University of Murcia, Campus Espinardo, 30100, Murcia, Spain

A R T I C LE I N FO

A B S T R A C T

Keywords: Effect of protection Spillover Larvae retention Habitat selection MPA network Enforcement

The role of Marine Protected Areas on distinct life stages of Mediterranean reef fish species (classified on the basis of their economic value and mobility categories) was assessed in a network of marine reserves in SE Spain. Only abundance and biomass of adult of both commercial and demersal species were positively affected by protection. Gradients across reserve boundaries (as a clue to the occurrence of spillover) were observed for fish abundance but not for biomass, indicating a protected fish assemblage with a predominance of small-sized individuals. Also, post-larvae of commercial species were negatively related to protected zones. Active selection of settlement preferred habitats, larval accumulation favoured by the geomorphological configuration of the coast or mixed effects has been proposed as possible explanations. Juveniles showed high spatial variability resulting in a lack of response to fishing protection measures. We highlight the need of including early life stages and overall suitable habitats for them when designing MPA networks due to the crucial importance of these stages to successful fulfillment of MPA objectives.

1. Introduction Marine protected areas (MPAs) have been advocated as a major tool for recovery and conservation of marine resources (Fenberg et al., 2012). Their multiple benefits in protecting ecosystems and ecological processes while enhancing fisheries through density-dependence spillover and larval dispersal of target species (Roberts et al., 2001) convert them as the most powerful tool for spatial management in the marine environment. Although growing evidence supports the real benefits of MPA such as biomass or abundance increase (Russ and Alcala, 2004; Claudet et al., 2010), population structure restoration (Guidetti, 2006), spillover (Harmelin-Vivien et al., 2008), exportation of eggs and larvae (Cudney-Bueno et al., 2009; Crec'hriou et al., 2010) among others, little is known about the MPAs effect on settlement and recruitment events (Planes et al., 2000; Sale et al., 2005). MPAs may have important effects on the resident population structure (Planes et al., 2000). Location of the MPA may determine whether nursery habitats will be protected and thus favour species settlement. In theory, mortality of juvenile fish is supposed to be higher inside MPAs due to the increase of predator abundance; however, compensatory effects such as the increase in recruitment due to high habitat quality provided by fishing protection may offset net differences (Syms and Carr, 2001). In this context, it is paramount to address the

∗

1

ecological effects of MPAs on the first life-history stages of fish in order to properly design and implement this management tool. In addition, it has been hypothesized that MPAs can act as a source of propagules due to increased density and fecundity (as a consequence of the recovery of larger size classes) of protected populations, thus replenishing unprotected areas by dispersal of eggs and larvae (Planes et al., 2000). An indirect method to estimate the magnitude and importance of such export of larvae, juveniles and adults fishes from MPA to neighbouring areas is to look for the likely existence of gradients of biomass of target species across MPA limits, under the rationale that, if spillover occurs, there would be more fishes near than far away from the MPA (Chapman and Kramer, 1999; Pérez-Ruzafa et al., 2008). This research strategy has been used in several studies in the Mediterranean (e.g. Harmelin-Vivien et al., 2008; Pérez-Ruzafa et al., 2008; Goñi et al., 2008, 2010; Hackradt et al., 2014) and worldwide (e.g., Russ and Alcala, 2004; Amargós et al., 2010). In this work, we used the beyond-ACI approach to investigate the effect of protection on distinct stages of Mediterranean reef fish life cycle (post-larvae, juveniles and adults). We addressed the following questions: Is the intensity of larval supply affected by protection? What is the response of juvenile and adult abundance and richness? Is the success of protection effects dependent on species mobility and economic importance? Moreover, we addressed the question of the

Corresponding author. E-mail address: [email protected] (C.W. Hackradt). Present address: Universidade Federal do Sul da Bahia, Centro de Formação em Ciências Ambientais, Campus Sosígenes Costa, Porto Seguro, BA, CEP: 45810–000, Brazil.

https://doi.org/10.1016/j.marenvres.2018.06.012 Received 7 August 2017; Received in revised form 11 June 2018; Accepted 13 June 2018 0141-1136/ © 2018 Elsevier Ltd. All rights reserved.

Please cite this article as: Félix-Hackradt, F.C., Marine Environmental Research (2018), https://doi.org/10.1016/j.marenvres.2018.06.012

Marine Environmental Research xxx (xxxx) xxx–xxx

F.C. Félix-Hackradt et al.

existence of gradients of abundance of distinct fish life-history stages (post-larvae, juveniles and adults) of species and species groups constructed according to their pattern of spatial occupation and their economic importance, as an indirect way to detect and quantify the likely occurrence of spillover from no-take sites to neighbouring unprotected zones. The answer to these questions aimed to fulfil an important gap in the marine reserves science, by adding light to a complex period of the fish life-cycle and assist in the proper management of marine resources.

(July to September) the cumulative rainfall was 30 mm, and the average wind velocity was 2 m s−1 mostly from south-east direction alternating with north-west winds (data collected from Nijar Meteorological station, provided by Consejeria de Agricultura, Pesca y Desarollo Rural, Junta de Andalucia, 2011). 2.2. Sampling design During the settlement peak months of most rock reef fishes species (such as C. chromis, C. julis, D. annularis and all Symphodus species), namely between July and September 2011 (see Félix-Hackradt et al., 2013a, 2014 for more details), the abundance of post-larvae, juvenile and adult fish was surveyed monthly in three locations situated thousands of meters apart: Loma Pelada (LP), La Polacra (PO) and Punta Javana (PJ), and represent the spatial replication of the “Reserve effect”. At each location, 3 zones were delimited: the no take-zone (Protected) and 2 zones where artisanal fishing is permitted (Unprotected), located upstream (North) and downstream (South) the MPA boundaries (Fig. 1). In each zone, 3 sites separated by hundreds of meters were randomly chosen each time.

2. Material and methods 2.1. Study site The Cabo de Gata–Níjar Natural Park covers 38,000 ha of both marine and terrestrial areas in the province of Almería, SE Spain and extends approximately 2 km offshore along about 60 km of coastline. The marine area under regulation extends over 12,200 ha, from which 4613.45 ha are within no-take zones, where all harvest and recreational activities are forbidden, but allowed within the park limits. There are 5 no-take zones (from South to North: Cabo de Gata, Loma Pelada, Polacra, Punta Javana and Media Naranja). The studied MPAs, namely Loma Pelada, Polacra and Punta Javana MPAs (Fig. 1), are particularly located in coastal promontories of volcanic formation, surrounded by coarse sand beaches or rocky embayments. The depth varied little among surveyed zones, reaching 20 m just tens of meters from the coast, however, slope is gentler near embayments and steeper close to the base of the promontories (Moreno, 2003). MPAs and adjacent zones are dominated by rocky reefs, which extends underwater to a depth of 60 m, surrounded by sandy and detritic bottoms interspersed with extensive patches of Cymodocea nodosa and Posidonia oceanica which forms a narrow belt following the coast extending down to 10 and 30 m, respectively (Luque and Templado, 2004). During the summer of 2011

