lable at ScienceDirect

Estuarine, Coastal and Shelf Science 154 (2015) 184e193

Contents lists avai

Estuarine, Coastal and Shelf Science

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

Large-scale dynamics of sandy beach ecosystems in transitionalwaters of the Southwestern Atlantic Ocean: Species turnover, stabilityand spatial synchrony

Diego Lercari a, b, *, Omar Defeo a, b

a UNDECIMAR, Facultad de Ciencias, Igu�a 4225, Montevideo, 11400, Uruguayb GEPEIA, Centro Universitario de la Regi�on Este, Ruta nacional N� 9 intersecci�on con Ruta N� 15, Rocha, Uruguay

a r t i c l e i n f o

Article history:Received 21 July 2014Accepted 4 January 2015Available online 10 January 2015

Keywords:Uruguaytransitional watersRío de la Platasandy beachesmorphodynamicsbenthic macrofaunaspecies richnessseasonal dynamics

* Corresponding author. UNDECIMAR, Facultad detevideo, 11400, Uruguay.

E-mail address: lercari@fcien.edu.uy (D. Lercari).

http://dx.doi.org/10.1016/j.ecss.2015.01.0110272-7714/© 2015 Elsevier Ltd. All rights reserved.

a b s t r a c t

Transitional waters (TW) are interfaces between the terrestrial and freshwater environments and the sea.These ecotones are characterized by highly dynamic physico-chemical and hydro-morphologic condi-tions, resulting in a mosaic of habitats in which species are particularly well adapted to variability.However, sandy beach ecotones occurring along estuarine gradients are rarely addressed from the TWperspective. We conducted a 2-yr study to assess the seasonal dynamics of environmental and macro-faunal descriptors in 16 sandy beaches of the Uruguayan coast in TW defined by the widest estuary of theworld (Rio de la Plata). A strong variability in environmental conditions was found at inner estuarinebeaches, reflecting the seasonal dynamics of the estuarine discharge. The greatest abundance and speciesrichness found in dissipative oceanic beaches were also characterized by their lowest temporal vari-ability, indicating that macrofaunal communities were more stable towards oceanic conditions, whereenvironmental variability was also lowest. Spatial synchrony was reflected in changes across seasons inthe species richness in the TW system. A high turnover of species along spatio-temporal gradientsoccurring within the TW ecotone was observed. Mollusca and Polychaeta were absent in highly-variableestuarine beaches, irrespective of the morphodynamic state. A functional equivalence between specieswas found at the extremes of the salinity gradient. The environmental variables that best explainedcommunity patterns differed among seasons: in summer and autumn, salinity, wave period and beachwidth were the main explanatory factors, whereas temperature had a primary influence in winter andmorphodynamic variables exerted a major influence in autumn. We highlight the need to considerconcurrent variations in estuarine and morphodynamic variables when assessing the spatial distributionof macrofaunal species richness and abundance in sandy beaches occurring along TW.

© 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Populations of plants and animals gradually change along gra-dients of physical conditions, shaping distinct ecological commu-nities within broadly defined ecosystems, such as forests,grasslands and estuaries (Ricklefs, 1990; Gaston, 2000). Everycoastal habitat exists within a multi-dimensional mesh of envi-ronmental gradients, notably related to four main sources of vari-ation: parameters that change across-shore (e.g. temperature,humidity), wave exposure, sand particle size and salinity (Raffaelliand Hawkins, 1996).

Ciencias, Igu�a 4225, Mon-

Transitional waters (TW) are ‘bodies of surface water in the vi-cinity of river mouths which are partially saline in character as aresult of their proximity to coastal waters but which are substan-tially influenced by freshwater flows’ (Directive, 2000/60/EC). Theycomprise estuaries, deltas and coastal lagoons, defining ecosystemswith unique functional and structural characteristics (Cognetti andMaltagliati, 2008; Basset et al., 2013). Indeed, TW are ecotonesbetween land, sea and freshwater, characterized by highly dynamicphysical, chemical and hydro-morphologic conditions, resulting ina mosaic of habitats in which species are particularly well adaptedto variability (Elliott and Whitfield, 2011). Ecotone dimensionscorrespond to major discontinuity boundaries between differentecosystem types, shaping TW as multi-dimensional, hierarchicallyorganised, ecotones. The hydrodynamic ecotone dimension (1storder) mainly reflects a spatial or temporal salinity gradient that

D. Lercari, O. Defeo / Estuarine, Coastal and Shelf Science 154 (2015) 184e193 185

limits the local number of species (a diversity) for some groups by afiltering process from the regional species pool (Zobel, 1997;Mouillot et al., 2007). The spatio-temporal patchiness found inTW is likely to support high taxonomic turnover among locations (bdiversity) and then a high regional g diversity (Barbone and Basset,2010). Additional ecotone dimensions occur at other TW interfaces,including the across-shore axis from the supralittoral to the sub-littoral margins (Basset et al., 2013). The relative size (spatialdimension) and the temporal presence of these ecotones vary withlocality and type of TW.

The dynamics of exposed sandy beach ecotones occurring alongestuarine gradients has been rarely studied from the TW perspec-tive. Exposed sandy beaches are relatively uncommon in TW such asestuaries, and large-scale macrofauna variations along estuarinegradients have been scarcely documented. Recent reviews in sandybeach ecology highlighted the need to consider a multiscaleapproach for ecosystem modelling, including the underlying pro-cesses and mechanisms (Defeo and McLachlan, 2005, 2011; Fujii,2007; Schlacher et al., 2008). Alongshore patterns in a single beachaffected by a salinity gradient showed an exponential decrease inspecies richness and abundance towards low salinities (Lercari et al.,2002; Lercari and Defeo, 2003). Temporal variations in freshwaterdischarges resulted in local short-term changes in community at-tributes (Defeo and Lercari, 2004). At the macroscale (hundreds ofkm), salinity range become an ecological master factor governingmacrofaunal distribution patterns in sandy beaches located along anestuarine gradient (Lercari and Defeo, 2006). Tolerance to salinitychanges varies among species according to the beach zone inhabited(intertidal species have lower tolerance than supralittoral ones) anddevelopment mode (species with planktonic larvae have lowertolerance than direct developers) (Barboza et al., 2012).

