Journal of Coastal Research 20 4 1155-1164 West Palm Beach, Florida Fall 2004

Seasonal Dynamics of a Microtidal Pocket Beach withPosidonia oceanica Seabeds (Mallorca, Spain)G. Basterretxeat, A. Orfilat, A. Jordit, B. Casast, P. Lynettt, P.L.F. Liu§, C. M. Duartet, and J. Tintorei

"!"Grupo de OceanografiaInterdisciplinar

IMEDEA (UIB-CSIC)Miquel Marques 2107190 Esporles, Baleares,

Spainvieagbo@uib.es

^Department of CivilEngineering

Texas A&M UniversityCollege Station, TX 77845.

U.S.A.

§School of Civil andEnvironmental Engineering

Cornell UniversityIthaca, NY 14853, U.S.A.

ABSTRACT I

BASTERRETXEA. G.; ORFILA, A.; JORDI, A.; CASAS, B.; LYNETT, P.; LIU, P.L.F.; DUARTE. CM., and TINTORfi,J., 2004. Seasonal dynamics of a microtidal pocket beach witb posidonia oceanica seabeds (Mallorca, Spain). Journalof Coastal Research. 20l4i, 1155-1164. West Palm Beacb (Florida). ISSN 0749-0208.

In this work, we analyze tbe .'seasonal evolution of a Mediterranean pocket beacb and its response to different stormepisodes. Magalluf, an intermediate medium sand beacb located in the Bay of Palma iBalearic Islands! was monitoredby topograpbic levelling during 14 months. Near tbe beach, a Posidonia oceanica meadow covers most of tbe seabedand appears to influence the cross-shore beach adjustment. The low variability observed during tbe sampling periodwas perturbed by two storm events that caused significant beacb evolution and sediment transport, Tbe first stormgave rise to waves from tbe SE, significant beigbt = 2.4 m, cross-shore sediment transport and along-sbore netsediment excbange tbat resulted in decreased dry beach extension to a minimum. The .second storm was characterizedby strong northeasterly winds and generated a set-up of 0.5 m and a nearshore drift reversal tbat redistributedsediment from the berm crest to the beacb face, thereby increasing beacb extension. Results from numerical simu-lations of wave propagation show tbe circulation patterns during botb events and their influence on tbe beacb mor-pbology. In general terms, the beach exhibited a bomeostatic bebaviour cbaracteristic of an equilibrium system,

ADDITIONAL INDEX WORDS: Morphodynamics, beach profile, sediment transport, Mediterranean Sea.

INTRODUCTION

The southern and eastern coasts of Mallorca (Balearic Is-lands, Spain I present numerous pocket beaches spread alongan irregular coastal morphology. The absence of significanttides in the Mediterranean Sea. and the generally weakcoastal currents, restrict significant heach morphology chang-es to severe weather episodes when wave related processesare enhanced. Indeed, GUILL£;N el ol. (20021 report that thesediment dynamics in the shallow part of the nearby Cata-lonian shelf is largely controlled hy wave-storm events.

One of the most characteristic features of Mallorcan heach-es is that, as in many Mediterranean coasts, the nearshoresandy bottoms are colonized by the endemic reef-buildingseagrass Posidonia oceanica which is known to affect near-coast sediment dynamics (MARBA et al, 2002), Seagrasses andother vegetation attenuate wave and current energy, there-fore diminishing sediment resuspension, erosion and trans-port ( FONSECA and CALAHAN, 1992; WORCESTER, 1995;MENDEZet al., 1999), Segrasses also affect and enhance particle de-position (GACIA et al., 1999; TERRADOS and DUARTK. 1999; GA-CIA and DUARTE, 2001). All these processes, although not yetfully understood, have obvious implications on the morphol-

03-0027R received 03 April 2003; accepted in revision 07 July 2003-1123.

ogy and variability of the adjacent heach on a seasonal timescale.

The upper growth limit of Posidonia oceanica is condi-tioned, among other factors, hy sediment stability and there-by this plant rarely grows in the very nearshore (less than3-4 m depth) where the vertical growth rate of the plant,around 1,5 cm year ', (BOIJDOURESQUE et al, 1984; PERGENTet al, 1989; MARBA and DI'ARTE, 1997) cannot cope with theintense sedimentary dynamics. The slow horizontal growthrate of the seagrass. about 5 cm year ' (MARBA and DL'ARTE,1998) precludes the rapid shoreward advances of the mead-ow.

On a longer time scale, the meadow plays an active role inthe sedimentary balance of the beach both supplying biogenicsand and/or trapping sediments in eventual offshore migra-tions. There is evidence that Posidonia oceanica. beds can pro-duce 60 to 70 gr m ^ year ' of calcium carbonate, whichtranslates to an important yield when scaled to the large areaoccupied by these meadows (CANALS and BALLESTEROS, 1997).In fact, Mallorcan beaches are composed by medium and finesand with a large percentage (more than 70*7 ) of bioclasticsediments (JAUME and FORNOS, 1992; RODRIGUEZ-PEREA e/ ai.2000) most of which derive from organisms associated withthe Posidonia oceanica meadow. The amount of'new'hiogenicsediment reaching the heach is nevertheless uncertain.

