Journal of Coastal Conservation https://doi.org/10.1007/s11852-018-0630-0

Effects of on early colonization stages of an artificial in North-East Atlantic

João Encarnação1 & Gonçalo Calado2,3

Received: 6 March 2018 /Revised: 12 May 2018 /Accepted: 29 May 2018 # Springer Science+Business Media B.V., part of Springer Nature 2018

Abstract Increasing interest for recreational worldwide is raising the concern about its potential effects on marine ecosystems. Available literature is still much focused either on impacts on coral reefs of tropical regions or on diver’s behaviour underwater. In this study we analysed, through photo-quadrats, the benthic community composition in a section of a decommissioned Portuguese navy ship that was sunk for touristic purposes. The ship broke down and became separated in two sections enabling a Control versus Impact sampling design, as one section is less attractive for diving. Gorgonians (mainly belonging to the species Leptogorgia sarmentosa and Eunicella verrucosa) were the taxa more negatively affected in the dived ship section, with smaller coverage and size. More resilient species such as the acorn barnacle Amphibalanus amphitrite were positively correlated with the Impact samples. In the case of the study area, according to the available data, 70% or more of the total amount of dives are now on the sunken ships. From these results, lessons can be taken to apply on natural reefs and related management plans.

Keywords Artificial reefs . Decommissioned ships . Coastal management . Diving effects . SCUBA diving .

Introduction More comprehensive studies on the effects of diving on benthic communities as a whole are still scarce. In the U.S. Extensive literature is available on human impacts and specif- Virgin Islands, diving showed to affect the structure and health ically on the effects of SCUBA diving activities on coral reefs of the reef community, with branching corals and erect all over the world. Direct action by divers, namely coral break- sponges being particularly affected (Chadwick-Furman age, has led to the degradation of coral reefs in the Red Sea 1997). In Bonaire’s reefs, a shift from massive old corals to (Hawkins and Roberts 1992; Hasler and Ott 2008), Caribbean more fast-growing opportunistic species of branching corals region (Tratalos and Austin 2001; Lyons et al. 2015)and was observed, while a reduction in structural complexity of Australia’s (Rouphael and Inglis 1995; the reef was also recorded (Hawkins et al. 1999; Lyons et al. Hardiman and Burgin 2010). Also, diver’s contact with the 2015). Lyons et al. (2015) also found that no significant im- bottom creates re-suspension of sediment that, over time, can pacts occurred on gorgonians and sponges, which the authors lead to higher stress levels for coral reefs and benthic commu- justify by the higher resistance and resilience of these organ- nities (Zakai and Chadwick-Furman 2002;Barkerand isms. A study in the Mediterranean Sea on bryozoan colonies Roberts 2004; Hasler and Ott 2008). showed that the impact of divers were much more significant on more exposed areas like boulders than on vertical walls, affecting erect and foliose species more readily than other * João Encarnação bigger encrusting species (Garrabou et al. 1998). [email protected]; [email protected] The deployment of artificial reefs (AR) was initially direct- ed towards fisheries management and fish stocks recovery 1 Subnauta, Portimão, Portugal programs, being deployed as part of governmental manage- ment plans since 1976 in Japan (see review by Bohnsack and 2 Department of Life Sciences, Lusophone University of Humanities and Technologies, Lisbon, Portugal Sutherland 1985). More recently, and with the increasing pop- ularity of recreational SCUBA diving worldwide, the deploy- 3 Departamento de Ciências e Engenharia do Ambiente, MARE NOVA Faculdade de Ciências e Tecnologia, Universidade Nova de ment of ARs and namely decommissioned ships has been Lisboa, Caparica, Portugal proposed and already recognized as a way to reduce diving J. Encarnação, G. Calado on natural reefs (Leeworthy et al. 2006), enhancing at more than 200 m apart from each other and, unexpectedly, the the same time local economies and creating jobs (Hasler and small stern section remained where it was sunk, while the longer Ott 2008). Still, to our present knowledge, studies on the ef- and heavier bow section was dragged away by ocean currents. fects of recreational diving on the colonisation of ARs are According to data obtained from the main diving center inexistent. operating at the Ocean Revival park, total number of dives We investigated the effects of recreational SCUBA diving per year ranged between 2700 and 3700 in the period of on the benthic assemblages of an AR established for touristic 2013–2016 (Table 1). During this four-year period, only at purposes. Previous studies on divers impact have focused the study site (Ex-NRP Oliveira e Carmo corvette), the aver- mostly on tropical ecosystems and are usually restricted to age was 1148 ± 126 dives per year, with a total of 4592 dives. the effects on specific species. This work provides insights To these numbers more dives would have to be added from at into anthropogenic effects (recreational diving in this case) least two other dive centers operating from nearby (no specific on a benthic community in its early stages of development data available). and settlement on sunken ships. Since the beginning of the Ocean Revival project, a bio- logical monitoring program has been carried out as a legal obligation. Technical reports are annually delivered to the Materials and methods Portuguese authorities (Calado G., unpublished). During the first two years the ships developed an initial cover mainly Study site composed by hydroids and filamentous algae, followed by serpulid polychaetes and acorn barnacles. At the third year The Ocean Revival underwater park consists of four (2015) the colonization began to stabilize being marked by Portuguese decommissioned navy ships deliberately sunk to the presence of large settlements of mussels (Mytilus sp.). create a unique AR system for touristic purposes, 2.5 nautical Fish assemblages also increased in numbers and biomass each miles of Alvor beach in the south of Portugal. A meticulous year, having on the last report from 2016 a list of 45 identified cleaning process was undertaker in order to remove all possi- fish species. After four years the colonization is still in a mid- ble pollutants, inside and out, meaning that the sunken ships successional stage, which means there are still some changes were only steel and aluminium. in the community to be expectable in the next years. ThisstudywasconductedintheBEx-NRP Oliveira e Nevertheless, to date, the biological succession is in compli- Carmo^ corvette (GPS: 37° 05.501’N / 008° 34.964’W), ance with other similar case studies like the ex-HMS BScylla^ sunken in October 30th 2012. This corvette weighted 1.430 shipwreck in south , England (Hiscock et al. 2010)or tons and measured 85 m long by 12 m beam. Positioned par- other AR systems in the Algarve region (Moura et al. 2006). allel to the coast in a West to East direction, it lies in a sand bottom around 32 m, having the main deck 24 m deep and its Sampling methods shallowest area in the mast with 13 m of water clearance. During some big storms in the first months of 2013 the ship The breaking of the ship provided the rare opportunity for this broke down close to the stern, resulting in two independent study. Now there is a section of the ship, the bow, where sections: the bow section is approximately 65 m long and the recreational divers keep diving (hereafter referred as Impact) stern section is approximately 20 m. They became separated and the stern section that, for its relative small size, stopped