Another sourceof variability in sandybeachecotones isdefinedbytheirmorphodynamic characteristics. Beachmorphodynamics play akey role in determining the structure of sandy beach communities,with species richness increasing from reflective (steep slope, coarsegrains) to dissipative (gentle slopes,fine grains) beaches (reviewed inDefeo andMcLachlan, 2013). In this context, sandy beaches occurringin TW are affected by the highly dynamic hydro-morphologic con-ditions that characterize these systems, resulting in weaker waveaction and a decrease in wave period and swash width at innerestuarine conditions (Lercari and Defeo, 2006). However, the dy-namics of these intertidal ecotones in TW has yet to be resolved. Thisuncertainty arises partly from twomain issues: firstly, the traditionalcomparison of only one pair of sites in sandy beach studies (e.g. highand low salinities, one dissipative vs. one reflective beach) cannotdistinguish between linear and nonlinear changes in increasingseverity of the gradients. Secondly, although salinity is a majorexplanatory driver of diversity patterns, little is known about howtemporal variations in complex TW systems affect structural andfunctional changes in sandy beach communities occurring inside TW.

This paper describes a 2-yr study to assess the spatial andtemporal dynamics of the environment and the macrobenthiccommunity on sandy beach ecotones occurring along the Rio de laPlata, the widest estuary of the world that defines a TW system inthe Southwestern Atlantic Ocean. According to the theoreticalframeworks developed for estuaries (Attrill, 2002; Basset et al.,2013) and sandy beaches (Defeo and McLachlan, 2005), we pre-dict: 1) a lower number of species and abundance in sandy beachesaffected by high salinity variability, and in those with reflectivecharacteristics; 2) a higher community stability in less variableenvironments (both in terms of salinity and beach morphody-namics); 3) a replacement (turnover) of species along the TWgradient maintaining functional redundancy, and 4) changes incommunity structure across seasons occurring synchronically(spatial synchrony) among beaches.

2. Materials and methods

The Rio de la Plata estuary, located at 35�S on the Atlantic coastof South America, constitutes a major TW system with uniquecharacteristics. It drains the second largest basin of the continentwith an average freshwater flow of 22,000 m3 s�1 (Simionato et al.,2001). This large and microtidal estuary covers more than 400 kmfrom the head to themouth (200 kmwide) with a broad permanentconnection to the sea. The estuary is characterized by seasonalvariations in river inputs, being highly sensitive to the atmosphericforcing, especially towind-induced turbulence. The strong seasonalvariability is controlled by the balance between onshore andoffshore winds, the river discharge, the tides and the Coriolis force(Simionato et al., 2001).

The study covered 400 km in the TW of the Uruguayan coast,where we selected eight beaches along 150 km of the Río de la Plataestuary (inner beaches) and eight beaches along 250 km of oceanicshores (outer beaches) outside the mouth of the estuary. Samplingsites were selected considering, whenever possible, close pairs ofbeaches with the same salinity range but markedly different mor-phodynamic features. Biological samples on each beach were takenevery 2 months from July 1999 to May 2001, along three transectsperpendicular to the shoreline, spaced 8 m apart, from the base ofthe dunes to the lower limit of the swash zone. Sampling units(SUs) on each transect were done every 4 m with a sheet metalcylinder, 27 cm in diameter and 40 cm deep, and the material wassieved through a 0.5 mm mesh. In order to cover the entire beachhabitat, the number of samples taken on each beach and samplingevent varied according to the beach width, which was measured asthe distance between the base of the dunes and the lower limit ofthe swash zone, where water moves over the beach face after abroken wave collapses on the sand (Lercari and Defeo, 2006). Theorganisms retained were fixed in 5% formalin, counted and iden-tified to the species level (whenever possible). Sampling on eachbeach was made at the same time of the day in order to minimizedaily effects associated with sampling conditions. Following Defeo(1996), macrofaunal abundance was expressed as individuals perstrip transect (ind m�1). The total number of species recorded oneach beach was considered as a measure of a diversity (Barbozaet al., 2012).

At each beach, salinity andwater temperature weremeasured atthe surf zone with an YSI 33 thermosalinometer, wave height wasvisually estimated and wave period was determined with a stop-watch. At each SU, we measured temperature at the surface and15 cm depth, sand compaction (kg cm�2), and beach slope (Emery,1961). Sediment samples were also collected at each SU to estimategranulometric variables (Folk, 1974), sediment moisture andorganic matter content (weight differences between wet, dry andincinerated samples, respectively). The amount of wrack or car-casses deposited on all the beaches can be considered negligible.Swash width was measured as the distance between upper andlower swash limits at sampling time. Mean estimates for eachbeach and sampling date were obtained by averaging individual SUestimates.

Alongshore patterns in salinity were modelled by season, andthe best model was selected according to the coefficient of deter-mination (r2) and statistical significance. Beach morphodynamicswere assessed by Dean's parameter U (Short, 1996):

U ¼ Hb$100Ws$T

whereHb is breaker height (m),Ws is sand fall velocity (cm s�1) andT is wave period (sec). Annual mean values of U were employed toclassify beaches according to their morphodynamic features.

Table 1Parameter estimates and associated statistics of the nonlinear models that relatesalinity (S) and distance from innermost site (D) in each season in TW of theSouthwestern Atlantic Ocean. All models and their parameters were highly signifi-cant (p << 0.001).

a b c R2

S ¼ bð1�eða�c$DÞ Þ

Summer 1.03 32.71 0.019 0.983Autumn 4.06 27.56 0.021 0.982S ¼ aþ b$ð1� CDÞWinter 0.81 28.44 0.9911 0.996Spring 5.70 24.49 0.9888 0.987

Fig. 1. Seasonal large-scale variations in a) mean salinity (nonlinear models areshown) and b) swash water temperature from the innermost beach site (0 km) tooceanic conditions in 16 Uruguayan sandy beaches.