While most of the literature on the influence of Posidonia

1156 Basterretxea et al.

460000 461000

10

WANA Point

40^

C. Enderrocat

C. Regana

456000 464000 472000 480000

Longitude (UTM)

Figure 1. Map of the study area displaying the coastal bathymetry and the location of the sampled profiles (PI and P2>.

oceanica and other seagrasses in the coastal areas has beendevoted to the analysis of the interaction between plants andcurrents, particles, turbulence, etc., (e.g. WARD et al., 1984,FONSECA and CALAHAN, 1992; GRANATA et al., 2001) the mor-phodynamic behavior of these beaches is poorly understood.Moreover, to the best of our knowledge there are no previouspublished studies of beacb evolution on Mediterranean pock-et beaches. In this paper we examine seasonal time and spaceevolution of Magalluf heach (SW Mallorcal, located in an em-bayment with substantial Posidonia coverage. The samphngperiod covers two storm events where the beach was signifi-cantly, albeit differently, modified.

REGIONAL SETTING

Magalluf is a 900 m sheltered urban beach located in asmall embayment on the southwestern coast of Mallorcawithin the bay of Palma (Figure 1). The coast embraces ashallow water inlet with substantial Posidonia oceanica cov-

erage (up to 78'7c], protected in its offshore opening by theisland of Sa Porrasa (Figure 21. A small pier in the middle ofthe beach that bides a concrete rainwater outfall acts as abarrier to the coastal alongshore flow. During 1996, the beachwas renourisbed with 85,000 m of medium sand (D o -0.35±0.06 mm, n = 79) with 70% content of calcareous ma-terial dredged from a coastal location off the Bay (Rafeubetx ].

Forcing by tides is almost negligible in the Mediterranean,with a spring tidal range of less than 0.25 m. Changes inatmospheric pressure and wind stress can account for a con-siderable portion of sealevel fluctuations. The embayment isopened to seas from the S and SE although most frequentwaves approach the coast from the southwest resulting in alittoral drift to the northeast. Southwesterly winds can gen-erate waves of 3 to 4 m which reach Punta Cala Figuera butwaves diminish as they propagate inside the bay. Indeed,wave estimations from Puertos del Estado at the nearbyWANA point 2069036 (lat 39.500 N Ion 2.625 E) located at

Journal of Coastal Research, Vol. 20, No. 4, 2004

Seasonal Dynamics of Beach with Posidonia Seabeds 1107

Figure 2. Aerial photograph of Magalluf beach in 2001. The submerged light patches indicate places without Posidonia oceanica coverage.

40 m depth east from Magalluf, indicate that the direction ofmore energetic waves are from the southwest rarely exceed-ing significant heights (Hj of 2,5 m.

Typical weather conditions in the Bay of Palma, duringsummer, exhibit a persistent sea breeze regime (RAMlSe/ al,1990) with weak onshore winds iless than 5 m/sl tbat onlyoccasionally increase up to 10 m/s. From late summer to earlywinter, storm episodes are frequent with viinds blowing witha SW or NE component.

METHODS

From October 2000 to December 2001 two heach profllesiPl and P2) and the shoreline position were sampled hi-monthly in Magalluf by precision topographic leveling usingan electronic total station. The topographic method was ex-tended to the sea by using a dinghy and an extendable rodI up to 9 meters) with a double prism to facilitate targeting.Measurement points where sampled at less than 6 m inter-vals in the subaerial beach and the swash zone, whereas mea-surements 15 m apart were carried out for the rest of theprofile. Errors in the measurement depend on the distance tothe instrument. For a distance of 400 m a vertical accuracywithin ± 0.03 m was estimated.

The flrst sampling was carried out the day after a strongstorm (12 '' October 2000) and hence, although no previoussurvey was undertaken, the effects on the beach, latter dis-cussed, are attributed to this event. The last sampling (De-cember 2001) was performed almost four weeks after a severestorm also affected the islands.

The spatial and temporal variations of shoreline and beachprofiles were statistically analyzed using the one dimensionalempirical orthogonal function (EOF) technique. This ap-

proach has been used in beacb evolution studies to analyzespatial and temporal variations in beach profiles and plants(e.g. HAYDEN et al., 1975; WiNANTe ai, 1975; AUBREY, 1979)and as other eigenvector techniques encompass the represen-tation of the observed data onto a set of shape functions (theEOFsl. It should be stated that although the EOFs are opti-mal in a statistical sense the eigenfunctions do not necessar-ily have a physical meaning. Nevertheless sucb interpreta-tions are possible in many cases.

We next briefly describe, for completeness, the methodused: consider a series of beach profiles i|/(x,,,, t,) of topograph-ic data collected at M locations, x J l s m s M ) at times t ^ t,(l<i<N). The goal of EOF analysis is to write the data seriesat any given location as a sum of M orthogonal functions suchthat

where 4( is offshore direction or height, depending if the shoreline or profile is analyzed, (b,,,, represent the spatial modesand a/t) are the amplitudes of the ith orthogonal mode attime tn. The weights a,(t) provide information of how the spa-tial modes 6,,,, vary with time. An orthogonality condition canbe imposed in the form

= S,

where ft,^ is the Kronecker delta

£ 1, J = I'0, J ^ i

Each profile was interpolated in one meter equidistant

Journal of Coastal Research, Vol. 20, No. 4, 2004

1158 Basterretxea et al.

points by linear interpolation whereas the shoreline was in-terpolated in 10 m intervals.