2.3. Collection of fish post-larvae Light-traps (CARE®, Ecocean, Montpellier, France) were used to sample the post-larval pool reaching the coastal zone. Post-larvae is defined here as a synonym of late-stage or competent larvae (FélixHackradt et al., 2013a). One light-trap per site (n = 9) was installed at sunset and retrieved at sunrise (∼10 h) during two consecutive nights, resulting in 18 samples for each locality. Although site location was not fixed we determined a deployment area, aparted from each other to avoid light interference, in which light traps were randomly installed in each sampling day. This procedure was repeated for all 3 studied MPAs in each campaign (i.e., July, August and September months), resulting

Fig. 1. Map of Cabo de Gata-Níjar Natural Park, Southeast Spain, showing the spatial delimitation of the three marine reserves studied, Loma Pelada (LP), Polacra (PO) and Punta Javana (PJ), and the location of light traps and underwater censuses within each zone (numbers 1–9). Shaded areas correspond to no-take zones. 2

Marine Environmental Research xxx (xxxx) xxx–xxx

F.C. Félix-Hackradt et al.

disturbed location(s) to multiple controls i.e. asymmetrical after control/impact design (ACI) (Glasby, 1997). A beyond-ACI design was constructed including three MPAs - LP, PO, and PJ (Factor Location (L), random, 3 levels), each with one no-take zone and 2 controls (Factor Zone (Z), 3 levels, random and nested in L). The asymmetrical analysis described in Glasby (1997) is a multiple stage procedure including three ANOVAs to construct the final asymmetrical design. Zone factor is only used in the construction of 2 contrasts: i) the contrast between the average of the three protected zones to the average of the six unprotected (control) zones (factor Protection (P), fixed, 2 levels) and the interaction between this factor P and factor Location (factor L × P), which contrasts the comparison of the protected area to the average of the two controls among the three locations; its significance would mean that, at each location, the difference between the protected area and the average of the controls will show different patterns; and ii) the contrast term comparing the two unprotected control zones that are nested within each location (factor C(L)); only the control zones are compared in this way because the no-take zones are not replicated within each location. See Glasby (1997) for a complete explanation of how to construct the asymmetrical analysis by combining the sum of squares values from separate analyses of variance and the appropriate F-tests, following Underwood (1993). Due to the above-mentioned loss of sampling devices, and also in order to homogenize the analysis for the three life stages surveyed, the time periods (nights for larvae and months for all life stages) and the sites (n = 9) were ruled out as experimental factors, and thus all replicates were included within each zone. This alteration led to a total sample number of 27 post-larvae, 135 juveniles and 81 adult samples, all times confounded. Prior to analyses, homogeneity of variance was checked using Cochran's test; in the few cases where variances were not homogeneous, we performed the analyses anyway since analysis of variance is quite robust to departures from their assumptions (Underwood, 1997). In these cases, tests have to be taken cautiously when the significance of the effect is 0.01 < P < 0.05. All analyses were done using “GAD” package in R software. Furthermore, a priori planned comparisons were used to evaluate the differences between the means of each Protection level (Control/Unprotected vs. Reserve) when interaction term L × P was significant, in other words, when effects of protection were not apparent or of the same sign in all MPAs studied, by comparing the protected area and the average of its two controls for each location separately; in that cases, to control the probability of type I error in multiple planned comparisons, the critical value of α was reduced to 0.05/ 3 = 0.02 (the Bonferroni correction) (see Glasby, 1997). Additionally, in order to evaluate the existence of a gradient of abundance and biomass (in the case of adults) across MPA limits, data from species categories in which a significant effect of protection was observed were tested using generalized additive modelling (GAM; Hastie and Tibshirani, 1990). GAMs were used due to the expectancy of a non-linear response of species abundance and predictor variables, as they performed better fitting over generalized linear models (Dixon, 1999). Models were constructed using a smooth cubic line and default degrees of freedom using “gam” package in R. We used a Gaussian error distribution which is more appropriate when dealing with averaged measures than others (i.e. Pearson, binomial).

in 54 replicate light-trap samples per month (but see below) (see FélixHackradt et al., 2013a, for a detailed description of trap installation procedure). To minimize moonlight influence as well as differences in settlement pulses phases field campaigns were concentrated within the 3 days after and before the new moon period in order to optimize larval catches (Sponaugle and Cowen, 1996). Additionally, to control for random local oceanographic conditions such as wind, turbidity or current effect on larval catch in each light trap, the sequence of which MPAs were sampled in each campaign was randomly chosen. Due to the loss of most light-traps at the beginning of the third campaign, the data issued from the last campaign (September) were not included in the analyses. Each trap sample was then sorted using a light stereomicroscope to the lowest identification level using the appropriate bibliography (e.g., Lo Bianco, 1931; Arias and Drake, 1990; Ré, 1999). Some individuals from the genus Atherina, Bothus, Trachurus, Lepadogaster, Symphodus, Diplodus and Serranus were not identified at the species level due to taxonomic difficulties; blennies and mugilids could only be identified to the Family level. 2.4. Juvenile and adult censuses In each site we counted visually by SCUBA diving all non-cryptic juveniles on five 10 × 2 m replicate transects set out haphazardly within a depth range of 0–5 m. As juveniles, we considered all settled individuals from a minimum size of 10 mm until the size of sexual maturity, defined a posteriori with proper bibliography. This minimum size was determined to guarantee that all settlers were counted, as evidenced by Félix-Hackradt et al. (2013b). Each observation was assigned to one of eleven abundance classes following Félix-Hackradt et al. (2013b). Geometric means of each fish abundance class were used for the calculation of abundances. The size of each observed fish was visually estimated to the nearest cm. For estimating the abundance of non-cryptic adults, three 50 × 5 m replicate transects were visually surveyed along a depth range of 10–15 m. Each observation was assigned to one of nine abundance classes following Harmelin (1987). Analogously to the case of juvenile data, the geometric mean of each class was considered for further calculations. Individual sizes were estimated in classes of 2 cm; to minimize bias in size estimation, 4 well trained and intercalibrated observers were responsible for all censuses, which were done between 10:00 and 15:00 h GMT, when water conditions (turbidity and swell) were optimal (Harmelin-Vivien et al., 1985). 2.5. Data analysis As fish species may respond to protection either by increasing abundance but also their body weight, biomass of each adult species recorded were estimated using the parameters of length-weight curves provided by online database Fishbase.org (Froese and Pauly, 2011). Due to significant variation in the response of different fish species to protection (Claudet et al., 2010), species were categorized into 4 groups. Their relative importance to fisheries was defined as commercial or non-commercial following Claudet et al. (2010), and spatial categories were defined as pelagic, which includes categories 1 and 2 of Harmelin (1987), and demersal, including categories 3 to 6 (Table 1). The effect of protection on fish abundance (log-transformed) and species richness of the different life stages (total and by commercial and spatial categories), and on biomass of adult categories only, was evaluated using an asymmetrical analysis of variance (ANOVA) based on Glasby (1997). The use of multiple control locations in an asymmetrical analysis of variance to test for differences between impact and control sites (beyond-BACI designs) allows sufficient power to detect changes owing to the impact source (Underwood, 1992). The asymmetrical beyond-BACI designs can be modified and used to identify environmental impacts when only 'after data' are available, by comparing the

3. Results 3.1. Fish assemblage A total of 74 different taxa were recorded among the 3 life stages surveyed (Table 1). Light-traps captured 6078 individuals belonging to 45 taxa. Blenniidae (34% of the total number of post-larval individuals captured), Oblada melanura (21%), Boops boops (18%) and Chromis chromis (10%) were the most representative taxa accounting for more than 80% of the total post-larval catch. A total of 16,351 juvenile individuals of 34 taxa were identified in visual censuses, from which C. 3

Marine Environmental Research xxx (xxxx) xxx–xxx

F.C. Félix-Hackradt et al.