D. Lercari, O. Defeo / Estuarine, Coastal and Shelf Science 154 (2015) 184e193186

Between-season differences in the number of species and totalabundance were tested by ANCOVA, using the distance from theinnermost site as a covariate. We tested for assumptions ofnormality (KolmogoroveSmirnov test), heteroscedasticity(Cochran's test) and parallelism, to assess whether the continuous(distance) and categorical (season) predictors interact in influ-encing responses (abundance and richness) (Underwood, 1997).Changes across seasons in environmental and community de-scriptors was measured by the coefficient of variation(CV ¼ standard deviation/mean) for each beach across all seasons.As CV standardizes for the mean, it is less dependent on the meanthan the standard deviation, and provides an index of temporalvariation relative to the mean (Tilman, 1996; McCann, 2000). CVswere calculated on a beach-by-beach basis, using data gathered oneach beach for all sampling dates. Intra-annual variability in envi-ronmental and biotic variables was modeled by the relationshipbetween the CV and the distance from the innermost beach. Modelswere adjusted by nonlinear least squares, using the Lev-enbergeMarquardt algorithm. Community stability was also eval-uated using the CV of total abundance with respect to speciesrichness. Increases in CV correspond to decreases in stability, andvice versa. When increasing species richness was associated withdecreasing (increasing) CV, we referred to this as a positive(negative) effect of richness on stability.

Temporal coherence or spatial synchrony refers to the tendencyof population, community or ecosystem to behave similarly amonglocations through time as a result of spatially-correlated environ-mental stochasticity (Huttunen et al., 2014). In order to assess thestrength of coherence in changes across seasons in abiotic (salinity,temperature) and community (species richness, total abundance)descriptors along the estuarine gradient, we used a simple linearregression approach. Thus, we tested the importance of changesacross seasons through linear regressions between all possibleseason pairs. Perfect coherence (i.e. concurrent changes in thevariable occurring in all beaches from one season to another) wasreflected by a significant linear regression between two contrastedseasons, suggesting that locations behave similarly through time,i.e., the biotic or abiotic variable changed at a same proportion fromone season to the other.

Changes in environmental conditions across beaches and sea-sons were assessed by 2-way Permutational Analysis of Variance(PERMANOVA; Anderson, 2001) included in the PRIMER 6.0 soft-ware package (Clarke and Gorley, 2006). This analysis was based onthe Euclidean distance matrix of log-transformed standardizedenvironmental variables. Two fixed factors (beach and season) and999 permutations of residuals under a reduced model wereconsidered in the design, which also assessed the beach � seasoninteraction. A 2-way PERMANOVA was also performed to assessdifferences in community structure using rooteroot transformeddata from each site and the BrayeCurtis similarity index, andfollowing the same design as in environmental variables. To assesswhich combination of environmental variables best explainedspatio-temporal variations in macrofaunal abundance, we used theroutine BIOENV (Clarke and Ainsworth, 1993). The degree of sig-nificance between matrices was tested with the RELATE routine ofPRIMER 6.0 (Clarke and Gorley, 2006).

3. Results

3.1. Environmental factors

Salinity increased from west to east in all seasons and beaches,and from autumn to summer. The salinity gradient was steeper overthe first 150 km from the innermost site than along the oceanicsector. In winter and spring, large-scale changes in salinity were

best explained by an asymptotic model, and by a logistic function inautumn and summer (Table 1; Fig. 1a). Seawater temperatureshowed two distinct patterns (Fig. 1b): 1) estuarine beaches pre-sented a warmer period (spring and summer) and a colder one(autumnewinter), and 2) oceanic beaches showedmarked changesacross seasons, being highest in summer and lowest in winter,whereas intermediate values were registered in spring andautumn.

Alongshore trends in morphodynamic variables reflected theselection of beaches during the sampling design (Supplementarymaterial, Fig. S1 and Table S1). Mean grain size was remarkablyconstant in all beaches throughout the period, whereas slope wassteepest on most beaches during autumn (erosion period) and

Fig. 2. Large-scale variations in the coefficient of variation of biotic (number of species, abundance) and environmental (temperature, salinity, beach and swash width) variables in16 Uruguayan sandy beaches located in TW of the Southwestern Atlantic Ocean. All models are highly significant (p < 0.001).

D. Lercari, O. Defeo / Estuarine, Coastal and Shelf Science 154 (2015) 184e193 187

flattest in summer (accretion period), particularly at oceanic bea-ches. Beaches with coarser grain size had steeper slopes (Fig. S2).Beach width tended to be greater at beaches with fine grains andgentle slopes and did not show evident changes across seasons.Swash width steadily increased from inner estuarine to oceanicconditions at all seasons. Sediment moisture was higher in widebeaches with fine grains and gentle slopes throughout the studyperiod. Dean's parameter U was higher in fine grain and gentleslope beaches than in coarse-grained beaches with steep slopes,particularly in oceanic beaches, but no changes across seasons weredetected. The large-scale patterns depicted above were reflected inhigher CV values towards the inner estuary in temperature, salinity,beach width and swash width (Fig. 2 aed ). By contrast, mean grainsize, beach slope, sediment organic matter and Dean's parameterdid not show evident seasonal variability.

Coherence in changes across seasons varied strongly amongabiotic and biotic response variables. Salinity change on sandybeaches was remarkably coherent when considering winter, springand summer, but this was not the case when comparing autumnwith the remaining seasons (Supplementary material Fig. S3).Temperature coherence was very low among all seasons(Supplementary material Fig. S4). Concerning biotic variables,species richness along the study area showed a remarkable spatialsynchrony (Fig. 3), while total abundance showed departures from

the coherence pattern when considering the summer season(Supplementary material Fig. S5).

Environmental variables significantly differed between beaches(2-way PERMANOVA: pseudo-F ¼ 7.63, p ¼ 0.001) and seasons(pseudo-F¼ 11.61, p¼ 0.001), while the beach � season interactionwas not (pseudo-F ¼ 0.78, p ¼ 0.96). This means independence ofthe effect of the factor ‘beach’ on the level of the factor ‘season’,reflecting again coherence in environmental changes for all bea-ches across seasons. Widespread differences between pair of bea-ches occurred in the four seasons, mainly reflecting unique abioticcharacteristics for each beach occurring all along the year(Table S2a, b).