Table 1 Number of dives per diving spot in the Portimão area between 2013 (first two ships sunk in October 2012) and 2016

Dive spot Year Grand Total

2013 2014 2015 2016

Natural reefs 502 1001 743 1210 3456 Other AR/Wrecks 160 287 73 161 681 Ocean Revival Oliveira e Carmo 1285 1223 1068 1016 4592 Zambeze 448 324 298 315 1385 H. Capelo 790 1314 1105 1164 4373 Almeida Carvalho 196 885 660 762 2503 Total Ocean Revival 2719 3746 3131 3257 12,853 Grand Total 3381 5034 3947 4628 16,990

Data provided by Subnauta diving center Effects of recreational diving on early colonization stages of an in North-East Atlantic being attractive for divers and there has been, virtually, no group possible, depending on image quality and water tur- diving in that section since winter 2013 (hereafter referred as bidity, which could affect image quality. All specimens Control). For the purpose of this study, both sections have from each taxonomic group were digitally selected (all the been underwater for the same amount of time; as the sections covered area) and cropped from the original image creating are from the same ship the characteristics of the sampling sites an independent layer. Each layer, corresponding to one tax- are identical; one is under the influence of recreational diving onomic group, was then saved in an independent JPEG activity (Impact) and the other is not (Control). image. In this study, we choose a more Bmanual^ species Sampling was carried out through photo-quadrats, using a selection, using BLasso^ or BElliptical Marquee^ tools 12 megapixel Canon G16 , in order to later available in most image-editing software’s, since organisms determining the cover percentage by digital image analysis. from completely different species and taxonomic groups For each photograph, camera was held perpendicularly over a can have similar shapes and/or colorations, which although 0.5 × 0.5 m quadrat that framed each sample. On each sam- more time-consuming, showed to be more accurate than pling area (horizontal and vertical flat surfaces) and ship sec- other automated computed procedures. Each JPEG image tion (Control and Impact), five randomly placed quadrats was then opened in the image processing Image-J software (replicates) were photographed. Together with each photo- (National Institutes of Health, New York). Here three steps quadrat some macro photographs were also taken in order to were taken in order to obtain the desired cover percentage: assist in later species identification. Photo-quadrat replicates 1) transform the image to 8-bit (black and white) by were always spaced half a meter from each other. selecting Image>Type>8-bit; 2) adjust the selection thresh- Field sampling occurred at four different dates: August 2015 old to all the covering area of the organism by selecting (t1), December 2015 (t2), May 2016 (t3) and March 2017 (t4). Image>Adjust>Threshold; and 3) calculation of coverage At both ship sections, horizontal photo-quadrats were always percentage by selecting Analyse>Measure, that Image-J close to the broken edge of the ship, while the verticals were shows as ‘%Area’. Besides the initial method for species on the first available vertical section near this area (Figs. 1 and selection, the procedure was similar to ones used by other 2). The horizontal zone (ship’s heliport) where horizontal sam- authors (i.e. Goodwin et al. 2014;Pechetal.2004;Preskitt pling was undertaken is on 22 m deep, while the vertical sam- et al. 2004). ples were in 23, maximum 24 m deep (Figs. 1 and 2). Differences in percentage of coverage were compared On March 2017 a height measurement of gorgonians was through a one-way analysis of variance (ANOVA) to test for also carried out in the same horizontal Control and Impact significance (α < 0.05) between Control/Impact sections and zones, using a ± 1 mm accuracy hand ruler. In three randomly between horizontal/vertical samples. Assumptions of normality placed 0.5 × 0.5 m quadrats (replicates), ten randomly chosen and equal variance were tested and a log(x + 1) data transforma- specimens were measured in height. tion was applied when required. Once significant differences were detected, the Tukey’s honest significant difference (HSD) Data analysis procedure was used to determine between which factors signif- icant changes occurred. In the cases of data failing to meet the Using image-editing software, digital images were at first normality assumption, the nonparametric Kruskal-Wallis test evenly cropped in order to eliminate areas outside the was used. The SigmaStat 3.5 software (Systat Software Inc., quadrat. Organisms were identified to the lowest taxonomic California) was used for the above statistical analysis.

