For Peer Review 19 20 03/11/2017 21 22 23 Dear Prof

Journal of Biogeography

Multipl e stressors facilitate the spread of a non -indigenous bivalve in the Mediterranean Sea

Journal:For Journal Peer of Biogeography Review Manuscript ID JBI-16-0620.R2

Manuscript Type: Research Paper

Date Submitted by the Author: n/a

Complete List of Authors: Sarà, Gianluca; University of Palermo, Earth and Marine Science Porporato, Erika; Università di Palermo, Dipartimento di Scienze della Terra e del Mare; Ca' Foscari University of Venice, Department of Environmental Sciences Mangano, Maria; Università di Palermo, Dipartimento di Scienze della Terra e del Mare Mieszkowska, Nova; Marine Biological Association of the U.K., Biodiversity and Ecology; University of Liverpool, Department of Earth Ocean and Ecological Sciences

Climate change, Habitat fragmentation, Maxent, Non-indigenous species, Key Words: Regional Climate Model, Sensitivity analysis, Species Distribution Model

Page 1 of 81 Journal of Biogeography

1 2 3 4 5 Patron: HRH The Prince Philip, Duke of Edinburgh The 6 Laboratory 7 President: Professor Sir John Beddington, CMG FRS Citadel Hill 8 Director: Professor Colin Brownlee Plymouth 9 PL1 2PB 10 United 11 Kingdom tel: +44 12 (0)1752 633207 13 fax: +44 14 (0)1752 633102 email: 15 [email protected] 16 www.mba. 17 ac.uk 18 For Peer Review 19 20 03/11/2017 21 22 23 Dear Prof. Ladle 24 25 Manuscript resubmission JBI-16-0620 Multiple stressors facilitate the spread of a 26 non-indigenous bivalve in the Mediterranean Sea. 27 28 Please extend our thanks to the reviewers for their detailed comments on our 29 manuscript. We have addressed all of their suggested edits that have helped to improve 30 the structure and text of the revised manuscript. Our responses are written below each 31 individual comment in the Editor’s comments to the author section below. 32 33 34 35 Yours sincerely 36 37 38 39 40 41 Dr Nova Mieszkowska 42 43 44 EDITOR'S COMMENTS TO AUTHOR 45 46 Editor: Chapman, Daniel 47 Comments to the Author: 48 Many thanks for your revised manuscript, which has addressed most of the previous 49 reviewer comments. In your revision please address the outstanding comment, including 50 51 visualising the responses and pairwise interactions (this could probably go in the 52 Appendices). 53 I also suggest improving the presentation by replacing Table 1 (which is very valuable 54 but overly detailed) with a 'heat map' figure showing the direction and magnitude of 55 56 57 58 59 60

Journal of Biogeography Page 2 of 81

1 2 3 changes with a colour scale. You could keep the layout of the table, but just colour the 4 cells. 5 6 To make space for this, I suggest plotting maps for only one of either 2030 or 2050 7 (Figs. 3-6) and putting the other year's maps in the Appendices. 8 Reply: Many thanks, we agreed and worked on it accordingly. Pairwise interactions are 9 now reported in Appendix 2 Figs. A4, A5, Table 1 is with colours, we preferred to leave 10 11 in the MS both 2030 and 2050 maps. 12 13 A minor comment from me - please consider changing the phrase 'a latere', in the 14 15 Abstract. To my knowledge it is not standard English. 16 Reply: Many thanks, we agreed and changed it accordingly. 17 18 REVIEWER COMMENTSFor TO AUTHORPeer Review 19 20 Referee 1 21 Second review of the manuscript by Sarà entitled “Multiple stressors facilitate the spread 22 of a non-indigenous bivalve in the Mediterranean Sea”. The manuscript is generally 23 improved, although some issues were not satisfactorily tackled. It would be far easier to 24 25 map author changes if they provide the lines where the modifications were performed. 26 27 Main comments 28 29 Introduction 30 L140 This sentence is still confusing. Why the native species increase with the invasive 31 species presence? Through which mechanism? Is it facilitation? Habitat improvement? 32 Please, clarify. 33 34 Reply: Yes, sorry it wasn’t clear. The sentence is referring to information reported by 35 other Authors in literature, we now added the specific references to this sentence and 36 we indicated the presence of a more accurate list in the table presented in Appendix 1. 37 38 The cited Authors suppose an increase in local species diversity, therefore we were 39 moved from your same curiosity, and also considering the scanty amount of only 40 qualitative papers already published focusing on this aspect, we are now trying to test it 41 in the field (data analyses are in progress). 42 43 44 Methods 45 L 257-260 Regarding comments on the first version, I suggest to add the spatial 46 resolution (1km) used to the main text and clarify what is exactly “threshold equal 47 48 training sensitivity and specificity” 49 Reply: Agree and added ( L 245-246 and L 256-265 ). 50 51 Results 52 Comment. I miss a section in the results section where the authors explain the 53 54 relationship of each predictive variable with HIS (positive, negative relationship, etc., 55 lineal/unimodal, etc.). I recommend plotting the response of HSI to pairwise stressor 56 combinations to show potential interactions. Currently, it is not possible to infer any of 57 58 them through the information provided. 59 60 Registered Charity No. 1155893 Incorporated by Royal Charter Page 3 of 81 Journal of Biogeography

1 2 3 Feld et al. (2016) provides a comprehensive description of how to analyse and 4 communicate multi-stressor interactions. 5 6 L345-353 should be appropriately reviewed 7 Feld, C. K., Segurado, P., & Gutiérrez-Cánovas, C. (2016). Analysing the impact of 8 multiple stressors in aquatic biomonitoring data: A Œcookbook¹with applications in R. 9 Science of the Total Environment. 10 11 12 Reply: We agree with the referee and we have added the Figure A4 and A5 in the 13 Supplementary materials Appendix 2 analysing the HSI 2010 relationships with the 14 15 environmental variables. First, we have added Figure A4, a graph produced by Maxent, 16 representing the response of B. pharaonis to each variable considered. Moreover, we 17 have added the percentage contribution of each variable within each graph. Second, 18 following the methodFor described Peerin Feld et al., 2016Review, we have added Figure A5 in order to 19 20 show potential interactions, representing the response of HSI to pairwise stressor 21 combinations ( L 294-296 and L 345-347 in the main text). Regarding the L345-353, we 22 agreed with the referee and we have changed it accordingly. 23 24 Comment. The authors may consider showing the marginal responses of B. pharaonis 25 26 to each of the environmental predictors in the model for 2010 (i.e. the modeled 27 probability of occurrence versus a range of values for a given environmental variable, 28 keeping all other variables constant, see Azzurro et al., 2013 Biol. Invasions for an 29 example). this would facilitate the identification of potential nonlinear responses (e.g. to 30 31 productivity, see below). 32 Reply: We agreed and we have added the partial dependence curves of the 2010 model 33 in the Supplementary materials Appendix 2, Figure A4. The importance of each variable 34 35 was reported within the graph and hence we have removed the Table A2. 36 37 Minor comments 38 L84: Do the authors mean “multiple anthropogenic factors”? 39 40 Reply: Agreed, checked and changed accordingly. 41 42 L277: I guess HSI was derived from the Maxent model. Authors should be more explicit 43 here and help reader to follow their methodology. 44 45 Reply: Agree and added ( L 253-256 ) 46 47 L308-315: These lines seem to be more appropriate for the discussion. 48 49 Reply: Agreed, checked and changed accordingly. 50 51 L317 It could be more informative to say: “Forecasted habitat suitability”. 52 Reply: Agreed, checked and changed accordingly. 53 54 55 L334 change “HIS” by “HIS”. 56 Reply: Agreed, checked and changed accordingly. 57 58 59 60 Registered Charity No. 1155893 Incorporated by Royal Charter Journal of Biogeography Page 4 of 81

1 2 3 A final tip from the Editorial Office: having dealt with all the comments above, please 4 work systematically through the attached author checklist prior to re-submitting your 5 6 paper. Failure to do so is likely to result in the paper being returned to your author 7 centre. 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Registered Charity No. 1155893 Incorporated by Royal Charter Page 5 of 81 Journal of Biogeography

1 2 3 1 Article Type: Original Article 4 5 2 6 7 3 Running head: Multiple stressor impacts on non-indigenous species 8 9 4 10 11 5 Multiple stressors facilitate the spread of a non-indigenous bivalve in the Mediterranean 12 13 14 6 Sea 15 16 7 17 18 8 Gianluca Sarà 1, ErikaFor M.D. Porporato Peer2, M. Cristina Review Mangano 1,3 and Nova Mieszkowska 4,5* 19 20 9 21 22 1 23 10 Dipartimento di Scienze della Terra e del Mare, Università di Palermo, V.le delle Scienze - 24 25 11 Ed. 16 - 90128 Palermo, Italy 26 27 12 28 29 13 2Department of Environmental Sciences, Informatics and Statistics, Ca' Foscari University of 30 31 14 Venice, Via Torino 155 - 30170 Venezia, Mestre, Italy 32 33 34 15 35 3 36 16 Consorzio Nazionale Interuniversitario per le Science del Mare (CoNISMa), Piazzale 37 38 17 Flaminio 9, 00196 - Roma, Italy 39 40 18 41 42 4 43 19 The Marine Biological Association of the UK, Citadel Hill, Plymouth, PL1 2PB, UK. 44 45 20 [email protected] Tel: +44 1752633333. *To whom correspondence should be addressed. 46 47 21 48 49 22 5Department of Earth, Ocean and Ecological Sciences, School of Environmental Sciences, 50 51 23 University of Liverpool, Nicholson Building, Brownlow Street, Liverpool, L69 3GP, UK 52 53 54 24 55 56 25 Word count abstract: 381 57 58 26 Word count main body of text: 7999 59 60 Journal of Biogeography Page 6 of 81

1 2 3 27 Abstract 4 5 28 Aim: The introduction of non-indigenous species (NIS) via man-made corridors connecting 6 7 29 previously disparate oceanic regions is increasing globally. The environmental and 8 9 10 30 anthropogenic factors facilitating invasion dynamics and their interactions are, however, still 11 12 31 largely unknown. This study compiles and inputs available data for the NIS bivalve 13 14 32 Brachidontes pharaonis across the invaded biogeographic range in the Mediterranean basin 15 16 33 into a species distribution model to predict future spread across the Mediterranean Sea under a 17 18 34 range of marine scenarios.For Peer Review 19 20 21 35 Location: Mediterranean Sea. 22 23 36 Methods: A systematic review produced the largest presence database ever assembled to 24 25 37 inform the selection of biological, chemical and physical factors linked to the spread of the 26 27 38 NIS bivalve B. pharaonis through the Suez Canal into the Mediterranean basin. After 28 29 30 39 comparing methodological approaches we elected to carry out a sensitivity analysis to 31 32 40 simulate current and future trophic and salinity scenarios. A species distribution model was 33 34 41 then run to determine key drivers of invasion, quantify interactive impacts arising from a 35 36 42 range of trophic states, salinity conditions and climatic scenarios, and forecast future 37 38 43 trajectories for the spread of NIS into new regions under multiple-parameter scenarios, based 39 40 41 44 on the main factors identified from the systematic review. 42 43 45 Results: Impacts on invasion trajectory arising from climate change and interactions with 44 45 46 increasing salinity from the new opening of the Suez Canal were the primary drivers of 46 47 47 expansion across the basin, the effects of which were further enhanced by eutrophication. 48 49 50 48 Predictions of the current distribution were most accurate when multiple stressors were used 51 52 49 to drive the model. An Habitat Suitability Index developed at a subcontinental scale from 53 54 50 model outputs identified novel favourable conditions for future colonization at specific 55 56 51 locations under 2030 and 2050 climatic scenarios. 57 58 59 60 Page 7 of 81 Journal of Biogeography

1 2 3 52 Main Conclusions: Future expansion of B. pharaonis will be enhanced by climate-facilitated 4 5 53 increased sea temperature, interacting with increasing pressures from salinity and 6 7 54 eutrophication. The spatially ‐explicit risk output maps of invasions, which also function as 8 9 10 55 risk/pest maps, represent a powerful visual product for use in communication of the spread of 11 12 56 NIS and decision-support tools for scientists and policymakers. The observed distribution 13 14 57 pattern and driving processes, as well the suggested approach, can be applied to other NIS 15 16 58 species and regions by providing novel forecasts of species occurrences under future multiple 17 18 59 stressor scenarios, andFor the location Peer of suitable reReviewcipient habitats with respect to anthropogenic 19 20 21 60 and environmental parameters. 22 23 61 24 25 62 Keywords : Climate change, Habitat fragmentation, Maxent, Non-indigenous species, 26 27 63 Regional Climate Model, Sensitivity analysis, Species Distribution Model 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 8 of 81

1 2 3 64 Introduction 4 5 65 Climate change is driving poleward range shifts in marine species across a wide range of 6 7 66 benthic taxa (Mieszkowska et al ., 2006; Helmuth et al ., 2006; Lima et al ., 2007; Poloczanska 8 9 10 67 et al ., 2013; Mieszkowska et al ., 2014) and is thought to be exacerbating the invasion success 11 12 68 of Non-Indigenous Species (NIS) (Pederson et al ., 2011). Climate change exerts a significant 13 14 69 and growing impact on global biodiversity and potential ‘globalization’ of marine fauna and 15 16 70 flora resulting in biodiversity loss, alteration of ecosystem function and degradation of 17 18 71 ecosystem servicesFor (Walther etPeer al ., 2009; Bradley Review et al., 2010; Gallien et al ., 2010; Sorte et 19 20 21 72 al ., 2010; Vilà et al ., 2011; Marzloff et al., 2016; Pecl et al., 2017). The majority of marine 22 23 73 NIS that have successfully colonized new regions beyond those accessible via natural modes 24 25 74 of dispersal are thought to proliferate in their introduced ranges due to their greater tolerances 26 27 75 for one or more environmental parameters when compared to native species present within 28 29 30 76 the invaded community (Sorte et al ., 2010; Lenz et al ., 2011; Zerebeki & Sorte 2011). 31 32 77 Advances in our ability to track biogeographic range shifts and invasions have increased 33 34 78 awareness of the enormous complexity of environmental and anthropogenic processes 35 36 79 involved in biological invasions in a changing world. Beyond a simplistic, unilateral response 37 38 80 to warming of the global oceans, scientists must seek new, integrated approaches to predict 39 40 41 81 future biogeographic shifts of NIS (Burrows et al ., 2014). The development of predictive 42 43 82 models that can be run for a range of multiple anthropogenic factors (hereafter termed 44 45 83 stressors; sensu Gunderson et al. , 2016) scenarios will increase the accuracy of the 46 47 84 quantitative forecasts for ecological and economic cost of invasion, and provide useful 48 49 50 85 guidance for planning management or control strategies that form part of the mitigation and 51 52 86 adaptation management processes (Ritcher et al ., 2013; Hamaoui-Laguel et al ., 2015; 53 54 87 Chapman et al ., 2016). 55 56 57 58 59 60 Page 9 of 81 Journal of Biogeography

1 2 3 88 A specific mode of invasion is that observed for ‘Lessepsian’ invasive species, those 4 5 89 utilizing the Suez Canal (a manmade corridor between previously unconnected seas) as a 6 7 90 pathway to colonize new environments far removed from their origin. This manmade 8 9 10 91 construction connects the Indo-Pacific and the Red Sea with the Mediterranean Sea and is 11 12 92 termed the “Eastern door”, through which NIS invade the Mediterranean basin by planktonic 13 14 93 larvae in a “stepping stone” fashion (as traditionally assumed) and via shipping vectors of hull 15 16 94 fouling and ballast water transport (Galil et al ., 2015). This situation is by no means unique, 17 18 95 with manmade corridorsFor providing Peer connective Review pathways for marine invasions around the 19 20 21 96 world, including the Panama Canal, White Sea – Baltic Sea Canal, Kiel Canal and Danube- 22 23 97 Black Sea Canal. Shipping and shipping-related constructions are thus contributing to the 24 25 98 movement of marine species around the world, shaping the origin, frequency and magnitude 26 27 99 of species movements by providing new introduction routes for Lessepsian invasions (Hulme 28 29 30 100 2009; Katsanevakis et al ., 2014; Ojaveer et al ., 2014). 31 32 101 The key to a successful invasion is the presence of suitable habitats ( sensu resistance 33 34 102 hypothesis - Ruiz et al ., 2000) with respect to physical, chemical and trophic conditions in 35 36 103 those areas where new NIS propagules arrive (Boudouresque et al ., 2004; Hulme et al ., 2008; 37 38 104 Galil 2009; Sarà et al ., 2013). Climatic and anthropogenic forcing of the marine environment, 39 40 41 105 coupled with an increase in shipping traffic from the Levantine Basin are thought to have 42 43 106 amplified both the Habitat Suitability (HS) and the propagule pressure for NIS within the 44 45 107 Mediterranean in recent years (Katsanevakis et al ., 2014), however, this has not been 46 47 108 quantitatively investigated for most invasive species recorded within the Mediterranean Sea. 48 49 50 109 Here we investigate how anthropogenically driven changes to the marine environment, 51 52 110 including those related to changes driven by manmade canals connecting separate seas, may 53 54 111 exacerbate the existing impacts of NIS on native species, communities and ecosystems, and 55 56 112 alter their trajectory of future spread. We use the recent expansion of the Suez Canal via a 57 58 59 60 Journal of Biogeography Page 10 of 81

