The impact of dredge-fill on Posidonia oceanica seagrass meadows: Regression and patterns of recovery

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

The impact of dredge-fill on Posidonia oceanica seagrass meadows:Regression and patterns of recovery

Fabio Badalamenti a, Adriana Alagna b,⇑, Giovanni D’Anna a, Antonio Terlizzi b, Giuseppe Di Carlo c

a Laboratorio di Ecologia Marina, IAMC-CNR, Castellammare del Golfo, 91014 Trapani, Italyb Laboratorio di Zoologia e Biologia Marina, DiSTeBA, Università del Salento, Prov.le Lecce-Monteroni, 73100 Lecce, Italyc Conservation International, Arlington, VA 22202, USA

a r t i c l e i n f o

Keywords:DredgingNatural recoverySubstratePercent coverPosidonia oceanicaMediterranean Sea

a b s t r a c t

Posidonia oceanica meadows can be severely damaged by dredge-fill operations. We report on the con-struction of gas pipelines that occurred between 1981 and 1993 in SW Sicily, Italy. A large portion ofthe meadow was mechanically removed, and the excavated trench was filled with a mosaic of substrates,ranging from sand to consolidated rock debris. Meadow loss and recovery were quantified over 7 yearsafter the end of operations. We recorded an overall loss of 81.20 ha of meadow. Substrate stronglyaffected recovery as the percent cover by P. oceanica consistently increased on calcareous rubble, reach-ing values of 44.37 ± 3.05% in shallow sites after 7 years, whereas no significant increase occurred onother substrates. As in the Mediterranean Sea exploitation of coastal areas continues to grow with con-sequent impacts on P. oceanica meadows, this case study illustrates how artificial rubble-like materialscould be employed to support the restoration of damaged meadows.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

In the last few decades, seagrasses have continued to declineglobally, due to both natural (Preen et al., 1995; Cardoso et al.,2008) and anthropogenic causes (Short and Wyllie-Echeverria,1996; Ruiz et al., 2001; Duarte, 2002; Orth et al., 2006a; Waycottet al., 2009). Fish trawling, industrial and sewage outfalls, fishfarms, boat anchoring and propeller scars are all sources of distur-bance that can impact seagrasses and cause their decline (Ruiz andRomero, 2003; González-Correa et al., 2005; Diaz-Almela et al.,2008a; Montefalcone et al., 2008; Perez et al., 2008). Most of thesesources of impact involve persistent deterioration in water qualityand can alter sediment dynamics (Cancemi et al., 2003). As a con-sequence, once the impact has been removed, seagrass recoverymay occur slowly over a period of decades or, in some cases, failaltogether (Delgado et al., 1999; González-Correa et al., 2005,2008). Other sources of anthropogenic disturbance, including portconstruction, the deployment of pipes and cables for gas, watertransport and communication, and beach replenishment involveimpacts occurring in single events (Guidetti and Fabiano, 2000;Erftemeijer and Robin Lewis, 2006). Although these events leadto local seagrass loss, they do not generally affect seawater proper-ties over the long term. Therefore, once the impact is removed,seagrasses may recover, as long as substrate and sediment condi-

tions are favorable. However, there is great variability in the recov-ery strategies of seagrasses following disturbance events, whichare dependent on plant size (rhizome diameter), life history, andlocal and regional environmental conditions (tropical vs. temper-ate) (Rollon et al., 1999; Olesen et al., 2004; Rasheed, 2004).

When the impacted area is small or not totally devoid of vegeta-tion, clonal growth might be sufficient to reach full recovery. How-ever, when a disturbance affects a large area, clonal growthbecomes insufficient, and a supply of propagules is required to fillthe gap. For some large species (e.g., Enhalus acoroides, Thalassiatestudinum, Posidonia angustifolia, P. sinuosa, and P. coriacea) thatproduce large numbers of seeds and have high dispersal potentialand good seedling survival, sexual recruitment can substantiallycontribute to recovery (Kendrick et al., 1999; Bryars and Neveraus-kas, 2004; Olesen et al., 2004; Whitfield et al., 2004). For theMediterranean seagrass Posidonia oceanica (L.) Delile, a large, slow-growing species with low sexual recruitment success (Diaz-Almelaet al., 2008b; Balestri et al., 2009), meadow recovery is expected tobe an extremely slow and difficult process (Duarte et al., 2006).

Whereas seagrass recovery through clonal propagation andseed production is well-documented, recovery through vegetativefragments has received little attention (Rasheed, 2004). Re-colonization through fragment dispersal has been considereduncommon for some species (Ewanchuk and Williams, 1996),whereas it is considered to be important for others (Campbell,2003; Hall et al., 2006; Diaz-Almela et al., 2008b). As pointed outby Duarte et al. (2006), the role of vegetative fragments in sea-grass re-colonization processes has been largely underestimated,

0025-326X/$ - see front matter � 2010 Elsevier Ltd. All rights reserved.doi:10.1016/j.marpolbul.2010.12.011

⇑ Corresponding author. Tel.: +39 0832 298885; fax: +39 0832 298702.E-mail addresses: [email protected], [email protected] (A.

Alagna).

Marine Pollution Bulletin 62 (2011) 483–489

Contents lists available at ScienceDirect

Marine Pollution Bulletin

journal homepage: www.elsevier .com/locate /marpolbul

Author's personal copy

probably due to the elusive nature of this type of event, whichmakes direct observation difficult.