Table 1 Total and relative abundance values per life stage: larvae, juvenile and adult, and spatial and economic categories of species registered at Cabo de Gata Natural Park. Spatial categories: Demersal (D) or pelagic (P) species; Economic category: commercial (C) or non-commercial (N). Family

Species

Spatial

Economic

Larvae

%

Anguilidae

Anguilla anguilla

D

C

1

0.02

Apogonidae

Apogon imberbis

D

N

Atherinidae

Atherina boyeri Atherina spp

P P

C C

4 18

0.07 0.30

Blennidae

Blenniidae

D

N

2078

34.19

Bothidae

Arnoglossus laterna Bothus spp

D D

N N

3 3

0.05 0.05

Lichia amnia Seriola dumerilli Trachinotus ovatus Trachurus mediterraneus Trachurus spp Trachurus trachurus

P P P P P P

C C C C C C

Centracanthidae

Spicara maena Spicara smaris

P P

N N

Clinidae

Clinitrachus argentatus

D

N

36

0.59

Clupeidae

Sardina pilchardus

P

C

149

2.45

Congridae

Ariosoma balearicum

D

N

41

0.67

Dactylopteridae

Dactylopterus volitans

D

N

4

0.07

Dasyatidae

Dasyatis pastinaca

D

C

Engraulidae

Engraulis encrasicolus

P

C

21

0.35

Gobiesocidae

Lepadogaster spp

D

N

4

0.07

Gobiidae

Gobius paganellus

D

N

1

0.02

Haemulidae

Parapristipoma octolineatum

D

C

Labridae

Coris julis Labrus merula Labrus viridis Symphodus cinereus Symphodus doderleine Symphodus mediterraneus Symphodus melanoceros Symphodus ocellatus Symphodus roissali Symphodus rostratus Symphodus spp Symphodus tinca Thalasoma pavo

D D D D D D D D D D D D D

N C C N N N N N N N N N N

Dicentrachus labrax Dicentrachus punctatus

D D

C N

Mugilidae

Mugilidae

P

C

7

Mullidae

Mullus barbatus Mullus surmuletus

D D

C C

15 16

Muraenidae

Muraena helena

D

Myliobatidae

Myliobatis aquila

Pomacentridae

Carangidae

183 5 14 36

Juvenile

%

Adult

%

230

1.41

671

3.76

936

5.72

1

0.01

1

0.01

9

0.06 20

0.11

143 98

0.80 0.55

1

0.01

72

0.40

1299 53 24 14 169 305 5 508 285 157

7.27 0.30 0.13 0.08 0.95 1.71 0.03 2.84 1.60 0.88

3.01 0.08 0.23 0.59

658 1 1 1 4 55

4.02 0.01 0.01 0.01 0.02 0.34

0.08

461 202

2.82 1.24

0.07 0.02 0.10

729 585

4.46 3.58

1038 78

5.81 0.44

259

1.58

7 18

0.04 0.10

0.12

195

1.19

100

0.56

0.25 0.26

58

0.35

246

1.38

C

10

0.06

D

N

2

0.01

Chromis chromis

P

N

Sciaenidae

Sciaena umbra

D

C

Scombridae

Auxis rochei

P

C

Moronidae

55

0.90

5 4 1 6

655

1

10.78

6213

38.00

2958

16.55

2

0.01

254

1.42

0.02

(continued on next page)

4

Marine Environmental Research xxx (xxxx) xxx–xxx

F.C. Félix-Hackradt et al.

Table 1 (continued) Family

Species

Spatial

Economic

Larvae

%

Scorpaenidae

Scorpaena notata Scorpaena porcus Scorpaena scrofa

D D D

C C C

4 7

0.07 0.12

Epinephelus costae Epinephelus marginatus Serranus cabrilla Serranus hepatus Serranus scriba Serranus spp

D D D D D D

C C C C C C

1 3 3

0.02 0.05 0.05

1

0.02

Boops boops Dentex dentex Diplodus annularis Diplodus cervinus Diplodus puntazzo Diplodus spp Diplodus vulgaris Litognathus mormyrus Oblada melanura Pagellus acarne Pagellus bogaraveo Pagrus pagrus Sarpa salpa Sparus aurata Spondyliosoma cantharus

P D D D D D D D P D D D D D D

N C C C C C C C N C C C N C C

1136

18.69

186

3.06

16 1

0.26 0.02

1273 1 2 35

20.94 0.02 0.03 0.58

Sphyraenidae

Sphyraena viridensis

D

C

Syngnathidae

Syngnathus thyphle

D

N

2

0.03

Trachinidae

Echiichthys vipera

D

N

1

0.02

Serranidae

Sparidae

1 39

0.02 0.64

Juvenile

%

Adult

%

2 1

0.01 0.01

9 5

0.06 0.03

1 17 84

0.01 0.10 0.47

149

0.91

309

1.73

779 52 22 7 336 533 764 7 1164

4.76 0.32 0.13 0.04 2.05 3.26 4.67 0.04 7.12

1778 45 567 57 105 623 1672 7 793

9.95 0.25 3.17 0.32 0.59 3.49 9.36 0.04 4.44

1909 9

11.68 0.06

5 3095 24

0.03 17.32 0.13

6

0.04

147

0.82

Total abundance

6078

16351

17868

Number of species/taxa

45

34

46

biomass (in the case of adults) and number of species (Table 2). A trend was observed in all larval categories to be more abundant outside the limits of the no-take reserves, although this pattern was found to be significant only in the case of commercial categories (Fig. 2a–d). High variability of larval abundance among control zones within each location was also recorded. For its part, we found a greater variation in abundance of juveniles among locations and zones, this variability being statistically significant in the case of pelagic noncommercial species that were more abundant inside the no-take areas (Fig. 2f), while the abundance of juveniles of commercial species varied greatly at the scale of location (Fig. 2e,g), and those of demersal noncommercial species among zones (Fig. 2h); it is noteworthy that the abundance of juveniles was lower in almost all unprotected zones located north of the reserves. Adult abundance (Fig. 2i-l) and biomass (Fig. 2m-p) followed the same pattern of variation between locations and zones, with non-commercial species presenting higher mean abundances outside the no-take reserves (except in PO for abundance), the opposite occurring for commercial species, pelagic commercial species showing also a significant variability at the scale of locations (especially at PJ) (Fig. 2).