One of the key points for shoreline evolution is the defini-tion of a common level for all samplings. The criterion fol-lowed in this work was to select a level located in the swashzone in all the surveys. This is only possible at places wheretidal oscillations are minimal and hence the vertical positionof the swash zone is almost invariant. Two measurements atthe approximate high and low points of the swash zone wereobtained at each alongshore point, and tbe shoreline was in-terpolated to a common level assuming a linear slope. Thepresence of a linear shaped swash zone is a well-acceptedfeature for most beaches.

RESULTS FROM THE EIELD EXPERIMENT

Atmospheric Conditions

During the sampling period two storms were observed toproduce significant effects on the heach. Just prior to the firstsampling (October 11) strong winds from the SW swept tbeislands. This event preceded a cold front crossing the IberianPeninsula, a typical situation for this region during fall.Strong winds of more than 10 m/s (Figure 3a) witb gusts of14.5m/s lasted for one day during which time winds steadilyhiew with a vector direction of 238±10'' generating waveswith significant heights IH ) of 2.4 m and peak spectral pe-riods (Tj,) of 8.2 s within the bay. Although the storm was notparticularly intense, the direction of the associated wave fielddirectly impacted the beacb.

A second severe meteorological event, with infiuence overmost local beaches, occurred during fall 2001. From Novem-ber 11th to 15th the synoptic situation in the western Medi-terranean revealed tbe presence a high pressure crest ex-tending from the Atlantic to central Europe (1038 to 1033hPa) and a depression (993 bPa) north of Africa. This situa-tion generated a strong isoharic gradient with winds form tbeN-NE. On November 11 the depression moved northwardsintensifying the gradient and causing winds of 15 m/s fromthe N and NE (Figure 3b I with maximum wind gusts of morethan 33 m/s. Secondary wind intensification occurred on No-vember 15 when the low, which bad progressed northeast-ward, retreated back to the Balearic Islands. The storm de-veloped strong seas from the N, a direction which the bay isprotected, and only H of 1.3 m and Tp 4.05 s was estimatedat the WANA point. The examination of time series of seal-evel reveals a setup of some 0.48 m during the passage of thedepression, largely exceeding the tidal range.

Longshore Variations

As in most headland bounded beaches, the shoreline shapein Magalluf is arcuated hut, in this case, the pier represents adiscontinuity tbat divides the shoreline in two sections restrict-ing the movement of sand across tbe shoreface (Figure 4l.

Most variations observed in shoreline shape were relativelysmall beach rotations that can he attributed to varying waveconditions; although erosive and accretive periods are stillevident. Beach rotation occurs around axes centered in nodalpoints at about the middle of each beach segment and con-

Dae

1000

0.4

E 0.2

0

-0.2

Seatevel

'V

1Sep

152001

29 13Od

27 10Nov

Figure 3. Tinie series nf low-passed wind velocity, (al and speed (bi dur-ing fall 2000 and 2001. Wind veLlors have been rotated 60° counter-clock-wise. The tidal component has been superimposed in Id) as a reference.

sequently, maximum shoreline displacements take place bothsides of the beach (A and B) and near the pier (B and Cl.Beach rotation caused a maximum annual shoreline oscilla-tion of 14.7 and 14.9 m at points A and C and of 7.8 and 5.6m at points B and D.

For most of the sampling period, sand accumulation wasobserved in the southern side of tbe pier IB) suggesting anorthward littoral drift. Wave records (not shown) supportthis evidence indicating northwestward propagation. The ex-ception is tbe December 2001 sampling in which the northernpart of the beach does not show rotation as in previous sur-veys and instead, subaerial beacb accretion occurred, in par-ticular near the pier. The gained beach surface in the north-em section was 1,237 m^ with a maximum offshore displace-ment of 8 m at point C. This sand accumulation is indicativeof alongshore flow reversal during tbe storm.

Minimum subaerial beach extension (.53,886 m ) occurred

Journal of Coastal Researcb, Vol. 20. No. 4. 2004

Seasonal Dynamics of Beach with Posidonia Seabeds 1159

Table 1. Subaerial beach extension and net surface changes.

os:1^ 300

100-

0 50 100 150 200Crossshore Distance (m)

Figure 4. Map showing shoreline variations at Magaliuf and approxi-mate location of nodal points. The grey line corresponds to the sur%'eyafter November 2001 storm. Dasbed lines indicate the position of beachprofiles PI and F2. The coast has been rotated for convenience.

on October 2000, just after a storm from the SW, when thebeach presented a minimum offshore distance at both flanks.Relevant beach accretion 12 m/m shoreline) followed this ep-isode until the beach stabilized in values between 55,000 and57,000 m (Table 1). In any case, variations did not exceed3.37f of total subaerial beach surface with average shorelineoscillations ranging from 0.6 to 1 m/month.