Fig. 1 Bow section (Impact) of the BEx-NRP Oliveira e Carmo^ corvette with location of horizontal and vertical quadrats J. Encarnação, G. Calado

Fig. 2 Stern section (Control) of the BEx-NRP Oliveira e Carmo^ corvette with location of horizontal and vertical quadrats

Results ramified hydroids and acorn barnacles. Percentage of cover of identified taxa remained below 10% in horizontal samples, Community structure while in vertical samples these values ranged, on average, between 15 and 21%. Nevertheless, the majority of the quad- The analysis of photo quadrats resulted in the identification of rat’s area was classified as ‘unidentified’, ranging in average a total of 10 taxonomic groups. Overall, four of the identified (±SD) from 55.18 ± 10.06% to 92.04 ± 2.07% (Table 2). groups dominated the samples with percentages of coverage These areas although it was not possible to identify its content above 1% in most samples (Table 2): Porifera, Bryozoa, in the digital images, had some type of biological coverage

Table 2 Average (±SD) coverage percentages of identified taxonomic groups

t1 t2 t3 t4

Control Impact Control Impact Control Impact Control Impact

Horizontal quadrats A. amphitrite 0.74 ± 0.21 3.16 ± 1.38 2.01 ± 0.97 4.95 ± 1.36 0.53 ± 0.17 1.07 ± 0.42 0.26 ± 0.14 1.33 ± 0.64 Bryozoa0 0000.27±0.000.41±0.000.44±0.390 C. smithii 0.14 ± 0.10 0.27 ± 0.35 0.29 ± 0.11 0.09 ± 0.04 0.21 ± 0.07 0.08 ± 0.08 0.28 ± 0.06 0.07 ± 0.04 Gorgonians 0.26 ± 0.07 0 0.29 ± 0.15 0.07 ± 0.02 2.28 ± 0.65 0.25 ± 0.04 3.17 ± 1.03 0.52 ± 0.22 Other Cnidaria 0 0 0.09 ± 0.01 0 0 0.34 ± 0.00 0.08 ± 0.00 0 Mytilus sp. 0 1.01 ± 1.14 0 0 0 0.58 ± 0.00 0 0 Polychaeta 0.41 ± 0.25 0 0.20 ± 0.00 0 0 0 0 0 Porifera 0.54 ± 0.57 7.58 ± 7.04 1.71 ± 1.62 1.80 ± 1.19 0.16 ± 0.05 1.98 ± 1.85 0.77 ± 0.53 3.02 ± 1.90 Ramified hydrozoans 3.63 ± 3.30 1.04 ± 0.31 0.64 ± 0.09 0.23 ± 0.00 0.85 ± 0.53 1.16 ± 0.57 0 0 Unidentified 71.73 ± 11.15 78.85 ± 7.67 89.03 ± 2.42 83.95 ± 3.41 88.86 ± 1.70 87.84 ± 1.91 92.04 ± 2.07 90.46 ± 4.15 Vertical quadrats A. amphitrite 0.61 ± 0.52 1.60 ± 0.99 0.63 ± 0.33 1.84 ± 1.69 0.12 ± 0.04 3.79 ± 0.00 0.28 ± 0.00 1.85 ± 0.9 Bivalvia 0 0.45 ± 0.00 0 0 0 0 0 0 Bryozoa 16.4 ± 6.48 4.15 ± 1.32 1.63 ± 0.41 0.79 ± 0.28 8.58 ± 4.48 7.62 ± 5.20 6.64 ± 1.88 0.56 ± 0.34 C. smithii 0.09 ± 0.00 0.06 ± 0.01 0.14 ± 0.11 0.08 ± 0.03 0.14 ± 0.14 0.13 ± 0.06 0.08 ± 0.03 0.10 ± 0.04 H. durotrix 0 000000.59±0.000.25±0.00 Other Cnidaria 0 0 0.17 ± 0.00 0 0.19 ± 0.17 0 0 0.06 ± 0.00 Polychaeta 0 0 1.00 ± 0.00 0.72 ± 0.00 3.42 ± 0.00 0 0 0 Porifera 2.43 ± 1.17 4.67 ± 5.36 5.34 ± 2.83 8.05 ± 9.24 9.08 ± 13.98 1.14 ± 0.58 8.07 ± 3.44 4.67 ± 3.36 Ramified hydrozoans 4.70 ± 2.52 3.39 ± 2.48 0 1.25 ± 0.11 0.26 ± 0.11 0 0 0.54 ± 0.28 Unidentified 55.18 ± 10.06 77.72 ± 8.88 72.95 ± 4.47 75.02 ± 9.51 56.27 ± 7.79 77.93 ± 8.89 72.23 ± 4.72 81.98 ± 4.06 Effects of recreational diving on early colonization stages of an artificial reef in North-East Atlantic and were not bare metal without any living content. This turf- Control, while on Impact the value increased to 2.63 ± 1.87% like cover showed to be difficult to specify its composition. with an absolute maximum of 6.60% in December 2015 (t3). In Still, in situ observations showed that hydroids, but also some two sampling dates (t1 and t4) acorn barnacles also showed a algae, other cnidarians or erect bryozoans, composed the most significant increase in Impact vertical samples. The cup coral C. of this ‘unidentified’ fouling. smithii, although with low contribution