1 2 3 113 second parallel seaway (officially inaugurated on the 06/08/2015; 4 5 114 http://newcanal.suezcanal.gov.eg) as a case study system to test the impacts of multiple 6 7 115 anthropogenic stressors on the invasion trajectory of a Lessepsian NIS within the 8 9 10 116 Mediterranean Sea, and address how this type of infrastructure can result in wider 11 12 117 implications for both Lessepsian and global species invasions. 13 14 118 The second Suez Canal waterway will have a large impact on the biological (e.g. the 15 16 119 increase of the propagule pressure for a wide variety of species), physical and chemical 17 18 120 characteristics of theFor Mediterranean Peer Sea. Biotic Review changes have already occurred as a result of 19 20 21 121 several previous enlargements of the existing canal, resulting in environmental changes 22

23 122 initiated in the Eastern basin and propagating across the entire basin (Katsanevakis et al ., 24 25 123 2014). Invasion dynamics are predicted to accelerate with exposure to these human- (e.g. 26 27 124 eutrophication; Nixon 2009) and climate-related factors (e.g. modification of temperature and 28 29 30 125 wind-driven hydrodynamics; Compton et al., 2010; Pachauri et al ., 2014; Adloff et al ., 2015) 31 32 126 which are already driving range shifts in the distributions of native species, altering 33 34 127 community structure, diversity and resilience, thus favouring biological invasions (Parmesan 35 36 128 & Yohe 2003). We predict that these drivers will interact with those deriving from the new 37 38 129 Suez Canal opening, exposing Mediterranean biodiversity to large modifications of chemical 39 40 41 130 and physical properties. 42 43 131 The focal species of this study is the Lessepsian mussel, Brachidontes pharaonis 44 45 132 (Bivalvia, Fischer 1870), a bivalve classified as a “pest model NIS” (Galil 2009). B. 46 47 133 pharaonis is widely reported to be invading the Mediterranean, however, knowledge of the 48 49 50 134 ecology and physiology of this species are lacking in comparison with other marine NIS. B. 51 52 135 pharaonis occurs on littoral rocky habitat. It has a planktonic larval phase, and a protracted, 53 54 136 year-round reproductive cycle. This species has a wide thermotolerance range (9-31 °C) and 55 56 137 can tolerate salinities from 35-53psu, traits typical of most Lessepsian NIS (Sarà et al ., 2008; 57 58 59 60 Page 11 of 81 Journal of Biogeography

1 2 3 138 Katsanevakis et al ., 2014; to see also references listed in Appendix 1). B. pharaonis exerts 4 5 139 strong local-scale effects on hard substrata biodiversity by creating biogenic habitat that 6 7 140 promotes local species richness, outcompeting native species for resources and space (Safriel 8 9 10 141 et al., 1980; 1988; Bonnici et al., 2012; to see also references listed in Appendix 1). The 11 12 142 biology and invasion ecology of this species is typical of NIS with respect to wide 13 14 143 ecophysiological tolerance ranges for environmental parameters including temperature, 15 16 144 salinity and water pH, making B. pharaonis a suitable model species with which to study how 17 18 145 human and climate-relatedFor factors Peer will drive Review biological invasions from the present day to 19 20 21 146 2050. 22 23 147 Whilst recent approaches based on mechanistic trait-based models ( e.g. Sarà et al., 24 25 148 2013) are able to reliably predict the current spatial distribution of NIS, they require huge 26 27 149 amounts of data in order to provide reliable predictions of NIS spread in the future when 28 29 30 150 assessing the effects of environmental change, including climate change. Unfortunately, the 31 32 151 investigation of impacts arising from multiple anthropogenic factors is still far from the 33 34 152 application’s range of mechanistic trait-based modelling. In contrast, to provide a valuable, 35 36 153 effective and immediate tool for decision making involved in NIS management, we employed 37 38 154 an integrated classical correlative approach to NIS modelling but bringing the novelty of the 39 40 41 155 interaction between multiple stressors (salinity and temperature as proxies of tropicalisation, 42 43 156 and eutrophication as a proxy of local urbanization) tested through a set of sensitivity 44 45 157 analyses. Thus we derived reliable and exploitable information to: i) generate risk maps of 46 47 158 future biological invasions ( sensu Hulme 2009) useful to feed strategic and tactical pest 48 49 50 159 management decisions; ii) forecast relevant outcomes to inform scientists and managers on 51 52 160 ecological and socio-economic potential impacts generated by ongoing invasions; iii) fulfil 53 54 161 emergent regulations, policy drivers and directives in the framework of European Parliament 55 56 162 and Council; and iv) provide strategies for managing NIS as part of a realistic, integrated, 57 58 59 60 Journal of Biogeography Page 12 of 81

1 2 3 163 ecosystem-based approach, which is a major challenge for the scientific community, 4 5 164 stakeholders and decision makers. 6 7 165 8 9 10 166 Materials and methods 11 12 167 Literature search 13 14 168 An extensive literature analysis was completed viz. a systematic review, validated as a 15 16 169 comprehensive, policy-neutral, transparent, reproducible, robust assessment and summary of 17 18 170 available evidenceFor to support Peer biodiversity conservaReviewtion and policy decision making on 19 20 21 171 environmental issues (Gurevitch & Hedges 2001; Bilotta et al ., 2014). The literature search 22 23 172 was designed to investigate the past and present distribution of B. pharaonis across the 24 25 173 Mediterranean basin and to identify factors potentially affecting its ability to colonise new 26 27 174 habitats. The search was carried out using prominent or substantial keywords forming a 28 29 30 175 simple search string (“ Brachidontes pharaonis ” AND “Mediterranean”). The search ranged 31 32 176 from the year 1900 to the present day and was restricted to the Mediterranean region. The 33 34 177 search string was entered into the following scientific computerised databases including: ISI 35 36 178 Web of Sciences, Scopus, BioOne, CAB Abstracts, Aquatic Sciences and Fisheries Abstracts 37 38 179 (since 1971), Directory of Open Access Journal and J-STOR. Additional general search 39 40 41 180 engines were used (Google and Google Scholar) limiting the search for appropriate data to the 42 43 181 Word, PDF and/or Excel documents and to the first 50 hits (CEE review guidelines, 2013; 44 45 182 Mangano & Sarà 2017a,b; Mangano et al., 2017). A hand search was also performed on the 46 47 183 bibliographies of relevant review articles to identify any additional references. Data on 48 49 50 184 presence records were searched for in specific database and information systems showing 51 52 185 current and past distribution maps ( e.g. shapefiles, polygons, points); Ocean Biogeographic 53 54 186 Information System [http://www.iobis.org/]; Global Biodiversity Information Facility 55 56 187 [http://www.gbif.org/]; AquaNIS [http://www.corpi.ku.lt/databases/index.php/aquanis]; 57 58 59 60 Page 13 of 81 Journal of Biogeography

1 2 3 188 DAISIE [http://www.europe-aliens.org], EASIN [http://easin.jrc.ec.europa.eu]; World 4 5 189 Register of Introduced Marine Species [http://www.marinespecies.org/introduced/aphia]. 6 7 190 Authors of relevant articles not readily available on-line were personally contacted via paper 8 9 10 191 request, providing any missing data and unpublished material or further recommendations 11 12 192 (search ended at 21/08/2015). Hits generated from the search were collated in a database, 13 14 193 examined for relevance and critically appraised (Appendix 1). Data and evidence extraction 15 16 194 from peer-review and grey literature were organised and synthesised according to specific 17 18 195 criteria, e.g. geographicFor area, habitatPeer preferences, Review associated species, with a complete list of 19 20 21 196 the collated studies for each Mediterranean sector (Appendix 1). All quantitative information 22 23 197 from each paper were extracted in order to draw up the most correct and precise picture of the 24 25 198 ecological status of B. pharaonis in the Mediterranean basin and on the environmental 26 27 199 parameters influencing its distribution. 28 29 30 200 31 32 201 Maxent modelling 33 34 202 Presence data that represented the current known distributions for B. pharaonis were 35 36 203 extracted through the systematic review process and used to build the occurrence dataset for 37 38 204 B. pharaonis within the Mediterranean basin. Data and evidence from the review process 39 40 41 205 were used to populate a presence-only Maximum entropy (Maxent) species-distribution 42 43 206 model (SDM) to forecast habitat suitability (Phillips et al., 2006) for B. pharaonis. Maxent 44 45 207 represents the most effective correlative modelling approach in context of SDM (Guisan & 46 47 208 Zimmermann 2000), providing an important ecological tool for the prediction of NIS 48 49 50 209 geographical distribution within the context of climate change (Elith & Leathwick 2009; 51 52 210 Walther et al ., 2009). These scenarios were used to forecast how the potential habitat 53 54 211 suitability for B. pharaonis will vary across three trophic statuses (oligotrophication, no- 55 56 212 change and eutrophication) in combination with 11 salinity conditions across a gradient from 57 58 59 60 Journal of Biogeography Page 14 of 81

1 2 3 213 decreasing (from -0.1 to -0.5 psu), no-change to increasing salinity (from +0.1 to +0.5 psu) 4 5 214 with respect to two climate scenarios (2030 and 2050; Med-Cordex Regional Climate Model, 6 7 215 Representative Concentration Pathways 4.5). Trophic and salinity conditions have been 8 9 10 216 simulated through sensitivity analyses, a very useful alternative tool to explore the robustness 11 12 217 of models’ outputs within an uncertain context (Payne et al ., 2015). 13 14 218 The data collated from the systematic review were used to determine the geographical 15 16 219 presence of B. pharaonis across the Mediterranean basin, to input into the Maxent modelling 17 18 220 processes. The currentFor distribution Peer of B. pharaonis Review in the invasive range was modelled 19 20 21 221 employing 98 presence records (Fig. 1) and 8 physical, chemical and biological variables 22 23 222 selected after the collinearity tests (Variance Inflation Factor – VIF, Table A2 in Appendix 2). 24 25 223 Data cleaning was required to remove data duplicates and incorrect records. Sampling 26 27 224 bias is a well-known factor influencing SDMs (Phillips et al., 2009), leading to spatial 28 29 30 225 autocorrelation of records and artificial spatial clusters of observations, violating the 31 32 226 assumption of independence (Dormann et al., 2007). This bias can be avoided by sampling 33 34 227 one point per cluster in the environmental space, which was carried out using the software 35 36 228 OccurrenceThinner (Verbruggen 2012). OccurrenceThinner identifies areas of high record 37 38 229 density based on the species occurrence records and a two-dimensional kernel surface grid file 39 40 41 230 representing the region of study to filter occurrence records (Verbruggen 2012; Verbruggen et 42 43 231 al., 2013). Ten pseudo-replicate datasets produced through OccurrenceThinner, each with 44 45 232 reduced sampling bias were run in order to reduce densely sampled regions. After this 46 47 233 procedure the occurrence dataset contained 98 unique presence records. 48 49 50 234 Environmental climatic NetCDF data (Network Common Data Form) were 51 52 235 downloaded from the Med-CORDEX (Mediterranean Coordinated Regional Climate 53 54 236 Downscaling Experiment) website (https://www.medcordex.eu/). NetCDF files were 55 56 237 extracted and manipulated employing the Climate Data Operator (CDO) software (1.6.4 57 58 59 60 Page 15 of 81 Journal of Biogeography

1 2 3 238 version; Max-Planck Institut für Meterologie). Layers of chlorophyll concentration data were 4 5 239 obtained from Copernicus project (http://marine.copernicus.eu/). For the future scenarios, we 6 7 240 assumed a magnitude of change in chlorophyll-a concentrations of +10%, coded as 8 9 10 241 eutrophication scenario, and -10%, coded as oligotrophication scenarios (sensu Nixon 2009). 11 12 242 Similarly, we generated 11 salinity scenarios: no change, 5 salinity decrease scenarios (from - 13 14 243 0.1 to -0.5) and 5 salinity increase scenarios (from +0.1 to +0.5). 15 16 244 Prior to analysis, all the environmental data were rescaled at 1 km applying the nearest 17 18 245 neighbour interpolation.For Eight environmentalPeer variabReviewles were selected (Table A1 in Appendix 19 20 21 246 2) based on their biological relevance, as potential predictors of habitat distribution for B. 22 23 247 pharaonis . Species distribution models were applied using Maxent software (version 3.3.3k; 24 25 248 Phillips et al ., 2006). The default settings including logistic output, regularization multiplier 26 27 249 1, and 10,000 background points were used. The model evaluation was carried out through 28 29 30 250 the random test percentage, splitting the whole dataset in training (70%) and test data (30%), 31 32 251 subsamples (equal to the number of observation) and 5000 iterations, using the easily 33 34 252 interpretable logistic output format with habitat suitability values. Indeed, Maxent generates 35 36 253 an estimate of species probability presence ranging from 0 (unsuitable habitat) to 1 (optimal 37 38 254 habitat) representing the distribution in geographic space of suitable habitat ( i.e. : Habitat 39 40 41 255 Suitability Index - HSI) (Elith et al. , 2006; Phillips & Dudík 2008). Subsample replicates 42 43 256 from Maxent were used as proxies for different single-models to reach a consensus scenario, 44 45 257 reduce model inter-variability and to avoid potentially compromising policy decisions. To 46 47 258 construct current distributions, we converted the continuous suitability predictions to binary 48 49 50 259 predictions using the “threshold equal training sensitivity and specificity” command in 51 52 260 Maxent. The sensitivity is the probability that the model correctly predicts an observation 53 54 261 (true positive rate), while specificity is the probability that a known absence is correctly 55 56 262 predicted (true negative rate). This is the most reliable threshold allowing to minimise the 57 58 59 60 Journal of Biogeography Page 16 of 81

1 2 3 263 absolute difference between sensitivity and specificity (Nenzén & Araújo 2011) and to 4 5 264 balance the accuracy of areas correctly modeled as present and absent in the training and test 6 7 265 data. Specifically, at this threshold the chance of missing suitable distribution and assigning 8 9 10 266 unsuitable distribution is the same. 11 12 267 Subsequently, the performance model was assessed by calculating the Area Under the 13 14 268 Curve (AUC) of the Receiver Operator Characteristic (ROC), a measure of discrimination 15 16 269 capacity of generated models (Delong et al., 1988). Models with an AUC of 0.5 corresponded 17 18 270 to the expected performanceFor ofPeer a random classifier, Review 0.7–0.8 are considered an acceptable 19 20 21 271 prediction, 0.8–0.9 are excellent and >0.9 are outstanding (Hosmer & Lemeshow 1989). 22 23 272 Prior to running the models, all environmental data were remapped using the nearest 24 25 273 neighbour interpolation, employing CDO in order to achieve the highest possible spatial 26 27 274 resolution (1 km), with the same extent and spatial projection used for all variables. 28 29 30 275 Subsequently, the entire environmental dataset was clipped with the Mediterranean coastline. 31 32 276 Collinearity between predictors was tested applying the vifstep function from the usdm 33 34 277 package in R (Naimi 2013) and the predictor variables were selected. 35 36 278 In order to assess the rate of expansion of B. pharaonis in response to recent and short- 37 38 279 term future climate change, the models were run for the time-steps 2010, 2030 and 2050. For 39 40 41 280 the future distribution models, 66 scenarios of salinity (11 scenarios) and chlorophyll-a 42 43 281 concentration parameters (3 scenarios) were calculated and the obtained results were divided 44 45 282 for the three principal Mediterranean basins: Eastern, Central and Western Mediterranean Sea. 46 47 283 From these scenarios, HSI was derived for each time step across the Mediterranean Sea basin 48 49 50 284 (Figs. 3 and 4) presented herein as geographical forecast maps according to the classical 51 52 285 geographical division of the Mediterranean basin: the Eastern, the Central (Sicilian Channel, 53 54 286 Ionian and Adriatic seas) and the Western (Tyrrhenian, Balearic and Alboran seas; Fig. 1). 55 56 287 The percentage of variation of mean HSI was estimated in comparison to 2010 (Fig. 2), 57 58 59 60 Page 17 of 81 Journal of Biogeography

1 2 3 288 within the three considered Mediterranean basins (Eastern, Central and Western) for each of 4 5 289 the 66 simulated scenarios, respectively for 2030 and 2050 under different trophic and salinity 6 7 290 scenario (Table 1). The invasion risk of this species was tested by calculating the frequency of 8 9 10 291 scenarios in which the HSI was greater than 0.7, in order to highlight the suitable areas for 11 12 292 colonization under the possible future environmental states. Potential interactions between 13 14 293 HSI and the environmental variables considered in 2010 were analysed in terms of absolute 15 16 294 Pearson correlations and kernel density overlays (Feld et al. , 2016). 17 18 295 For Peer Review 19 20 21 296 Results 22 23 297 Studies published on B. pharaonis during the 2000s (Fig. 1) show the introduced range 24 25 298 of B. pharaonis spreading westwards throughout the Mediterranean Sea, following the 26 27 299 anticlockwise direction taken by most Lessepsian NIS (Katsanevakis et al ., 2014). The 28 29 30 300 outputs of the Systematic Review identified the ability of this NIS to compensate for large 31 32 301 changes in temperature and salinity regimes, with wider thermo-tolerance ranges and 33 34 302 increased tolerance of higher salinities in comparison to native Mediterranean bivalves 35 36 303 (Appendix 1). No evidence exists within the current literature on the effect of trophic status 37 38 304 on the presence of B. pharaonis , although as most filter feeder diets comprise of fresh 39 40 41 305 particulate organic matter and detritus, B. pharaonis will likely be affected by trophic 42 43 306 condition shifts as expressed by changes in suspended chlorophyll-a. 44 45 307 Under future scenarios for 2030 and 2050, eight predictors were identified, with 46 47 308 chlorophyll-a, salinity and surface temperature exerting the greatest influence on the invasion 48 49 50 309 pathway for B. pharaonis (Tables A1-A3; Fig. A2, A3 jackknife test in Appendix 2). 51 52 310 Chlorophyll-a (adopted as a proxy of local urbanisation; Nixon 2009), salinity and surface 53 54 311 temperature (as a proxy of tropicalisation; Azzurro et al., 2013) were the predictors 55 56 312 accounting for the highest percentage of the modelled current distribution (Table A1 and Fig. 57 58 59 60 Journal of Biogeography Page 18 of 81