Oct 2000Dec 2000Feb 2001Apr 2001Jun 2001Aug 2001Oct 2001Dec 2001

Subaeriai

Surface x 10^(m l

53.8955.6656.2356.1956.0255.0755.1056.77

Beach Extension

^surface X 10-' Ini^i

—

1.770.57

-0-04-0.17-0.95

0.031.67

CROSS-SHORE VARIATIONS

Beach profiles, extending some 750 m offshore, show thegeneral topographic patterns of Magalluf bay (Figure 5). Thebeach {sandy bottom I extends about 100 m from the shore toa depth of 4 m where sparse Posidonia oceanica patches pro-gressively appear interleaved with sandy bottoms. The drybeach extends for about 50 meters with a slope of 1% acrossthe berm, markedly sloping in the beach face (lO'/r). This con-figuration is typical of medium to coarse-sand beaches. Thevalue at the beach face agrees with the theoretical value ofBASCOM(1951) fora D. ^ of 0.35mm in a low energy beach (seeKoMAR 1998). The general across-shore trend of the beachprofile is very similar in both of the profiles (PI and P2) withdifferences of less than 1.59f. The inshore, between 5 and 7meters depth, is covered almost completely by a Pnsidoniaoceanica meadow with a profile structure that resembles theshape of reef-protected beaches.

The set of beach profiles in Magalluf exhibits little vari-ability, mainly concentrated on the swash zone (Figure 6).Profile 1 shows minor changes, with the exception of a smallregion at 3.5 m depth where an incipient sand bar or step isformed during some periods. Profile 2 presents larger fluc-tuations in the heach face. These differences between profilesare attributable to the relative position of the profile to thenearby nodal point. Profile 1 is only a few tenths of metersaway from Node 1, whereas P2 is located in an area of highervariability, far from Node 2. The most conspicuous featuresof the profiles are (1) the loss of berm sediment in December

200 1000 1400 leoo

Ping nunnber2200 2600 3000

Figure 5. Acoustic cross-shore profiles of the seahed along Fl and F2. The grey bars indicate the position of the vegetated areas. The represented recordroughly corresponds to a distance of 750 m.

Journal of Coa.stal Re.search, Vol. 20, No. 4, 2004

1160 Basterretxea et al.

40 60 80 100Offshore Distance (m)

120

Figure 6. Beach profiles at PI and P2 for the period October 2000-Oc-tober 2001

2001 profile and (2) the lateral profile migration on October2000. Both features are related to storm events, nevertheless,the effects on the beach are different prohably due to thedistinct nature of the storms.

The small variahility of the profiles suggests a relativeequilibrium that can be parameterised by nonlinear fitting ofthe equation of BRUUN (1954, 1962) and DEAN (1977, 1991,1997):

h = (1)

where A is an adimensional shape parameter related to themean grain diameter of the beach sediment (MOORE, 1982)and to the corresponding grain settling velocity (DEAN, 1987),and m is related to the concavity of the profile. Values of mfor Magalluf heach, obtained hy non-linear fitting, indicatean intermediate beach profile (0.57±0.02 and 0.60±0.04 forPI and P2) with a slightly steeper value than the 2/3 obtainedBRUUN and DEAN. However, the values of A (0.27±0.05 and0.35±0.03) lie far from the value (about 0.18J estimated fromthe empirical relation

A - 0.067UJ " (2)

where w^ is the sediment settling velocity, in our case, ob-tained following GiBSs et al., (1971). Theses agreements anddifferences are discussed in section: discussion and conclu-sions.

The comparison of profiles, excluding Oct 2000 sampling,suggests an annual depth of closure depth (ft,,) of 4 m at PIand even shallower at P2. This value is consistent with thetheoretical value obtained witb the expression after (HALLER-

MEIER, 1981):

h.. l i d

where H^ is the mean annual significant wave 0.5m and a isthe standard deviation.

Volume changes computed for PI and P2 are shown in Ta-ble 2. According to these calculations, most important accre-

Table 2. Net and cumulative volumetric changes in PI and P2 per unitlongshore length.

AccretionImVm)

Erosion(mVm)

Balance Cumulative(m'/m) imVni)

Profile 1Oct-Dec 2000Dec-Feb 2001Feb-Apr 2001Apr^un 2001Jun-Aug 2001Aug-Oct 2001Oct-Dec 2001

Profile 2Oct-Dec 2000Dec-Feb 2001Feb-Apr 2001Apr-Jun 2001Jun-Aug 2001Aug-Oct 2001Oct-Dec 2001

20.411.80.91.12.48.75.8

29.45.84.22.32.01.84.6

0.92.4

12.93.25.04.47.9

0.02.13.97,75.65.1

10.4

+ 19.5+9.4

-12.0-2.1-2.7+4.3-2.1

+ 29.4+3.8+0.3-5.4-3.6-3.2-5.7

19.528.916.914.812.116.414.3

29.433.233.528.124.521.315.6

tion occurred in both profiles after Octoher 2000 storm. Thisnet sediment gain implies sediment transport from other ar-eas. Profile 1 also displays relevant gains and losses ( +11.81and -12.87 m-') in the subsequent periods that correspond tothe appearance and erosion of a sandbar. For the rest of thesampling period the budgets for both profiles are essentiallyin halance. Note that November 2001 storm reorganized thesediments without dramatic net sediment losses.

EOF Analysis

Table 3 summarizes the results of EOF heach surface andprofile analysis. It shows that the first eigenfunction accountsfor more than 99.8% of the variance. The high 'representativ-ity' of the first mode can mask the results of the second andthird EOF. The first EOF modes describe the mean profilesand the related temporal functions describe fiuctuations inmean steepness, where higher values imply a steeper thanaverage profile (Figure 7). The second and third eigenfunc-tion describe small modulations of the mean profile shape.The second EOF modes relate to the upper shoreface andspecifically, to the sediment exchange between the herm crestand the nearshore. The temporal functions of this EOF revealenhanced sediment transference after both storms. The thirdEOF identifies the occurrence of sandhars which is particu-larly noticeable at PI (100 m). No systematic variation is ob-served in this function.