chromis (38% of total abundance), Sarpa salpa (11%) and O. melanura (7%) accounted for 56% of all observed fishes. Finally, 17,868 adult individuals from 46 species were censused, the species S. salpa (17%), C. chromis (16%), B. boops (10%) and Diplodus vulgaris (9%) accounting for 52% of total abundance. About 77% of all taxa were categorized as demersal species, and ∼60% belonged to commercially important species. A total of 15 taxa were observed in the three life history stages surveyed; whereas twenty-three were exclusively observed in the lighttrap catches, while Lichia amnia was censused only as a juvenile, and 12 species were only seen as adults (Table 1). 3.2. Effect of protection The effect of protection measures was evident for adult stages, confounded by spatial variability in post-larval samples and absent for juveniles (Table 2). Regarding post-larvae, total post-larval richness and the abundance of the post-larvae of both pelagic and demersal commercial species responded significantly to the effect of the interaction term L × P, indicating that the effect of protection was not homogeneous among locations. Post-hoc comparisons indicate that mean post-larval species richness was found to be higher in unprotected zones except at PO location, where no differences were found. For its part, the mean abundance of post-larvae of commercial species was higher in unprotected zones at PO and PJ reserves for pelagic species, and only at PJ location for demersal ones. (Table 2, Fig. 2a–c). In the case of adults, the richness, abundance and biomass of both demersal and commercial species responded significantly to protection (Table 2; Fig. 2k,o), with higher mean values inside protected sites in all locations. Additionally, significant variability among unprotected controls sites within locations [C(L)] was found for demersal juveniles (Fig. 2g–h), demersal commercial adults (Fig. 2k,o) and all adult species, both for abundance,

3.3. Gradients of abundance of adult fish across no-take zones limits Gradients across MPA boundaries were only investigated in adult demersal commercial species as they were the only ones in which densities and biomass were significantly different between protected and unprotected areas (see above). Significant fits of GAMs were obtained for all locations regarding species abundance, in which a clear reduction with increasing distances from MPA core was observed, except for Loma Pelada reserve, where the southern unprotected control zone harboured densities as high as the corresponding no-take area 5

6

2 1 2 3 72

Df

F

N

0.08 9.36 0.52 1.52

2 1 2 3 126

Df

2 1 2 3 18

ns ns ns ns

P

1.24 16.35 0.42 2.81

F

0.53 6.25 0.04 4.65

F

F

S P ns ns ns ***

Adults

ns ns ns *

P

F

B

S

S

0.91 17.13 0.10 2.99

F

0.12 0.17 0.31 16.30

F

5.01 2.26 10.02 0.20

Juveniles

ns ns ns *

P

0.27 25.17 0.13 7.46

N

N

Larvae

ns ns ns *

P

ns ns ns ***

P

ns ns * ns

P

0.26 6.43 2.85 0.58

F

N

9.50 2.23 5.68 0.46

F

ns ns ns ns

P

103.27 0.82 233.71 0.01

F

N

N

3.44 4.40 0.72 0.80

F

S

ns ns ns ns

P

Ad pel com

ns ns ns ns

P

F

B

S

S

0.37 14.53 0.52 1.24

F

4.27 1.69 2.49 1.12

F

4.42 30.03 0.85 0.19

Juv pel com

** ns *** ns

P

Lar pel com

ns ns ns ns

P

ns ns ns ns

P

* ns ns ns

P

3.86 0.06 1.14 0.48

F

N

N

ns ns ns ns

P

0.69 8.03 0.24 1.93

F

0.53 3.33 0.97 1.45

F

N

0.29 0.43 3.00 2.63

F

0.21 0.39 1.82 1.26

F

S

F

ns ns ns ns

P

F

S

S

B

5.20 0.28 3.20 0.25

3.24 15.55 0.26 1.34

Ad pel nc

ns ns ns ns

P

Juv pel nc

ns ns ns ns

P

Lar pel nc

ns ns ns ns

P

ns ns ns ns

P

ns ns ns ns

P

0.38 30.13 0.15 4.60

F

N

1.08 0.00 1.00 29.10

F

22.11 0.18 25.50 0.18

F

ns * ns **

P

N

N

4.40 0.00 1.85 0.61

F

0.27 5.43 0.18 6.95

F

0.34 61.28 0.09 6.07

F

S

ns * ns **

P

Ad dem com

ns ns ns ***

P

Juv dem com

* ns * ns

P

Lar dem com

F

B

ns ns ns ***

P

ns ns ns ns

P

1.21 36.46 0.18 3.89

S

S

ns * ns *

P

0.46 2.74 0.12 2.39

F

N

N

0.11 1.51 0.57 11.27

F

8.52 2.59 0.45 0.84

F

N

ns ns ns ns

P

1.01 5.55 1.06 0.48

F

0.07 0.25 0.76 13.65

F

F

S

ns ns ns **

P

Ad dem nc

0.25 2.55 0.30 5.22

ns ns ns ***

P

Juv dem nc

ns ns ns ns

P

Lar dem nc

Significance: *** - P < 0.001; ** - P < 0.01; * - P < 0.05; ns - not significant. Codes: lar – larvae, juv – juvenile, ad – adult, pel - pelagic, dem - demersal, com - commercial and nc - non-commercial. a Biomass was only calculated for adults categories.

Location Protection (R vs C) LxP C(L) Res

Location Protection (R vs C) LxP C(L) Res

Location Protection (R vs C) LxP C(L) Res

Df

1.55 2.14 0.25 1.73

F

S

S

B P ns ns ns ns

ns ns ns ***

P

ns ns ns ns

P

Table 2 Results of ANOVA on total abundance (N), species richness (S) and biomassa (B) of fish life stages – larvae (lar), juvenile (juv) and adult (ad) - for random factors Location (L), Protection (P: contrast of Reserve vs Controls levels, R vs C), and the orthogonal and nested interactions L × P and C(L).

F.C. Félix-Hackradt et al.

Marine Environmental Research xxx (xxxx) xxx–xxx

Marine Environmental Research xxx (xxxx) xxx–xxx

F.C. Félix-Hackradt et al.

Fig. 2. Mean abundance of larval (a–d), juvenile (e–h), adult (i–l) and biomass of adult (m–p) categories within zones inside reserves studied. LP-Loma Pelada, PJPunta Javana and PO-Polacra. Error bars indicate standard mean error.