1 2 3 313 A1 jackknife test in Appendix 2), which was in accordance with the values in published 4 5 314 literature (AUC = 0.816 ± 0.052; Fig. 2 year 2010 current scenario). For the 2010 model, the 6 7 315 marginal response of B. pharaonis to the selected environmental variables is reported in Fig. 8 9 10 316 A4 (Appendix 2). Of all stressor predictor variables tested, salinity change will be the most 11 12 317 important driver modifying Lessepsian NIS distribution pathways (Rilov & Galil 2009; Nagar 13 14 318 et al., 2016; Galil et al., 2017). As is already well-known from the literature (Sarà et al., 15 16 319 2008) this species shows a marked hypersaline affinity and the forecasted increase in salinity 17 18 320 will likely promoteFor the spatial spreadPeer of the propa Reviewgules toward the Western basin. 19 20 21 321 22 23 322 Habitat Suitability 24 25 323 In 2030, the contribution of climate-related factors will increase the forecasted habitat 26 27 324 suitability HSI by 47%, 25% and 14% for the Eastern, Central and Western basins 28 29 30 325 respectively (Table 1) in comparison to 2010 (Fig. 2 and Fig. 3). Estimating the HSI 31 32 326 likelihood under increasing salinity scenarios arising from the new Suez Canal opening (from 33 34 327 +0.1 to +0.5 psu; Table 1) and under no-change trophic conditions ( i.e. no change; Table 1), 35 36 328 in 2030 HSI will increase by 161%, 53% and 27% for the Eastern, Central and Western 37 38 329 basins, respectively. Similarly, in 2050 under no change trophic conditions, HSI will increase 39 40 41 330 by 182%, 75% and 40% respectively. If the trophic conditions of the three Mediterranean 42 43 331 basins become more eutrophic, the HSI will also dramatically increase (see Table 1 right 44 45 332 panels; Figs. 3, 4). Conversely, an overall reduction in HSI values will occur for the whole 46 47 333 Mediterranean Sea under conditions of decreased salinity and oligotrophication (of Table 1 48 49 50 334 left panels; Fig. 3, Fig. 4). The effect of changes in chlorophyll-a concentrations depends on 51 52 335 the basin considered. For the scenario with no change in salinity in 2030, eutrophication leads 53 54 336 to increase in HSI by 16.30%, while decrease by -2.08% is expected for more oligotrophic 55 56 337 conditions in the Eastern basin. Exactly the opposite pattern, however, occurs in the Western 57 58 59 60 Page 19 of 81 Journal of Biogeography

1 2 3 338 basin with a decrease by -2.03% in case of eutrophication and increase of 26.02% for 4 5 339 oligotrophication. Similarly, when the salinity is increasing by +0.5 psu, HSI in the Western 6 7 340 basin increases more strongly for oligotrophication (+48.84%) than for eutrophication 8 9 10 341 scenario (+27.28%), but the opposite response occurs in the East (+129.15% vs +153.69%). 11 12 342 The interactions between HSI and the environmental variables considered in 2010 13 14 343 were reported in Fig. A5 (Appendix 2). The results highlighted that the salinity (0.45) and the 15 16 344 SST (0.38) were the most correlated environmental variables with the HSI. 17 18 345 In Figures 5For and 6 are representedPeer the numberReview of scenarios in which the HSI resulted 19 20 21 346 greater than 0.7 for the 2030 and 2050 respectively. The frequency of agreement between 22 23 347 models outputs, in both time periods, resulted higher in the Eastern portion of the 24 25 348 Mediterranean basin, while the area of the Aegean Sea and the South Eastern Italy presented 26 27 349 lowest values. HSI never resulted > 0.7 within the Western Mediterranean basin in the 28 29 30 350 scenarios analysed. 31 32 351 33 34 352 Multiple stressors 35 36 353 In 2030 and 2050 the combined effects of multiple stressors will generate a synergistic 37 38 354 effect under increased salinity scenarios. In contrast, the effects of multiple stressors will 39 40 41 355 generate an antagonistic response under decreasing salinity scenarios. In particular, the 42 43 356 climate change stressors (SST and wind stress) will increase HSI by 11%, 18% and 8% in 44 45 357 2030 and 32%, 41% and 32% in 2050 for the Eastern, Central and Western basins, 46 47 358 respectively (Table 1). A synergistic effect (Table 1) is evident with changes in trophic status 48 49 50 359 conditions (HSI will range from -2% to 26% in 2030 and 19% to 48% in 2050). Under future 51 52 360 decreasing salinity scenarios there will be a decrease in the mean HSI (Table 1), whereas HSI 53 54 361 will increase under increasing salinity scenarios, with the effect being more marked in 2050. 55 56 57 58 59 60 Journal of Biogeography Page 20 of 81

1 2 3 362 Potential interactions, representing the response of HSI to pairwise stressor combinations are 4 5 363 reported in Appendix 2 (Figs. A4, A5). 6 7 364 8 9 10 365 Discussion 11 12 366 The forecast scenarios show unexpected consequences when climate change interacts 13 14 367 with increasing salinity derived from the new Suez Canal opening, which will be further 15 16 368 altered by changing trophic conditions produced by local human pressures ( sensu Nixon 17 18 369 2009). The main predictedFor effect Peer of the doubling Review of the Suez Canal will be the increase of the 19 20 21 370 propagule pressure for a great variety of species that are likely to colonize the Levantine 22 23 371 waters. The future spread of the Lessepsian bivalve B. pharaonis westwards in the 24 25 372 Mediterranean Sea is forecast for both periods 2030 and 2050 under a 10% increase in 26 27 373 eutrophication scenario, when both climate change and salinity increase are modelled 28 29 30 374 together, with interaction effects evident. The pathway from the current, localized distribution 31 32 375 within the introduced range in the western Mediterranean basin is predicted to predominantly 33 34 376 follow a north-westerly trajectory, with colonisation of new sites forecast along the northern 35 36 377 coastline. Some colonization of the southern coastline is also predicted, but to a far lesser 37 38 378 extent. Secondary introductions ( e.g. through ballast waters) in a westerly direction from 39 40 41 379 Levantine waters are a more complex phenomenon, potentially related to the warming of the 42 43 380 sea and also to additional hydrographic changes that are a consequence of the global climate 44 45 381 change ( e.g. salinity and trophic factors). The interregional differences showed by our 46 47 382 modelling are predominantly related to the spatial variation in the productivity across the 48 49 50 383 Mediterranean Sea, with more oligotrophic conditions prevailing in the Eastern sector. 51 52 384 Considering the preference of B. pharaonis for average levels of productivity, the increase in 53 54 385 productivity by 10% should facilitate the spread throughout the Eastern basin, but at the same 55 56 386 time hinder the invasion in Western waters that are already eutrophic. 57 58 59 60 Page 21 of 81 Journal of Biogeography

1 2 3 387 Synergistic interactions cause the equilibrium of native communities to shift, favoring 4 5 388 NIS invasions. A climate facilitated expansion will likely result from future increases in 6 7 389 temperature, as the spread of B. pharaonis will be promoted under conditions of interacting 8 9 10 390 stressors of trophic and salinity changes. 11 12 391 To date, however, there is a lack of biogeochemical models and the availability of 13 14 392 projected datasets for salinity and eutrophication is still scant. Thus, the reliability of 15 16 393 projected scenarios is widely debated and considered less confident than that of other 17 18 394 variables consideredFor in this study Peer ( e.g. sea surface Review temperature, wind stress etc.). Following 19 20 21 395 the precautionary principle, a set of sensitivity analysis was performed to cover all possible 22 23 396 expected changes. 24 25 397 A number of criticisms have been advanced against the use of SDMs, e.g. these tools 26 27 398 not consider biotic interactions, evolutionary change and species dispersal (Pearson & 28 29 30 399 Dawson 2003). However, the SDM approach can provide useful first results, giving an 31 32 400 approximation of the impact of environmental change including climate on species 33 34 401 distribution (Pearson & Dawson 2003; Guisan & Thuiller 2005; Wiens et al., 2009; Guisan et 35 36 402 al., 2013). Despite limitations, SDMs may be useful to assist in conservation planning by 37 38 403 contributing to strategic decisions about environmental change impacts, and can play a key 39 40 41 404 role by highlighting likely shifts of suitable habitat of NIS invasion (Thuiller et al., 2005; 42 43 405 Araujo et al., 2011). One of the main problems of SDMs could be due to the potential 44 45 406 underestimation of the potential spread of these species and consequentially the suitable 46 47 407 habitat predicted can represent only a conservative estimate (Parravicini et al., 2015). 48 49 50 408 The use of SDM tools remains challenging, but the potential to assess future invasion 51 52 409 risk by identification of areas vulnerable to invasion demonstrates the value of this method for 53 54 410 predicting potential NIS distributions. 55 56 411 57 58 59 60 Journal of Biogeography Page 22 of 81

1 2 3 412 Future range expansion 4 5 413 Although a degree of uncertainty inherent in all modelling approaches may complicate 6 7 414 projections of future biodiversity (Guisan & Thuiller 2005; Thuiller et al ., 2005; Walther 8 9 10 415 2007), SDMs represent the best approach to date with which to forecast biological invasions 11 12 416 (Walther et al ., 2011). Our predictions represent an excellent test to evaluate invasive 13 14 417 distribution shifts within marine systems using models developed and validated for terrestrial 15 16 418 ecosystems (Fernández & Hamilton 2015). When coupled with functional trait-based 17 18 419 approaches based onFor the fundamental Peer niche (Sarà Review et al., 2013), such results may improve the 19 20 21 420 ability to predict changes from current to future spatial distributions of NIS. The model 22 23 421 outputs support hypothesis that the NIS B. pharaonis will proliferate from its current invasive 24 25 422 distribution to 2050 under future salinity and eutrophication scenarios, being able to colonise 26 27 423 hard substrata across the Mediterranean (Fig. 4). This Lessepsian NIS will expand its invasive 28 29 30 424 range by more than 1,000 km in a westerly trajectory with respect to hydrographic conditions, 31 32 425 reaching the Spanish coasts and the Gibraltar Strait by 2050. This is consistent with 33 34 426 predictions made by Sarà et al. (2013) based on mechanistic functional trait based models, 35 36 427 which were performed to test the reliability of that approach to predict current physiologically 37 38 428 suitable habitats for this species. Although there were no historical records of occurrence 39 40 41 429 further west in the North Atlantic in the literature, the likelihood of B. pharaonis being 42 43 430 present was also predicted along the African ( e.g. Libya), Southern Italian Peninsula ( e.g. 44 45 431 Calabria) and Northern Sardinia coasts both during 2030 and 2050 (Fig. 3 and 4; Table 1). 46 47 432 48 49 50 433 Policy implications of invasion 51 52 434 Our results demonstrate that current European management actions and marine spatial 53 54 435 planning frameworks that are based solely on measures to manage ballast waters and hull- 55 56 436 fouling as the primary vector of invasion (see Ojaveer et al ., 2014 for an updated and detailed 57 58 59 60 Page 23 of 81 Journal of Biogeography

1 2 3 437 list) may be rendered ineffective by the construction of manmade corridors such as the new 4 5 438 Suez Canal opening. The SDM approach presented here provides a new tool with which to 6 7 439 realistically predict habitat suitability for NIS, via a worked example for one NIS model 8 9 10 440 species, B. pharaonis . 11 12 441 Our invasion maps (Figs. A6 and A7 in Appendix 2) will assist managers to identify 13 14 442 areas of vulnerability for native ecosystems. These geospatial model outputs will facilitate the 15 16 443 development and implementation of new, effective mitigation actions to prevent novel, 17 18 444 favourable conditionsFor for the introductionPeer of NIS,Review and subsequently address the related risks 19 20 21 445 and cost to policymaking and administration at the national and European levels (Hulme et 22 23 446 al., 2008). In addition, these models can be more widely applied to coastal marine systems 24 25 447 globally to forecast invasion dynamics for benthic marine species under a range of multiple 26 27 448 stressor scenarios. 28 29 30 449 At the European scale, these quantitative, geo-referenced, spatially ‐explicit risk maps 31 32 450 of biological invasions under changing environmental conditions provide powerful visual 33 34 451 communication and decision-support tools with which to describe where and when species 35 36 452 might invade, and provide geo-spatial trajectories of future spread. The need to predict the 37 38 39 453 NIS distribution pathways within the context of multiple stressors, including environmental 40 41 454 and anthropogenic drivers is a primary and essential step to identify and evaluate management 42 43 455 options and decisions to regulate and prevent new introductions. SDMs are useful statistical 44 45 456 toolkits with which to geospatially map past, current and future biogeographic ranges for NIS, 46 47 457 accommodating multiple stressor scenarios to enable more realistic forecasts of environmental 48 49 50 458 and anthropogenic stressors and the resultant impacts on the range and spread of NIS into new 51 52 459 areas. 53 54 460 Risk/pest maps are a valuable tool for the accurate assessment of Good Environmental 55 56 461 Status (GES) for member state compliance with the EU Marine Strategy Framework 57 58 59 60 Journal of Biogeography Page 24 of 81

1 2 3 462 Directive. GES targets are set against a background of “prevailing physiographic, geographic 4 5 463 and climatic conditions” for MSFD Descriptor 1, Biological diversity is maintained, and 6 7 464 Descriptor 2, Non-indigenous species (NIS) introduced by human activities are at levels that 8 9 10 465 do not adversely alter the ecosystems (European Union 2008). At the present time, there are 11 12 466 no standard methods to calculate the future distribution and status of native and invasive 13 14 467 species in European waters with respect to uncontrollable environmental drivers such as 15 16 468 climate change, salinity, eutrophication etc. termed “prevailing… conditions” (European 17 18 469 Union 2008). For Peer Review 19 20 21 470 Pest maps will provide additional information on the ecological and economic impacts 22 23 471 of invasion, predicting future risks areas, discussing and addressing related costs relevant to 24 25 472 policy and management practices. By combining and overlapping our maps with outputs from 26 27 473 a mechanistic trait-based approach ( e.g. Sarà et al., 2013) and human use layers (aquaculture 28 29 30 474 farms; e.g. Brigolin et al. , 2017; major ports, hubs of international maritime transport; e.g. the 31 32 475 Milazzo harbour in Sicily; D'Alessandro et al., 2016 and to see Appendix 1 for more 33 34 476 references around the Mediterranean basin) or layers of spatial management measures ( e.g. 35 36 477 Marine Protected Areas to see Galil et al. , 2017; Special Protection Areas to see the Stagnone 37 38 478 di Marsala in Sicily Sarà et al. , 1999), we show that locations predicted to be highly suitable 39 40 41 479 for colonization by B. pharaonis may overlap with protected or highly anthropic areas that are 42 43 480 highly likely to receive new propagules. The same exercise can be done using the major 44 45 481 shipping routes within the Mediterranean Sea or using the circulation patterns or more other 46 47 482 local maritime use layer. 48 49 50 483 This approach can be more widely applied to any marine species for which current 51 52 484 distributional, ecological and environmental tolerance data are available, and thus has wide 53 54 485 potential applications for quantitatively determining the changes in GES with respect to the 55 56 486 relative contributions of a range of drivers, including uncontrollable environmental change 57 58 59 60 Page 25 of 81 Journal of Biogeography

1 2 3 487 and local/regional anthropogenic stressors. Interestingly, this species shares similar 4 5 488 distribution patterns with other marine invasive species whose biogeographic ranges are 6 7 489 driven by temperature ( e.g. the European green crab, Carcinus maenas ; Compton et al., 2010; 8 9 10 490 and other benthic invertebrate; de Rivera et al., 2011) and specifically with other Lessepsian 11 12 491 invaders clearly showing climatic niche expansion in the Mediterranean (Azzurro et al., 2013; 13 14 492 Weinmann et al. , 2013; Parravicini et al., 2015). Given the vastly different ecology of these 15 16 493 species (fish, large benthic foraminifera and mussels) the consistency of the obtained results 17 18 494 might represent a Formajor step inPeer making broader Review generalizations about the future spread of 19 20 21 495 Lessepsian species under changing climate conditions. More specifically, combination of both 22 23 496 chlorophyll-a and salinity were the most influential environmental variables explaining the 24 25 497 expansion or abiotic resistance to this the invasion of the bluespotted cornetfish Fistularia 26 27 498 commersonii (Osteichtyes, Fistulariidae; Azzurro et al., 2013). Both for fish and large 28 29 30 499 foraminifera the main future areas of environmental suitability were identified along the 31 32 500 northern coasts of the Levantine Sea, Egypt, Turkey Dodecanese, Sicily Strait and Tyrrhenian 33 34 501 Sea, with suitability continuously increasing towards the Central and Western Mediterranean 35 36 502 Sea along the coasts of Italy, Croatia, Montenegro and Albania in the Adriatic Sea, in Sicily 37 38 503 and along the western coast of Italy (Azzurro et al., 2013; Weinmann et al., 2013). 39 40 41 504 An additional finding of the literature assessment suggests that in situ monitoring is 42 43 505 the most effective option to support biological invasion management, via the provision of 44 45 506 early detection warnings, and a rapid response derived from field data. Active and ongoing 46 47 507 NIS monitoring programs should be continued to track new introductions and spread of NIS, 48 49 50 508 evaluate changes in species composition and assess the status of both vulnerable and resilient 51 52 509 ecosystems (Butchart et al., 2010). Prevention seems to be the only feasible management 53 54 510 alternative when facing the need to take post-invasion adaptive management actions to control 55 56 511 biological invasions in marine ecosystems. Information on future invasion spread combined 57 58 59 60 Journal of Biogeography Page 26 of 81