As for other Mediterranean sub-tidal habitats, the recovery pat-terns of P. oceanica can be driven and shaped by elements of hab-itat structure, including substrate availability, type and complexity(Faimali et al., 2004; Davis, 2009; Guarnieri et al., 2009). However,the influence of substrate type and complexity on P. oceanica colo-nization and recovery patterns has been scarcely investigated,probably due to the long time span required for plant establish-ment and growth.

The deployment of a gas pipeline between Cape Bon (NE Tuni-sia) and Cape Feto (SW Sicily, Italy) required dredging of a pristineseagrass bed. As a consequence, a great portion of the meadow wasdestroyed or severely damaged by direct removal or sediment bur-ial. When dredging and filling operations were completed, the im-pacted area contained a mosaic of substrates, ranging fromcalcarenitic boulders, coming from the excavation of the originalsubstrate on which the meadow grew, to residual material suchas dead matte, sand, gravel, and exogenous calcaureous rubbles,used as back-fill material. As a result of the dumping activities,consolidated calcareous rocks formed rubble mounds (Di Carloet al., 2005). In addition, dredge-fill operations caused meadowsurrounding the dredged area to die off (Badalamenti et al.,2006) through sediment plumes and altered sedimentation rates.Previous studies documented the recruitment of Posidonia oceanicavegetative fragments on rubble mounds one decade after the endof dredging operations (Di Carlo et al., 2005, 2007).

In light of the potential increase in the installation of oil and gaspipelines in the Mediterranean Sea (IEA, 2009; Tubb, 2010), ouraim was to use this dredge fill event as a case study to understandof how P. oceanica meadows can be impacted by and recover fromthis type of disturbance. Specifically, the objectives of this studywere to (a) quantify seagrass loss as a result of dredge-fill opera-tions, (b) assess whether P. oceanica recovery occurred within the

dredged area over the 7 years following the disturbance event,and (c) compare recovery patterns across different substrates pres-ent within the dredged area.

2. Materials and methods

2.1. Study site

The study area is located on the SW coast of Sicily, Italy(41�730N, 18�120E) and extends SW to a depth of approximately30 m (Fig. 1). Geologically, the area is a wide calcareous plateau(Ruggeri et al., 1975), atop which P. oceanica forms a densemeadow. The meadow is part of a wider seagrass formation thatextends almost continuously along the Western Sicilian coast,representing the largest P. oceanica formation in Sicily (SI.DI.MAR.,2002).

The construction of a new pipeline required two dredging oper-ations, occurring in 1981 and 1993. In 1981, the trench was back-filled with residual materials from the excavation, part of whicheventually eroded into sand and gravel. In 1993, a second dredgingevent was conducted, enlarging the damaged area (Badalamentiet al., 2006). Consolidated calcareous rubbles from a nearby quarrywere used to fill in the damaged area.

2.2. Substrate mapping and assessment of meadow loss

Three maps of substrate distribution were produced and com-pared to assess meadow loss and substrate variation in the studyarea from 1981 to 1995. Maps were built by compiling informationfrom aerial photographs, Side Scan Sonar (SSS) and Remotely Oper-ated Vehicles (ROV), which were ground-truthed with underwaterSCUBA observations. Aerial photographs at a scale of 1:10.000 weretaken in black and white in 1979, and color images were taken in

Fig. 1. Geographic location of the study area.

484 F. Badalamenti et al. / Marine Pollution Bulletin 62 (2011) 483–489

Author's personal copy

1987 and 1990. The SSS used was an Edge Tech (EG&G) 260equipped with a transducer transmitting an acoustic signal at a fre-quency varying from 100 to 500 Hz, with a resolution at least of0.20 m. Ground truthing on a set of randomly selected pointswas conducted with SCUBA divers in areas where interpretationof substrate presence was problematic.

The whole data set was imported into a GIS environment usingArcGis 9.0, georeferenced and synthesized into three vectorialmaps at a scale of 1:10.000, showing the distribution of substratesbefore (1979 map), after the first (1993 map) dredging event, andafter the second dredging event (1995 map). The maps covered atotal area of 166 ha, with a depth range between 0 and 30 m,extending approximately 300 m around the dredged area andincluding portions of the surrounding meadow. Seven categoriesof substrate were distinguished from the maps within the im-pacted area: P. oceanica cover, Cymodocea nodosa cover, deadmatte, fine sand, sand and gravel, calcarenitic boulders and calcar-eous rubbles. The area was divided and analyzed separately into ashallow (0 to �15) section and a deep (�16 to �30) section.

2.3. Assessment of P. oceanica recovery

ROV images were used to estimate the percent cover byP. oceanica within the dredged area from 1993 to 1999 on five‘‘substrates’’: dead matte, sand, calcarenitic boulders and calcare-ous rubbles. The rubbles were divided in two distinct areas: moundvalley and mound crest, which differed in hydrodynamic andsedimentary conditions, as valleys were more sheltered and expe-rienced higher sedimentation rates than crests (Di Carlo et al.,2005). ROV surveys started immediately after the conclusion ofbackfilling operations (November/December 1993) and then ranyearly from 1994 to 1999 (always in the fall season). No surveywas conducted in 1996. The sampling design included 120 tran-sects and 480 video frames. Each transect, 2 m wide and 600 mlong, ran parallel to the coastline from a depth of �5 m to a depthof �25 m. Transects were subdivided into two depth ranges: shal-low (�5 to �15 m) and deep (�16 to �25 m). Within each year,two interspersed replicated transects were independently chosen

for each substrate at the two depth ranges. Within each transect,four ROV video frames were chosen at random, and the percentageof seagrass cover was recorded.