to the total coverage, also In some taxonomic groups, it was not possible to identify showed significant differences (p < 0.05) between ship sections down to species level all the organisms in the digital images. in t2, t3 and t4 in horizontal samples (Fig. 3). In these three dates Still, for some of them, we were able to confirm the identifica- coverage was always higher in Control (Table 2) with average of tions with further in situ observations. This was the case of 0.23 ± 0.10% and 0.12 ± 0.16% in Impact. Although being also gorgonians that mainly belonged to the species Leptogorgia common in vertical samples, ANOVA did not reveal significant sarmentosa and Eunicella verrucosa, acorn barnacles differences between sections. Amphibalanus amphitrite and Bryozoa mainly belonging to Bryozoa showed for two times (t1 and t4) a significant Schizobrachiella sanguinea and Watersipora aff. Subtorquata. increase in Control, but only in vertical samples. Mussels (Mytilus sp.) were scarce but were only observed in Impact samples, although no statistical significance was identified. Differences between sections

The group of organisms with more consistent differences be- Differences between sample orientation tween Control and Impact sections were the gorgonians. In terms of coverage percentage gorgonians showed a significant de- Considering the position of the quadrats (horizontal or verti- crease in Impact in all sampling dates (Fig. 3), being absent in cal), there were also some significant changes identified by Impact on t1 and were only identified in horizontal samples. On one-way ANOVA. March 2017 (t4) gorgonians already showed an average (±SD) Bryozoa showed significant differences in almost all in- percentage of coverage of 3.17 ± 1.03% on Control, while on stances (except for t3 Control), either in Control or Impact, Impact the value was still less than 1% (Table 2). These differ- always with higher cover observed in vertical samples. On av- ences were considered significant in all sampling dates by erage, (±SD) Bryozoa had 6.17 ± 6.08% coverage in vertical ANOVA analysis, being very significant at t3 and t4 (p < samples, while on horizontal the average was 0.40 ± 0.29%. 0.001). The in situ measurement of gorgonians showed that Porifera also seemed to have more affinity with vertical surfaces specimens were higher in all three Control replicates (Fig. 4). as it showed a significant increase in this sample orientation at t3 One-way ANOVA test indicated that these differences were very (p =0.013)andatt4(p < 0.001), on the Control samples. significant (p < 0.001). The average (±SD) height was 24.63 ± In horizontal samples at t2, t3 and t4, C. smithii showed 4.23 mm for Control, while in Impact was 15.97 ± 3.76 mm. significantly higher percentages, but only in the Control sam- One-way ANOVA also showed significant differences for ples which is consistent with the previous comparisons be- two other taxonomic groups: acorn barnacles A. amphitrite and tween sections. At t4 in Control this increase in horizontal the cup coral Caryophyllia smithii. In the four sampling dates, samples was considered as very significant (p < 0.001). acorn barnacles always had significantly higher cover percent- Gorgonians were only identified in horizontal samples, which ages in Impact, on all horizontal samples (Fig. 3). In horizontal reflected naturally in a significant increase from vertical to samples they showed, on average, 0.95 ± 0.89% coverage in horizontal samples identified by ANOVA tests.