1 2 3 512 with data of propagule pressure, and the roles that climatic and anthropogenic drivers play in 4 5 513 altering invasion dynamics will be crucial in informing prevention and monitoring strategies 6 7 514 suggesting where to focus monitoring plans and target management options at appropriate 8 9 10 515 scales and frequencies (from local to regional) to successfully mitigate invasions and 11 12 516 minimize their impact on native biodiversity, ecosystem services and human activities 13 14 517 (McDonald-Madden et al ., 2011; Norton et al ., 2015; Marzloff et al., 2016; Pecl et al., 2017). 15 16 518 Using data collected from observational and experimental research on multiple 17 18 519 stressors, and targetFor species asPeer input variables Review ensures scientific rigor of the SDM outputs 19 20 21 520 encourages a “call for collection of …” species occurrence and environmental data, at both 22 23 521 finer scales and in additional spatial regions. Such research-based, integrated approaches are a 24 25 522 priority over the coming decades as climate-facilitated biological invasions will create new 26 27 523 and unexpected challenges for biodiversity conservation. 28 29 30 524 31 32 525 Acknowledgements 33 34 526 PRIN TETRIS 2010 grant n. 2010PBMAXP_003, funded by the Italian Minister of Research 35 36 527 and University (MIUR) supported this research. NM was funded by a Marine Biological 37 38 528 Association of the UK Research Fellowship and the University of Liverpool. The authors 39 40 41 529 declare no competing financial interests. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 27 of 81 Journal of Biogeography

1 2 3 530 References 4 5 531 Adloff F, Somot S, Sevault F et al . (2015) Mediterranean Sea response to climate change in 6 7 532 an ensemble of twenty first century scenarios. Climate Dynamics, 4, 2775-2802. DOI 8 9 10 533 10.1007/s00382-015-2507-3. 11 12 534 Araujo MB, Alagador D, Cabeza M, Nogues-Bravo D, Thuiller W (2011). Climate change 13 14 535 threatens European conservation areas. Ecology Letters, 14, 484–492. 15 16 536 Azzurro E, Soto S, Garofalo G, Maynou F (2013) Fistularia commersonii in the 17 18 537 MediterraneanFor Sea: invasion Peer history and Reviewdistribution modeling based on presence-only 19 20 21 538 records. Biological Invasions, 15 (5), 977-990. 22 23 539 Bilotta GS, Milner AM, Boyd, I (2014) On the use of systematic reviews to inform 24 25 540 environmental policies. Environmental Science and Policy, 42 , 67–77. 26 27 541 Bonnici L, Evans J, Borg JA, Schembri PJ (2012) Biological aspects and ecological effects of 28 29 30 542 a bed of the invasive non-indigenous mussel Brachidontes pharaonis (Fischer P., 1870) 31 32 543 in Malta. Mediterranean Marine Science 13 (1):153-161. 33 34 544 Boudouresque C (2004) Marine biodiversity in the Mediterranean: status of species, 35 36 545 populations and communities . Scientific Reports of the Port-Cros National Park, 20 , 97- 37 38 546 146. 39 40 41 547 Bradley BA , Blumenthal DM, Wilcove DS, Ziska LH (2010) Predicting plant invasion in an 42 43 548 era of global change. Trends in Ecology and Evolution, 25 , 310-318. 44 45 549 Brigolin D, Porporato EMD, Prioli G, Pastres R (2017) Making space for shellfish farming 46 47 550 along the Adriatic coast. ICES Journal of Marine Science, 74 (6), 1540-1551. 48 49 50 551 Burrows MT, Schoeman DS, Richardson AJ et al . (2014) Geographical limits to species- 51 52 552 range shifts are suggested by climate velocity. Nature, 507 , 492-495. 53 54 553 Butchart SHM, Walpole M, Collen B et al . (2010) Global biodiversity: indicators of recent 55 56 554 declines. Science, 328 , 1164-1168. 57 58 59 60 Journal of Biogeography Page 28 of 81

1 2 3 555 Chapman DS, Makra L, Albertini R et al . (2016) Modelling the introduction and spread of 4 5 556 non ‐native species: International trade and climate change drive ragweed invasion. 6 7 557 Global Change Biology DOI: 10.1111/gcb.13220. 8 9 10 558 Collaboration for Environmental Evidence (CEE) (2013) Guidelines for systematic review 11 12 559 and evidence synthesis in environmental management. 2013. Version 4.2. 13 14 560 Environmental Evidence: 15 16 561 [http://environmentalevidence.org/Documents/Guidelines/Guidelines4.2.pdf ] 17 18 562 Compton TJ, LeathwickFor JR, Inglis Peer GJ (2010) ThermogeReviewography predicts the potential global 19 20 21 563 range of the invasive European green crab ( Carcinus maenas ). Diversity and 22 23 564 Distributions, 16 (2), 243-255. 24 25 565 D'Alessandro M, Esposito V, Giacobbe S, Renzi M, Mangano MC, Vivona P, Consoli P, 26 27 566 Scotti G, Andaloro F, Romeo T (2016) Ecological assessment of a heavily human- 28 29 30 567 stressed area in the Gulf of Milazzo, Central Mediterranean Sea: an integrated study of 31 32 568 biological, physical and chemical indicators. Marine Pollution Bulletin, 106 (1), 260- 33 34 569 273. 35 36 570 Delong ER, Delong DM, Clarke-Pearson DL (1988) Comparing the areas under two or more 37 38 39 571 correlated receiver operating characteristic curves: a nonparametric approach. 40 41 572 Biometrics, 44 , 837-845. 42 43 573 Dormann CF, McPherson JM, Araújo MB, Bivand R, Bolliger J et al. (2007) Methods to 44 45 574 account for spatial autocorrelation in the analysis of species distributional data: a 46 47 575 review. Ecography, 30 , 609-628. 48 49 50 576 de Rivera CE, Steves BP, Fofonoff PW, Hines AH, Ruiz GM (2011) Potential for high_ 51 52 577 latitude marine invasions along western North America. Diversity and Distributions, 53 54 578 17 (6), 1198-1209. 55 56 57 58 59 60 Page 29 of 81 Journal of Biogeography

1 2 3 579 Elith J, Graham CH, Anderson RP, Dudik M, Ferrier S, Guisan A et al. (2006) Novel methods 4 5 580 improve prediction of species distributions from occurrence data. Ecography, 29, 129- 6 7 581 151. 8 9 10 582 Elith J, Leathwick JR (2009) Species Distribution Models: Ecological Explanation and 11 12 583 Prediction Across Space and Time. Annual Reviews in Ecology and Evolutionary 13 14 584 Systems , 40, 677-697. 15 16 585 European Union (2008) Directive 2008/56/EC of the European Parliament and of the Council 17 18 586 of 17 June 2008For establishing Peer a framework Review for community action in the field of marine 19 20 21 587 environmental policy (Marine Strategy Framework Directive). Official Journal of the 22 23 588 European Union L 164/19 pp. 22. 24 25 589 Feld CK, Segurado P, Gutiérrez-Cánovas C (2016) Analysing the impact of multiple stressors 26 27 590 in aquatic biomonitoring data: A ‘cookbook’ with applications in R. Science of the 28 29 30 591 Total Environment, 573 , 1320-1339. 31 32 592 Fernández M, Hamilton H (2015) Ecological Niche Transferability Using Invasive Species as 33 34 593 a Case Study. PLoS ONE , 10, e0119891. 35 36 594 Floerl O, Rickard G, Inglis G, Roulston H (2013) Predicted effects of climate change on 37 38 595 potential sources of non indigenous marine species. Diversity and Distributions , 19, 39 40 41 596 257-267. 42 43 597 Galil BS (2009) Taking stock: inventory of alien species in the Mediterranean sea. Biological 44 45 598 Invasions 11, 359-372. 46 47 599 Galil BS, Boero F, Campbell ML et al. (2015) ‘Double trouble’: the expansion of the Suez 48 49 50 600 Canal and marine bioinvasions in the Mediterranean Sea. Biological Invasions , 17, 973- 51 52 601 976. 53 54 55 56 57 58 59 60 Journal of Biogeography Page 30 of 81

1 2 3 602 Galil B, Marchini A, Occhipinti-Ambrogi A, Ojaveer H (2017) The enlargement of the Suez 4 5 603 Canal-Erythraean introductions and management challenges. Management of Biological 6 7 604 Invasions 8. 8 9 10 605 Gallien L , Münkemüller T, Albert CH, Boulangeat I., Thuiller W (2010) Predicting potential 11 12 606 distributions of invasive species: where to go from here? Diversity and Distributions, 13 14 607 16, 331-342. 15 16 608 Gurevitch J, Hedges LV (2001) Meta-analysis: Combining the Results of Independent 17 18 609 experiments. ForIn: Design andPeer analysis of ecologicalReview experiments (eds Scheiner, SM and 19 20 21 610 Gurevitch, J), pp. 347-369. Oxford University Press, Oxford, United Kingdom. 22 23 611 Guisan A, et al. (2013) Predicting species distributions for conservation decisions. Ecology 24 25 612 Letters, 16, 1424-1435. 26 27 613 Guisan A, Thuiller W (2005) Predicting species distribution: offering more than simple 28 29 30 614 habitat models. Ecological Letters , 8, 993-1009. 31 32 615 Guisan A, Zimmermann NE (2000) Predictive habitat distribution models in ecology. 33 34 616 Ecological Modelling , 135 , 147-186. 35 36 617 Hamaoui-Laguel L, Vautard R, Liu L et al . (2015) Effects of climate change and seed 37 38 618 dispersal on airborne ragweed pollen loads in Europe. Nature Climate Change 5, 766- 39 40 41 619 771. 42 43 620 Helmuth B, Mieszkowska N, Moore P, Hawkins SJ (2006) Living on the edge of two 44 45 621 changing worlds: forecasting the impacts of climate change on rocky intertidal 46 47 622 ecosystems. Annual Review of Ecology Systematics and Evolution 37 , 373-404. 48 49 50 623 Hosmer DW, Lemeshow S (1989) Applied logistic regression. Wiley, New York. 51 52 624 Hulme PE, Bacher S, Kenis M et al . (2008) Grasping at the routes of biological invasions: a 53 54 625 framework for integrating pathways into policy. Journal of Applied Ecology , 45.2 , 403- 55 56 626 313. 57 58 59 60 Page 31 of 81 Journal of Biogeography

1 2 3 627 Hulme PE (2009) Trade, transport and trouble: managing invasive species pathways in an era 4 5 628 of globalization. Journal of Applied Ecology , 46 , 10-1. 6 7 629 Katsanevakis S , Zenetos A, Belchior C, Cardoso AC (2014) Invading European Seas: 8 9 10 630 assessing pathways of introduction of marine aliens. Ocean and Coastal Management , 11 12 631 76 , 64-74. 13 14 632 Lenz M, da Gama BA, Gerner NV et al . (2011) Non-native marine invertebrates are more 15 16 633 tolerant towards environmental stress than taxonomically related native species: results 17 18 634 from a globallyFor replicated Peer study. Environmental Review Research, 111 (7), 943-952. 19 20 21 635 Lima FP, Ribeiro PA, Queiroz N, Hawkins SJ, Santos AM (2007) Do distributional shifts of 22 23 636 northern and southern species of algae match the warming pattern? Global Change 24 25 637 Biology, 13(12), 2592-2604. 26 27 638 Mangano MC, Sarà G (2017a) Collating science-based evidence to inform public opinion on 28 29 30 639 the environmental effects of marine drilling platforms in the Mediterranean Sea. Journal 31 32 640 of Environmental Management, 188 , pp.195-202. 33 34 641 Mangano MC Sarà G (2017b) The author's reply to NR Haddaway. Journal of Environmental 35 36 642 Management, 197 , pp.114-116. 37 38 643 Mangano MC, Sarà G, Corsolini S (2017) Monitoring of persistent organic pollutants in the 39 40 41 644 polar regions: knowledge gaps & gluts through evidence mapping. Chemosphere, 172 , 42 43 645 pp.37-45. 44 45 646 Marzloff MP, Melbourne-Thomas J, Hamon KG, Hoshino E, Jennings S, Putten IE, Pecl GT 46 47 647 (2016) Modelling marine community responses to climate-driven species redistribution 48 49 50 648 to guide monitoring and adaptive ecosystem-based management. Global Change 51 52 649 Biology, 22 (7): 2462-2474. 53 54 55 56 57 58 59 60 Journal of Biogeography Page 32 of 81

1 2 3 650 McDonald-Madden E , Runge MC, Possingham HP, Martin TG (2011) Optimal timing for 4 5 651 managed relocation of species faced with climate change. Nature Climate Change , 1(5), 6 7 652 261-265. 8 9 10 653 Mieszkowska N, Kendall MA, Hawkins SJ, Leaper,R, Williamson P, Hardman-Mountford NJ 11 12 654 Southward AJ (2006) Changes in the range of some common rocky shore species in 13 14 655 Britain - a response to climate change? Hydrobiologia, 55, 241-251. 15 16 656 Mieszkowska N, Sugden H, Firth L & Hawkins SJ (2014) The role of sustained observations 17 18 657 in tracking Forimpacts of environmentalPeer changeReview on marine biodiversity and ecosystems. 19 20 21 658 Philosophical Transactions of the Royal Society A, 372.2025 , 20130339 22 23 659 http://dx.doi.org/10.1098/rsta.2013.0339 24 25 660 Nagar LR, Shenkar N (2016) Temperature and salinity sensitivity of the invasive ascidian 26 27 661 Microcosmus exasperatus Heller, 1878. Aquatic Invasions, 11 (1), 33-43. 28 29 30 662 Naimi B (2013) usdm: Uncertainty analysis for species distribution models. R package 31 32 663 version 1.1-12. http://CRAN.R-project.org/package . 33 34 664 Nenzén HK & Araújo MB (2011) Choice of threshold alters projections of species range 35 36 665 shifts under climate change. Ecological Modelling, 222 (18), 3346-3354. 37 38 666 Nixon S (2009) Eutrophication and the macroscope. Hydrobiologia, 629 , 5-19. 39 40 41 667 Norton DA, Warburton B (2015) The potential for biodiversity offsetting to fund effective 42 43 668 invasive species control. Conservation Biology , 29.1 , 5-11. 44 45 669 Ojaveer H, Galis BS, Minchin D et al . (2014) Ten recommendations for advancing the 46 47 670 assessment and management of non-indigenous species in marine ecosystems. Marine 48 49 50 671 Policy , 44 , 160-165. 51 52 672 Pachauri RK, Allen MR, Barros VR et al . (2014) Climate Change 2014: Synthesis Report. 53 54 673 Contribution of Working Groups I, II and III to the Fifth Assessment Report of the 55 56 57 58 59 60 Page 33 of 81 Journal of Biogeography

1 2 3 674 Intergovernmental Panel on Climate Change (eds.R. Pachauri, L. Meyer) IPCC, pp. 4 5 675 151, ISBN: 978-92-9169-143-2, Geneva, Switzerland. 6 7 676 Parmesan C, Yohe G (2003) A globally coherent fingerprint of climate change impacts across 8 9 10 677 natural systems. Nature , 421 , 37-42. 11 12 678 Parravicini V, Azzurro E, Kulbicki M, Belmaker J (2015) Niche shift can impair the ability to 13 14 679 predict invasion risk in the marine realm: an illustration using Mediterranean fish 15 16 680 invaders. Ecology Letters, 18, 246-253. 17 18 681 Pearson RG, DawsonFor TP (2003) Peer Predicting the Reviewimpacts of climate change on the distribution 19 20 21 682 of species: are bioclimatic envelope models useful? Global Ecology and Biogeography 22 23 683 12 , 361-371. 24 25 684 Pederson J, Mieszkowska N, Carlton JT, Gollasch S, Jelmert A, Minchin D, Occhipinti- 26 27 685 Ambrogi A, Wallentinus I (2011) ICES Position Paper on Climate Change Chapter 11: 28 29 30 686 Climate Change and Non-Native Species in the North Atlantic. In: CES Status Report 31 32 687 on Climate Change in the North Atlantic (eds. Reid, PC, Valdés L), 174-190. 33 34 35 688 Payne MR, Barange M, Cheung WW et al . (2015) Uncertainties in projecting climate change 36 37 689 impacts in marine ecosystems ICES Journal of Marine Science: Journal du Conseil, 38 39 690 fsv231. 40 41 42 691 Phillips SJ, Anderson RP, Schapire RE (2006) Maximum entropy modeling of species 43 44 692 geographic distributions. Ecological Modelling , 190 , 231-259. 45 46 693 Phillips SJ. Dudík M (2008) Modeling of species distributions with Maxent: new extensions 47 48 694 and a comprehensive evaluation. Ecography, 31 , 161-175. 49 50 51 695 Pecl GT, Araujo MB, Bell J, Blanchard J, Bonebrake TC, Chen I, Clark TD, Colwell RK, 52 53 696 Danielsen F, Evengard B, Robinson S et al., (2017) Biodiversity redistribution under 54 55 697 climate change: Impacts on ecosystems and human well-being. Science 35 (6332): 1-9. 56 57 58 59 60 Journal of Biogeography Page 34 of 81