3.2. Macrobenthic community

Total species richness was highest in summer and lowest inwinter and increased towards oceanic beaches, being highest inBeach 16 (Barra del Chuy) throughout the study period (Table S1;Fig. 4a). However, the number of species in the two innermostestuarine Beaches 1 and 2 was higher than those found in innerestuarine Beaches 3 and 4, which showed the lowest species rich-ness during the four seasons. Considering contiguous beach pairs,dissipative or intermediate beaches harboured more species thanreflective ones. In particular, estuarine beaches presented lower

Fig. 3. Spatial synchrony analysis by pairwise linear regressions between the number of species recorded on each beach on every season in 16 Uruguayan sandy beaches.***p < 0.001, **p < 0.01, ns p > 0.05.

D. Lercari, O. Defeo / Estuarine, Coastal and Shelf Science 154 (2015) 184e193188

species richness during autumn andwinter (Fig. 4a). The number ofspecies significantly differed between seasons (ANCOVA: Table S3):for a same distance from the innermost beach (covariate), winterhad significantly lower number of species than the remaining threeseasons (LSD test: p < 0.01). The number of species was signifi-cantly lower in spring than in summer, whereas autumn onlydiffered with winter.

Total macrofauna abundance followed the same patternobserved for species richness (Table S1; Fig. 4b), increasing towardsboth extremes of the TW system. The highest abundance occurredon the innermost Beach 2 in winter and autumn, decreasing inmore than an order of magnitude fromwinter (105,000 ind m�1) tosummer (8000 ind m�1). The outermost Beach 16 showed thehighest abundance among oceanic beaches throughout the 2-yrstudy period, particularly in spring (26,000 ind m�1). The highvariability in abundance precluded the fulfilment of basic ANCOVAassumptions (e.g. homogeneity of variances). Nonlinear modellingof the CV showed a higher intra-annual variability of species

richness and total abundance towards the inner estuary (Fig. 2eef).The CV of both biological descriptors exponentially decreased to-wards oceanic conditions, irrespective of the beach morphody-namic state. Moreover, a negative exponential trend was foundbetween abundance variability (as denoted by its CV) and speciesrichness, meaning less variable communities in oceanic sandybeaches (Fig. 5).

A high species turnover was found along spatio-temporal gra-dients occurring within the TW ecotone. Changes across seasons inspecies dominance in the 16 sandy beaches showed that (Fig. 6),firstly, species dominance was lowest at both extremes of thestudied area, whereas Mollusca and Polychaeta were absent inouter estuarine beaches, irrespective of the morphodynamic state.These Phyla were always found in all oceanic beaches, and in theinner estuary (Beaches 1 and 2). However, different species char-acterized both extremes of the TW system: a) concerning Mollusca,Donax hanleyanus and Mesodesma mactroides were dominant inoceanic beaches, while Erodona mactroides (Bivalvia) and Heleobia

Fig. 4. Seasonal large-scale distribution of: a) number of species and b) total abun-dance in 16 Uruguayan sandy beaches located in TW of the Southwestern AtlanticOcean.

Fig. 5. Nonlinear model between the coefficient of variation (CV) in communityabundance and the mean number of species found in each Uruguayan beach(p < 0.001). Beaches are numbered fromwest to east: 1: Pascual (km 0); 2: Penino (km2); 3: Honda (km 51); 4: Verde (km 52); 5: Costa Azul (km 108); 6: Baguala (km 112);7: Hermosa (km 125); 8: Negra (km 139); 9: S M�onica (km 197); 10: J Ignacio (km 203);11: Aguada (km 257); 12: Arachania (km 260); 13: S Isabel (km 267); 14: P Diablo (km350); 15: Achiras (km 356); 16: B Chuy (km 378).

D. Lercari, O. Defeo / Estuarine, Coastal and Shelf Science 154 (2015) 184e193 189

sp. (Gastropoda) dominated at inner estuarine beaches; b) con-cerning Polychaeta, Euzonus furciferus, Scololepis gaucha and Hem-ipodus olivieri occurred in oceanic beaches, whereas Laeonereisacuta and Nephtys fluviatilis only appeared in Beaches 1 and 2.Secondly, the cirolanid isopod Excirolana armata dominated estu-arine and oceanic dissipative beaches, excepting the two innermostBeaches 1 and 2. E. armata was the only species found in innerestuarine beaches during autumn and spring, and showed a cleardominance in oceanic beaches during autumn, winter and spring.Thirdly, the isopod Excirolana braziliensis and the talitrid amphipodAtlantorchestoidea brasiliensis were dominant all year round inouter estuarine and oceanic reflective beaches. Fourthly, the molecrab Emerita brasiliensiswas a typical inhabitant of oceanic beachesin all seasons. On inner estuarine beaches, this crab was only foundduring summer (Fig. 6a) in the form of recently settled megalopae.

Differences between beach pairs occurred in all the four sea-sons, highlighting distinctive benthic communities for each beach.Results of PERMANOVA Pair-wise tests are shown in Table S4b.

3.3. Biotic e abiotic matching

Multivariate linking between macrofauna and its habitat (BIO-ENV analysis) was highly significant for the four seasons(0.797 < Rho < 0.915, Table 2). The analysis highlighted salinity asthe main explanatory variable of the observed large-scale trends insummer, autumn and spring, where wave period and beach andswash width also played a role as explanatory variables. In winter,temperature was the most important abiotic variable, followed bysalinity and beach width.

4. Discussion

This study demonstrated important changes across seasons insandy beach communities in the TW system defined by the Río dela Plata, as well as the role of the morphodynamic and estuarinegradients in structuring these communities. Sandy beaches occur-ring along the Uruguayan coast showed environmental and bio-logical similarities with other TW, notably the presence of strongabiotic gradients producing environmental filtering that result inheterogeneity of species composition (Barbone and Basset, 2010).Our analysis illustrates how environmental gradients may drive thedynamics of the sandy beach communities, as reflected by thestrong species turnover among sites. Community stability waspositively related with species richness and negatively with envi-ronmental variability. Moreover, spatial synchrony in salinity andspecies richness was observed along the study area.