Table 3. EOF explained variance.

Profile 1

Profile 2

EOF

123123

Explained Variante

MeanIncluded

99.920.030.02

99.800.150.02

MeanRemoved

65.1725.555.25

68.8025.833.82

Journal of Coastal Researeb. Vol. 20, No. 4, 2004

Seasonal Dynamics of Beach with Posidonia Seaheds 1161

20 40 60 80 100Cross-shore distance (m)

0.36

2 3 4 5 6 7 8Sampling number

Figure 7. First, second and third EOF (a, b, c). Amplitudes for the firstmode (d) and for the second (e) and third mode (dotted hnei. Profile 1(dark linel and Profile (2) grey line.

NUMERICAL SIMULATIONS

With the aitn of exploring possible coastal drift modifica-tions during the storms, a wave propagation model (L'YTV'ETT,

2002) was used. The model was forced with wave parameters(H.. and T ) obtained from the nearby WANA point corre-sponding to the above mentioned storms. The numerical al-gorithm uses a high order, finite difference scheme and isbased on assumptions of an inviscid fluid and irrotationalflow. The waves are assumed to be highly nonlinear andweakly dispersive, although the model has good linear andnonlinear accuracy up to kb~6. The incident waves are gen-erated using an internal source, which creates the correctStokes drift as well as the regular wave train. No restrictionis placed on the bathymetry or the dominant wave propaga-tion direction. No attempt has been made to incorporate theeffect of the meadow on bottom friction.

Examination of the wave field simulation for October 2000storm shows the important effect of the island and bottombathymetry on wave height and related current velocity (Fig-ure 8). The 15 m isobath is an approximate boundary sepa-rating the offshore field, with waves of 2.4 m from the SW.from the complex pattern over the Posidonia oceanica mead-

ow that is consequence of wave refraction in the proximity ofthe island and coastal reflexion.

At the emhayment boundary wave-generated currents areenhanced (~ 0.8 m/s) and forced to flow following bottom to-pography. Further onshore, the island seems to induce cy-clonic circulation, although there is no clear pattern with theexception of the northernmost area where wave refractionyields a weak cyclonic circulation and enhanced free surfaceoscillations. A weak and variable alongshore fiow in thesouthern section of the beach strongly intensifies and devi-ates in an offshore direction in the northern side, in accor-dance with the observed morphological changes.

The simulation for November 2001 storm (Figure 8b andd) shows a boundary wave propagating from the North andstrongly attenuated waves in Magalluf originated by wavediffraction and subsequent refraction. This sheltering and re-fraction result in a longshore transport directed southward(Fig 8d).

DISCUSSION AND CONCLUSIONS

We have shown that Magalluf beach is characterized by asmall variability typical of low-energy areas. The seasonalsignal in the profile and shoreline variability is weak owingto tbe bay geomorphology that restricts alongshore transport,the lack of strong tidal currents and the presence of Posidoniaoceanica meadows that hamper near bottom transport. In thecase of Magalluf, this lack of variability is further enhancedby the protective effect of Sa Porrasa island and by the pres-ence of a concrete structure in the beach restricting the move-ments to small adjustments mainly attributed to changingwave field conditions. The sediment transport is highly con-strained by this configuration, ensuring the conservation oftotal volume of sand. Only when the nearshore energy is in-creased during episodic storm events is the energy supplysufficient as to promote significant alongshore or/and cross-shore exchanges with the meadow.

One of the peculiarities that the meadow confers to thesebeaches is a relative stability even though a complete beachprofile, down to h,., is never developed. The beach is sustainedby the reef-like structure of the meadow and tbe offshore partof tbe profile extends inside the shoots being in some casesincorporated to the consolidated substrate created by the rhi-zome structure of the plant. Nevertheless, this is a long last-ing process, as reef formation by Posidonia oceanica requiresmillenary timescales. At shorter time scales (months), Posi-donia oceanica meadows should not be viewed as sedimentflxing organisms. Substantial amounts of sediment can mi-grate throughout the meadow, by resuspension and throughbedform migi'ation, and eventually reincorporate to the beachface dynamics. For example, the increase in subaerial beachextension after the storm registered in October 2000 involvedan important onshore sand migration (up to 24 m'Vm). Fur-thermore, Posidonia meadow.s can contribute to beacb sus-tainment by exporting calcium carbonate material derivedfrom fragmentation and abrasion of calcifying organisms in-habiting tbe meadow.