(Fig. 3). In the case of the adult fish biomass of demersal commercial species, however, significantly lower values were observed within the limits of Loma Pelada reserve, which were increased beyond those boundaries both to the north and to the south. No significant differences were found for biomass values in the other two locations studied (Fig. 3, Table 3). The coefficients of determination of the models that were significant were relatively low, ranging 0.18–0.24 (Table 3).

recruitment success in MPAs. On the other hand, sampling design is considered the major limiting factor to fully evaluate MPAs effect (Guidetti, 2002). Lack of proper replication, pseudo-replication, and uncontrolled confounding factors (such as habitat features) are the main obstacles to accurately assess the effects of protection (GarcíaCharton et al., 2000; Goñi et al., 2008). In this work we assessed the effect of protection by using a beyondACI design (Underwood, 1992, 1993; Glasby, 1997) incorporating a replication of the effect of protection (in 3 distinct locations) and their contrast with multiple nearby unprotected controls, as well as finerscale spatial replication, in order to control for habitat heterogeneity, thus allowing comparative and reliable estimates of MPA effects. Moreover, this is the first contribution to date that evaluates simultaneously the response of different life stages – post-larvae, juveniles and adults – to protection measures. This study was able to register about 45 taxa of fish post-larvae, 34 juvenile and 46 adult species, of which 15 taxa were recorded in all life

4. Discussion A growing number of studies highlighted the benefits of MPAs on adult populations by evidencing increases in abundance and mean size as well as density-dependent effects (García-Charton et al., 2008; PérezRuzafa et al., 2017). Since fecundity is exponentially related to fish size, increased production of eggs and larvae within MPAs should result in enhanced recruitment both inside and outside protected areas (Planes et al., 2000). However, few empirical studies have addressed so far the 7

Marine Environmental Research xxx (xxxx) xxx–xxx

F.C. Félix-Hackradt et al.

Fig. 3. GAMs output for the relationship abundance and biomass of adult demersal commercial species to distance from MPA core of studied reserves; Punta Javana (a,d), Polacra (b,e) and Loma Pelada (c,f). Shaded area delimit no take zones within each locality, dashed lines correspond to 95% confidence limits. * - significant, ns – not significant.

interaction with other individuals (Abesamis and Russ, 2005). In any case, the patchy nature of the marine environment can act as barriers to fish movement and will vary according to species and their relationship with habitat, thus influencing spillover effect (Forcada et al., 2009). Indeed, a significant variation among unprotected controls within locations was detected for adult species (see Fig. S1), mainly for demersal ones. This heterogeneous distribution could be related to habitat preferences and availability in the area; no-take zones have a greater proportion of rocky bottoms than adjacent unprotected controls. Interestingly, we found a pattern of adult biomass distribution across Loma Pelada reserve boundaries which was opposite to the one found for fish abundance. One possible explanation to the increase in biomass observed outside this no-take area was mainly due to the accumulation of small-sized fish inside the no-take area, rather than an increase in the mean size of the individuals of protected populations. Recently, the Spanish Ministry for the Environment has strengthened the conditions of use inside the Cabo de Gata Natural Park due to countless cases of transgression of park rules as illegal spearfishing, poaching by professional fishing boats inside park limits, etc., as well as the enforcement of monitoring and in situ surveillance. It is well known that without enforcement, positive effects of marine reserves could not be distinguished from fished areas (Guidetti et al., 2008; Giakoumi et al., 2017). Although no data of fishing pressure (legal or illegal) inside the Marine Park are available, considering the testimonies of users and regular visitors of the Cabo de Gata Marine Park, it is more than likely that the continued removal of the largest fish from inside the no-take zones, mainly by spearfishers, whose activity is very size-selective, have prevented individuals from reaching larger sizes; this would explain why especially small-sized fish were found inside the Loma Pelada reserve. Another expected benefit of MPAs is a greater production of eggs and larvae and their export to nearby areas as a consequence of higher adult fecundity (García-Charton et al., 2008), which has only recently received some empirical support (Harrison et al., 2012; Almany et al., 2013). In our study, the larval supply was studied using light traps, which are selective, but still useful devices to sample post-larval stages that are able to settle. By this method, we detected a negative effect of protection on post-larval commercial species, however, it varied greatly among MPAs studied (significant LxP). With the increased evidence of

Table 3 GAM results of the relationship abundance and biomass of adult demersal commercial species for each studied marine reserve – Loma Pelada, Polacra and Punta Javana. F-statistic (F), degrees of freedom (Df), coefficient of determination (R2), deviance explained (Deviance), P - significance at α = 0.05,* P < 0.05,** - 0.05 < P < 0.001,*** - P < 0.001. Variable

Location

Df

F

P

Deviance

R2

Abundance

Loma Pelada Polacra Punta Javana Loma Pelada Polacra Punta Javana

76 76 76 76 76 76

3.772 6.373 3.108 7.904 2.578 1.938

* *** * *** ns ns

11.624 9.913 11.547 25.237 32.942 32.617

0.18 0.23 0.19 0.24 0.19 0.25

Biomass

stages. Other studies conducted in larger temporal scales recorded 36–43 post-larval taxa (Félix-Hackradt et al., 2013a,b; 2014), 36–42 juveniles (Félix-Hackradt et al., 2014) and 40-45 adult species (GarcíaCharton and Pérez-Ruzafa, 2001, 2004) indicating that, although temporally concentrated, the effort employed was high enough to sample adequately the local rocky fish species diversity. Only adults belonging to both commercial and demersal categories were positively affected by protection. On average, higher abundance, biomass and species richness were found inside the no-take zone in all reserves studied. In a meta-analytical study of European marine reserves, Claudet et al. (2010) found that the effect of protection on fish will depend firstly on their commercial value and to a lesser extent to their habitat preference. It is expected that target species will be the first to respond to a fishing ban by an abundance increase inside reserve limits, and the greater is the relationship of a species with the bottom, the greater and faster will be the response to protection; this was the case of demersal economically important species in this work. On the other hand, a clear reduction in densities with increasing distances from the reserve cores was consistent at both southern and northern boundaries for all reserves analysed, with the exception of Loma Pelada, in which similar abundances were detected within the notake area and across the southern boundary. Spillover could be the result of random movements of individuals from MPAs to outside their borders (Chapman and Kramer, 1999) or by ontogenetic habitat shifts (Nagelkerken and van der Velde, 2002) or also by density-dependent 8

Marine Environmental Research xxx (xxxx) xxx–xxx

F.C. Félix-Hackradt et al.

