Large-scale spatial patterns of benthic assemblages in the SW Atlantic: the Rio de la Plata estuary and adjacent shelf waters

of 13

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
13 pages
0 downs
This study deals with the spatial distribution of macrobenthic communities (biomass and abundance) in the Rı´o de la Plata estuary, Argentina-Uruguay, and the adjacent shelf waters. The benthic invertebrates were caught with an epibenthic dredge (41
  Large-scale spatial patterns of benthic assemblagesin the SW Atlantic: the Rı ´o de la Plataestuary and adjacent shelf waters D.A. Giberto a,b, ) , C.S. Bremec a,b , E.M. Acha b,c , H. Mianzan a,b a Consejo Nacional de Investigaciones Cientı´  ficas y Te´ cnicas (CONICET), Argentina b Instituto Nacional de Investigacio´ n y Desarrollo Pesquero (INIDEP), Po. V. Ocampo no. 1,B7602HSA Mar del Plata, Argentina c Universidad Nacional de Mar del Plata, Mar del Plata, Argentina Received 18 July 2003; accepted 4 March 2004 Abstract This study deals with the spatial distribution of macrobenthic communities (biomass and abundance) in the Rı ´o de la Plataestuary, Argentina e Uruguay, and the adjacent shelf waters. The benthic invertebrates were caught with an epibenthic dredge (41samples). Multivariate analysis (cluster, MDS), SIMPER and BIO-ENV analyses were applied to analyze benthic communitystructure and their relationships with environmental variables. A consistent large-scale diversity pattern was found: a distinctiveestuarine zone could be distinguished, with muddy sediments and a wide range of salinity, characterized by higher abundance of fauna and lower diversity, dominated by the deposit-feeding bivalve  Mactra isabelleana ; and a marine zone, with sandy-shell debrissediments and higher and less variable salinity values, with higher values of diversity. Major presence of deposit feeders was relatedto higher particulate organic matter in the estuarine area. Bottom type, salinity and the presence of a turbidity front are consideredthe main physical variables in structuring benthic communities of the Rı ´o de la Plata estuary.   2004 Elsevier Ltd. All rights reserved. Keywords:  macrobenthos; spatial scale; diversity; estuarine; marine; SW Atlantic 1. Introduction Benthic estuarine animals are commonly thought tobe distributed along gradients of physiological stressaccording to their environmental tolerance (Remaneand Schlieper, 1971). Spatial differences in the compo-sition of the benthic communities along estuarinegradients have been related mainly to changes insalinity, depth, sediment grain size, and organic content(Day et al., 1989). These environmental variables restrictthe range of organisms that are able to survive with theconsequence that estuaries are considered areas of lowdiversity, with high abundances, when compared withbordering marine systems (Day et al., 1989; Atrill et al.,1996; Constable, 1999).The Rı ´o de la Plata is an extensive and shallowcoastal plain estuary on the western South Atlantic,between Argentina and Uruguay (35 e 36 ( S) (Fig. 1).The main characteristics of the estuary are its largegeographic extent and the occurrence of a quasi-permanent salt wedge, which generates bottom andsurface salinity fronts, important in fish reproductiveprocesses and where high concentrations of zooplanktonoccur (Mianzan et al., 2001). A turbidity front is present in the inner estuary (Mianzan et al., 2001).While there is mounting evidence that factors drivingecological forces could be affected by ecosystem size(Post, 2002), available information on benthic assemb-lages generally comes from studies covering small areas(see Remane and Schlieper, 1971; Day et al., 1989; Atrill ) Corresponding author. Present address: Benthos Laboratory,Instituto Nacional de Investigacio ´n y Desarrollo Pesquero (INIDEP),V. Ocampo no. 1, B7602HSA Mar del Plata, Argentina. E-mail address: (D.A. Giberto).Estuarine, Coastal and Shelf Science 61 (2004) 1 e 13 0272-7714/$ - see front matter    2004 Elsevier Ltd. All rights reserved.doi:10.1016/j.ecss.2004.03.015  et al., 1996; Platell and Potter, 1996; Constable, 1999). A similar situation is found in studies about benthicdiversity, where data over different spatial scales areneeded worldwide (Gray, 2001). It is the case of our study area, where the present knowledge about subtidalbenthos comes from scarce studies with limited spatialcoverage (see references in Mianzan et al., 2001).This contribution provides information about thebenthic assemblages from a large system, includingestuarine and marine regions, from temperate Southernlatitudes. The aims of this work are to characterize thebenthic assemblages of the Rı ´o de la Plata estuary andadjacent shelf waters, and to assess their relationshipwith the gradient of environmental variables (salinity,depth, temperature, bottom type and turbidity frontposition). 2. Study area The Rı ´o de la Plata estuary extends for over 280 kmfrom the head (25 km wide) to the 230-km-wide mouthbetween Punta Rasa and Punta del Este, with a mixoha-line area of 38,000 km 2 (Fig. 1). Depth in the estuaryranges from 5 to 25 m and salinity values between 0 and30 PSU, while in the adjacent marine zone depth rangesfrom 25 to 40 m and salinity between 30 and 34 PSU.The freshwater discharge (annual mean 22,000 m 3 s  1 )from the Parana ´ and Uruguay rivers into the estuaryexhibits little seasonality. Also significant horizontaltemperature gradients are absent (Mianzan et al., 2001).The average annual suspended sediment load fromParana ´ and Uruguay rivers, 79.8 ! 