Fig. 3 Averagepercentageof coverage from horizontal samples of species/taxonomic groups that showed the most significant changes between Control and Impact J. Encarnação, G. Calado

through the planulae and polyp stages is already very low (Linares et al. 2008). Additionally, an excess of sediment re- suspension is also a key factor in polyp development and sur- vival success (Neil 1990; Abidin and Mohamed 2014), which in our study case may be an additional stress factor and therefore delay the development of gorgonians. Another species that is positively correlated with the non- dived ship section (Control) was the cup coral C. smithii.Bell and Turner (2000) discussed whether this coral could indirectly benefit from sedimentation, as it may be more tolerant than other organisms, thus gaining in direct competition for space. Nevertheless, in that study carried out in Ireland, the corals were Fig. 4 Measurements of gorgonians heights in Control and Impact in under natural sedimentation conditions, allowing them to adapt March 2017 to those local sedimentation regimen, at least until a certain sedimentation threshold, as stated by the authors. Although this Discussion species has been documented as sedimentation resilient and equipped with a sediment removal mechanism (Hiscock and Our results show that recreational divers are having some Howlett 1976), the continuous disturbance caused by divers is effect in shaping the biological community settling at the possibly causing excessive stress to these corals. Ocean Revival underwater park. The present study was main- Our study area has three main sources of sediments in a ly focused in understanding if there is, or not, an influence in relatively close range that contribute to the overall sediment the benthic community, therefore it is not intended to be an load: a small creek in Lagos (Ribeira de Bensafrim), the Alvor AR colonization study. A particular point of interest in this AR lagoon and the Arade river in Portimão, with the latter being was that its biological community was under influence of the largest contributor to a sedimentary composition that con- diving activity since day-one, thus the interest in understand- sists mainly of fine to very fine sand (Martins et al. 2012; ing how the community reacted during this early colonization Cúrdia et al. 2013). This way, species in the present study area stage. Although sampling took place at different time periods, are adapted to deal with a relatively moderate amount of across different times of the year, the four sampling periods suspended material. Although in the two previous cases, gor- were not intended to study temporal patterns or settlement, but gonians and cup coral, moderate natural sedimentation itself as an extra replication to compare Control versus Impact sam- might not be a problem, the combined impact of both physical ples and assess if results were consistent over time. damage and constant sediment re-suspension from divers is Depending on their level of experience, divers interact with the most likely cause of reduction in coverage percentage of the environment in different ways. In most of the occurrences, these two organisms in Impact. damages to the bottom occur by the action of ’sfins As stated by Baynes and Szmant (1989) in a study also (Rouphael and Inglis 1995; Zakai and Chadwick-Furman 2002; carried out on a shipwreck, vertical surfaces (namely the sides Barker and Roberts 2004;Lunaetal.2009). This direct physical of the hull) had higher diversity and greater live cover, which action will therefore affect first the more fragile, erect and has been attributed to higher velocity flow that reduced sedi- branching organisms. In our study this was clearly observed mentation stress to species. Although divers already showed to with gorgonians - they were always more abundant in Control; have less impact in life inhabiting vertical surfaces than horizon- before December 2015 (t2) they were not even observed in tal ones (Garrabou et al. 1998;Lloretetal.2006), in our study Impact; and in subsequent sampling events they always case the differences might also be due to natural hydrodynamic displayed lower coverage in Impact than in Control. Although factors. Bryozoa showed a clear affinity with vertical samples, not present in the quadrats of t1 in Impact, some gorgonians both in Control and Impact. The same happened in Control with were already observed in some protected places such as Porifera at t3 and t4. These two types of encrusting biota are corners in the ship hull or under some structures where more readably affected by sedimentation and thus more easily disturbance from divers is reduced to a minimum. Coma et al. colonize vertical surfaces than horizontal ones (Baynes and (2004) found that the main source of mortality was due to col- Szmant 1989; Wendt et al. 1989; Cocito et al. 2002). ony detachment and high diving activity caused a raise in natural The amount of dives per year in the present study site can mortality by three times. At the present study site, an AR in its be considered as moderate. These numbers are not has high as first years of colonization, constant disturbance is probably mak- in other more popular dive sites worldwide such as the Red ing it more difficult for the settlement of polyps on the ship, Sea where number of dives per year can exceed 30.000 (Zakai which was also visible by a smaller size of the gorgonians in and Chadwick-Furman 2002; Hasler and Ott 2008). Still, they the impacted section. Under natural conditions, survivorship are comparable with some Caribbean sites with 6.000 dives Effects of recreational diving on early colonization stages of an artificial reef in North-East Atlantic per year (Hawkins et al. 1999) or Mediterranean sites with We recall that this former war vessel, together with the 8.000–10.000 dives per year (Lloret et al. 