The ROV had a video camera set at a fixed angle and connectedto a monitor and recording device. Video frame dimensions werecalculated according to Bourgoin et al. (1985):

W ¼ 2H1

sinatan

b2

where W is the width, H is the height of the camera above the sea-bed, a is the camera angle from the horizontal plane and b is thelens angle. Using this formula, a single video frame covered an esti-mated area of approximately 4 m2.

2.4. Data analysis

Variation in the percent cover by P. oceanica over time on thedifferent substrates was tested using a four-way ANOVA with thefollowing design: year (Ye, 6 levels: 1993, 1994, 1995, 1997,1998 and 1999), depth (De, 2 levels, shallow and deep) and sub-strate (Su, 5 levels, dead matte, sand, calcarenitic boulders, moundcrest and mound valley) as fixed and orthogonal factors, and tran-sect (Tr, 2 levels) as a random factor nested within the interaction(SuxDe). Four replicates per transect were randomly chosen, eachof which correspond to a video frame. Cochran’s test was used tocheck for homogeneity of variances (Winer et al., 1991). When sig-nificant differences were found in the ANOVA, they were compareda posteriori using Student–Newman-Keuls (SNK) test (Underwood,1997).

3. Results

3.1. Substrate mapping and meadow loss

In the shallower section of the study area (0 to �15), the extentof P. oceanica meadows decreased from 70.84 to 50.14 ha (31.8%loss) after the first disturbance event (Fig. 2a and b; Table 1). Ofthe surface lost, 12.26 ha turned into dead matte, 7.37 ha was

Fig. 2. Maps of substrate distributions in the study area: (a) in 1979, before trenches excavation; (b) in 1993, after the first trench excavation; and (c) in 1995, after the secondtrench excavation.

F. Badalamenti et al. / Marine Pollution Bulletin 62 (2011) 483–489 485

Author's personal copy

replaced by bare sand and gravel, and 2.26 ha was replaced by cal-carenitic boulders. The second disturbance (Fig. 2c) caused afurther decrease in meadow surface area to 44.63 ha (39.9% lossof the initial meadow extent), which was replaced by dead matte(6.88 ha), sand and gravel (8.88 ha) and the rubble used to fill inthe trench (6.34 ha) (Table 1).

In the deeper section of the study area, meadow extent de-creased from 72.11 to 39.39 ha after the first disturbance (41.0%loss) (Fig. 2a and b; Table 1). At this depth, the P. oceanica meadowwas replaced by fine sand (24.37 ha), dead matte (12.35 ha),C. nodosa (1.80 ha) and sand and gravel (1.73 ha) (Table 1). Thesecond disturbance (Fig. 2c) decreased the meadow surface area

to 17.13 ha (69.0% loss of the initial meadow extent), coincidingwith an increase in dead matte (25.59 ha), sand and gravel(2.68 ha), and rubble (1.35 ha) (Table 1).

Total seagrass loss in the study area was calculated at 37%(53.42 ha) after the first disturbance and 57% (81.20 ha) after thesecond.

3.2. Assessment of P. oceanica recovery

A significant interaction term of P. oceanica percent cover be-tween year, depth and substrate was detected through the analysisof variance (Table 2). P. oceanica percent cover significantly varied

Table 1Percentage of meadow loss and substrate variation in the study area after the first (1993) and the second (1995) impact at shallow (0–15 m) and deep (16–30 m) sites. Hahectares.

Category Shallow (0–15 m) Deep (�16–30)

1979 1993 1995 1979 1993 1995

ha (%) ha (%) ha (%) ha (%) ha (%) ha (%)

P. oceanica 70.84 (100.0) 50.14 (69.3) 44.63 (60.1) 72.11 (90.5) 39.39 (49.5) 17.13 (21.5)Fine sand 0.00 (0.0) 0.37 (0.5) 2.73 (3.7) 7.53 (9.5) 24.37 (30.6) 28.02 (35.2)Calcarenitic boulders 0.00 (0.0) 2.26 (3.1) 4.86 (6.5) 0.00 (0.0) 0.00 (0.0) 0.00 (0.0)Sand and gravel 0.00 (0.0) 7.37 (10.2) 8.88 (11.9) 0.00 (0.0) 1.73 (2.2) 2.68 (3.4)Dead matte 0.00 (0.0) 12.26 (16.9) 6.88 (9.3) 0.00 (0.0) 12.35 (15.5) 25.59 (32.1)C. nodosa 0.00 (0.0) 0.00 (0.0) 0.00 (0.0) 0.00 (0.0) 1.80 (2.3) 4.86 (6.1)Calcareous rubble 0.00 (0.0) 0.00 (0.0) 6.34 (8.5) 0.00 (0.0) 0.00 (0.0) 1.35 (1.7)Total 70.84 (100.0) 72.40 (100.0) 74.31 (100.0) 79.64 (100.0) 79.64 (100.0) 79.64 (100.0)

Table 2Four-way ANOVA showing variation in percent cover by P. oceanica from 1993 to 1999 on five different substrates at shallow (5–15 m) and deep (16–25 m) sites. Ye year, Dedepth, Su substratum, Tr transect. ⁄⁄⁄p < 0.001, ⁄⁄p < 0.01, ⁄p < 0.05. SNK Student–Newman-Keuls test. NS, not significant; MV, mound valley; MC, mound crest; CB, calcareniticboulders; DM, dead matte; SA, sand.