We found that not only adults but also post-larvae were related to spatial protection measures. Adults of commercial and demersal species were more abundant inside reserve limits, as well as their biomass. However, as no clear biomass gradient was observed across reserve boundaries, we conclude that higher abundance values were caused by an increase in small-sized fishes, which did not contribute with proportionally high biomass values. Lack of adequate enforcement is suggested as the principal cause for these results, highlighting the need for a high level of surveillance to fulfil MPA objectives. On the other hand, post-larvae of commercially important species (both pelagic and demersal) were more abundant outside protected areas although differing greatly among MPAs studied. Active selection of preferred habitats for settlement, as well as larval retention favoured by the geomorphological configuration of the coast (i.e. predominance of embayments outside no-take areas), has been proposed to explain such findings. The observed high spatial variability in the distribution of juveniles may explain their lack of response to fishing protection measures. By changing mortality rates, increasing habitat complexity, enhancing predator number, and other community/ecosystem effects, protection can exert a huge influence on population dynamics. Nevertheless, much more research has to be undertaken in order to understand the factors explaining larval spatial distribution, their reflection on year-class strength and therefore on adult stocks. We strongly suggest the inclusion of knowledge on early life-history stages of reef fishes on the design of MPA networks, due to the direct implications of these phases on the successful achievement of MPA objectives.

self-recruiting mechanisms for a number of different species (see also Jones et al., 2009), higher abundance in unprotected control zones could be reflecting the enhancement of production of larvae in adjacent no-take zones (see also Almany et al., 2013). Nevertheless, as we were unable to relate larval species to corresponding juveniles and adults samples due to distinct methodologies applied, further studies focusing on the identification of larval sources are fundamental to support the hypothesis of self-replenishment (Calò et al., 2017). Additionally, the singular configuration of local currents and winds could also have caused larval accumulation at unprotected control zones, thus contributing to larval retention (Jones et al., 2009). Regardless of the origin of larvae, the studied no-take zones were all located around protruding projections of the rocky coastline highly exposed to prevailing winds (mostly from SW and NE direction) and currents, which may have influenced the distribution of larvae along the coast. For their part, adjacent unprotected control zones were usually situated in sheltered embayments, in which local oceanographic conditions may well have contributed to the concentration of post-larvae in these zones, especially those of pelagic species. Embayment areas are often dominated by seagrass patches formed mainly by P. oceanica (see Fig. S1), which is considered the nursery habitat for a number of Mediterranean species (Francour, 1997) such as sparids, syngnathids, and labrids, among others (García-Rubies and Mcpherson, 1995). Besides of being passively accumulated into embayments, larvae of some species are capable to perceive reef habitats from distances of at least 1 km (Leis et al., 1996), and they could be actively choosing their preferred habitat on which to settle at the time of capture, as these zones may harbour higher diversity of settlement habitats than protected zones, which are mainly rocky (see Fig. S1). Given the patchy distribution of settlement habitat and species-specific patterns of larval retention or dispersal as well as behavioural differences among species, results may be interpreted with caution as it could be a consequence of multiple effects, thus confounding the outcomes of protection measures. This suggests that only MPAs harbouring a high diversity of habitats will be effective in protecting the initial stages of the life cycle of coastal fishes, by providing high-quality habitats for their settlement (Lindholm et al., 2001). Besides, the abundance of juvenile fish did not vary significantly between levels of protection. Instead, a high spatial variability in juvenile abundance and species richness among unprotected control zones within locations was observed for demersal species and all species together. The heterogeneous distribution and low densities of many species may have contributed to reducing the power of statistical comparisons of individual species at the smaller spatial scale (∼3 km) studied (Chapman and Kramer, 1999). It is known that variation in recruitment, survival, competition, and habitat structure all result in high spatial variability in fish density (Williams, 1991; García-Charton et al., 2004). For instance, studies which investigated the effect of protection on juveniles of Diplodus species have found high spatial and temporal variability at both small and large scales, but no differences in settlement intensity and mortality rates inside MPAs compared to unprotected areas were found (Macpherson et al., 1997; Vigliola et al., 1998). Juvenile habitat-mediated survival can be important for shaping fish community structure (Johnson, 2007; Félix-Hackradt et al., 2013b). Much of the post-settlement mortality is attributed to predation (Doherty et al., 2004) and may be strongly affected by local habitat features such as structural complexity (Johnson, 2007). In that sense, MPAs could not only enhance juvenile mortality by the higher abundance of predators but also increase their survival probability by providing high-quality habitat (D'Alessandro et al., 2007), however further studies are needed to solve this question.

Acknowledgements Authors want to thank the Environmental Council of Andalucía for authorizing the research campaigns inside marine reserves of Cabo de Gata Natural Park (Authorization no 201110100000325, Junta de Andalucía). Also, we would like to attest our gratitude to Alejandra Irasema and Ramón A. Hernández for field assistance and sample sorting. Authors FCFH and CWH wish to thank the Ministry of Education (CAPES, Process no IBEX 0123-9/09) and Ministry of Science and Technology of Brazil (CNPq, Process no 200630/2008-3), respectively, for their personal economic support. This study was financed in part by projects AECID —PCI A/024758/09 and A/030397/10. Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx. doi.org/10.1016/j.marenvres.2018.06.012. References Abesamis, R.A., Russ, G.R., 2005. Density-dependent spillover from a marine reserve: long-term evidence. Ecol. Appl. 15, 1798–1812. Almany, G.R., Hamilton, R.J., Bode, M., Matawai, M., Potuku, T., Saenz-Agudelo, P., Planes, S., Berumen, M.L., Rhodes, K.L., Thorrold, S.R., Russ, G.R., Jones, G.P., 2013. Dispersal of grouper larvae drives local resource sharing in a coral reef fishery. Curr. Biol. 23, 626–630. Amargós, P.F., Sanson, G.G., Castillo, A.J., Fernandez, A.Z., Blanco, F.M., Red, W.A., 2010. An experiment of fish spillover from a marine reserve in Cuba. Environ. Biol. Fish. 87, 363–372. Arias, A.M., Drake, P., 1990. Estados juveniles de la ictiofauna en los caños de las salinas de la Bahía de Cádiz. Consejería de Gobernación, Junta de Andalucía, Sevilla. Bianco, Lo, 1931. Fauna e flora del golfo di Napoli. 38 monografía: Uova, larve e staidi giovanili di teleostei. Stazione Zoologica di Napoli, Napoli. Calò, A., Lett, C., Mourre, B., Pérez-Ruzafa, A., García-Charton, J.A., 2017. Use of Lagrangian simulations to hindcast the geographical position of propagule release zones in a Mediterranean coastal fish. Mar. Environ. Res. 134, 16–27. Chapman, M.R., Kramer, D.L., 1999. Gradients in coral reef fish density and size across the Barbados Marine Reserve boundary: effects of reserve protection and habitat characteristics. Mar. Ecol. Prog. Ser. 181, 81–96. Claudet, J., Osenberg, C.W., Domenici, P., Badalamenti, F., Milazzo, M., Falcón, J.M., Bertocci, I., Benedetti-Cecchi, L., García-Charton, J.A., Goñi, R., Borg, J.A., Forcada, A., Lucia, G.A., Pérez-Ruzafa, A., Afonso, P., Brito, A., Guala, I., Le Diréach, L., Sanchez-Jerez, P., Somerfield, P.J., Planes, S., 2010. Marine reserves: fish life history and ecological traits matter. Ecol. Appl. 20, 830–839.