1 2 3 698 Poloczanska ES, Brown CJ, Sydeman WJ et al . (2013) Global imprint of climate change on 4 5 699 marine life. Nature Climate Change, 3(10), 919-925. 6 7 700 Richter R Dullinger S, Essl F, Leitner M, Vogl G (2013) How to account for habitat 8 9 10 701 suitability in weed management programmes? Biological Invasions, 15 , 657-669. 11 12 702 Rilov G, Galil B (2009) Marine bioinvasions in the Mediterranean Sea – History, distribution 13 14 703 and ecology. In Biological Invasions in Marine Ecosystems (Rilov, G. & Crooks, J. 15 16 704 A., eds.), pp. 549-575. Berlin Heidelberg: Springer. 17 18 705 Ruiz GM, Ruiz GM,For Fofonoff PeerPW, Carlton JT, Review Wonham MJ, Hines AH (2000) Invasion of 19 20 21 706 Coastal Marine Communities in North America: Apparent Patterns, Processes, and 22 23 707 Biases. Annual Reviews in Ecology and Evolutionary Systems , 31 , 481-531. 24 25 708 Safriel UN, Gilboa A, Felsenburg T (1980) Distribution of Rocky Intertidal Mussels in the 26 27 709 Red Sea Coasts of Sinai, the Suez Canal and the Mediterranean Coast of Israel, with 28 29 30 710 Special Reference to Recent Colonizers. Journal of Biogeography 7(1): 39-62. 31 32 711 Safriel UN, Sasson-Frostig Z (1988) Can colonizing mussel outcompete indigenous mussel? 33 34 712 Journal of Experimental Marine Biology and Ecology 117 (3): 221-226. 35 36 713 Sarà G, Romano C, Widdows J, Staff FJ (2008) Effect of salinity and temperature on feeding 37 38 714 physiology and scope for growth of an invasive species ( Brachidontes pharaonis - 39 40 41 715 Mollusca: Bivalvia) within the Mediterranean Sea. Journal of Experimental Marine 42 43 716 Biology and Ecology , 363 : 130-136. 44 45 717 Sarà G, Leonardi M, Mazzola A (1999) Spatial and temporal changes of suspended matter in 46 47 718 relation to wind and vegetation cover in a Mediterranean shallow coastal environment. 48 49 50 719 Chemistry and Ecology, 16 (2-3), pp.151-173. 51 52 720 Sarà G, Palmeri V, Rinaldi A, Montalto V, Helmuth B (2013) Predicting biological invasions 53 54 721 in marine habitats through eco-physiological mechanistic models: a case study with the 55 56 722 bivalve Brachidontes pharaonis . Diversity and Distributions , 19 , 1235-1247. 57 58 59 60 Page 35 of 81 Journal of Biogeography

1 2 3 723 Sorte CJ, Williams SL, Carlton JT (2010) Marine range shifts and species introductions: 4 5 724 comparative spread rates and community impacts. Global Ecology and Biogeography, 6 7 725 19(3) , 303-316. 8 9 10 726 Thuiller W, Richardson DM, Pyšek P, Midgley GF, Hughes GO, Rouget M (2005) Niche 11 727 based modelling as a tool for predicting the risk of alien plant invasions at a global 12 13 728 scale. Global Change Biology, 11(12) , 2234-50. 14 15 729 Verbruggen H (2012) OccurrenceThinner 1:04. 16 730 http://www.phycoweb.net/software/OccurrenceThinner/. Accessed: 2016 May. 17 18 731 Verbruggen H, TybergheinFor L, BeltonPeer GS, Mineur Review F, Jueterbock A, Hoarau G, De Clerck O 19 20 732 (2013) Improving transferability of introduced species’ distribution models: new tools 21 733 to forecast the spread of a highly invasive seaweed. PLoS ONE, 8(6), e68337. 22 23 734 doi:10.1371/journal.pone.0068337 24 25 735 Vilà M, Espinar JL, Hejda M, Hulme PE, Jarošík V, Maron JL, Pergl J, Schaffner U, Sun Y, 26 27 736 Pyšek P (2011) Ecological impacts of invasive alien plants: a meta-analysis of their 28 29 737 effects on species, communities and ecosystems. Ecology Letters, 14 , 702-708. 30 31 738 Walther GR (2007) ECOLOGY: Tackling Ecological Complexity in. Science, 315 , 606. 32 33 739 Walther GR, Roques A, Hulme PE et al . (2009) Alien species in a warmer world: risks and 34 35 36 740 opportunities, Trends in Ecology and Evolution , 24 , 686–693. 37 38 741 Weinmann AE, Rödder D, Lötters S, Langer MR (2013) Traveling through time: the past, 39 40 742 present and future biogeographic range of the invasive foraminifera Amphistegina spp. 41 42 743 in the Mediterranean Sea. Marine Micropaleontology, 105 , 30-39. 43 44 45 744 Wiens JA, Stralberg D, Jongsomjit D, Howell CA, Snyder MA (2009) Niches, models, and 46 47 745 climate change: assessing the assumptions and uncertainties. Proceedings of the 48 49 746 National Academy of Sciences of the United States of America; 106 (2), 19729-19736. 50 51 747 Zerebecki RA & Sorte CJ (2011) Temperature tolerance and stress proteins as mechanisms of 52 53 748 invasive species success. PLoS One , 6(4) , e14806. 54 55 56 749 57 58 750 Supporting Information may be found in the online version of this article: 59 60 Journal of Biogeography Page 36 of 81

1 2 3 751 Appendix 1 Systematic review outcomes dataset 4 5 752 Appendix 2 Tables detailing variables used, ROC curves and spatial distributions in 2030 and 6 7 753 2050 for all 33 scenarios considered in the Species Distribution Model. 8 9 10 754 11 12 755 Data Accessibility: Rasters derived from the habitat suitability models will be available as 13 14 756 raster grids from the Pangaea database. 15 16 757 17 18 758 Biosketch: GianlucaFor Sarà (Ph.D., Peer 1994) is Professor Review of Ecology at University of Palermo 19 20 21 759 (Italy) and coordinates the Laboratory of Experimental Ecology of the Department of Earth 22 23 760 and Marine Science. He graduated for his PhD in 1994 at University of Messina (Italy) 24 25 761 discussing a thesis dealing with bioenergetics and growth performance of cultivated bivalves 26 27 762 in the Southern Mediterranean Sea. Through over 110 peer-reviewed papers published in the 28 29 30 763 last 20 years, his research focuses on the effect of anthropogenic influence on ecosystems and 31 32 764 the study of structures and ecosystem functioning through its influence on the rates of 33 34 765 synthesis of biological structures, chemical compositions, energy and material fluxes, 35 36 766 population processes, species interactions and thereby biodiversity. 37 38 767 39 40 768 41 42 769 Editor: Richard Pearson 43 44 770 Author contributions: G.S. conceived the idea and addressed the objective of analyses, and 45 46 771 with N.M. equally led the writing, provided funds, hardware and software facilities; E.M.D.P. 47 48 772 carried out the predictive modelling and led the modelling writing; M.C.M. performed the 49 50 51 773 systematic review, led the review and the management issues writing and with other authors 52 53 774 equally contributed to draft this manuscript. 54 55 775 56 57 776 58 59 60 Page 37 of 81 Journal of Biogeography

1 2 3 777 Table 1. Percentage of variation of mean Habitat Suitability Index (HSI), in comparison to 4 5 778 2010, within the three considered Mediterranean basins (Eastern, Central and Western) for 6 7 779 each of the 66 simulated scenarios, respectively for 2030 and 2050 under different trophic and 8 9 10 780 salinity scenario (OLIGOTROPHIC = OLIGOTROPHICATION; EUTROPHIC = 11 12 781 EUTROPHICATION). 13 14 782 15 16 DECREASE 2030 INCREASE

17 SALINITY 18 For0.5 0.4Peer 0.3 0.2 Review 0.1 0 0.1 0.2 0.3 0.4 0.5 19 20 Eastern -26.44 -17.48 -18.12 -13.26 -10.05 -2.08 44.43 68.73 135.26 129.59 129.15 OLIGOTROPHIC 21 Central -6.62 5.16 2.65 10.57 14.05 14.73 17.11 20.53 30.87 42.80 106.07 22 Western 1.92 13.24 12.00 18.35 23.77 26.02 27.99 27.02 35.33 48.31 48.84 23 Eastern -12.66 -13.19 -8.03 -1.26 0.66 10.91 61.73 116.12 116.84 122.62 139.13 24 NO CHANGE Central 0.36 0.47 1.21 12.45 11.77 18.23 19.92 29.67 30.00 45.71 111.71 25 Western -1.87 -5.21 -2.28 6.47 5.58 7.74 13.56 20.51 19.74 26.71 36.08 26 Eastern -8.96 1.49 0.65 -0.87 3.94 16.30 67.72 107.24 145.66 147.54 153.69 27 EUTROPHIC Central -2.73 6.46 5.80 6.82 6.51 15.91 21.49 32.32 27.63 48.51 115.16 28 Western -16.47 -9.35 -7.65 -6.33 -8.67 -2.03 6.55 19.68 7.60 19.99 27.28 29 DECREASE 2050 INCREASE

30 SALINITY 31 0.5 0.4 0.3 0.2 0.1 0 0.1 0.2 0.3 0.4 0.5 32 Eastern -1.58 3.09 5.57 8.54 11.46 18.88 74.91 113.07 154.10 160.10 181.08 33 OLIGOTROPHIC Central 20.74 22.98 25.91 30.05 37.90 41.39 43.21 48.61 62.12 70.00 134.15 34 Western 25.14 25.96 35.27 30.34 41.31 48.01 49.60 51.49 61.96 64.04 67.59 35 36 Eastern 7.99 15.40 18.36 24.32 27.76 31.83 90.10 137.55 136.39 167.72 185.92 NO CHANGE 37 Central 18.98 25.07 32.71 37.33 34.94 40.69 55.54 58.17 61.25 68.12 136.67 38 Western 7.53 11.77 21.91 24.60 21.39 31.63 41.01 37.26 39.76 34.83 53.24 39 Eastern 19.12 23.41 35.41 31.29 32.80 44.61 94.62 133.33 150.93 177.08 192.60 40 EUTROPHIC Central 24.29 27.82 40.95 35.45 38.31 46.81 48.83 52.88 66.50 74.81 136.41 41 Western 0.38 7.34 17.64 15.86 17.04 22.11 22.22 27.05 33.71 34.21 35.32 42 783 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 38 of 81

1 2 3 784 Figure 1. Temporal trend of the number of scientific publications across the three considered 4 5 785 Mediterranean basin (total number of plotted hits 98 except global and regional reviews; 6 7 786 search performed at 20/07/2015). Presence records of B. pharaonis extracted from literature 8 9 10 787 and employed to run Maxent models (yellow points on the map). The main evolutions of the 11 12 788 Suez Canal are reported (yellow labels on the graph, data derived from 13 14 789 "http://www.suezcanal.gov.eg/" \t "_blank"). 15 16 790 17 18 791 Figure 2. Spatial For distributions Peer of predicting suitableReview habitat of B. pharaonis under 2010 19 20 21 792 current scenario (AUC = 0.809 ±0.043). 22 23 793 24 25 794 Figure 3. Spatial distributions of predicting suitable habitat of B. pharaonis under 2030 future 26 27 795 scenario, considering the scenarios with a decrease (-0.5 psu, panels a and b) and increase 28 29 30 796 (+0.5 psu panels c and d) of salinity, and both oligotrophication (-10%; on the left, panels a 31 32 797 and c) and eutrophication (+10%; on the right, panels b and d) conditions. 33 34 798 35 36 799 Figure 4. Spatial distributions of predicting suitable habitat of B. pharaonis under 2050 future 37 38 800 scenario, considering the scenarios with a decrease (-0.5 psu, panels a and b) and increase 39 40 41 801 (+0.5 psu panels c and d) of salinity, and both oligotrophication (-10%; on the left, panels a 42 43 802 and c) and eutrophication (+10%; on the right, panels b and d) conditions. 44 45 803 46 47 804 Figure 5. Spatial distributions of frequency of predicted Habitat Suitability of B. pharaonis 48 49 50 805 under 2030 for all the 33 scenarios. 51 52 806 53 54 807 Figure 6. Spatial distributions of frequency of predicted Habitat Suitability of B. pharaonis 55 56 808 under 2050 for all the 33 scenarios. 57 58 59 60 Page 39 of 81 Journal of Biogeography

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 For Peer Review 21 22 23 24 25 26 27 28 1 29 30 2 Figure 1 31 32 3 33 34 35 4 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 40 of 81

1 2 3 5 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 6 For Peer Review 21 22 23 7 Figure 2 24 25 8 26 27 9 28 29 30 10 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 41 of 81 Journal of Biogeography

1 2 3 11 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 For Peer Review 21 12 22 23 13 Figure 3 24 25 14 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 42 of 81

1 2 3 15 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 For Peer Review 21 16 22 23 17 Figure 4 24 25 18 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 43 of 81 Journal of Biogeography

1 2 3 4 5 6 7 8 9 10 11 12 13 For Peer Review 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 19 30 31 20 Figure 5 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 44 of 81

1 2 3 4 5 6 7 8 9 10 11 12 13 For Peer Review 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 21 30 31 22 Figure 6 32 33 23 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 45 of 81 Journal of Biogeography

1 2 3 1 APPENDIX 1, SYSTEMATIC REVIEW OUTCOMES DATASET (search ended at 21/08/2015) 4 5 2 6 7 Min / Max mea mean mean Ref Associated species - Density Species Basin Country Locality Study aims Habitat length n T Salinity CHL-a 8 Code assemblages (Ind./m 2) 9 (mm) (°C) (‰) (µg/l) 10 Itanieh, Jiyeh, Bioindicators 11 Beirut, Byblos, measures of trace Bp_1 B. variabilis E Lebanon ------12 Batroun, Tripoli, metals (Pb, Cd, and 13 andFor Arida PeerV) Review 14 Aspidosiphon 15 B. Calcareous Bp_2 E Turkey Mersin Bay Epibiosis study (Aspidosiphon) - - - - - 16 pharaonis rocks 17 elegans 18 Calcareous Sicily (Augusta) Beds 19 rocks Volcanic 20 Sicily (Catania) Patches 7.1 / 14.9 21 rocks Calcareous 22 Bp_3 B. variabilis W Italy Sicily (Riposto) Biometric study 23 rocks Calcareous 24 Sicily (Siracusa) 25 rocks 26 Sicily (Milazzo) 27 Measures of 28 B. Sicily (Stagnone di temperature Rocks at Bp_4 C Italy 29 pharaonis Marsala) dependentresponses waterline 30 (mesocosm) 31 B. Bp_5 E Palestine Checklist 32 pharaonis 33 B. Rocks at Bp_6 E Syria Ras El Bassit Checklist 100200 20 38.5 34 pharaonis waterline 35 B. Bp_7 E Lebanon Checklist 36 pharaonis 37 Bittium spp., Tanaidae beds 38 Birzebbugia Bay Effects of mussel Globigerina sp., Rissoa sp., Hyale small B. 39 Bp_8 C Malta (Marsaxlokk bed establishment on limestone sp., Elasmopus sp., clusters 05 / 2030 pharaonis 40 Harbour) the associated biota bedrock Dynamene sp., mixed to 41 Acanthochitona sp., marco 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 46 of 81

1 2 3 Osilinus turbinatus , algae 4 Nereis ?rava , (16550 ± 5 Amphitoe sp. 2051) 6 Tel Aviv (Shemen) 7 B. Cellular and Intertidal Bp_9 E Israel Patella cerulea 20 38.5 8 pharaonis molecular responses rocky shore 9 Tel Aviv (Akko) 10 11 B. Bp_10 W Italy Corsica Checklist 12 pharaonis 13 B. Bp_11 E Turkey For Peer Checklist Review - 14 pharaonis B. 15 Bp_12 E Greece Cyprus Checklist - 16 pharaonis 17 18 B. SalsesLeucate Food web isotopic Bp_13 W France - Isolated 19 pharaonis Lagoon characterisation 20 21 22 Zostera noltii - Chaetomorpha sp. 23 Ulva sp. - B. SalsesLeucate Food web isotopic 24 Bp_14 W France Enteromorpha sp. 32 2 pharaonis Lagoon characterisation 25 Gammarus 26 aequicauda - Idotea 27 sp. 28 Sicily (Trapani, 21.3 29 Saline Ettore 46.8 ± 5.6 Cerastoderma ± 8 30 B. Infersa) Rocks at glaucum - Abra Bp_15 C Italy Filter feeding rate Patches 31 pharaonis waterline segmentatus - Loripes Sicily (Stagnone di 19 ± 32 lacteus 39.2 ± 3.6 Marsala) 6.2 33 34 B. Bp_16 E Greece Checklist 35 pharaonis 36 Intertidal 37 B. Bp_17 E Turkey Iskenderun Bay New record arificial hard Other bivalves 25 38.5 38 pharaonis 39 subsyratum 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 47 of 81 Journal of Biogeography