4.1. The environment

The estuarine discharge showed sharp intra-annual oscillations.Indeed, salinity was highest in all beaches during the warm periodand lowest during the cold season. Moreover, the influence offreshwater discharge from Uruguay and Paran�a rivers into the Ríode la Plata increased in autumn, generating a decrease in salinity inestuarine beaches (Fig. 1a). In this system, salinity is controlled by anumber of variables interacting in a big and shallow basin (Mianzanet al., 2001). Seasonally, the surface horizontal salinity gradient ismainly controlled by winds, which determine two major seasons:springesummer (winds from the NE) and autumnewinter (windsfrom the SW) (Guerrero et al., 1997). However, in intertidal envi-ronments, the horizontal salinity gradient may be dominated byriver runoff, wind patterns, tidal range, the proximity of secondaryfreshwater discharges and the local shore configuration (e.g.orientation, closeness to rocky heads). The interaction among thesefactors at a local scale is hardly known, but our results highlight a

Fig. 6. Relative contribution in abundance of the main macrofaunal species recorded in 16 Uruguayan sandy beaches: a) summer; b) autumn; c) winter; d) spring.

D. Lercari, O. Defeo / Estuarine, Coastal and Shelf Science 154 (2015) 184e193190

higher intra-annual salinity variability in estuarine than in oceanicbeaches.

Seawater temperature also showed strong changes across sea-sons, which markedly differed along the coast. Inside the estuarywe found two main periods (cold autumnewinter and warm

Table 2Results of the BIOENV analysis showing the group of environmental variables thatbest explained variations in sandy beach community structure in each season, in 16Uruguayan sandy beaches located in TW of the Southwestern Atlantic Ocean.

Season r Variable combination

Summer 0.797 Salinity; wave period; sediment compactionAutumn 0.848 Wave period; grain size; sediment temperature; sediment

water contentWinter 0.828 Water temperature; salinity; compaction; sediment water

contentSpring 0.822 Salinity; beach width; sediment sorting; sediment organic

matter

r ¼ Spearman rank correlation coefficient.

springesummer), whereas three periods were found in the oceaniccoast (cold winter, temperate autumnespring and warm summer).This is in agreement with the trends observed by satellite imageryin this coastal zone, where two periods (autumnewinter andspringesummer) were identified in relation to the solar radiationcycle, dominant wind patterns and the influence of the freshwaterplume (Simionato et al., 2010). Intra-annual variations in temper-ature were also higher under estuarine conditions and couldrepresent an additional stressor to the macrofaunal speciesinhabiting the inner estuary. Patterns of water salinity and tem-perature variation along the study area showed that the salinitygradient was the most important descriptive component of TW,with the widest variation occurring at the freshwater e brackishwater and brackish wateremarine interfaces (Beaches 8 and 9).This area may correspond to the actual discontinuity betweenfreshwater and marine ecosystems (Basset et al., 2013).

Large-scale variations in beach morphodynamics largely re-flected the sampling design (Lercari and Defeo, 2006). Grain size

D. Lercari, O. Defeo / Estuarine, Coastal and Shelf Science 154 (2015) 184e193 191

and Dean's parameter displayed blur changes across seasons,without showing seasonal trends along the coast. Grain size hasbeen previously defined as an enduring variable in sandy beaches(Ortega et al., 2013) because it does not exhibit substantial changesover time (Valesini et al., 2010). By contrast, beach and swash widthshowed a well-defined exponential decrease in the CV towardsoceanic beaches, while slope showed variations according toerosion/accretion processes. These variables have been mentionedas key factors influencing the macrofauna, where beach squeezefrom marine to estuarine regions constitutes a direct evidence ofhabitat reduction (Lercari and Defeo, 2006). Therefore, in additionto salinity and temperature variability, species inhabiting estuarinebeaches have to cope with a wide variation in habitat expansion/reduction cycles.

4.2. Abundance and species richness

Independently of the season of the year, species richness andabundance increased from reflective to dissipative domains(McLachlan, 1990; Defeo et al., 1992; Brazeiro, 1999; Defeo andMcLachlan, 2005). Species richness on most beaches rose sharplyin spring, peaked in summer and then strikingly declined duringautumn and winter (excepting Beach 2). Therefore, differencesamong seasons in the number of species were found whenremoving the effect of the position of beaches along the estuary.

Total abundance did not show a consistent seasonal pattern inall beaches. Different beaches attained their highest abundance indifferent seasons, regardless their location along the estuary. Var-iations among seasons in sandy beach communities are common intemperate and subtropical sandy beaches (Dexter, 1979, 1984;Haynes and Quinn, 1995; Jaramillo et al., 1996, 2001), and havebeen mainly ascribed to the clear timing in recruitment periodsthat mainly occurred during warm seasons (Silva et al., 2008). Inaddition, temporal variability in abundance was highest whenspecies richness was lowest, particularly in inner estuarine bea-ches, whereas dissipative oceanic beaches with the highest speciesrichness (Beaches 13, 15 and 16) had the lowest abundance vari-ability. These results suggest that macrofaunal communities aremore stable at the ocean edge of the TW system, where environ-mental variability is also lowest.

4.3. Species turnover

Species turnover, i.e., the replacement of species along spatio-temporal gradients occurring within the TW ecotone, was veryhigh and substantially higher than those found in large-scalestudies conducted on exposed ocean beaches with different mor-phodynamics, meaning an unambiguous effect of TW systemcharacteristics in structuring macrofaunal communities. Moreover,our patterns were similar to those depicted for other TW systems,including estuaries, deltas and coastal lagoons (Basset et al., 2013).