2006;Lunaetal. other ships that form the Ocean Revival project, were specif- 2009). The concept of diver carrying capacity, expressed as ically prepared and sunken for recreational diving and there- dives per site per year, can be defined as the threshold beyond fore this impact we report is on these ARs and not on natural what an area will become degraded due to diving activity reefs. Also, these structures are still in its early years of colo- (Abidin and Mohamed 2014) and can be a valuable tool for nization, in a mid-successional stage, which means still going comparisons between regions. Nevertheless, in most pub- through changes in its composition and species domination. lished work containing these dive counts it is not completely The high values of cover that we were not able to further clear if the numbers concern either one area of diving activity specify its living content, and therefore placed in the ‘uniden- or a specific diving spot. In the latter case, numbers could be tified’ class, is also a reflection of the AR’s young age, where computed in dives per site per year, making comparisons eas- the various species are still on an early development stage ier. For these reasons, comparisons in diving must making it more difficult to identify them from digital images. be taken with caution. In our present study case, the definition It remains to be seen if in the following years species that are of these AR’s carrying capacity is still premature, mainly due now more affected by divers can overcome these disturbances to their young age and the need for more sampling effort on and adapt in order to evolve into a community more similar to the remaining ships. To our present knowledge, studies the one in the section where diving is absent. assessing recreational diving pressure in Portugal waters are In several parts of the world the use of ARs for diving inexistent. Still, precautionary measures can be taken by local purposes has already been proposed also as a way to reduce diving operators in order to mitigate divers impact. Among the pressure on natural habitats (Wilhelmsson et al. 1998; these may be the inclusion in the on the impor- Zakai and Chadwick-Furman 2002; Hasler and Ott 2008). tance of maintaining a good control throughout the Diving tourism has been increasing in the Algarve region dive or a previous buoyancy check dive for the less experi- and natural reefs of Portimão could benefit from the Ocean enced divers. Although these ships were sunk with diving Revival project, as major operating diving centers are now tourism as primary focus, they created a new fixation point focused on the sunken ships, where since 2013 at least 70% for marine life, which for the majority of divers is a main of the dives occur. As happened in other places where ships attraction factor when . Thus, is also of the diving and other ARs were deployed for diving tourism purposes, not operators’ best interest to promote sustainable diving practices only the natural reefs can benefit from them but also local in order to have healthy and diverse dive spots. economy can grow without a direct increase in impacts on Still, persistent disturbance from human activities may, natural reefs and habitats (Leeworthy et al. 2006; Stolk et al. overtime, cause shifts in the community structure, increas- 2007). Through these results, some lessons may be learned ing stress and testing the resilience of organisms in the and applied on restoration projects and management plans in ecosystem (Nyström et al. 2000; Barker and Roberts natural ecosystems, such as restriction periods for visitors 2004; Lyons et al. 2015;). Our results showed that acorn while benthic communities are being established. barnacles (mainly A. amphitrite) had significantly higher cover percentages in Impact. Their hard shell provides nat- Acknowledgements This project was funded by Associação para a ural protection for the barnacle’s soft parts against predators promoção e desenvolvimento do turismo subaquático (Musubmar) and as well as against adverse conditions and physical G.C. had the support of Fundação para a Ciência e Tecnologia (FCT), through the strategic project UID/MAR/04292/2013 granted to MARE. exerted on it (see review in Barnes 1999). This relatively The authors would also like to acknowledge the assistance from Subnauta higher resistance to physical impacts (e.g., a diver’sfin)can diving center in logistics regarding data collection as in providing valu- give barnacles competitive advantage over more fragile spe- able information about the number of dives in the study sites and adjacent cies, such as gorgonians. Barnacles may then persist and diving spots. even expand their coverage, occupying gorgonians’ ecolog- ical niche. Several studies revealed that erect, foliose or Compliance with ethical standards branching species are the ones primarily affected by divers Conflict of interest The authors declare that they have no conflict of (Rouphael and Inglis 1995; Chadwick-Furman 1997; interest. Garrabou et al. 1998), which in our case is specially reflected on the decrease in gorgonians coverage and size, Ethical approval This article does not contain any studies with animals in the ship section under influence of recreational diving. performed by any of the authors. Only non-invasive methods were ap- Giving that the two sampled sections of the ship are under plied to acquire the biological data. the same environmental conditions, distanced apart only Sampling and field studies All necessary permits for sampling and 200 m, epifauna is given the same conditions to settle and observational field studies have been obtained by the authors from the develop. The only difference between the two is the use of competent authorities and are mentioned in the acknowledgements, if the bow (Impact) for recreational diving. applicable. J. Encarnação, G. Calado