5. Conclusions In this work, we assessed the effects of a network of no-take areas on different stages of the life cycle of Mediterranean reef fish – post-larvae, juveniles and adults - by using an asymmetrical beyond-ACI approach. 9

Marine Environmental Research xxx (xxxx) xxx–xxx

F.C. Félix-Hackradt et al.

and around marine protected areas in the western Mediterranean Sea. PLoS One 9 (6), e98206. Harmelin, J.G., 1987. Structure and variability of the ichthyofauna in a Mediterranean protected rocky area (National park of port-cros, France). Mar. Ecol. 8, 263–284. Harmelin-Vivien, M.L., Harmelin, J.G., Chauvet, C., Duval, C., Galzin, R., Lejeune, P., Barnabe, G., Blanc, F., Chevalier, R., Declerc, J., Laserre, G., 1985. Evaluation des peuplements et populations de Poissons. Méthodes et problémes. Rev. Ecol. Terre Vie 40, 467–539. Harmelin-Vivien, M., Le Diréach, L., Bayle-Sempere, J., Charbonnel, E., García-Charton, J.A., Ody, D., Pérez-Ruzafa, Á., Reñones, O., Sánchez-Jerez, P., Valle, C., 2008. Gradients of abundance and biomass across reserve boundaries in six Mediterranean marine protected areas: evidence of fish spillover? Biol. Conserv. 141, 1829–1839. Harrison, H.B., Williamson, D.H., Evans, R.D., Almany, G.R., Thorrold, S.R., Russ, G.R., Feldheim, K.A., van Herwerden, L., Planes, S., Srinivasan, M., Berumen, M.L., Jones, G.P., 2012. Larval export from marine reserves and the recruitment benefit for fish and fisheries. Curr. Biol. 22, 1023–1028. Hastie, T., Tibshirani, R., 1990. Generalized Additive Models. Chapman and Hall, London. Johnson, D.W., 2007. Habitat complexity modifies post-settlement mortality and recruitment dynamics of a marine fish. Ecology 88, 1716–1725. Jones, G.P., Almany, G.R., Russ, G.R., Sale, P.F., Steneck, R.S., van Oppen, M.J.H., Willis, B.L., 2009. Larval retention and connectivity among populations of corals and reef fishes: history, advances and challenges. Coral Reefs 28, 307–325. Leis, J.M., Sweatman, H.P.A., Reader, S.E., 1996. What the pelagic stages of coral reef fishes are doing out in blue water: daytime field observations of larval behavioural capabilities. Mar. Freshw. Res. 47, 401–411. Lindholm, J.B., Auster, P.J., Ruth, M., Kaufman, L., 2001. Modelling the effects of fishing and implications for the design of marine protected areas: juvenile fish responses to variations in Seafloor habitat. Conserv. Biol. 15, 424–437. Luque, A.A., Templado, J., 2004. Praderas y bosques marinos de Andalucía. Consejería de Medio Ambiente, Junta de Andalucía, Sevilla. Macpherson, E., Biagi, F., Francour, P., Garcia-Rubies, A., Harmelin, J., Harmelin-Vivien, M., Jouvenel, T., Planes, S., Vigliola, L., Tunesi, L., 1997. Mortality of juvenile fishes of the genus Diplodus in protected and unprotected areas in the western Mediterranean Sea. Mar. Ecol. Prog. Ser. 160, 135–147. Moreno, D., 2003. El espectacular patrimonio natural litoral y sumergido de Cabo de Gata-Níjar (España). In: Moreno, D., Frías, A. (Eds.), Actas de las I Jornadas sobre Reservas Marinas y I Reunión de la Red Iberoamericana de Reservas Marinas (RIRM). Cabo de Gata-Níjar 17-23 de septiembre de 2001, pp. 25–46. Nagelkerken, I., van der Velde, G., 2002. Do non-estuarine mangroves harbour higher densities of juvenile fish than adjacent shallow-water and coral reef habitats in Curaçao (Netherlands Antilles)? Mar. Ecol. Prog. Ser. 245, 191–204. Pérez-Ruzafa, A., Martín, E., Marcos, C., Zamarro, J., Stobart, B., Harmelin-Vivien, M., Polti, S., Planes, S., García-Charton, J.A., González-Wangüemert, M., 2008. Modelling spatial and temporal scales for spill-over and biomass exportation from MPAs and their potential for fisheries enhancement. J. Nat. Conserv. 16, 234–255. Pérez-Ruzafa, A., Marcos, C., García-Charton, J.A., 2017. North East vs. Mediterranean marine protected areas as fisheries management tool. Front. Mar. Sci. 4, 245. Planes, S., Galzin, R., García-Rubies, A., Goñi, R., Harmelin, J.G., Le Diréach, L., Lenfant, P., Quetglas, A., 2000. Effects of marine protected areas on recruitment processes with special reference to Mediterranean littoral ecosystems. Environ. Conserv. 27, 126–143. Ré, P., 1999. Ictioplânton estuarino da peninsula Ibérica. Guia de identificação de ovos e estados larvares planctônicos. Universidade de Lisboa, Lisboa. Roberts, C.M., Bohnsack, J.A., Gell, F., Hawkins, J.P., Goodridge, R., 2001. Effects of marine reserves on adjacent fisheries. Science 294, 1920–1923. Russ, G.R., Alcala, A.C., 2004. Marine reserves: long-term protection is required for full recovery of predatory fish populations. Oecologia 138, 622–627. Sale, P.F., Cowen, R.K., Danilowicz, B.S., Jones, G.P., Kritzer, J.P., Lindeman, K.C., Planes, S., Polunin, N.V., Russ, G.R., Sadovy, Y.J., Steneck, R.S., 2005. Critical science gaps impede use of no-take fishery reserves. Trends Ecol. Evol. 20, 74–80. Sponaugle, S., Cowen, R.K., 1996. Nearshore patterns of coral reef fish larval supply to Barbados, West Indies. Mar. Ecol. Prog. Ser. 133, 13–28. Syms, C., Carr, M.H., 2001. Marine Protected Areas: Evaluating MPA Effectiveness in an Uncertain World. Prepared for a workshop sponsored by the Commission for Environmental Cooperation, held May 2001 in Monterey, California. Underwood, A.J., 1992. Beyond BACI: the detection of environmental impacts on populations in the real, but variable, world. J. Exp. Mar. Biol. Ecol. 161, 145–178. Underwood, A.J., 1993. The mechanics of spatially replicated sampling programmes to detect environmental impacts in a variable world. Aust. J. Ecol. 18, 99–116. Underwood, A.J., 1997. Experiments in Ecology: Their Logical Design and Interpretation Using Analysis of Variance. Cambridge University Press, Cambridge. Vigliola, L., Harmelin-Vivien, M.L., Biagi, F., Galzin, R., Garcia-Rubies, A., Harmelin, J.G., Jouvenel, J.Y., Le Direach-Boursier, L., Macpherson, E., Tunesi, L., 1998. Spatial and temporal patterns of settlement among sparid fishes of the genus Diplodus in the northwestern Mediterranean. Mar. Ecol. Prog. Ser. 168, 45–56. Williams, D.McB., 1991. Patterns and processes in the distribution of coral reef fishes. In: Sale, P.F. (Ed.), The Ecology of Fishes on Coral Reefs. Academic Press, San Diego, pp. 437–474.