The general heach morphology of Magalluf fits in the lowenergy extreme (reflective statej of the wave dominated mod-

Joumal of Coastal Research. Vol. 20, No. 4, 2004

1162 Basterretxea et al.

ii;

Figure 8. Free surface oscillations snapshot (in) and wave induced circulation during Octoher 2002 (a and cl and November 2001 (b and d) stormepisodes.

d

iiil

• i l l

III1illii i I

i 1

,ij

el of WRIGHT and SHORT (1984) characterized by a steepbeachface slope, narrow surf zone and absence of bar. Nev-ertheless, the presence of a Posidonia oceanica in the nearshore conditions the shape of equilibrium profiles. It is plau-sible that if the equilibrium profile represents a state inwhich flows and morphology adjust to produce a uniform dis-sipation of energy, the 'extra" amount of energy dissipated bytbe meadow results in a cbange in the beacb sbape towardssteeper forms. MUNOZ-PEREZ ef al., (1999) stress that for tbesame grain size, in reef-protected beaches, tbe shape param-eter will not be the same as the A value used in the usualDean equilibrium proflle. The relationship between the fittedvalues A , and the theoretical ones A,,, yields values of 1.6-1.8 that are close to those obtained for reef-protected beaches(1.48). The higher values in our case could be indicative ofhigher energy dissipation over the Posidonia meadow thanover reef-like structures, as shoaling waves travelling overhard or cohesive bottoms loss less energy than waves rollingover vegetated seafloors. Consequently, losses of seagrass ex-tent would lead to sediment redistribution towards more dis-

sipative shapes and probably, subaerial beach erosion wouldfollow, as observed in the past [e.g. WILSON, 1949; CHRISTIAN-SEN et al, 1981; HiNE et al., 1987).

Regarding the storms, NORDSTROM (1992) essentially ob-served two types of profile response on sheltered sandybeacbes: (1) transfer of sediment from the upper to tbe lowershoreface and (2) parallel retreat of the foreshore related toan alongshore current flow. In the present study, we docu-ment two different ways by which enhanced energy episodescan influence beach geomorphology.

The October 2000 storm generated seas from the prevailingwave direction resulting in an intensified cross-shore trans-port. Low alongshore transport during this storm indicates agood adjustment of the beacb sbape to these conditions.Hence the increased wave energy is mainly dissipatedthrough changes in beach profile. The lack of sampling pre-vious to the storm weakens conclusions on the effects of thestorm, but the absence of energetic events (H^ > 2 m) in theprevious months suggests that most of the erosion can beattributed to this storm. Numerical simulations confirmed

Journal of Coastal Research. Vol. 20. No. 4, 2004

Seasonal Dynamics of Beach with Posidonia Seabeds 1163

the observed strong cross-shore component in the near-shoreflow patterns. During this episode of beach erosion sand wastransferred beyond the theoretical annual h,. The consequentsediment loss resulted in a minimum subaerial beach exten-sion that was recovered in the subsequent weeks (less thana sampling period). This rapid recovery of sand is indicativeof a high degree of resilience, or tendency to re-establishmentof steady state with short relaxation times. It is also a keyprocess for the survival o^Posidonia oceanica shoots that can-not bear oversedimentation for long periods IMANZANERA e/oL 1998).

Rapid beach accretion requires sand migration and resus-pension that depends on oscillatory wave motions, hence, itis only possible under certain amount of energy input. Inci-dent waves tend to transport the sediment shoreward; thistransport to some extent being balanced by offshore meanflows and oscillatory transport at infragravity frequencies(OsBORNE and GREENWOOD, 1992a,h). Infragravity motionsappear to assume significance under storm conditions (AA-(iAARl) and GREENU'OOD, 1995) and especially in the inner surfand swash zones IWRIOHT and SHORT, 1983; BEACH et al,1991; RiLSSEL, 1993) but beach recovery may begin at the laststages of a storm event.

The set-up during November 2001 storm varied the verticalposition where wave processes act and made available sandfrom the berm crest to the nearshore drift causing longshoresand redistribution. Surges, particularly when surge levelsexceed the tidal range, as observed, may play a key role inlow energy coastal environments (JACKSONP/ a/., 2002), HEG-CE et ai. (1996) state that as wave and tidal energy decreasethe relative importance of other phenomena, such as non-tidal sealevel fluctuations, in controlling the nearshore mor-phology increases. Nevertheless, in our case, the ohservedsealevel uprise cannot be solely explained hy static responseof the sea to atmospheric variations (-29 cm) and, hence,dynamic sealevel responses and wind set-up should be re-garded as mechanisms for the observed variations, particu-larly during November 15 when the pressure was 13hPahigher than that on November 11 but the upraise was simi-lar. Although it is usual to consider surges in terms of theset-up they generate at the coast, the currents which are gen-erated can be also important iPi'GH. 1987). Indeed, in thesampling after this storm, the southern zone of the beachpresented anticlockwise rotation and lower, although signif-icant, dry beach increase (432 m-). These beach changes canbe attributed to a flow reversal during the storm. Nearshorecurrents based on the model results support these observa-tions.

In summary, our results show the seasonal variability of a'low energy heach' (not under the specifications of JACKSON efai. 2002) and the morphodynamic effects of two differentstorms. The EOF technique was found to adequately repro-duce the main patterns of the observed evolution. Moreover,beach profiles suggest a relationship between the presence ofa Posidonia oceanica meadow in the vicinity of the beach andits morphodynamic adjustment providing further evidence onthe importance of the conservation of this endangered eco-system for coastal stability.

ACKNOWLEDGMENTS

This work has been carried out as part of POSICOST(1FD97-1654) and MEET projects. C. Homs assisted withfieldwork. We are indebted to J.L. Lopez-Jurado from IEO,wbo kindly supplied sealevel records and to Clima Maritimo(Puertos del Estado) who provided wave estimations. Wethank Fernando Garrido and Carlos Jover (Demarcacion deCostas, Ministerio de Medio Ambiente) for useful informationand data on Magalluf renourishment. Thanks also to Dr. R.Paskoff for his valuable review of the manuscript.