Source of variation Posidonia oceanica percent cover

df MS F

Year 5 601.6455 61.99⁄⁄⁄

Depth 1 0.3508 0.01Substrate 4 5583.1661 148.37⁄⁄⁄

Transect(DeXSu) 10 37.6302 2.21⁄

YeXDe 5 16.3271 1.68YeXSu 20 495.2579 51.03⁄⁄⁄

YeXTr(DeXSu) 50 9.7045 0.57⁄⁄

DeXSu 4 34.6143 0.92YeXDeXSu 20 20.2467 2.09⁄

Residual 360 17.008Total 479

Cochran’s test C = 0.0271 (NS)SNK testInteractionYeXDeXSu

(a) Su (YexDe) Shallow Deep1993 NS NS1994 MV > CB = DM = MC = SA MV > MC = CB = DM = SA1995 MV > MC > CB = DM = SA MV > MC > CB = DM = SA1997 MV > MC > CB = DM = SA MV > MC > CB = DM = SA1998 MV > MC = CB = DM = SA MV > MC > CB = DM = SA1999 MV > MC = CB = DM = SA MV > MC > CB = DM = SA

(b) Ye (SuxDe) Shallow DeepCB NS NSDM NS NSMC NS 93 < 94 = 95 = 97 = 98 = 99MV 93 < 94 < 95 < 97 = 98 < 99 93 < 94 < 95 < 97 = 98 < 99SA NS NS

(c) De (YexSu) CB DM SA MC MV1993 NS NS NS NS NS1994 NS NS NS NS NS1995 NS NS NS NS NS1997 NS NS NS NS NS1998 NS NS NS NS NS1999 NS NS NS NS Shallow > deep

486 F. Badalamenti et al. / Marine Pollution Bulletin 62 (2011) 483–489

Author's personal copy

between substrates from 1994 to 1999 at both depths (SNK test,Table 2). No significant differences were detected in 1993, immedi-ately after the second impact, demonstrating that any potential ini-tial differences did not affect the experiment, as initial cover wassimilar for all transects and depths. From 1993 to 1999, percentcover by P. oceanica was always significantly higher on rubblemounds than on any other substrate at both depths (SNK test a)Table 2). No significant differences were found between sand, deadmatte and calcarenitic boulders at both depth ranges for all years(SNK test, Table 2).

Within mounds, percent cover by P. oceanica was higher on val-leys than on crests for all years at both depths. On valleys, percentcover continuously increased over time, except between 1997 and1998, from a value of 0.75 ± 0.41 (SE) in 1993 to a value of44.38 ± 3.05 (SE) in 1999 (Fig. 3) at shallow sites and from a valueof 0.75 ± 0.41 (SE) in 1993 to a value of 26.88 ± 2.30 (SE) in 1999(Fig. 3) at deep sites (SNK test b) (Table 2). On crests, a significantincrease in P. oceanica cover was detected only between 1993 and1994 at deep sites, whereas values increased from 0.50 ± 0.33 (SE)to 1.75 ± 0.37 (SE) (Fig. 3). No significant differences were recordedbetween the following years or in shallow sites. A significant differ-ence in the percent cover by P. oceanica among transects was de-tected, confirming the high heterogeneity of the environment(Table 2).

4. Discussion

Besides mechanical removal and the burial of vegetation, dredg-ing entails a number of potential ‘‘side-effects’’ to seagrasses,including increased turbidity and sedimentation, current alter-ation, the release of organic matter and the introduction of con-taminants (Erftemeijer and Robin Lewis, 2006). Few cases ofseagrass loss due to dredge-fill activities have been quantifiedworldwide; reported impacts range from a few to several thousandhectares (15,000 ha in Laguna Madre, Texas; Onuf, 1994). Mostcases remain in gray literature or have not been documented(Erftemeijer and Robin Lewis, 2006). To our knowledge, this isthe first study quantifying P. oceanica meadow loss after pipelinedeployment in Mediterranean Sea. These dredge-fill operationscaused a total loss of 81.20 ha of P. oceanica meadows, equivalentto 57% of the initial extent of meadows in the study area, and to

0.2% of the whole seagrass extent of the Western Sicily sub-tidalenvironment (approximately 42,350 ha, SI.DI.MAR, 2002). Theextent of the meadow loss was clearly greater than the trench area,where P. oceanica was directly removed, and wide digitations ofdead matte stretched out in the continuous meadow (Fig. 2),denoting the presence of additional indirect impacts fromdredge-fill activities.

There is a large degree of bias between studies identifyinglosses of seagrasses and those reporting recovery, due also to a lackof long-term monitoring programs (Erftemeijer and Robin Lewis,2006). To our knowledge, few records have documented instancesof natural recovery of P. oceanica meadows after disturbanceevents, commonly reporting vegetative growth from the surround-ing, well-established patches still present in the damaged area(González-Correa et al., 2005, 2008; Diaz-Almela et al., 2008b).Data on recovery from wide, totally un-vegetated areas, involvingpropagule supply, are more scarce (Meinesz and Lefèvre, 1984;Balestri et al., 1998; Balestri and Lardicci, 2008). However, fullrecovery has never been observed and is thought to require centu-ries (Meinesz and Lefèvre, 1984; Duarte, 2002; González-Correaet al., 2005), especially considering the slow horizontal elongationrate of P. oceanica rhizomes, which is estimated, on average, at2–4 cm yr�1 (Duarte, 1991; Marbà and Duarte, 1998).