Mollusca and Polychaetawere absent in sandy beaches located inthe middle of the main estuarine axis, whereas they were found inoceanic beaches at all morphodynamic conditions, and in theinnermost estuarine Beaches 1 and 2. By contrast, crustaceans wereconspicuous components that inhabited the whole salinity gradientin TW. Excirolana armata was systematically recorded in estuarinebeaches, suggesting a high tolerance to estuarine conditions (Lozoyaand Defeo, 2006; Lozoya et al., 2010). It was dominant in all finegrain size and gentle slope beaches throughout the coast, andtherefore could be defined as a high substrate-specific species (Defeoet al., 1997; Petracco et al., 2010). The congeneric isopod Excirolanabraziliensis and the talitrid amphipod Atlantorchestoidea brasiliensiswere always dominant in outer estuarine and oceanic reflectivebeaches, which are considered as more stable and safer

environments for supralittoral species (Defeo and G�omez, 2005).Estuarine reflective beaches may also prevent these supratidal spe-cies to be submerged into waters with highly fluctuating salinity,causing an osmotic stress and potential lethal effects (G�omez andDefeo, 2012). The mole crab Emerita brasiliensis was a typicalinhabitant of oceanic beaches during all seasons, but it was alsoregistered in inner estuarine beaches in late summer and autumn,because of the occurrence of recently settled megalopae. This couldbe explained by the highly-dispersive larval phase of the species,which in the summer of 2000 was favoured by a marked decrease inthe freshwater runoff of the Río de la Plata (Piola et al., 2005),together with a higher influence of the warm Brazil current. Thisenvironmental scenario disappeared in autumn, and no E. brasiliensisindividuals overwintered inner estuarine beaches. Therefore, innerestuarine beaches act as sink habitats in the metapopulation dy-namics of this species (Celentano et al., 2010). Regardless of thispunctual observation, no settling organisms of other marine specieswith pelagic larvae (e.g. Donax hanleyanus, Mesodesma mactroides)were observed in inner beaches. This suggests an important role ofsmall to mesoscale oceanographic features in larval accumulation,recruitment and circulation patterns close to reproduction areas(Pineda et al., 2007; Porri et al., 2014). Further surveys includingplankton sampling, coastal currents studies and genetic aspects willclarify the processes andmechanisms involved in the connectivity ofsandy beach populations in this TW system.

Although high turnover may affect the structural properties ofbeach communities, a functional equivalence between species werefound at the extremes of the salinity gradient. For example, theoceanic filter feeder bivalves (e.g.Donax hanleyanus andMesodesmamactroides) were replaced by brackish species as Erodona mac-troides in the inner estuary; similarly, the marine deposit-feedingpolychaete Euzonus furciferus was replaced in brackish waters byits functional equivalent Laeonereis acuta. Thus, different speciessustained similar functions and therefore fill equivalent niches,reinforcing the concept of functional redundancy (Basset et al.,2013). The capacity of these species to perform the same functionin markedly different parts of the TW system (i.e., in sandy beacheslocated in freshwater and marine waters) substantially increasesthe level of redundancy within functional groups (Basset et al.,2013). This functional redundancy provides some sort of ecolog-ical insurance against environmental fluctuations (Yachi andLoreau, 1999), increasing the ecological resilience of ecosystemsoccurring in TW.

An ecotone may be defined as a narrow ecological zone dividingtwo different and relatively homogeneous community types (Attrilland Rundle, 2002). The fact that different species of Mollusca andPolychaetawere only registered at both extremes of the TW systemanalysed here reinforces the concept that an ecotone model couldbe used to describe the variation of species composition along theUruguayan coast. Indeed, the two innermost Beaches 1 and 2 (innerestuary) showed unique faunal components, which presentedmarked differences in species composition among seasons. Thetemporal dynamics of species dominance also behaved quitedifferently than in the remaining beaches. For example, higherspecies dominance was attained during winter and spring. Inaddition, the distribution of most species recorded in inner estua-rine beaches goes beyond the intertidal fringe, having an extensivesubtidal distribution on the Río de la Plata estuary (Cortelezzi et al.,2007). In contrast, the habitat of all species that occurred in outerestuarine and oceanic beaches is restricted to a narrow intertidalfringe of sand between terrestrial and marine systems. Thesetrends suggest an ecological shift in sandy beaches towards theinner estuary, thus developing a narrow ecotone, which is incontrast to the expected gradual transition defined by an ecoclinemodel (Attrill and Rundle, 2002; Elliott and Whitfield, 2011).

D. Lercari, O. Defeo / Estuarine, Coastal and Shelf Science 154 (2015) 184e193192

4.4. Matching environmental and faunal dynamics

The strong correlations between the biotic and abiotic multi-variate matrices in all seasons remarked the outstanding role of theenvironment in controlling sandy beach macrofaunal assemblages(Defeo and McLachlan, 2005; Ortega Cisneros et al., 2011). How-ever, the set of environmental variables that best explained thecommunity structure differed between seasons, suggesting that thesandy beach fauna could be controlled by a different suit of con-ditions depending on the time of the year. In warmer periods(summer and autumn), salinity, wave period and beach width werethe main environmental factors influencing sandy beach assem-blages, whereas temperature had a primary influence inwinter andmorphodynamic variables exerted a major influence in autumn.This highlights the need to consider morphodynamic informationwhen treating sandy beaches occurring in TW systems. Thesevariables (e.g. wave period) may act in addition to the stress pro-duced by salinity variability to filtering the regional pool of speciesand limiting local diversity values. Our study demonstrated that insandy beaches occurring inside TW, spatio-temporal patchiness(discreteness) is likely to support high taxonomic turnover amonglocations (b diversity) and then high regional diversity (g) (Barboneand Basset, 2010).

In summary, environmental changes across seasons in sandybeaches occurring in TW plays a fundamental role in structuringmacrofaunal communities. Beaches located towards freshwaterconditions showed a major environmental dynamism, whichgenerated high variability in species richness and abundancemainly driven by a higher intra-annual salinity variability in estu-arine than in oceanic beaches. Decreasing abiotic and biotic vari-ability towards oceanic conditions led to more stable communities,which could explain the spatial synchrony found in species richnessin the TW system. Taxonomic turnover was observed among bea-ches with contrasting morphodynamics and also along the salinitygradient, but a functional equivalence between species were foundat the extremes of the salinity gradient. Long-term studies willelucidate the importance of climatic factors in deciphering thestrength of coherence through time in biotic and abiotic variables ofthese dynamic ecosystems occurring in TW.

Acknowledgements

We would like to express our gratitude to all the people thathelped in sampling and laboratory analyses. We are grateful for thefinancial support provided by CONICYT (FCE 4034), GEF-FAO-DINARA (GCP/URU/030/GFF), PEDEClBA and ANII.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.ecss.2015.01.011.

References

Anderson, M.J., 2001. A new method for non-parametric multivariate analysis ofvariance. Austral Ecol. 26, 32e46.