References Leeworthy VR, Maher T, Stone EA (2006) Can artificial reefs alter user pressure on adjacent natural reefs. Bull Mar Sci 78(1):29–37 Linares C, Coma R, Mariani S, Díaz D, Hereu B, Zabala M (2008) Early Abidin SZ, Mohamed B (2014) A review of SCUBA diving impacts and life history of the Mediterranean gorgonian Paramuricea clavata: implication for coral reefs conservation and tourism management. implications for population dynamics. Invertebr Biol 127(1):1–11. SHS Web of Conf 12:01093. https://doi.org/10.1051/shsconf/ https://doi.org/10.1111/j.1744-7410.2007.00109.x 20141201093 Lloret J, Marín A, Marín-Guirao L, Carreño MF (2006) An alternative Barker NH, Roberts CM (2004) Scuba diver behaviour and the manage- approach for managing scuba diving in small marine protected ment of diving impacts on coral reefs. Biol Conserv 120(4):481– areas. Aquatic Conserv: Mar Freshw Ecosyst 16:579–591. https:// 489. https://doi.org/10.1016/j.biocon.2004.03.021 doi.org/10.1002/aqc.734 Barnes M (1999) The mortality of intertidal cirripedes. Oceanogr Mar Luna B, Pérez CV, Sánchez-Lizaso JL (2009) Benthic impacts of recre- Biol Annu Rev 37:153–244 ational divers in a Mediterranean Marine Protected Area. ICES J Baynes TW, Szmant AM (1989) Effect of on the sessile benthic Mar Sci 66(3):517–523. https://doi.org/10.1093/icesjms/fsp020 community structure of an artificial reef. Bull Mar Sci 44(2):545– 566 Lyons PJ, Arboleda E, Benkwitt CE, Davis B, Gleason M, Howe C, Bell J, Turner JR (2000) Factors influencing the density and morphomet- Mathe J, Middleton J, Sikowitz N, Untersteggaber L, Villalobos S rics of the cup coral Caryophyllia smithii in Lough Hyne. J Mar Biol (2015) The effect of recreational SCUBA divers on the structural Ass UK 80:437–441. https://doi.org/10.1017/S0025315400002137 complexity and benthic assemblage of a Caribbean . – Bohnsack JA, Sutherland DL (1985) Artificial reef research: A review Biodivers Conserv 24(14):3491 3504. https://doi.org/10.1007/ with recommendations for future priorities. Bull Mar Sci 37:11–39 s10531-015-1009-2 Chadwick-Furman NE (1997) Effects of SCUBA diving on coral reef Martins R, Azevedo MR, Mamede R, Sousa B, Freitas R, Rocha F, invertebrates in the U.S. Virgin Islands: implications for the man- Quintino V, Rodrigues AM (2012) Sedimentary and geochemical agement of diving tourism. In: den Hartog JC (ed) Proceedings of characterization and provenance of the Portuguese continental shelf – the sixth international conference on coelenterate biology. Nationaal soft-bottom sediments. J Mar Syst 91:41 52. https://doi.org/10. Naturhistorisch Museum, The Netherlands, pp 91–100 1016/j.jmarsys.2011.09.011 Cocito S, Bedulli D, Sgorbini S (2002) Distribution patterns of the sub- Moura A, Boaventura D, Cúrdia J, Santos MN, Monteiro CC (2006) littoral epibenthic assemblages on a rocky shoal in the Ligurian Sea Biomass production of early macrobenthic communities at the (NW Mediterranean). Sci Mar 66(2):175–181. https://doi.org/10. Faro/Ancão artificial reef (Portugal): effect of depth and reef layer. 