1 2 3 Miger Ilma, below Rocks at 4 Dingli Cliffs waterline 5 Detritus from Marsaskala Bay 6 4 m 7 Juveniles Bahar ic˙C˙aghaq 8 from tidepool Quajjenza, Fish farm 9 10 Marsaxlokk Bay nets Old tyre near 11 Rinella Bay 12 shore 13 Weathered/er Dwejra,For Gozo Peeroded concreteReview Isolated 14 platform 15 Lower and 16 Qbajjar Bay, Middle ~10 17 Marsalforn,Gozo Globigerina 18 Limestone 19 Upper 20 Blue Lagoon, Globigerina Isolated 21 Comino Limestone B. Checklist and 22 Bp_18 C Malta coastline 23 pharaonis mapping Upper 24 Ghadira, Mellieha Coralline Isolated 25 Limestone 26 Upper Isolated 27 Bahar icCaghaq– Globigerina ~10 Salini Limestone 28 ~100 29 coastline Artificially 30 modified 31 St George′s Bay Lower Isolated 32 area, St Julian′s Coralline 33 Limestone 34 coastline 35 Lower 36 Coralline 37 Limestone Isolated 38 Spinola Bay area, coastline with 39 Sliema concrete 40 patches Wooden 41 <10 42 fishing boat 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 48 of 81

1 2 3 Balluta Concrete wall <10 4 Lower Sliema Seafront, Globigerina 5 ~50 6 Sliema Limestone 7 coastline 8 Lower Quisisana, Globigerina ~10 to 9 Sliema Limestone ~100 10 coastline 11 Lower 12 Tignè Point, Globigerina ~10 to 13 ForSliema PeerLimestone Review ~100 14 coastline 15 Lower ~100 16 Globigerina Tà Xbiex 17 Limestone <10 18 coastline 19 Pietà Safront Concrete wall <10 20 Xatt itTiben, Sa Concrete ~50 21 Maison slope 22 Lower Fort St Elmo area Globigerina 23 (Marsamxett), ~10 Limestone 24 Valletta 25 coastline 26 Isolated Lower 27 Tà Liesse area, Globigerina ~10 28 Valletta Limestone 29 patch 30 Lower ~10 to 31 Globigerina ~100 32 Limestone Isolated 33 coastline 34 Weathered/er Xghajra– 35 oded concrete Scattered Marsaskala stretch 36 platform 37 Lower Globigerina ~10 m2 to 38 39 Limestone ~100 m2 40 coastline 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 49 of 81 Journal of Biogeography

1 2 3 Middle/Lowe TalMig˙nuna r Globigerina ~10 to 4 5 Cliffs, Marsakala Limestone ~100 coastline 6 Middle/Lowe 7 r Globigerina >1000 8 Limestone 9 St. Thomas’ Bay area, Birzebbuga coastline 10 Concrete wall ~100 11 Sand ~10 12 13 Upper For PeerGlobigerina Review 14 Limestone 15 coastline, 16 with Xrobb lGhagin, 17 occasional Scattered Delimara 18 Middle and 19 Lower 20 Globigerina, 21 Limestone 22 outcrops 23 Concrete wall 24 (relatively Scattered 25 smooth) 26 Upper Globigerina 27 Limestone 28 coastline, 29 with 30 San Lucian occasional Scattered 31 promontory, Middle and 32 Qajjenza, Lower 33 Birzebbuga Globigerina, 34 Limestone 35 outcrops 36 >1000 37 <10 38 Middle and 39 Upper 100 to 40 Globigerina 1000 41 Limestone 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 50 of 81

1 2 3 platform 4 5 Middle/Lowe ~100 6 Munxar Point, r Globigerina ~10 7 Birzebbuga Limestone 8 coastline Scattered 9 Upper 10 Globigerina 11 Limestone 12 coastline, with 13 For Peeroccasional Review ~10 14 St George Bay Middle and 15 Lower 16 Globigerina, 17 Limestone 18 outcrops 19 ~100

20 >1000

21 Weathered/er Ghar Lapsi main 22 oded concrete ~10 area 23 platform 24 Upper 25 Ghajn Tuffieha to coralline 26 IlMixquqa area, Limestone ~10 27 Mellieha boulder and 28 pebbles 29 Serpulids ( Hydroides 30 brachyacanthus , H. 31 diramphus , H. 32 elegans , H. B. Intertidal heterocerus , H. 33 Bp_19 E Turkey Meydan Köy Epibiosis pharaonis rock homoceros , H. minax , 34 H. operculatus , 35 Pomatoleios kraussii 36 and Spirobranchus 37 tetraceros 38 B. Bp_20 E Turkey Iskenderun Bay Associated species Pseudonereis anomala 39 pharaonis 40 B. Calabria (Vibo Artificial 41 Bp_21 W Italy Checklist pharaonis Valentia) hard substrate 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 51 of 81 Journal of Biogeography

1 2 Gulf of Naples (harbour 3 4 Gulf of Taranto wall) 5 Artificial Mytilaster minimus - 6 hard substrate Mytilus 7 B. Artificial galloprovincialis- Bp_22 C Italy Calabrian shores Checklist 200 8 pharaonis hard substrate Corallina, 9 Fish Chaetomorpha and 10 farming nets Ulva species B. 11 Bp_23 E Lebanon Checklist 12 pharaonis 13 B. Shallow Bp_24 E Israel TelFor Shiqmona AssociatedPeer species ReviewFlatworm 14 pharaonis intertidal 15 B. Bp_25 C Croatia Checklist 38.5 16 pharaonis 17 B. Adriatic Sea 18 Bp_26 C Italy Checklist pharaonis (Trieste) 19 20 Bp_27 B. variabilis C Italy Eastern Sicily First record 21 22 Bp_28 B. variabilis C Italy Eastern Sicily Checklist 23 Sicily Calabrian 24 Bp_29 B. variabilis C Italy Checklist coasts 25 26 Bay of Đskenderun 27 Mersin Bay

28 Mugla Province Shallow 29 [Iztuzu Dalyan] intertidal B. 30 Bp_30 E Turkey AnamurAydincik First record rocks (natural >50 20 38.5 31 pharaonis Antalya hard 32 substrata) Meydan Koyu 33 Karabur um 34 35 Peninsula 36 Iskenderun Bay Pachygrapsus 37 B. Epibiosis alien Bp_31 E Turkey marmoratus - beds 27.5 38.5 pharaonis species 38 Pilumnus hirtellus 39 Antalya Bay 40 B. Bp_32 C Malta Checklist 41 pharaonis 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 52 of 81

1 2 3 B. Biomonitoring MDR Rocks at Bp_33 E Israel 4 pharaonis transporters waterline 5 6 Bp_34 B. variabilis E Israel Checklist 7 8 Comparative B. morphology and 9 Bp_35 E Israel 10 pharaonis cytology (mesocosm) 11 12 Brachidontes 13 pharaonis- For Peer ReviewDendropoma 14 15 petraeum community - Osilinus turbinatus - 16 Intertidal Pachygrapsus 17 B. Bp_36 E Israel assemblages marmoratus - Chiton pharaonis 18 description olivaceus-Actinia 19 aequina - Mytilaster 20 minimus - Mytilus 21 galloprovincialis - 22 Ostrea edulis and 23 patellids 24 Bp_37 B. variabilis E Egypt Checklist 25 B. Review 26 Bp_38 27 pharaonis (state of the art) 28 B. Review Bp_39 29 pharaonis (state of the art) 30 B. Review Bp_40 31 pharaonis (state of the art) 32 33 B. Distribution, Bp_41 E Israel Rock Patch 34 pharaonis associated fauna list 35 Bioassays for 36 antifouling treatment 37 optimization and B. 38 Bp_42 C Italy Sicily, Siracusa management, pharaonis 39 Biofouling in 40 industrial water 41 systems 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 53 of 81 Journal of Biogeography

1 2 3 Ventrosia ventrosa - 4 Pirenella conica 5 - Loripes lacteus - Cerastoderma 6 B. Sicily (Salina di Bp_43 C Italy Associated species glaucum - Nassarius pharaonis Marsala) 7 costulatus - Conus 8 mediterraneus - 9 Gibbula adansonii - 10 Cerithium vulgatum 11 B. Sicily Bp_44 C Italy First record 12 pharaonis (Salina di Marsala) 13 B. Sicily Bp_45 C Italy For PeerFirst record Review 14 pharaonis (Salina di Marsala) 15 Ventrosia ventrosa - 16 Pirenella conica 17 - Loripes lacteus - Cerastoderma 18 B. Bp_46 C Italy Sicily (Trapani) Associated species glaucum - Nassarius 19 pharaonis 20 costulatus - Conus mediterraneus - 21 Gibbula adansonii - 22 Cerithium vulgatum 23 B. SouthWestern 24 Bp_47 C Italy Associated species pharaonis Sicily 25 B. 26 Bp_48 C Italy Western Sicily First record 27 pharaonis B. Review 28 Bp_49 29 pharaonis (state of the art) 30 B. Bioaccumulation of Intertidal Bp_50 E Turkey 10 / 20 31 pharaonis heavy metals rocky shore 32 B. 33 Bp_51 E Lebanon Checklist 34 pharaonis 35 Intertidal subtidal 36 B. Turkey, Mersin >500 Bp_52 E Turkey Bioerosive effect natural hard Patella sp 10 / 20 20 38.5 37 pharaonis Bay, Mezitli Park (bed) substrata rock 38 (limestone) 39 B. Bp_53 E Israel Checklist 40 pharaonis 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 54 of 81

1 2 B. 3 Bp_54 E Cyprus Checklist 4 pharaonis 5 Intertidal B. subtidal >500 6 Bp_55 E Turkey Mersin Bay Heavy metals effects 30 / 40 25 35.64 7 pharaonis natural hard (bed) 8 substrata Inventory alien 9 B. Bp_56 E Greece Cyprus species 10 pharaonis state of the art 11 B. Syria 12 Bp_57 E Checklist 13 pharaonis Turkey For Peer Review 14 Bp_58 B. variabilis E Greece Checklist 15 Study of sympatric Modiolus auriculatus- 16 Bp_59 B. variabilis C Malta Checklist beds 17 species Mytilaster minimus B. 18 Bp_60 E Turkey Heavy metals effects 19 pharaonis 20 B. Sicily, Stagnone di Btbased pesticide Bp_61 C Italy / 15 21 pharaonis Marsala Lagoon effetcs 22 Intertidal 23 B. Venice Lagoon Bp_62 C Italy Checklist natural hard 15 28.5 24 pharaonis Grado substrata 25 Inventory alien 26 B. Sicily, Stagnone di Bp_63 C Italy species state of the 27 pharaonis Marsala Lagoon 28 art 29 Shallow Green Algae 30 B. Rosh Haniqra Intertidal Bp_64 E Israel Checklist Vermetus Barnacles 20 38.5 pharaonis Akhziv MPA natural 31 Limpets 32 substrata B. 33 Bp_65 W Italy Corsica Checklist 34 pharaonis B. 35 Bp_66 E Israel Predation 36 pharaonis B. 37 Bp_67 E Israel Competition 38 pharaonis B. 39 Bp_68 E Israel Checklist pharaonis 40 B. Birzebbugia Bay, Distribution, Rocky shore Lepidochitona Dense Bp_69 C Malta 3 / 25 41 pharaonis within Marsaxlokk associated fauna (Globigerina caprearum - Osilinus cluster 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 55 of 81 Journal of Biogeography

1 2 3 Harbour limestone turbinatus - Tricolia 4 bedrock) pullus pullus - 5 fishfarm rigs Gibbula adansonii - Gibbula rarilineata - 6 Gibbula umbilicaris - 7 Rissoa similis - Setia 8 turriculata - Alvania 9 mamillata - Caecum 10 trachea - Cerithium 11 vulgatum - Bittium 12 reticulatum - 13 For Peer ReviewColumbella rustica - 14 Hexaplex trunculus - 15 Stramonita 16 haemastoma - 17 Nassarius cuvierii - 18 Gibberula philippi - Granulina marginata - 19 Mytilus 20 galloprovincialis - 21 Retusa truncatula - 22 Chrysallida 23 interstincta - 24 Odostomia unidentata 25 - Arca noae - Pinctada 26 radiata - Nucula 27 nitidosa - Chlamys 28 varia - Parvicardium 29 scriptum - Venerupis 30 aurea - Holothuria 31 polii - Paracentrotus lividus - Anemonia 32 viridis - Clibanarius 33 erythropus - 34 Chthamalus stellatus - 35 Crangon crangon - 36 Ligia italica - 37 Pachygrapsus 38 marmoratus - 39 Modiolus barbatus 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 56 of 81

1 2 Dynamic energy 3 B. Bp_70 C Italy Sicily budget 4 pharaonis 5 parameterisation 6 Shallow 7 intertidal B. Egypt (Beirut Biological indicators subtidal 8 Bp_71 E Lebanon 100500 20 / 35 20 38.5 pharaonis Batroum) for cadmium natural hard 9 substrata 10 (rock) 11 Hydrocarbon B. 12 Bp_72 E Syrian Chlorinated pharaonis 13 For PeerCompounds Review 14 Inventory alien B. 15 Bp_73 species state of the pharaonis 16 art 17 B. Bp_74 E Cyprus Checklist 18 pharaonis 19 B. Bp_75 E Egypt Checklist 20 pharaonis 21 Horizontal flat rocks 22 Stramonita with many 23 B. haemastoma Bp_76 E Israel Akhziv incisions and 24 pharaonis predatory effect holes 25 mesocosm (infralittoral 26 boulders) 27 Narrow and 28 offshore 29 platforms 30 beachrock 31 Subtidal Beds Palmachim 010 / 3040 32 bedrock patches Subtidal 33 Green Algae B. Distribution patterns platform 34 Bp_77 E Israel Vermetus Barnacles 20 38.5 pharaonis density walls and 35 Limpets 36 boulders 37 Narrow and 38 offshore platforms Beds 39 BatYam Subtidal patches 40 bedrock 41 Subtidal 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 57 of 81 Journal of Biogeography

1 2 3 platform 4 walls and 5 boulders 6 Intermediate 7 platform Individual 8 Subtidal Mikhmoret small platform 9 patches 10 walls and 11 boulders 12 Narrow and 13 offshore For Peerplatforms Review 14 Incisioned 15 rock Beds 16 Beachrock patches 17 Intertidal Akhziv Individual 18 walls small 19 Subtidal patches 20 bedrock 21 Subtidal 22 platform 23 walls and 24 boulders 25 Narrow and 26 offshore 27 platforms Horizontal 28 Beds vermetid patches 29 ledge HaBonim Individual 30 Intertidal small 31 walls patches 32 Subtidal 33 platform 34 walls and 35 boulders 36 Beachrock 37 Subtidal TelBaruch platform Patches 38 39 walls and 40 boulders 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 58 of 81

1 2 3 Intermediate platform 4 Individual Subtidal 5 Shiqmona small platform 6 patches walls and 7 boulders 8 Intermediate 9 platform Individual 10 Subtidal Atlit small 11 platform patches 12 walls and 13 For Peerboulders Review 14 Beachrock 15 Subtidal Rishpon platform Patches 16 17 walls and 18 boulders 19 Intertidal 20 boulders Subtidal Small 21 Olga platform patches 22 walls and 23 boulders 24 Intertidal 25 boulders 26 Subtidal Small NofYam 27 platform patches 28 walls and 29 boulders 30 Subtidal 31 bedrock Subtidal 32 BatYam Patches 33 platform walls and 34 boulders 35 Inventory alien B. 36 Bp_78 species state of the pharaonis 37 art 38 Tlul Port Said Intertidal Mytilaster minimus - 39 B. [All other Israeli Competition field natural hard Green Algae Bp_79 E Israel Beds 20 38.5 40 pharaonis sites are the same study substratum Vermetus Barnacles 41 of Rilov et al (rocks) Limpets 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 59 of 81 Journal of Biogeography

1 2 3 2004] 4 5 6 7 seaward Individual Competition field dipping cluster 8 Bp_80 B. variabilis E Israel Palmachim Mytilaster minimus 14 / 16 20 38.5 9 study beachrock patch 10 slabs beds Laurencia papillosa - 11 submerged B. Sicily (Stagnone di Clearance, filtration Padina pavonica - 18.6 12 Bp_81 C Italy hard patches / 30 47 ± 4.3 0.8 ± 0.4 pharaonis Marsala) and ingestion rate Acetabularia ± 7.4 13 substrates For Peer Reviewacetabulum 14 15 mediolittoral– 16 upper Sources of carbon infralittoral Chaetomorpha linum, 17 B. Sicily (Stagnone di Bp_82 C Italy and dietary habits hard Cystoseira sp., 20 / 30 18 pharaonis Marsala) (isotopes analysis) substrates Laurencia papillosa 19 (both natural 20 and artificial) 21 Intertidal 22 Density and B. natural and clusters Bp_83 C Italy Sicily biometrical Mytilaster minimus 20 23 pharaonis artificial hard patches measures 24 substrata 25 26 Integrated isotopic, biochemical and 9387 ± 27 B. Sicily (Stagnone di transplant study on 4366 Bp_84 C Italy ~1.0 28 pharaonis Marsala) the origin and (annual 29 quality of organic mean) 30 matter 31 Mytilaster minimus - 32 Ammonium, Nitrates Mytilus B. Sicily (Gulf of Fouling (fish 33 Bp_85 W Italy and Orthophosphates galloprovincialis - 20 / pharaonis Castellammare) farm nets) 34 responses Modiolus barbatus - 35 macroalgae 36 Cymodeocea nodosa - 37 Density Laurencia papillosa - B. Sicily, Stagnone di 38 Bp_86 C Italy demography Padina pavonica - pharaonis Marsala Lagoon 39 resource allocation Acetabularia 40 acetabulum 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 60 of 81