Attrill, M.J., 2002. A testable linear model for diversity trends in estuaries. J. Anim.Ecol. 71, 262e269.

Attrill, M.J., Rundle, S.D., 2002. Ecotone or ecocline: ecological boundaries in estu-aries. Estuar. Coast. Shelf Sci. 55, 929e936.

Barbone, E., Basset, A., 2010. Hydrological constraints to macrobenthic faunabiodiversity in transitional waters ecosystems. Rend. Lincei 21 (4), 301e314.

Barboza, F.R., G�omez, J., Lercari, D., Defeo, O., 2012. Disentangling diversity patternsin sandy beaches along environmental gradients. PLoS ONE 7 (7), e40468.

Basset, A., Barbone, E., Elliott, M., Li, B.-L., Jorgensen, S.E., Lucena-Moya, P., Pardo, I.,Mouillot, D., 2013. A unifying approach to understanding transitional waters:fundamental properties emerging from ecotone ecosystems. Estuar. Coast. ShelfSci. 132, 5e16.

Brazeiro, A., 1999. Community patterns in sandy beaches of Chile: richness,composition, distribution and abundance of species. Rev. Chil. Hist. Nat. 72,99e111.

Celentano, E., Guti�errez, N.L., Defeo, O., 2010. Effects of morphodynamic and estu-arine gradients on the demography and distribution of a sandy beach molecrab: implications for sourceesink habitat dynamics. Mar. Ecol. Prog. Ser. 398,193e205.

Clarke, K.R., Ainsworth, M., 1993. A method of linking multivariate communitystructure to environmental variables. Mar. Ecol. Prog. Ser. 92, 205e219.

Clarke, K.R., Gorley, R.N., 2006. PRIMER V6: User Manual/Tutorial. PRIMER-E, Ply-mouth, UK.

Cognetti, C., Maltagliati, F., 2008. Perspectives on the ecological assessment oftransitional waters. Estuar. Coast. Shelf Sci. 56, 607e608.

Cortelezzi, A., Rodrígues-Capítulo, A., Boccardi, L., Arocena, R., 2007. Benthic as-semblages of a temperate estuarine system in South America, transition from afreshwater to an estuarine zone. J. Mar. Syst. 68, 569e580.

Directive 2000/60/EC of the European parliament and of the council of 23 October2000 establishing a framework for community action in the field of waterpolicy.

Defeo, O., 1996. Recruitment variability in sandy beach macroinfauna: much tolearn yet. Rev. Chil. Hist. Nat. 69, 615e630.

Defeo, O., Lercari, D., 2004. Testing taxonomic resolution levels for ecologicalmonitoring in sandy beach macrobenthic communities. Aquat. Conserv. Mar.Freshw. Ecosyst. 14, 65e74.

Defeo, O., G�omez, J., 2005. Morphodynamics and habitat safety in sandy beaches:life-history adaptations in a supralittoral amphipod. Mar. Ecol. Prog. Ser. 293,143e153.

Defeo, O., McLachlan, A., 2005. Patterns, processes and regulatory mechanisms insandy beach macrofauna: a multi-scale analysis. Mar. Ecol. Prog. Ser. 295,1e20.

Defeo, O., McLachlan, A., 2011. Coupling between macrofauna community structureand beach type: a deconstructive meta-analysis. Mar. Ecol. Prog. Ser. 433,29e41.

Defeo, O., McLachlan, A., 2013. Global patterns in sandy beach macrofauna: speciesrichness, abundance, biomass and body size. Geomorphology 199, 106e114.

Defeo, O., Jaramillo, E., Lyonnet, A., 1992. Community structure and intertidalzonation of the macroinfauna in the Atlantic coast of Uruguay. J. Coast. Res. 8,830e839.

Defeo, O., Brazeiro, A., de Alava, A., Riestra, G., 1997. Is sandy beach macroinfaunaonly physically controlled? Role of substrate and competition in isopods. Estuar.Coast. Shelf Sci. 45, 453e462.

Dexter, D.M., 1979. Community structure and seasonal variation in intertidal Pan-amanian sandy beaches. Estuar. Coast. Mar. Sci. 9, 543e558.

Dexter, D.M., 1984. Temporal and spatial variability in the community structure ofthe fauna of four sandy beaches in south-eastern New South Wales. Aust. J. Mar.Freshw. Res. 35, 663e672.

Elliott, M., Whitfield, A.K., 2011. Challenging paradigms in estuarine ecology andmanagement. Estuar. Coast. Shelf Sci. 94, 306e314.

Emery, K.O., 1961. A simple method of measuring beach profiles. Limnol. Oceanogr.6, 90e93.

Folk, R.L., 1974. Petrology of Sedimentary Rocks. Hemphill Publishing Company,Austin, Texas.

Fujii, T., 2007. Spatial patterns of benthic macrofauna in relation to environmentalvariables in an intertidal habitat in the Humber estuary, UK: developing a toolfor estuarine shoreline management. Estuar. Coast. Shelf Sci. 75, 101e119.

Gaston, K.J., 2000. Global patterns in biodiversity. Nature 405, 220e227.G�omez, J., Defeo, O., 2012. Predictive distribution modeling of the sandy-beach

supralittoral amphipod Atlantorchestoidea brasiliensis along a macroscale estu-arine gradient. Estuar. Coast. Shelf Sci. 98, 84e93.

Guerrero, R.A., Acha, E.M., Frami~nan, M.B., Lasta, C.A., 1997. Physical oceanographyof the Río de la Plata estuary, Argentina. Cont. Shelf Res. 17, 727e742.

Haynes, D., Quinn, G.P., 1995. Temporal and spatial variability in communitystructure of a sandy intertidal beach, Cape Paterson, Victoria, Australia. Mar.Freshw. Res. 46, 931e942.

Huttunen, K.-L., Mykr€a, H., Huusko, A., M€aki-Pet€ays, A., Vehanen, T., Muotka, T.,2014. Testing for temporal coherence across spatial extents: the roles of climateand local factors in regulating stream macroinvertebrate community dynamics.Ecography 37, 1e10.