LITERATURE CITED

AAGAARII, T . and GREENWOOD, B., 1995. Suspended sediment trans-port and morphologica] response on a dissipative beach. Continen-tal Shelf Research, 1519), 1061-1086.

AUBREY, D.G., 1979. Seasonal patterns of onshore/ofTshore sedimenttransport. Journal of Geophysical Research. 84(C10), 6347-6354.

BASCOM, W.H.. 1951. The relationship between sand size and beachface slope. Transactions of the American Geophysical Union. 32,866-874.

BEACH, R.A.; BRUI'N, P., and STERNBERG. R.W., 1991. Infra^avity

driven suspended sediment transport in the swash, inner and out-er-surf zone. Proceedings of Coastal Sediments '91 I New York,USA, ASCE), pp. 114-128.

BouDOURESQUE, C.F.; JEUDY DE GRISSAC, A., and MKINESZ. A., 1984.

Relations entre la sedimentation e Tallongement des rinhomeK or-thotropes de Posidonia oceanica dans la Baie d'EIbu (Corse) In:BouDouRESQUE, C.F.; JEUPY DE GRISAC, A., and OLFVER, O . (eds.l.

International workshop on Posidonia oceanica beds. Marseille,France: G.I.S Posidonie Publ., pp. 185-191.

BRLIN, P., 1954. Migrating .sand waves and sand humps, with spe-cial reference to investigations carried out on the Dutch North Seacoast. Proceedings of the 5" Conference on Coastal Engineering,pp. 269-295.

BR[f[!N, P., 1962. Sea-level rise as a cause of storm erosion. Proceed-ings of the American Society of Civil Engineering. J. WaterwaysHarbors Division. 88lWWIl,'pp, 117-130.

CANALS, M. and BALLESTEROS, E. , 1997. Production of carbonate par-ticles by phytobenthic communities on the Mallorca-Menorcashelf, northwestern Mediterranean Sea. Deep-Sea Research, 44(3-41,611-629.

CHRISTIANSEN, C ; CHRISTOFFERSEN, H . ; DAI-SGAAD, J . , and NORBERC,

R., 1981. Coastal and nearshore changes correlated with die-hack-in eel-grass (Zostera marina L.l. Sedimentary Geology, 28, 163-173.

DEAN, R.G.. 1977. Equilibrium heach profiles: US Atlantic and GulfCoast Ocean. Engineering Report No. 12, University of Delaware,45p.

DEA,\, R.G.. 1987. Coastal sediment processes: Toward engineeringsolutions. Proceedings of Coastal Sediments '87 Conference, I NewOrleans, USA, ASCE), pp. 1-24.

DEAN, R.G., 1991. Equilibrium heach profiles: characteristics and ap-plications. Journal of Coastal Research. 7, 53-84.

DEAN, R.G., 1997. Models of harrier island restoration. Journal ofCoastal Research. 13, 694-703.

FoNSECA, M.S. and CALAHAN, J.A., 1992. A preliminary evaluation ofwave attenuation hy four species of seagrasses. Estuarine Coastaland Shelf Science. 35. 565-576.

GACIA. E. ; GRA_\ATA, T . C . and DITARTE, C-M.. 1999. An approach to

measurement of particle flux and sediment retention within sea-grass {Posidonia oceanica} meadows. Aquatic Botany. 65, 255—268.

GACIA, E . and DrARTE, CM., 2001. Elucidating sediment retentionby seagrasses: sediment deposition and resuspension in a Medi-terranean I Posidonia oceanica) meadow. Estuarine Coastal andShelf Science, 52. 505-514.

GIB.S.S, R.J.; MATTHEWS, M.D., and LINK, D.A., 1971. The relationshiphetween sphere size and settling velocity. Journal of SedimentaryPetrology. 41(1), 7-18.

Journal of Coastal Research, Vol. 20. No. 4, 2004

1164 Baaterretxea el at.

GRANATA, T . C ; SERRA. T.; COLOMER, J . ; CASAMITJAKA, X.; DUARTE,

CM., and GACIA, E. , 2001. Plow and particle distributions in anearshore seagrass meadow before and after a storm. Marine Ecol-ogy Progress Series. 218, 95-106.

GUILLEN, J . ; JIMSXEZ, J.A.; PALANQUES, A.; GRACIA, V.; PUIG. P., and

SANCHEZ-AKCILLA, A., 2002. Sediment resuspension across a mi-crotidal low-energy inner shelf. Continental Shelf Research, 22,305-325.

HALLEKMEIER. R.J., 1981, A profile zonation for seasonal sand beach-es from wave climate. Coastal Engineering, 4, 253—277.

HAYDEN, B.; FELDRR, W,; FISHER, J.; RESIO, D,; VINCENT, L., and Do-

LAN, R,, 1975, Systematic variations in inshore bathymetry. Tech-nical Report No 10. Charlotteaville, University of Virginia.

HEGGK, B,; ELIOT, L, and Hsr, J., 1996. Sheltered sandy beaches ofSouthern Australia. Journal of Coastal Research, 12(3), 748-760.