Our analysis suggests that percent cover by P. oceanica on rub-ble mound valleys increased from 0 to 44.38 ± 3.05 (SE) between�5 and �15 m and from 0 to 26.88 ± 2.30 (SE) between �16 and�25 m over the 7 years following the end of the dredge-fill opera-tion. Full recovery on mound valleys can be estimated to occurafter 13.5 years between �5 and �15 m and after 22.32 yearsbetween �16 and �25 m, on a total surface of approximately6.8 ha. On the basis of substrate maps, we estimate that a total of3.24 ha had already recovered after the 7 years following the endof dredge activities. These results point out that recruitmentthrough vegetative fragments, already documented for Posidoniaaustralis and P. oceanica (Campbell, 2003; Di Carlo et al., 2005),can be a successful strategy for P. oceanica to achieve successfuland fast recovery in suitable conditions. Our results also provideevidence on the importance of substrate type on the recoverypatterns of P. oceanica. Previous studies on post-dredging recoveryand transplantations showed how species able to recover on sandyor muddy substrates (Williams, 1990; Gallegos et al., 1994;Kenworthy et al., 2002; Di Carlo and Kenworthy, 2008) failed to

0

5

10

15

20

25

30

35

40

45

50

1993

1994

1995

1997

1998

1999

1993

1994

1995

1997

1998

1999

P. o

cean

ica p

erce

nt c

over

CB = calcarenitic boulders

DM = dead matte

SA = sand

MC = mound crest

MV = mound valley

DeepShallow

Fig. 3. Variation in percent cover by P. oceanica from 1993 to 1999 on five substrates identified at shallow (5–15 m) and deep (16–25 m) sites in the trench. Bars = standarderror.

F. Badalamenti et al. / Marine Pollution Bulletin 62 (2011) 483–489 487

Author's personal copy

recruit on debris mounds due to sediment instability (Brown-Peterson et al., 1993; Kaldy et al., 2004; Sheridan, 2004), suggest-ing that this feature prevents seagrass re-colonization beyond thehabitat requirements of the species. Posidonia oceanica propaguleshave been Di observed on different substrates (Meinesz andLefèvre, 1984; Piazzi et al., 1999; Balestri and Lardicci, 2008), butonly those that settled on consolidated substrates survived andestablished successfully (Meinesz and Lefèvre, 1984; Balestriet al., 1998; Piazzi et al., 1999). Meinesz and Lefèvre (1984)observed P. oceanica recovery on dead matte through vegetativefragments, suggesting that dead rhizomes emerging from mattefacilitated the entanglement and establishment of propagules.Balestri et al. (1998) and Piazzi et al. (1999) reported seedling sur-vival on dead matte and rock but not on gravel or pebbles, probablydue to substrate instability and the abrasive action of particles onunconsolidated substrates.

If we assume that bottom morphology and substrate distribu-tions fully stabilized in the studied area site within the observationperiod, the differences observed in P. oceanica recovery betweensubstrates must reflect the substrate preference of this species.Calcareous rubbles are made of a consolidated and stable material,which is motionless even in harsh hydrodynamic conditions anddoes not erode over time. Moreover, crevices between adjacentrubbles create a pattern of substrate complexity on the same scaleas the propagules (centimeters, tens of centimeters). Our interpre-tation is that these features made the rubble mounds a suitableenvironment for P. oceanica recovery from vegetative fragments,unlike the other substrates, which lack sufficient complexity (deadmatte, C. nodosa) and/or stability (calcarenitic boulder, sand). Thedifferences in water movement regimes and sedimentation ratesbetween mound crests and mound valleys documented in previouswork (Di Carlo et al., 2005), account for the strong variation inrecovery between these two locations.

Posidonia oceanica seagrass beds are a protected habitat (EEC,1992), and increasing efforts have been made in protection, trans-plantation and restoration (Duarte, 2002; González-Correa et al.,2005; Airoldi and Beck, 2007). Until now, P. oceanica transplanta-tions have been developed mostly on sand or dead matte and makeuse of artificial supports, such as grids, nets or even bioengineeringmaterials, to secure propagules to the substrate (Molenaar andMeinesz, 1995; Gobert et al., 2005; Cinelli et al., 2007). However,transplanting methodologies are still in an experimental phaseand produce highly variable results (Molenaar et al., 1993;Molenaar and Meinesz, 1995; Balestri et al., 1998; Piazzi et al.,1998; Sanchez-Lizaso et al., 2009; Vangeluwe, 2007). In additionto the uncertain success of transplantation projects and the lackof long-term monitoring data, restoration initiatives always re-quire considerable labor and incur significant costs, reducing theapplicability of these techniques in large-scale contexts (Gobertet al., 2005; Orth et al., 2006a; Sanchez-Lizaso et al., 2009). Currenttrends in seagrass restoration ecology aim to develop cost- andlabor-effective technologies, taking advantage of species’ naturalrecovery abilities and of the presence of former populations thatmay act as sources of propagules, accelerating the overall recoveryof these systems (Orth et al., 2006a,b; Wear et al., 2006; Lee andPark, 2008). As growing portions of the Mediterranean coastlineare being engineered, involving the introduction of artificial struc-tures, there is an increasing need to understand the impacts ofthese materials on the marine environment to, in turn, mitigateany negative effects (Burt et al., 2009) or take advantage of positiveones. In this scenario, artificial rubble-like materials, which provedto provide a suitable substrate for P. oceanica recovery, could beemployed in developed coastal areas to support restoration ofdamaged meadows, allowing natural recovery to occur if there isa sufficient propagule supply from adjacent meadows, or toenhance the success of transplantation initiatives.

References

Airoldi, L., Beck, M.W., 2007. Loss, status and trend for coastal marine habitats ofEurope. Oceanogr. Mar. Biol. Annu. Rev. 45, 345–405.