Jaramillo, E., Contreras, H., Quij�on, P., 1996. Macroinfauna and human disturbance ina sandy beach of south-central Chile. Rev. Chil. Hist. Nat. 69, 655e663.

Jaramillo, E., Contreras, H., Duarte, C., Quij�on, P., 2001. Relationships betweencommunity structure of the intertidal macroinfauna and sandy beach charac-teristics along the Chilean coast. Mar. Ecol. Pubbl. Stn. Zool. Napoli 22, 323e342.

Lercari, D., Defeo, O., 2003. Variation of a sandy beach macrobenthic communityalong a human-induced environmental gradient. Estuar. Coast. Shelf Sci. 58S,17e24.

Lercari, D., Defeo, O., 2006. Large-scale diversity and abundance trends in sandybeach macrofauna along full gradients of salinity and morphodynamics. Estuar.Coast. Shelf Sci. 68, 27e35.

Lercari, D., Defeo, O., Celentano, E., 2002. Consequences of a freshwater canaldischarge on the benthic community and its habitat on an exposed sandybeach. Mar. Pollut. Bull. 44, 1392e1399.

Lozoya, J.P., Defeo, O., 2006. Effects of a freshwater canal discharge on an ovovi-viparous isopod inhabiting an exposed sandy beach. Mar. Freshw. Res. 57,421e428.

D. Lercari, O. Defeo / Estuarine, Coastal and Shelf Science 154 (2015) 184e193 193

Lozoya, J.P., G�omez, J., Defeo, O., 2010. Modelling large-scale effects of estuarine andmorphodynamic gradients on distribution and abundance of the sandy beachisopod Excirolana armata. Estuar. Coast. Shelf Sci. 87, 472e478.

McCann, K.S., 2000. The diversityestability debate. Nature 405, 228e233.McLachlan, A., 1990. Dissipative beaches and macrofauna communities on exposed

intertidal sands. J. Coast. Res. 6, 57e71.Mianzan, H., Lasta, C., Acha, E., Guerrero, R., Macchi, G., Bremec, C., 2001. The Río de

la Plata Estuary, Argentina-Uruguay. In: Seeliger, U., Kjerfve, B. (Eds.), CoastalMarine Ecosystems of Latin America. Springer Verlag, Heidelberg, pp. 185e204.

Mouillot, D., Dumay, O., Tomasini, J.A., 2007. Limiting similarity, niche filtering andfunctional diversity in coastal lagoon fish communities. Estuar. Coast. Shelf Sci.71, 443e456.

Ortega, L., Celentano, E., Finkl, C., Defeo, O., 2013. Effects of climate variability on themorphodynamics of Uruguayan sandy beaches. J. Coast. Res. 29, 747e755.

Ortega Cisneros, K., Smit, A.J., Laudien, J., Schoeman, D.S., 2011. Complex, dynamiccombination of physical, chemical and nutritional variables controls spatio-temporal variation of sandy beach community structure. PLoS ONE 6 (8), e23724.

Petracco, M., Cardoso, R.S., Corbisier, T.N., 2010. Population biology of Excirolanaarmata (Dana, 1853) (Isopoda, Cirolanidae) on an exposed sandy beach inSoutheastern Brazil. Mar. Ecol. 31, 330e340.

Pineda, J., Hare, J.A., Sponaugle, S., 2007. Larval dispersal and transport in the coastalocean and consequences for population connectivity. Oceanography 20, 22e39.

Piola, A.R., Matano, R.P., Palma, E.D., M€oller Jr., O.O., Campos, E.J., 2005. The influ-ence of the Plata River discharge on the western South Atlantic shelf. Geophys.Res. Lett. 32. L01603.

Porri, F., Jackson, J.M., Von der Meden, C.E.O., Weidberg, N., McQuaid, C.D., 2014. Theeffect of mesoscale oceanographic features on the distribution of mussel larvaealong the south coast of South Africa. J. Mar. Syst. 132, 162e173.

Raffaelli, D.G., Hawkins, S.J., 1996. Intertidal Ecology. Chapman and Hall, London.Ricklefs, R.E., 1990. Ecology, third ed. W.H. Freeman, New York.Schlacher, T.A., Schoeman, D.S., Dugan, J., Lastra, M., Jones, A., Scapini, F.,

McLachlan, A., 2008. Sandy beach ecosystems: key features, sampling issues,management challenges and climate change impacts. Mar. Ecol. 29, 70e90.

Silva, P.S.R., Neves, L.P., Bemvenuti, C.E., 2008. Temporal variation of sandy beachmacrofauna at two sites with distinct environmental conditions on Cassinobeach, extreme southern Brazil. Braz. J. Oceanogr. 56, 257e270.

Simionato, C.G., Nu~nez, M.N., Engel, M., 2001. The salinity front of the Río de la Plata- a numerical case study for winter and summer conditions. Geophys. Res. Lett.28, 2641e2644.

Simionato, C.G., Tejedor, M.L.C., Campetella, C., Guerrero, R., Moreira, D., 2010.Patterns of sea surface temperature variability on seasonal to sub-annual scalesat and offshore the Río de la Plata estuary. Cont. Shelf Res. 30, 1983e1997.

Short, A.D., 1996. The role of wave height, period, slope, tide range and embay-mentisation in beach classifications: a review. Rev. Chil. Hist. Nat. 69, 589e604.

Tilman, D., 1996. Biodiversity: population versus ecosystem stability. Ecology 77,350e363.

Underwood, A.J., 1997. Experiments in Ecology: Their Logical Design and Interpre-tation Using Analysis of Variance. Cambridge University Press, Cambridge, U.K.

Valesini, F., Hourston, M., Wildsmith, M., Coen, N., Potter, I., 2010. New quantitativeapproaches for classifying and predicting local-scale habitats in estuaries.Estuar. Coast. Shelf Sci. 86, 645e664.

Yachi, S., Loreau, M., 1999. Biodiversity and ecosystem functioning in a fluctuatingenvironment: the insurance hypothesis. Proc. Natl. Acad. Sci. 96, 1463e1468.

Zobel, M., 1997. The relative role of species pools in determining plant speciesrichness: an alternative explanation of species coexistence? Trends Ecol. Evol.12, 266e269.