Crec’hriou, R., Alemany, F., Roussel, E., Chassanite, A., Marinaro, J., Mader, J., Rochel, E., Planes, S., 2010. Fisheries replenishment of early life taxa: potential export of fish eggs and larvae from a temperate marine protected area. Fish. Oceanogr. 19, 135–150. Cudney-Bueno, R., Lavın, M.F., Marinone, S.G., Raimondi, P.T., Shaw, W.W., 2009. Rapid effects of marine reserves via larval dispersal. PLoS One (4), e4140. Dixon, P.A., 1999. Episodic fluctuations in larval supply. Science 283, 1528–1530. Doherty, P.J., Dufour, V., Galzin, R., Hixon, M.A., Meekan, M.G., Planes, S., 2004. High mortality during settlement: a population bottleneck for a tropical surgeonfish. Ecology 85, 2422–2428. D'Alessandro, E., Sponaugle, S., Lee, T., 2007. Patterns and processes of larval fish supply to the coral reefs of the upper Florida Keys. Mar. Ecol. Prog. Ser. 331, 85–100. Félix-Hackradt, F.C., Hackradt, C.W., Treviño-Otón, J., Segovia-Viadero, M., PérezRuzafa, A., García-Charton, J.A., 2013a. Environmental determinants on fish postlarval distribution in coastal areas of south-western Mediterranean Sea. Estuar. Coast Shelf Sci. 129, 59–72. Félix-Hackradt, F.C., Hackradt, C.W., Treviño-Otón, J., Pérez-Ruzafa, Á., García-Charton, J.A., 2013b. Temporal patterns of settlement, recruitment and post-settlement losses in a rocky reef fish assemblage in the South-Western Mediterranean Sea. Mar. Biol. 160, 2337–2352. Félix-Hackradt, F.C., Hackradt, C.W., Treviño-Otón, J., Pérez-Ruzafa, Á., García-Charton, J.A., 2014. Habitat use and ontogenetic shifts of fish life stages at rocky reefs in South-western Mediterranean Sea. J. Sea Res. 88, 67–77. Fenberg, P.B., Caselle, J., Claudet, J., Clemence, M., Gaines, S., García-Charton, J.A., Gonçalves, E., Grorud-Colvert, K., Guidetti, P., Jenkins, S., Jones, P.J.S., Lester, S., McAllen, R., Moland, E., Planes, S., Sørensen, T.K., 2012. The science of European marine reserves: status, efficacy, and future needs. Mar. Pol. 36, 1012–1021. Forcada, A., Valle, C., Bonhomme, P., Criquet, G., Cadiou, G., Lenfant, P., Sánchez-Lizaso, J.L., 2009. Effects of habitat on spillover from marine protected areas to artisanal fisheries. Mar. Ecol. Prog. Ser. 379, 197–211. Francour, P., 1997. Fish assemblages of Posidonia oceanica beds at port-cros (France, NW Mediterranean): assessment of composition and long-term fluctuations by visual census. Mar. Ecol. 18, 157–173. Froese, R., Pauly, D. (Eds.), 2011. FishBase. World Wide Web electronic publication version (08/2011). www.fishbase.org. García-Charton, J.A., Pérez-Ruzafa, A., 2001. Spatial pattern and the habitat structure of a Mediterranean rocky reef fish local assemblage. Mar. Biol. 138, 917–934. García-Charton, J.A., Williams, I., Pérez-Ruzafa, A., Chemello, R., Marcos, C., Kitsos, M.S., Koukouras, A., Riggio, S., 2000. Evaluating the ecological effects of Mediterranean marine reserves: habitat, scale and the natural variability of ecosystems. Environ. Conserv. 27, 159–178. García-Charton, J.A., Pérez-Ruzafa, A., Sánchez-Jerez, P., Bayle-Sempere, J.T., Reñones, O., Moreno, D., 2004. Multi-scale spatial heterogeneity, habitat structure, and the effect of marine reserves on Western Mediterranean rocky reef fish assemblages. Mar. Biol. 144, 161–182. García-Charton, J.A., Pérez-Ruzafa, Á., Marcos, C., Claudet, J., Badalamenti, F., Benedetti-Cecchi, L., Falcón, J.M., Milazzo, M., Schembri, P.J., Stobart, B., Vandeperre, F., Brito, A., Chemello, R., Dimech, M., Domenici, P., Guala, I., Le Diréach, L., Maggi, E., Planes, S., 2008. Effectiveness of European AtlantoMediterranean MPAs: do they accomplish the expected effects on populations, communities and ecosystems? J. Nat. Conserv. 16, 193–221. García-Rubies, A., Mcpherson, E., 1995. Substrate use and temporal pattern of recruitment in juvenile fishes of the Mediterranean littoral. Mar. Biol. 124, 35–42. Giakoumi, S., Scianna, C., Plass-Johnson, J., Micheli, F., Grorud-Colvert, K., Thiriet, P., Claudet, J., Di Carlo, G., Di Franco, A., Gaines, S.D., García-Charton, J.A., Lubchenco, J., Reimer, J., Sala, E., Guidetti, P., 2017. Ecological effects of full and partial protection in the crowded Mediterranean Sea: a regional meta-analysis. Sci. Rep. 7, 8940. Glasby, T.M., 1997. Analysing data from post-impact studies using asymmetrical analyses of variance: a case of study of epibiota on marinas. Aust. J. Ecol. 22, 448–459. Goñi, R., Adlerstein, S., Alvarez-Berastegui, D., Forcada, A., Reñones, O., Críquet, G., Polti, S., Cadiou, G., Valle, C., Lenfant, P., Bonhomme, P., Pérez-Ruzafa, Á., SánchezLizaso, J.L., García-Charton, J.A., Bernard, G., Stelzenmüller, V., Planes, S., 2008. Evidence of biomass export from six Western Mediterranean marine protected areas as measured from artisanal fisheries. Mar. Ecol. Prog. Ser. 366, 159–174. Goñi, R., Hilborn, R., Díaz, D., Mallol, S., Adlerstein, S., 2010. Net contribution of spillover from a marine reserve to fishery catches. Mar. Ecol. Prog. Ser. 400, 233–243. Guidetti, P., 2002. The importance of experimental design in detecting the effects of protection measures on fish in Mediterranean MPAs. Aquat. Conserv. 12, 619–634. Guidetti, P., 2006. Marine reserves re-establish lost predatory interactions and cause community changes in rocky reefs. Ecol. Appl. 16, 963–976. Guidetti, P., Milazzo, M., Bussotti, S., Molinari, A., Murenu, M., Pais, A., Spano, N., Balzano, R., Agardy, T., Boero, F., Carrada, G., Cattaneo-Vietti, R., Cau, A., Chemello, R., Greco, S., Manganaro, A., Sciara, G.N., Russo, G.F., Tunesi, L., 2008. Italian marine reserve effectiveness: does enforcement matter? Biol. Conserv. 141, 699–709. Hackradt, C.W., García-Charton, J.A., Harmelin-Vivien, M., Pérez-Ruzafa, A., Le Diréach, L., Bayle-Sempere, J.T., Charbonnel, E., Ody, D., Reñones, O., Sánchez-Jerez, P., Valle, C., 2014. Response of rocky reef top predators (Serranidae: Epinephelinae) in

10