Badalamenti, F., Di Carlo, G., D’Anna, G., Gristina, M., Toccaceli, M., 2006. Effectsof dredging activities on population dynamics of Posidonia oceanica (L.) Delile inthe Mediterranean Sea: the case study of Capo Feto (SW Sicily, Italy).Hydrobiologia 555 (1), 253–261.

Balestri, E., Gobert, S., Lepoint, G., Lardicci, C., 2009. Seed nutrient content andnutritional status of Posidonia oceanica seedlings in the northwesternMediterranean Sea. Mar. Ecol. Prog. Ser. 388, 99–109.

Balestri, E., Lardicci, C., 2008. First evidence of a massive recruitment event inPosidonia oceanica: spatial variation in first-year seedling abundance on aheterogeneous substrate. Estuar. Coast. Shelf Sci. 76 (3), 634–641.

Balestri, E., Piazzi, L., Cinelli, F., 1998. Survival and growth of transplanted andnatural seedlings of Posidonia oceanica (L.) Delile in a damaged coastal area. J.Exp. Mar. Biol. Ecol. 228 (2), 209–225.

Bourgoin, A., Guillou, M., Morvan, C., 1985. Etude préliminaire de épifaune dessédiments meubles de la rade de Brest (Finistere, France) à l’aide d’une caméravidéo sus-marine. Ann. Inst. Océanogr. 61, 39–50.

Brown-Peterson, N.J., Peterson, M.S., Rydene, D.A., Eames, R.W., 1993. Fishassemblages in natural versus well-established recolonized seagrassmeadows. Estuaries 16, 177–189.

Bryars, S., Neverauskas, V., 2004. Natural recolonisation of seagrasses at a disusedsewage sludge outfall. Aquat. Bot. 80 (4), 283–289.

Burt, J., Bartholomew, A., Bauman, A., Saif, A., Sale, P.F., 2009. Coral recruitment andearly benthic community development on several materials used in theconstruction of artificial reefs and breakwaters. J. Exp. Mar. Biol. Ecol. 373 (1),72–78.

Campbell, M.L., 2003. Recruitment and colonisation of vegetative fragments ofPosidonia australis and Posidonia coriacea. Aquat. Bot. 76 (2), 175–184.

Cancemi, G., De Falco, G., Pergent, G., 2003. Effects of organic matter input from afish farming facility on a Posidonia oceanica meadow. Estuar. Coast. Shelf Sci. 56,961–968.

Cardoso, P.G., Raffaelli, D., Pardal, M.A., 2008. The impact of extreme weather eventson the seagrass Zostera noltii and related Hydrobia ulvae population. Mar. Pollut.Bull. 56 (3), 483–492.

Cinelli, F.L., Boccalaro, F., Burgassi, M., Rende, F., Cinelli, F., Piazzi, F., Zanella, M.,2007. Utilizzo sperimentale in mare di sistemi tecnici già impiegatidall’ingegneria naturalistica terrestre. Biol. Mar. Medit. 14 (2), 342–343.

Davis, R.D., 2009. The role of mineral, living, and artificial substrata in thedevelopment of subtidal assemblages. In: Wahl, M. (Ed.), Marine Hard BottomCommunities. Springer-Verlag, Berlin, Heidelberg, pp. 19–37.

Delgado, O., Ruiz, J., Perez, M., Romero, J., Ballesteros, E., 1999. Effects of fish farmingon a segrass (Posidonia oceanica) in a Mediterranean bay: seagrass decline afterorganic loading cessation. Oceanol. Acta 22 (1), 109–117.

Di Carlo, G., Badalamenti, F., Jensen, A., Koch, E., Riggio, S., 2005. Colonisationprocess of vegetative fragments of Posidonia oceanica (L.) Delile on rubblemounds. Mar. Biol. 147 (6), 1261–1270.

Di Carlo, G., Badalamenti, F., Terlizzi, A., 2007. Recruitment of Posidonia oceanica onrubble mounds: substratum effects on biomass partitioning and leafmorphology. Aquat. Bot. 87, 97–103.

Di Carlo, G., Kenworthy, W., 2008. Evaluation of aboveground and belowgroundbiomass recovery in physically disturbed seagrass beds. Oecologia 158, 285–298.

Diaz-Almela, E., Marba, N., Alvarez, E., Santiago, R., Holmer, M., Grau, A., Mirto, S.,Danovaro, R., Petrou, A., Argyrou, M., Karakassis, I., Duarte, C.M., 2008a. Benthicinput rates predict seagrass (Posidonia oceanica) fish farm-induced decline. Mar.Pollut. Bull. 56, 1332–1342.

Diaz-Almela, E., Marbà, N., Álvarez, E., Santiago, R., Martínez, R., Duarte, C.M., 2008b.Patch dynamics of the Mediterranean seagrass Posidonia oceanica: implicationsfor recolonisation process. Aquat. Bot. 89, 397–403.

Duarte, C.M., 1991. Allometric scaling of seagrass form and productivity. Mar. Ecol.Prog. Ser. 77, 289–300.

Duarte, C.M., 2002. The future of seagrass meadows. Environ. Conserv. 29, 192–206.Duarte, C.M., Fourqurean, J.W., Krause-Jensen, D., Olesen, B., 2006. Dynamics of

seagrass stability and change. In: Larkum, A.W.D., Orth, R.J., Duarte, C.M. (Eds.),Seagrasses: Biology Ecology and Conservation. Springer, Dordrecht, TheNederlands, pp. 271–294.

EEC, 1992. Council Directive 92/43/EEC on the conservation of natural habitats andof wild fauna and flora. Official Journal of European Community. L206 of 22/July.