1 2 Absorbtion efficency 3 B. Sicily, Stagnone di Bp_87 C Italy Temp Sal effects 4 pharaonis Marsala Lagoon 5 (mesocosm) Sicily (Capo Plaia, 6 37.4 7 Cefalù) Heart beat rate B. adaptations to Rocks at 22.0 Bp_88 W Italy Mytilaster minimus Beds 8 pharaonis varying salinity waterline ± 0.5 9 Marsala Lagoon (mesocosm) 40 10 11 B. Bp_89 Modelling 12 pharaonis 13 B. Bp_90 For Peer Modelling Review 14 pharaonis 15 Modelling Climate 16 B. change Bp_91 17 pharaonis Ecophysiological 18 study review Inventory alien 19 B. Bp_92 C Malta species state of the 20 pharaonis 21 art 22 Achziv 23 B. Shemen Bp_93 E Israel Molecular study 24 pharaonis Tel Shiqmona 25 Ashqelon 26 27 Intertidal B. natural and 28 Bp_94 E Lebanon Beirut area 10 / 20 20 38.5 29 pharaonis artificial hard 30 substrata 31 B. Bp_95 E Greece Checklist 32 pharaonis 33 34 35 Genetic variation B. Sicily (Stagnone di 36 Bp_96 W Italy study (molecular pharaonis Marsala) 37 phylogeography) 38 39 40 B. Sicily (Salina di Analysis of 41 Bp_97 W Italy 42 pharaonis Marsala) molecular variance 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 61 of 81 Journal of Biogeography

1 2 Sicily (Termini (mitocondrial COI 3 4 Imerese) sequences) Sicily (Torre 5 6 Normanna) Sicily (Capo 7 8 D’Orlando) 9 Heavy metal 10 Bp_98 B. variabilis E Turkey accumulation Beds 11 (mesocosm) B. 12 Bp_99 C Italy Trieste Checklist 13 pharaonis B. For Peer Review 14 Bp_100 C Italy Trieste Checklist pharaonis 15 Sicily (Trapani, Cytogenetic Sandstone 16 B. Bp_101 C Italy Saline Ettore characterization cubic 17 pharaonis 18 Infersa) (molecular study) boulders (Artifical 19 Ascidians, bryozoans, hard 20 B. IlQajjenza algae, sessile Bp_102 C Malta species coexistence substrate) fish 21 pharaonis (Marsaxlokk Bay) polychaetes and farm buoys 22 bivalves 23 fouling 24 Bp_103 B. variabilis C Italy Calabrian coast First record 25 B. Bp_104 Review 26 pharaonis 27 Inventory alien B. 28 Bp_105 E Greece species state of the pharaonis 29 art 30 B. Bp_106 E Lebanon Checklist 31 pharaonis 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 62 of 81

1 2 3 AbiGhanem C., Khalaf G., and Najjar E. 2014. Distribution of Lead, Cadmium, and Vanadium in Lebanese 4 Bp_1 5 Coastal Sediments and Mussels. Journal of Coastal Research 30(5):10741080 Açik . 2008. Occurrence of the alien species Aspidosiphon (Aspidosiphon) elegans (Sipuncula) on the 6 Bp_2 7 levantine and aegean coasts of Turkey. Turkish Journal of Zoology 32(4):443448 Arcidiacono A, and Di Geronimo I. 1976. Studio biometrico di alcuni campioni di Brachidontes variabilis 8 Bp_3 9 (Krauss). Conchiglie 12(34):6174 Arizza A., Zenone A., Giaramita F.T., Rinaldi A., Sarà G. 2008. Heat shock proteins (HSP) in Brachidontes 10 Bp_4 11 pharaonis (Mollusca, Bivalvia) at varying temperatures. Biologia Marina Mediterranea 15(1):404405 12 Barash A. and Danin, Z. 1992. Fauna Palaestina: Mollusca I. Annotated list of Mediterranean molluscs of Bp_5 13 Israel and Sinai. The Israel Academy of Sciences and Humanities, Jerusalem 14 Bitar G., Dupuy de la Grandrive R., Foulquié M. 2003. Second mission relating to the Development of 15 Bp_6 Marine Protected Areas on Syrian coasts, 118 August 2003. Mission Report. Regional Project for the 16 Development of Marine and Coastal Protected Areas in the Mediterranean Region. UNEP pp. 40 17 Bitar G. 2014. Exotic molluscs from the Lebanese coast. Bulletin de la Société zoologique de France 139(1 Bp_7 18 4):3745 For Peer Review 19 Bonnici L., Evans J., Borg J.A:, Schembri P.J. 2012. Biological aspects and ecological effects of a bed of the 20 Bp_8 invasive nonindigenous mussel Brachidontes pharaonis (Fischer P., 1870) in Malta. Mediterranean Marine 21 Science 13(1):153161 22 Bresler V., Abelson. A., Feldstein T., Mokady O., Fishelson L., Rosenfeld M. 2003. Marine molluscs in Bp_9 23 environmental monitoring I. Cellular and molecular responses. Helgoland Marine Research 57(34):157165 24 Boudouresque C.F. 1999. The Red Sea Mediterranean link: unwanted effects of canals. Invasive Species Bp_10 25 and Biodiversity Management, Kluwer Academic Publishers, Dordrecht, the Netherlands 26 Buzzurro G., Greppi E. 1996. The Lessepsian molluscs of Taucu (SouthEast Turkey). La Conchiglia Bp_11 27 28(279):322 28 Buzzurro G., Greppi E. 1997. Note e considerazioni sui molluschi di Cipro con particolare riguardo alle Bp_12 29 specie alloctone. La Conchiglia 29(283):2131 30 Carlier A. 2007. Apports des isotopes stables a la description de l’architecture et du fonctionnement des 31 Bp_13 réseaux trophiques benthiques de plusieurs environnements côtiers du Golfe du Lion (Méditerranée Nord 32 Occidentale). PhD Thesis, Université Pierre et Marie Curie Paris 6, Paris. 201 pp 33 Carlier A., Riera P., Amouroux J.M., Bodiou J.Y., Desmalades M., Grémare A. 2009. Spatial heterogeneity 34 Bp_14 in the food web of a heavily modified Mediterranean coastal lagoon: stable isotope evidence. Aquatic 35 Biology 5:167179 36 Caruso M., Romano C., Sarà G. 1999. Seston food availability for suspension feeder molluscs in two areas Bp_15 37 with different hidrodinamic features. Biologia Marina Mediterranea 6(1) 38 Cecalupo A., Quadri P. 1996. Contributo alla conoscenza malacologica per il nord dell’isola di Cipro (Terza Bp_16 39 e ultima parte). Bollettino Malacologico 31(58):95118 40 Çevik C., Erkol I.L., Toklu B. 2006. A new record of an alien jellyfish from the Levantine coast of Turkey 41 Bp_17 Cassiopea andromeda (Forsskål, 1775) [Cnidaria: Scyphozoa: Rhizostomea]. Aquatic Invasions 1:196197 42 Cilia D.P., Deidun A. 2012. Branching out: Mapping the spatial expansion of the lessepsian invader mytilid 43 Bp_18 Brachidontes pharaonis around the Maltese Islands. Marine Biodiversity Records e(e28):18 44 Çinar M.E. 2006. Serpulid species (Polychaeta: Serpulidae) from the Levantine coast of Turkey (eastern 45 Bp_19 Mediterranean), with special emphasis on alien species. Aquatic Invasions 1(4):223240 46 47 Çinar M.E., Altun C 2007. A preliminary study on the population characteristics of the Lessepsian species 48 Bp_20 Pseudonereis anomala (Polychaeta: Nereididae) in Đskenderun Bay (Levantine Sea, Eastern Mediterranean). 49 Turkish Journal of Zoology 31(4):403410 Crocetta F., Renda W., Colamonaco G. 2009. New distributional and ecological data of some marine alien 50 Bp_21 51 molluscs along the southern Italian coasts. Marine Biodiversity Records 2(e23):17 Crocetta F., W. Renda, Vazzana A. 2009. Alien Mollusca along the Calabrian shores of the Messina Strait 52 Bp_22 53 area and a review of their distribution in the Italian seas. Bollettino Malacologico 45(1):1530 Crocetta F., Bitar G., Zibrowius H., Oliverio M. 2013. Biogeographical homogeneity in the eastern 54 Bp_23 55 Mediterranean Sea. II. Temporal variation in Lebanese bivalve biota. Aquatic Biology 19(1):75109 CuriniGalletti M., Campus P. 2007. Boninia neotethydis sp. nov. (Platyhelminthes: Polycladida: Cotylea)— 56 Bp_24 57 the first lessepsian flatworm. Journal Marine Biolpgical Association UK 87:435–442 58 59 60 Page 63 of 81 Journal of Biogeography

1 De Min R., Vio E. 1997. Molluschi conchiferi del litorale sloveno. Annals for Istrian and Mediterranean 2 Bp_25 3 Studies, Koper, Historia Naturalis 11:241258 4 Bp_26 De Min R., Vio E. 1998. Molluschi esotici nell’Alto Adriatico. Annales. Series historia naturalis 5 Di Geronimo I. 1971. Prima segnalazione sulle costa italiane di Brachidontes variabilis (Krauss). Bollettino Bp_27 6 delle sedute dell’Accademia Gioenia di scienze naturali in Catania 10:847852 7 Bp_28 Di Geronimo I. 1971. Molluschi rari o nuovi per le coste orientali della Sicilia. Conchiglie 7(56):6172 8 Di Natale A. 1982. ExtraMediterranean species of Mollusca along the Southern Italian Coasts. Malacologia Bp_29 9 22(12):578580 10 Doǧan A., Önen M., Öztürk B. 2007. A new record of the invasive Red Sea mussel Brachidontes pharaonis Bp_30 11 (Fischer P., 1870) (Bivalvia: Mytilidae) from the Turkish coasts. Aquatic Invasions 2(4):461463 12 Dogan A., Azcan T., Bakir K., Kataga T. 2008. Crus tacea Decapoda associated with Brachidontes pharaonis 13 Bp_31 (P. Fischer, 1870) (Mollusca, Bivalvia) beds from the Levantine coasts of Turkey. Crustaceana 81(11):1357 14 1366 15 Evans J., Barbara, J., Schembri P.J. 2015.Updated review of marine alien species and other 'newcomers' Bp_32 16 recorded from the Maltese Islands (Central Mediterranean). Mediterranean Marine Science 16(1):225244 17 Feldstein T., Nelson N., Mokady O. 2006. Cloning and expression of MDR transporters from marine Bp_33 18 bivalves, and theirFor potential use Peer in biomonitoring. ReviewMarine Environmental Research 62:118121 19 Felsenburg T., Safriel U. 1974. Colonization of eastern Mediterranean intertidal zone by Indopacific mussel, Bp_34 20 Brachidontes variabilis. Israel Journal of Zoology 23:212213 21 Fishelson L. 2000. Comparative morphology and cytology of siphons and siphonal sensory organs in Bp_35 22 selected bivalve molluscs. Marine Biology 137(3):497509 23 Fishelson L. 2000. Marine animal assemblages along the littoral of the Israeli Mediterranean seashore: The Bp_36 24 Red Mediterranean Seas communities of species. Italian Journal of Zoology 67(4):393415 25 Fuchs Th. 1978. Die geologische Beschaffenheit der Landenge von Suez. Denkschriften der Kaiserkichen Bp_37 26 Akademie der Wissenschaften, MathematischNaturwissenschaftliche Classe 38:25 27 Galil B.S. 2007. Loss or gain? Invasive aliens and biodiversity in the Mediterranean Sea. Marine Pollution Bp_38 28 Bulletin 55(7–9):314–322 29 Galil B.S. 2007. Seeing Red: Alien species along the Mediterranean coast of Israel. Aquatic Invasions 30 Bp_39 2(4):281312 31 Galil B.S. 2008. Alien species in the Mediterranean Sea Which, when, where, why? Hydrobiologia 32 Bp_40 606(1):105116 33 34 Galil B., Gertman I., Nurit G., Barak H., Israel A., Lubinevsky H., Rilov G., Rinkevich B. 2013. Biodiversity 35 Bp_41 monitoring along the Israeli coast of the Mediterranean activities and accumulated data. Israel 36 Oceanographic and Limnological Research contribution (IOLR Report) 37 Garaventa F., Corra C., Piazza V., Giacco E., Greco G., Pane L., Faimali M. 2012. Settlement of the alien 38 Bp_42 mollusc Brachidontes pharaonis in a Mediterranean industrial plant: Bioassays for antifouling treatment 39 optimization and management. Marine Environmental Research 76:906 Gianguzza P., Chemello R., Cicciari A., Riggio S. 1997. Struttura del popolamento a molluschi della vasca 40 Bp_43 41 di fredda di una salina marsalese. Biologia Marina Mediterranea 4(1):396398 42 Gianguzza P., Chemello R., Riggio S. 1998. Segnalazione di Brachidontes pharaonis (P. Fischer, 1870) 43 Bp_44 (Bivalvia, Mytilidae) nella salina di Marsala e considerazioni sulla distribuzione della specie in 44 Mediterraneo. Bollettino Malacologico 33(912):169172 45 Gianguzza P., Sarà G., Chemello R., Riggio S. 1998. Note su una popolazione a Brachidontes pharaonis Bp_45 46 (Fischer P.) (Bivalvia, Mytilidae) in una Salina marsalese. Biologia Marina Mediterranea 5(1):561562 47 Gianguzza P., Zava B., Riggio S. 2001. Descrizione del popolamento a molluschi della salina "Grande" di 48 Bp_46 Trapani e Paceco (Tp, Sicilia). XIII S. It. E. Congress. Como, Italy. Ecologia, Edited by: Casagrandi, R. and 49 Melià, P. Gianguzza P., Chemello R., Riggio S. Composizione e struttura della malacofauna di una salina della Sicilia 50 Bp_47 51 occidentale. Bollettino Malacologico 36(912):201207 Gianguzza P. 2001. Nuova segnalazione di Brachidontes pharaonis (Fischer P., 1870) (Bivalvia, Mytilidae) 52 Bp_48 53 lungo le coste della Sicilia occidentale. Biogeographia XXII:259262 Gofas S., Zenetos A. 2003. Exotic molluscs in the Mediterranean basin: Current status and perspectives. 54 Bp_49 55 Oceanography and Marine Biology 41:237277 56 Goksu M.Z.L., Akar M., .Cevik F., Findik O. 2005. Bioaccumulation of some heavy metals (Cd, Fe, Zn, Cu) 57 Bp_50 in two Bivalvia species (Pinctada radiata Leach, 1814 and Brachidontes pharaonis Fischer, 1870). Turk J Vet 58 Anim Sci 29:8993 59 60 Journal of Biogeography Page 64 of 81