HiNE. A,C.; EVANS, M.W.; DAVIS, R.A., and BELKNAP. D,F, , 1987, De-

positional response to seagrass mortality along a low-energy, bar-rier island coast: west-centra] Florida, Journal of Sedimentary Pe-trology. 57,431-439,

JACKSON, N,L . : NURDSTKIIM, K.F.; EI.IOT, I., and MASSELINK, G., 2002,

'Low energy' sandy beaches in marine and estuarine environ-ments: A review, Geoniorphology. 48, 147-162.

JAUME, C . and FoRNOs, J.J,, 1992, Composicio i textura dels sedi-ments de platja del Htoral niallorqui, Bolleti de la Societatd'Historia Natural de lex Balears, 35. 93-109,

KOMAR, P.D,, 1998, Beach proces.-ies and sedimentation. PrenticeHall, New Jersey. 544p.

LYNETT, P,, 2002. A multi-layer approach to modeling generation,propagation, and interaction of water waves. Ph.D. Thesis, CornellUniversity,

MANZANERA, M. : PEREZ, M, , and ROMERO, J., 1998. Seagrass Mortality

due to oversedimentation: an experimental approach. Journal ofCoastal Con.seri>ation, 4, 67-70.

MARBA, N , and DuARTE, CM., 1997, Interannual changes in seagrass(Posidonia oceanicai growth and environmental change in theSpanish Mediterranean littoral. Limnology and Oceanography 42,800-810.

MARBA, N . and DL'ARTE, CM., 1998, Rhizome elongation and sea-grass clonal growth. Marine Ecology Progresfi Series, 174, 269-280.

MARBA, N , ; DUARTE, CM,; HOLMER, M.; MARTINEZ, R,: BASTEKRKTXEA,

G,; ORFILA, A.; JORDI, A,, and TINTORE, J,, 2002, Assessing the ef-

fectiveness of protection on Posidonia oceanica populations in theCabrera National Park (Spain). Environmental Conservation,29(41, 29, 509-518.

MENDEZ, F , J , ; LOSADA, I,J-, and LosADA. M.A,, 1999, Hydrodynamicsinduced by wind waves in a vegetation field. Journal of Geophys-ical Research. 104(08), 18,383-18,396,

MOORE, B.D, , 1982, Beach profile evolution in response to changesin water level and wave height. Newark, Delaware: University ofDelaware, Master's thesis, 164p.

MUNOZ-PKRKZ, J , J . ; TEJEDOR, L,, and MEDINA, E, , 1999. Equilibriumbeach profile model for reef-protected beaches. Journal of CoastalResearch, 15(4), 950-957.

NORDSTROM, K,F, , 1992. Estuarine beaches, Elsevier, 225p,OSBORNE, P.D. and GREENWOOD, B., 1992a, Frequency dependent

cross-shore suspended sediment transport, 1, A non-barred sbor-eface. Marine Geology, 106, 1—24.

OsBORNE, P,D. and GREENWOOD, B,, 1992b. Frequency dependentcross-shore suspended sediment transport. 2, A barred shoreface.Marine Geology, 106, 25-51.

PERGENT, G,; Boi:DOiTRESQUE, C.F.: CROUZET, A,, and MEINESZ, A,,1989. Cyclic changes along Posidonia oceanica rhizomes (lepidoch-ronology): present state and perspectives. Marine Ecology: 10.221-230.

PuGH, D.T., 1987. Tides, surges and mean sea-level. Chichester, UK:Wiley & Sons, 472p.

RAMIS, C ; JANSA, A., and ALONSO, S., 1990. Sea breeze in Mallorca.A numerical study. Meteorology and Atmosphere Physics, 42, 249-258.

RODRIGUEZ-PEREA, A.. SERVEKA-NICOLAU, J, and MARTIN-PRIETO, J ,A , ,2000. Alternatives a la dependencia de les platges de Les Balearsde la regeneracio artificial continuada: infornie Metadona. Palma:Universitat de Les Illes Balears, 108p,

RussEL, P., 1993, Mechanisms for beach erosion during storms. Con-tinental Shelf Research, 13, 1243-1265,

TERRADOS, J , and Dl'ARTE, CM,, 1999. Experimental evidence of re-duced particle resuspension within a seagrass (Posidonia oceanicaL.) meadow. Journal of Experimental Marine Biology and Ecology,243, 45-53,

WARD, L.G, : KEMP, W.M, , and BOVTOM, W,R, , 1984- The infiuence ofwaves and seagrass communities on suspended particulates in anestuarine embayment. Marine Geology. 59. 85-103.

WILSON, D,P , , 1949. The decline of Zostera marina L, at Salcombeand its effects on the shore. Journal of the Marine Biology Asso-ciation of the United Kingdom, 28, 395-412.

WiNANT, CD,; INMAN, D.L., and NORDSTROM, C.E., 1975. Descriptionof seasonal beach changes using empirical eigenfunctions. Journalof Geophysical Research, 80, 1,979-1,986,

WORCESTER, S,E,, 1995, EfFects of eelgrass beds on advection andturbulent mixing in low current and low shoot density environ-ments. Marine Ecology Progress Series. 126. 223-232,

WRIGHT, L .D, and SHORT, A.D. , 1983, Morphodynamics of beachesand surf zones in Australia. In: KOMAR P.D, (ed.l. Handbook ofCoastal Processes and Erosion. Boca Raton, Florida: CRC Press,pp 35-64.

Journal of Coastal Research, Vol, 20, No, 4, 2004