Erftemeijer, P.L.A., Robin Lewis III, R.R., 2006. Environmental impacts of dredging onseagrasses: a review. Mar. Pollut. Bull. 52 (12), 1553–1572.

Ewanchuk, P., Williams, S.L., 1996. Survival and re-establishment of vegetativefragments of eelgrass (Zostera marina). Can. J. Bot. 74, 1584–1590.

Faimali, M., Garaventa, F., Terlizzi, A., Chiantore, M., Cattaneo-Vietti, R., 2004. Theinterplay of substrate nature and biofilm formation in regulating Balanusamphitrite Darwin, 1854 larval settlement. J. Exp. Mar. Biol. Ecol. 306 (1), 37–50.

Gallegos, M., Merino, M., Rodríguez, A., Marbà, N., Duarte, C.M., 1994. Growthpatterns and demography of pioneer Caribbean seagrasses Halodule wrightii andSyringodium filiforme. Mar. Ecol. Prog. Ser. 109, 99–104.

Gobert, S., Lepoint, G., Bouquegneau, J.M., Vangeluwe, D., Eisinger, M., Paster, M.,Schuhmaker, H., van Treeck, P., 2005. Restoration of seagrass meadows: meansand limitation. In: Ozhan, E. (Ed.), MedCoast. Ankara, Turkey, pp. 1323–1334.

488 F. Badalamenti et al. / Marine Pollution Bulletin 62 (2011) 483–489

Author's personal copy

González-Correa, J.M., Bayle, J.T., Sánchez-Lizaso, J.L., Valle, C., Sánchez-Jerez, P.,Ruiz, J.M., 2005. Recovery of deep Posidonia oceanica meadows degraded bytrawling. J. Exp. Mar. Biol. Ecol. 320 (1), 65–76.

González-Correa, J.M., Torquemada, Y.F., Sánchez-Lizaso, J.L., 2008. Long-term effectof beach replenishment on natural recovery of shallow Posidonia oceanicameadows. Estuar. Coast. Shelf Sci. 76 (4), 834–844.

Guarnieri, G., Terlizzi, A., Bevilacqua, S., Fraschetti, S., 2009. Local vs regional effectsof substratum on early colonization stages of sessile assemblages. Biofouling 25(7), 593–604.

Guidetti, P., Fabiano, M., 2000. The use of lepidochronology to assess the impact ofterrigenous discharges on the primary leaf production of the Mediterraneanseagrass Posidonia oceanica. Mar. Pollut. Bull. 40 (5), 449–453.

Hall, L., Hanisak, M.D., Virnstein, R.W., 2006. Fragments of the seagrasses Halodulewrightii and Halophila johnsonii as potential recruits in Indian River Lagoon.Florida. Mar. Ecol. Prog. Ser. 310, 109–117.

IEA, 2009. Word Energy Outlook 2009. OECD/IEA. International Energy Agency,Paris, France. pp. 17.

Kaldy, J.E., Dunton, K.H., Kowalski, J.L., Lee, K.-S., 2004. Factors controlling seagrassrevegetation onto dredged material deposits: a case study in lower lagunaMadre, Texas. J. Coast. Res. 20 (1), 292–300.

Kendrick, G.A., Eckersley, J., Walker, D.I., 1999. Landscape-scale changes in seagrassdistribution over time: a case study from Success Bank, Western Australia.Aquat. Bot. 65 (1–4), 293–309.

Kenworthy, W., Fonseca, M., Whitfield, P.E., Hammerstrom, K., 2002. Analysis ofseagrass recovery in experimental excavation and propeller-scars disturbancein the Florida keys national marine sanctuary. J. Coast. Res. 37, 75–85.

Lee, K.-S., Park, J.-I., 2008. An effective transplanting technique using shells forrestoration of Zostera marina habitats. Mar. Pollut. Bull. 56, 1015–1021.

Marbà, N., Duarte, C.M., 1998. Rhizome elongation and seagrass clonal growth. Mar.Ecol. Prog. Ser. 174, 269–280.

Meinesz, A., Lefèvre, J.R., 1984. Régénération d’un herbier de Posidonia oceanicaquarante années après sa destruction par une bombe dans la rade deVillefranche (Alpes-Maritimes, France). In: Boudouresque, C.F., Jeudy deGrissac, A., Olivier, J. (Eds.), International Workshop Posidonia oceanica Beds.GIS Posidonie, France, pp. 39–44.

Molenaar, H., Meinesz, A., 1995. Vegetative reproduction in Posidonia oceanica:survival and development of transplanted cuttings according to differentspacings, arrangements and substrates. Bot. Mar. 38, 313–322.

Molenaar, H., Meinesz, A., Caye, G., 1993. Vegetative Reproduction in Posidoniaoceanica. Survival and development in different morphological types oftransplanted cuttings. Bot. Mar. 36, 481–488.

Montefalcone, M., Chiantore, M., Lanzone, A., Morri, C., Albertelli, G., Bianchi, C.N.,2008. BACI design reveals the decline of the seagrass Posidonia oceanica inducedby anchoring. Mar. Pollut. Bull. 56 (9), 1637–1645.

Olesen, B., Marba, N., Duarte, C., Savela, R., Fortes, M., 2004. Recolonizationdynamics in a mixed seagrass meadow: the role of clonal versus sexualprocesses. Estuar.Coast. 27 (5), 770–780.

Onuf, C.P., 1994. Seagrasses, dredging and light in Laguna Madre, Texas, USA. Estuar.Coast. Shelf Sci. 39, 75–91.