1 2 Gruvel A., Moazzo G. 1931. Contribution à la faune malacologique marine des côtes LibanoSyriennes. 3 Bp_51 Gruvel A. (ed.), Les états de Syrie. Richesses marines et fluviales. Société des Editions Géographiques, 4 Maritimes et Coloniales, Paris Guel M., Oezbek A., Karayakar F., Kurt A. 2008. Biodegradation effects over different types of coastal 5 Bp_52 6 rocks. Environmental Geology 55:16011611 Haas G. 1937. Mollusca marina, in F.S. Bodenheimer, Prodromus Faunae Palestinae. Mem. Inst. Egypte 7 Bp_53 8 33:275280 9 Hadjichristophorou M., Argyrou M., Demetropulos A., Bianchi T.S. 1997. A species list of the sublittoral Bp_54 10 softbottom macrobenthos of Cyprus. Acta Adriatica 38(1):332 11 Karayakar F., Erdem C., Cicik B. 2007. Seasonal variation in copper, zinc, chromium, lead and cadmium 12 Bp_55 levels in hepatopancreas, gill and muscle tissues of the mussel Brachidontes pharaonis Fischer, collected 13 along the Mersin coast, Turkey. Bulletin of Environmental Contamination and Toxicology 79(3):350355 14 Katsanevakis S., Tsiamis K., Ioannou G., Michailidis N., Zenetos A. 2005. Inventory of alien marine species Bp_56 15 of Cyprus. Mediterranean Marine Science 10(2):109113 16 Kinzelbach R. 1985. Lesseps'sche Wanderung: neue stationen von Muscheln (Bivalvia: Anisomyaria). Bp_57 17 Archiv fur Molluskenkunde 115(46):273278 18 Bp_58 Koroneos J. 1979.For Les Mollusques Peer de la Grèce. Athèn Reviewes 36:48 19 Bp_59 Lanfranco G. 1975. Some additions to the local Mollusca. The Maltese Naturalist 2,27 20 Leach P., Fischer B. 2005. Bioaccumulation of Some Heavy Metals ( Cd , Fe , Zn , Cu ) in Two Bivalvia 21 Bp_60 Species. Turk J VetAnim Sc 29:8993 22 Manachini B., Arizza V., Rinaldi A., Montalto V., Sarà G. 2013. Ecophysiological response of two marine 23 Bp_61 24 bivalves to acute exposition to commercial Btbased pesticide. Marine Environmental Research 83:2937 Marchini A., Marchini C. 2006. A fuzzy logic model to recognise ecological sectors in the lagoon of Venice 25 Bp_62 26 based on the benthic community. Ecological Modelling 193:105–118 27 Mazzola A., Vizzini S. 2005. Caratteristiche ecologiche, fattori di pressione antropica e sviluppo sostenibile 28 Bp_63 di un ambiente costiero mediterraneo (Stagnone di Marsala, Sicilia Occidentale). Naturalista Siciliano IV, 29 XXIX (12), 3765. 30 MedMPA. 2004. Marine Biodiversity Study of the Rosh HaniqraAkhziv Nature Reserves (Israel) to the 31 Bp_64 Establishment of a Management Plan. Regional Project for the Development of Marine and Coastal 32 Protected Areas in the Mediterranean Region (MedMPA). Final report 33 Merella P., Porcheddu A., Casu S. 1994. La malacofauna della riserva naturale di Scandola (Corsica Nord Bp_65 34 occidentale). Bollettino Malacologico 30(59):111128 35 Bp_66 Mienis H.K. 2002. A case of predation on mussels by the hite Sea Bream. Spirula 325:27 36 Bp_67 Mienis H.K. 2003. Native marine molluscs replaced by Lessepsian migrants. Tentacle 11:1516 37 Mienis H.K. 2004. New data concerning the presence of Lessepsian and other indopacific migrants among 38 Bp_68 the mollusca in the Mediterranean Sea with emphasize on the situation in Israel. 1st National Malacology 39 Congress 40 Mifsud C., Cilia D.P. 2009. On the presence of a colony of Brachidontes pharaonis (P. Fischer, 1870) Bp_69 41 (Bivalvia : Mytilidae) in maltese waters (Central Mediterranean). Triton 20:12 42 Montalto, V., Palmeri V., Rinaldi A., Kooijman S.A.L.M and G. Sarà. 2014. Dynamic energy budget 43 Bp_70 parameterisation of Brachidontes pharaonis, a Lessepsian bivalve in the Mediterranean Sea. Journal of Sea 44 Research 94:4751 45 Nakhle K. F., Cossa D., Khalaf G., Beliaeff B. 2006. Brachidontes variabilis and Patella sp as quantitative 46 Bp_71 biological indicators for cadmium, lead and mercury in the Lebanese coastal waters. Environmental 47 Pollution 142(1):7382 48 Noureddin S., Ali H.K., Ammar I., Abbass G., Baker M., Arabiah I., Abdow O. 2011. Using an International 49 Bp_72 Monitoring Net for Hydrocarbon Chlorinated Compounds at Syrian Sea Water. Tishreen University Journal 50 for Research and Scientific Studies Basic Sciences Series 33(1):121138 51 Occhipintiambrogi A., Galil B. 2010. Marine alien species as an aspect of global change. Advances in Bp_73 52 Oceanography and Limnology 1(1): 199218 53 Öztürk B., Buzzurro G., Avni Benli H. 2004. Marine molluscs from Cyprus: new data and checklist. 54 Bp_74 Bollettino Malacologico 39(58):4978 55 Pallary P. 1912. Catalogue des mollusques du littoral méditerranéen de l’Egypte. Mém. Inst. Egypte 7:69 Bp_75 56 207 57 58 59 60 Page 65 of 81 Journal of Biogeography

1 Rilov G., Gasith A., Benayahu Y. 2002. Effect of an exotic prey on the feeding pattern of a predatory snail. 2 Bp_76 3 Marine Environmental Research 54(1):8598 Rilov G., Benayahu Y., Gasith A. 2004. Prolonged lag in population outbreak of an invasive mussel: a 4 Bp_77 5 shiftinghabitat model. Biological Invasions 6:347364 6 Rilov G., Galil B. 2009. Marine Bioinvasions in the Mediterranean Sea – History, Distribution and Ecology. Chapter 31 in Biological Invasions in Marine Ecosystems (book) Olyarnik S.V., Bracken M.E.S., Byrnes 7 Bp_78 8 J.E., Hughes R., Hultgren K.M., Stachowicz J.J., Rilov G., Crooks J., Caldwell M.M., Heldmaier G., Jackson 9 R.B., Lange O.L., Mooney H., Schulze E.D., Sommer U. eds 10 Safriel U.N., Gilboa A., Felsenburg T. 1980. Distribution of Rocky Intertidal Mussels in the Red Sea Coasts 11 Bp_79 of Sinai, the Suez Canal and the Mediterranean Coast of Israel, with Special Reference to Recent Colonizers. 12 Journal of Biogeography 7(1):3962 13 Safriel U.N., SassonFrostig Z. 1988. Can colonizing mussel outcompete indigenous mussel? Journal of Bp_80 14 Experimental Marine Biology and Ecology 117(3):221226 15 Sarà G., Romano C. 2000. The new Lessepsian entry Brachidontes pharaonis (Fischer P., 1870) (Bivalvia, 16 Bp_81 Mytilidae) in the western Mediterranean: A physiological analysis under varying natural conditions. Journal 17 of Shellfish Research 19:967977 18 Sara G., Vizzini S., Mazzola A. 2003. Sources of carbon and dietary habits of new Lessepsian entry Bp_82 For Peer Review 19 Brachidontes pharaonis (Bivalvia, Mytilidae) in the western Mediterranean. Marine Biology 143(4):713722 20 Sarà G., Buffa G. 2004. Density and biometrical features of two cooccurring bivalves (Mytilaster minimus 21 Bp_83 and Brachidontes pharaonis) in Western Sicily (South Tyrrhenian). Proceedings 4° National Symposium 22 CONISMAAIOL, Terrasini (PA) 2004, p 132 23 Sarà G. 2006. Hydrodynamic effects on the origin and quality of organic matter for bivalves: an integrated Bp_84 24 isotopic, biochemical and transplant study. Marine Ecology Progress Series 328:6573 25 Sarà G., Lo Martire M., Buffa G., Mannino A.M., Badalamenti F. 2007. The fouling community as an Bp_85 26 indicator of fish farming impact in Mediterranean. Aquaculture Research 38(1):6675 27 Sará G., Romano C., Mazzola A. 2008. A new lessepsian species in the western Mediterranean ( 28 Bp_86 Brachidontes pharaonis Bivalvia: Mytilidae): density, resource allocation and biomass. Marine Biodiversity 29 Records 1(e8) 30 Sarà G., Romano C., Widdows J. Staff F.J. 2008. Effect of salinity and temperature on feeding physiology 31 Bp_87 and scope for growth of an invasive species (Brachidontes pharaonis MOLLUSCA : BIVALVIA) within 32 the Mediterranean sea. Journal of Experimental Marine Biology and Ecology 363(12):130136 33 Sara G., de Pirro M. 2011. Heart beat rate adaptations to varying salinity of two intertidal Mediterranean 34 Bp_88 bivalves: The invasive Brachidontes pharaonis and the native Mytilaster minimus. Italian Journal of Zoology 35 78(2):193197 36 Sara G., Palmeri V., Montalto V., Rinaldi A., Widdows J. 2013. Parameterisation of bivalve functional traits 37 Bp_89 for mechanistic ecophysiological dynamic energy budget (DEB) models. Marine Ecology Progress Series 38 328:6573 39 Sara G., Palmeri V., Montalto V., Rinaldi A., Helmuth B. 2013. Predicting biological invasions in marine 40 Bp_90 habitats through ecophysiological mechanistic models: a case study with the bivalve Brachidontes 41 pharaonis. Diversity and Distributions 19(10):12351247 42 43 Sara G., Milanese M., Prusina I., Sarà A., Angel D.L., Glamuzina B., Nitzan T., Freeman S., Rinaldi A., Palmeri V., Montalto V., Lo Martire M., Gianguzza P., Arizza V., Lo Brutto S., De Pirro M. , Helmuth B., 44 Bp_91 Murray J. , De Cantis S., Williams G.A. 2004. The impact of climate change on mediterranean intertidal 45 communities: losses in coastal ecosystem integrity and services. Regional Environmental Change 14:S517 46 Sciberras M., Schembri P.J. 2007. A critical review of records of alien marine species from the Maltese 47 Bp_92 48 Islands and surrounding waters (Central Mediterranean). Mediterranean Marine Science 6(1):4166 Shefer S., Abelson A., Mokady O., Geffen E. 2004. Red to Mediterranean Sea bioinvasion: natural drift 49 Bp_93 50 through the Suez Canal, or anthropogenic transport? Molecular Ecology 13(8):23332343 Shiber J.G., Shatila T.A. 1978. Lead, cadmium, copper, nickel and iron in limpets, mussels and snails from 51 Bp_94 52 the coast of ras beirut, lebanon. Marine Envronmental Research 1: 125134 53 Bp_95 Tenekides N.S. 1989. On a collection of shells from the Greek seas. Athens: Protopapa Press pp 187 54 Terranova M.S., Lo Brutto S., Arculeo M., Mitton J.B. 2006. Population structure of Brachidontes pharaonis 55 Bp_96 (P. Fisher, 1870) (Bivalvia, Mytilidae) in the Mediterranean Sea, and evolution of a novel mtDNA 56 polymorphism. Marine Biology 150(1):89101 57 58 59 60 Journal of Biogeography Page 66 of 81

1 2 Terranova M.S., Lo Brutto S., Arculeo M., Mitton J.B. 2006. A mitochondrial phylogeography of 3 Bp_97 Brachidontes variabilis (Bivalvia : Mytilidae) reveals three cryptic species. Journal of Zoological 4 Systematics and Evolutionary Research 45(4):289298 5 Bp_98 Ünsal M. 1984. Accumulation and loss of tin by the mussel. Oceanologica Acta 7(4): 493498 6 Vio E., De Min R. 1996. Contributo alla conoscenza dei Molluschi marini del Golfo di Trieste. Atti del Bp_99 7 Museo Civico di Storia Naturale di Trieste 47:173232 8 Bp_100 Vio E., De Min R. 1999. Ritrovamenti malacologici nel Golfo di Trieste. Hydrores Information 16(17):2934 9 Vitturi R., Gianguzza P., Colomba M.S., Riggio S. 2000. Hybridization (fish) (Mollusca: Bivalvia: Bp_101 10 Mytilidae). Ophelia 52(3):213220 11 Zammit P.P., Longo C., Schembri P.J. 2009. Occurrence of Paraleucilla magna Klautau et al., 2004 Bp_102 12 (Porifera: Calcarea) in Malta. Mediterranean Marine Science 10(2):135138 13 Zanca M. 1976. Rinvenimento di esemplari di Brachidontes variabilis (Krauss, 1848) lungo la costa ionica Bp_103 14 della Calabria. Conchiglie 12(78):161162 15 Zenetos A., Gofas S., Verlaqu e M., Cinar M.E., Garcia Raso J,E., Bianchi C.N., Morri C., Azzurro E., 16 Bilecenoglu M., Froglia C., Siokou I., Violanti D., Sfriso A., San Martin G., Giangrande A., Katagan T., 17 Ballesteros E., RamosEsplà A., Mastrototaro F., Ocana A., Zingone A., Gambi M.C., Streftaris N. 2010. Bp_104 18 Alien species in Forthe Mediterranean Peer Sea by 2010. AReview contribution to the application of European Union’s 19 Marine Strategy Framework Directive (MSFD). Part I. Spatial distribution. Mediterranean Marine Science 20 11(2):381493 21 Zenetos A., Koutsogiannopoulos D., Ovalis P., Poursanidis D. 2013. The role played by citizen scientists in Bp_105 22 monitoring marine alien species in Greece. Cahiers de Biologie Marine 54(3):419426 23 Zibrowius H., Bitar, G. 2003. Invertébrés marins exotiques sur la cote du Liban. Lebanese Science Journal Bp_106 24 4(1):6774 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 67 of 81 Journal of Biogeography

1 2 3 4 APPENDIX 2 5 6 7 Table A1. Predictor environmental variables used in the models, variables short name, data sources, model and spatial resolution. 8 9 10 11 Variables Variables short name Source Model Resolution 12 Surface temperature mean ts_mean Medcordex MED11 CNRM ALADIN52 ~ 12 km 13 Surface temperature S.D.For Peerts_std ReviewMedcordex MED11 CNRM ALADIN52 ~ 12 km 14 Surface Downward Eastward Wind Stress mean tauu_mean Medcordex MED11CNRMALADIN52 ~ 12 km 15 Surface Downward Eastward Wind Stress S.D. tauu_std Medcordex MED11CNRMALADIN52 ~ 12 km 16 tauv_mean MED11CNRMALADIN52 ~ 12 km 17 Surface Downward Northward Wind Stress mean Medcordex 18 Surface Downward Northward Wind Stress S.D. tauv_std Medcordex MED11CNRMALADIN52 ~ 12 km 19 Salinity mean sos_mean Medcordex NEMOMED 8 v1 ~ 10 km 20 Chlorophyll a mean chl_mean Copernicus OPATMBFM ~ 6 km 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 68 of 81

1 2 3 4 Table A2. Variables considered for use as input into the distribution model and Variance Inflation Factor (VIF) values of the subset of the variables 5 6 selected for model development. 7 8 9 Variables Variables short name VIFs 10 11 Salinity mean sos_mean 3.523947 12 Surface Downward Eastward Wind Stress mean tauu_mean 2.413771 13 ForSurface Downward Peer Eastward Wind Stress Review S.D. tauu_std 2.490110 14 Surface Downward Northward Wind Stress mean tauv_mean 1.993681 15 Surface Downward Northward Wind Stress S.D. tauv_std 2.323726 16 17 Sea Surface temperature mean ts_mean 4.065276 18 Sea Surface temperature S.D. ts_std 2.038435 19 Chlorophyll a mean chl_mean 1.769682 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 69 of 81 Journal of Biogeography

1 2 3 4 Table A3. Relative contributions of each predictor variable to the B. pharaonis distribution model for 2030 Scenario 1 and 2050 Scenario 33. 5 6 2030 2050 7 Percent Percent Variable Variable 8 contribution contribution 9 Salinity mean 0.5 psu 81.2 Salinity mean +0.5 psu 81.2 10 11 Chlorophyll a mean 10% 5.3 Chlorophyll a mean +10% 4.6 12 Surface temperature mean 5.2 Surface temperature S.D. 4.5 13 Surface temperature S.D.For Peer 4.0 Review Surface temperature mean 3.7 14 15 Surface Downward Eastward Wind Stress S.D. 1.8 Surface Downward Northward Wind Stress S.D. 1.9 16 Surface Downward Northward Wind Stress S.D. 1.7 Surface Downward Eastward Wind Stress S.D. 1.8 17 Surface Downward Eastward Wind Stress mean 0.6 Surface Downward Northward Wind Stress mean 1.3 18 19 Surface Downward Northward Wind Stress mean 0.2 Surface Downward Eastward Wind Stress mean 0.8 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 70 of 81

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 Figure A1. A) ROC (Receiver Operating Characteristic) curves for B. pharaonis current models 2010; 46 47 AUC: Area Under the Curve. B) Jackknife tests of variable importance for the B. pharaonis current 48 49 2010 distribution model. See Table A2 for full names of environmental variables. 50 51 52 53 54 55 56 57 58 59 60 Page 71 of 81 Journal of Biogeography

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 Figure A2. A) ROC (Receiver Operating Characteristic) curves for B. pharaonis 2030 Scenario 1 46 47 model; AUC: Area Under the Curve. B) Jackknife tests of variable importance for the B. pharaonis 48 49 2030 distribution model. See Table A2 for full names of environmental variables. 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 72 of 81

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 Figure A3. A) ROC (Receiver Operating Characteristic) curves for B. pharaonis 2050 Scenario 33 48 49 model; AUC: Area Under the Curve. B) Jackknife tests of variable importance for the B. pharaonis 50 51 2050 distribution model. See Table A2 for full names of environmental variables. 52 53 54 55 56 57 58 59 60 Page 73 of 81 Journal of Biogeography

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 Figure A4. Partial dependence curves of the marginal response of B. pharaonis and the importance in 53 54 percentage of each variable for the 2010 model. 55 56 57 58 59 60 Journal of Biogeography Page 74 of 81

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 Figure A5. Scatterplot matrix of HSI and the environmental variables considered in 2010 represented 27 28 to detect correlations between variables. From the top of the panel, are represented absolute Pearson 29 30 correlations, and significance asterisks (* p < 0.05, ** p < 0.01, *** p < 0.001), histograms and kernel 31 32 density overlays. Variables short name are reported in Table A1. 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 75 of 81 Journal of Biogeography

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 76 of 81

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 77 of 81 Journal of Biogeography

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 Figure A6. Spatial distributions of predicting suitable habitat of B. pharaonis in 2030 for all 33 45 46 47 scenarios considered. 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 78 of 81

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 79 of 81 Journal of Biogeography

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Biogeography Page 80 of 81

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 Figure A7. Spatial distributions of predicting suitable habitat of B. pharaonis in 2050 for all 33 44 45 46 scenarios considered. 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 81 of 81 Journal of Biogeography

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60