3989/scimar.2002.66n2175 Bull Mar Sci 78:83–92 Coma R, Pola E, Ribes M, Zabala M (2004) Long-term assessment of Neil D (1990) Potential for coral stress due to sediment resuspension and temperate octocoral mortality patterns, protected vs. unprotected deposition by reef walkers. Biol Conserv 51:221–227. https://doi. areas. Ecol Appl 14(5):1466–1478. https://doi.org/10.1890/03-5176 org/10.1016/0006-3207(90)90127-B Cúrdia J, Monteiro P, Afonso CM, Santos MN, Cunha MR, Gonçalves Nyström M, Folke C, Moberg F (2000) Coral reef disturbance and resil- JM (2013) Spatial and depth-associated distribution patterns of shal- ience in a human-dominated environment. Trends Ecol & Evol low gorgonians in the Algarve coast (Portugal, NE Atlantic). Helgol 15(10):413–417. https://doi.org/10.1016/S0169-5347(00)01948-0 Mar Res 67:521–534. https://doi.org/10.1007/s10152-012-0340-1 Pech D, Condal AR, Bourget E, Ardisson P (2004) Abundance estimation Garrabou J, Sala E, Arcas A, Zabala M (1998) The impact of diving on of rocky shore invertebrates at small spatial scale by high-resolution rocky sublittoral communities: A case study of a bryozoan popula- digital photography and digital image analysis. J Exp Mar Biol Ecol tion. Conserv Biol 12(2):302–312. https://doi.org/10.1111/j.1523- 299:185–199. https://doi.org/10.1016/j.jembe.2003.08.017 1739.1998.96417.x Preskitt LB, Vroom PS, Smith CM (2004) A rapid ecological assessment Goodwin C, Rodolfo-Metalpa R, Picton B, Hall-Spencer JM (2014) (REA) quantitative survey method for benthic algae using Effects of ocean acidification on sponge communities. Mar Ecol photoquadrats with scuba. Pac Sci 58(2):201–209 – 35:41 49. https://doi.org/10.1111/maec.12093 Rouphael T, Inglis G (1995) The effects of qualified recreational SCUBA Hardiman N, Burgin S (2010) Recreational impacts on the fauna of divers on coral reefs. Technical Report No. 4. CRC Reef Research Australian coastal marine ecosystems. J Environ Manag 91(11): Centre, Townsville 39 pp 2096–2108. https://doi.org/10.1016/j.jenvman.2010.06.012 Stolk P, Markwell K, Jenkins JM (2007) Artificial reefs as recreational Hasler H, Ott JA (2008) Diving down the reefs? Intensive diving tourism scuba diving resources: A critical review of research. J Sustain Tour threatens the reefs of the northern Red Sea. Mar Pollut Bull 56: 15(4):331–350. https://doi.org/10.2167/jost651.0 1788–1794. https://doi.org/10.1016/j.marpolbul.2008.06.002 Tratalos JA, Austin TJ (2001) Impacts of recreational SCUBA diving on Hawkins JP, Roberts CM (1992) Effects of recreational SCUBA diving coral communities of the Caribbean island of Grand Cayman. Biol on fore-reef slope communities of coral reefs. Biol Conserv 62(3): Conserv 102:67–75. https://doi.org/10.1016/S0006-3207(01) 171–178. https://doi.org/10.1016/0006-3207(92)91045-T 00085-4 Hawkins JP, Roberts CM, Hof TV, De Meyer K, Tratalos J, Aldam C (1999) Effects of recreational scuba diving on Caribbean coral and Wendt PH, Knott DM, Van Dolah RF (1989) Community structure of the fish communities. Conserv Biol 13(4):888–897. https://doi.org/10. sessile biota on five artificial reefs of different ages. Bull Mar Sci – 1046/j.1523-1739.1999.97447.x 44(3):1106 1122 Hiscock K, Howlett RM (1976) The ecology of Caryophyllia smithi Wilhelmsson D, Öhman MC, Ståhl H, Shlesinger Y (1998) Artificial Strokes and Broderip on South-western coasts of the British Isles. reefs and dive tourism in Eilat, Israel. Ambio 27(8):764–766 In: Drew EA (ed) Underwater Research. Academic Press, New Zakai D, Chadwick-Furman NE (2002) Impacts of intensive recreational York, 438 pp diving on reef corals at Eilat, northern Red Sea. Biol Conserv Hiscock K, Sharrock S, Highfield J, Snelling D (2010) Colonization of an 105(2):179–187. https://doi.org/10.1016/S0006-3207(01)00181-1 artificial reef in south-west England - ex-HMS ‘Scylla’.JMarBiol Ass UK 90:69.–94 https://doi.org/10.1017/S0025315409991457