Orth, R.J., Carruthers, T.J.B., Dennison, W.C., Duarte, C.M., Fourqurean, J.W., Heck Jr.,K.L., Hughes, A.R., Kendrick, G.A., Short, F.T., Waycott, M., Williams, S.L., 2006a. Aglobal crisis for seagrass ecosystems. BioScience 56 (12), 987–996.

Orth, R.J., Luckenbach, M.L., Marion, S.R., Moore, K.A., Wilcox, D.J., 2006b. Seagrassrecovery in the Delmarva Coastal Bays, USA. Aquat. Bot. 84 (1), 26–36.

Perez, M., Garcia, T., Invers, O., Ruiz, J.M., 2008. Physiological responses of theseagrass Posidonia oceanica as indicators of fish farm impact. Mar. Pollut. Bull.56 (5), 869–879.

Piazzi, L., Balestri, E., Magri, M., Cinelli, F., 1998. Experimental transplanting ofPosidonia oceanica (L.) Delile into a disturbed habitat in the Mediterranean Sea.Bot. Mar. 41, 593–601.

Piazzi, L., Acunto, S., Cinelli, F., 1999. In situ survival and development of Posidoniaoceanica (L.) Delile seedlings. Aquat. Bot. 63 (2), 103–112.

Preen, A.R., Lee Long, W.J., Coles, R.G., 1995. Flood and cyclone related loss, andpartial recovery, of more than 1000 km2 of seagrass in Hervey Bay, Queensland,Australia. Aquat. Bot. 52 (1–2), 3–17.

Rasheed, M.A., 2004. Recovery and succession in a multi-species tropical seagrassmeadow following experimental disturbance. the role of sexual and asexualreproduction. J. Exp. Mar. Biol. Ecol. 310 (1), 13–45.

Rollon, R.N., Van Steveninck, E.D.D.R., Van Vierssen, W., Fortes, M.D., 1999.Contrasting recolonization strategies in multi-species seagrass meadows. Mar.Pollut. Bull. 37 (8–12), 450–459.

Ruggeri, G., Unti, A., Unti, M., Moroni, M.A., 1975. La calcarenite di Marsala(Pleistocene inferiore) ed i terreni contermini. Boll. Soc. Geol. It. 94, 1623–1657.

Ruiz, J.M., Pérez, M., Romero, J., 2001. Effects of fish farm loadings on seagrass(Posidonia oceanica) distribution, growth and photosynthesis. Mar. Pollut. Bull.42 (9), 749–760.

Ruiz, J.M., Romero, J., 2003. Effects of disturbances caused by coastal constructionson spatial structure, growth dynamics and photosynthesis of the seagrassPosidonia oceanica. Mar. Pollut. Bull. 46 (12), 1523–1533.

Sanchez-Lizaso, J.L., Fernandez-Torquemada, Y., Gonzalez-Correa, J.M., 2009.Evaluation of the viability of Posidonia oceanica transplants associated with amarina expansion. Bot. Mar. 52, 471–476.

Sheridan, P., 2004. Recovery of floral and faunal communities after placement ofdredged material on seagrasses in Laguna Madre, Texas. Estuar. Coast. Shelf Sci.59, 441–458.

Short, F.T., Wyllie-Echeverria, S., 1996. Natural and human-induced distrurbance ofseagrasses. Environ. Conserv. 23, 17–27.

SI.DI.MAR, 2002. Mappatura delle praterie di Posidonia oceanica lungo le coste dellaSicilia e delle isole minori circostanti. Dati Ministero dell’Ambiente e dellaTutela del Territorio e del Mare -Direzione Protezione Natura- SI.DI.MAR.

Tubb, R., 2010. Pipeline & Gas Journal’s 2010. International Pipeline ConstructionReport. Pipeline & Gas Journal 237(2), pp. 12.

Underwood, A.J., 1997. Experiments in ecology: their logical design andinterpretation using analysis of variance. Cambridge University Press,Cambridge.

Vangeluwe, D., 2007. Effets de la transplantation sur la biométrie et sur ladynamique des nutriments, du carbone et de la chlorophylle de Posidoniaoceanica (L.) Delile. PhD Thesis, Faculté des Sciences Département des Scienceset Gestion de l’Environnement Océanographie biologique. Université de Liège.

Waycott, M., Duarte, C.M., Carruthers, T.J.B., Robert, J.O., Dennison, W.C., Olyarnik,S., Calladine, A., Fourqurean, J.W., Heck, K.H., Hughes, A.R., Kendrick, G.A.,Kenworthy, W.J., Short, F.T., Williams, S.L., 2009. Accelerating loss of seagrassesaccross the globe threatens coastal ecosystems. PNAS 106 (30), 12377–12381.

Wear, R., Tanner, J.E., Venema, S., 2006. Seagrass Rehabilitation in AdelaideMetropolitan Coastal Waters III. Development of Recruitment FacilitationMethodologies. SARDI Research Report Series. South Australian Research andDevelopment Institution, 48, pp. 44.

Whitfield, P.E., Kenworthy, W.J., Michael, J.D., Kamille, K.H., Manuel, F.M., 2004.Recruitment of Thalassia testudinum seedlings into physically disturbedseagrass beds. Mar. Ecol. Prog. Ser. 267, 121–131.

Winer, B.J., Brown, D.R., Michels, K.M., 1991. Statistical Principles in ExperimentalDesign. Mc Grow-Hill, Boston.

Williams, S.L., 1990. Experimental studies of Caribbean Seagrass bed development.Ecol. Monogr. 60 (4), 449–469.

F. Badalamenti et al. / Marine Pollution Bulletin 62 (2011) 483–489 489