10 6 tons yr  1 , con-tains 75% coarse to medium silt, 15% fine to very finesilt, and 10% clay. Muddy sediments (predominance of silt and clay, modal diameter = 44  m m e 3.9  m m) coverthe mixohaline zone. Most of the river discharge flowsout through the northern channel bordering theUruguayan coast. This channel extends out of theestuary onto the marine continental shelf. The marineenvironment, except for the channel zone, is character-ized by a large sand body very uniform in grain size(modal diameter = 250  m m e 125  m m) and composition,extending from the continental shelf into the mouth of the estuary. The contact of the sands with the muddysediments makes a transitional zone of sandy silt andsandy clay textures. A coarse sedimentary texture isfound in the inner shelf (Urien, 1972).A variable particularly considered in this study is thepresence of the turbidity front, a region of locallyelevated suspended matter concentrations, which occurnear the bottom salinity front (see Framin ˜an et al., Fig. 1. The Rı ´o de la Plata estuary and adjacent shelf waters, with location of sampling sites ( n  ¼  41) ( C ) and isobaths (m).2  D.A. Giberto et al./Estuarine, Coastal and Shelf Science 61 (2004) 1 e 13  1999). The mean distribution of the turbidity front hasa high degree of variability at the northern coast of theestuary. In this region, the frontal position variesbetween 57 ( 00 #  and 54 ( 12 # W, a distance of approxi-mately 200 km (Framin ˜an et al., 1999). At the southerncoast the modal position of the front coincides with the5m isobath, although great variations also occur. Thedistribution has three areas with maximum values of frontal density: the southern and northern area of theSamborombo´n Bay (centered in 36 ( 13 # S, 56 ( 50 # W and35 ( 40 # S, 57 ( 00 # W, respectively) and the coastal areasouth of Montevideo (35 ( 00 # S, 56 ( 20 # W) (Framin ˜anet al., 1999). 3. Materials and methods 3.1. Sampling Benthos samples and environmental data came from41 stations sampled in spring 1993, by the R/V ‘‘EduardoL. Holmberg’’ (INIDEP) (Fig. 1). Faunal samples weretaken with an epibenthic dredge (see Rothlisberg andPearcy, 1977; one trawling at each station, frame Z 200 cm ! 50 cm, mesh size Z 1 cm, mean ship speed Z 1.8 knots, duration of hauls Z 10 min, mean areasampled Z 1.5 km 2 ). The dredge remained open whenit was lowered to the bottom and brought to the surface.Samples were sieved through a 1 mm screen on boardthe ship and preserved for analysis in 5% formalin. In thelaboratory, all invertebrates were sorted out from thesamples, identified and counted under a stereo micro-scope; wet weight was measured in grams for each taxonwith an analytical balance (accuracy 0.01 g). Samplingwas considered semi quantitative: we estimated the sweptarea of every sample using the ship speed, frame size andduration of every tow. This procedure permitted toestimate densities in order to compare the data of thewhole sampling area. The abundance and biomass werestandardized to km 2 .Oceanographic sampling was performed with a SeaBird-19 CTD (Conductivity e Temperature e Depth pro-filer). The salinity is reported following the PracticalSalinity Scale. Information of the sediment type at eachstation was taken from local sediment maps (Urien,1972; Giberto, 2001) and assigned to the following mainfraction found in the different sectors of the estuary:mud, mud e sand, fine medium sand, and shell debris(predominance of shells, calcareous fragments andpebbles). Density values (using pixel values) of theturbidity front were assigned to the sampling stationsfrom Framin ˜an and Brown (1996), who used a four-yearspan of NOAA-AVHRR daily images to estimate thedistribution of the frontal density, a probabilisticmeasure of frontal occurrence in a given area. 3.2. Data analysis3.2.1. Benthic assemblages A non-parametric multivariate analysis (Field et al.,1982; Clarke, 1993; Clarke and Warwick, 2001) based on community structure were performed to definebiological grouping of sampling sites. Classification(group average sorting of the Bray e Curtis similaritymeasures based on 4th root transformed abundance andbiomass data) and ordination (multi-dimensional scal-ing (MDS) on the above similarity matrices) methodswere used. In the analysis among species the referreddata matrix was used, considering species with abun-dance or biomass  O 1% of the total of any sample inorder to have any chance of an interpretative clusteringand ordination analysis. The composition of theassemblages was analyzed at two levels: taxonomicidentity and feeding guilds. Feeding type includeddeposit feeders (DF), filter and suspension feeders(F/S), carnivores and/or scavengers (C/S), omnivores(OM) and herbivores (HE). Macrobenthic taxa wereassigned to trophic groups based on data documented inthe literature (Fauchald and Jumars, 1979; Wolff, 1992;Barnes and Morton, 1997 and references included inMianzan et al., 2001).The samples that showed a marked tendency tocluster (similarity O 20% in the abundance and biomassclassification analysis) were used to perform a SIMPERanalysis (‘‘similarity percentages’’) in order to identifythose species which contributed most to similaritieswithin groups (‘‘typical’’ and ‘‘discriminating’’ species,Clarke, 1993).Based on these biological patterns, distinct areas withdifferent species composition were established. Differ-ences in abundance and biomass (standardized to meanpercentage per station) among areas (defined in themultivariate analysis) and group composition (usingtaxonomic identity and feeding guilds) were analyzed bya two-way ANOVA followed by a Tukey test modifiedfor unequal  n  (Zar, 1984). An arcsine or log-trans-formation was applied when data displayed significantdepartures from homoscedasticity (detected with theBarlett’s test) (Zar, 1984).Following Hill (1973) and Gray (2000), we analyzed for each site: (a) the species richness (Margalef’s index d   ¼  S     1 = log  N  , where  S   = total number of species and N   = total number of individuals); (b) the diversity bythe index Hill  N  1  ¼  exp ð H  # Þ , were  H  #  is the commonlyused index Shannon e Wiener; (c) the Pielou’s index of evenness  J  #  ¼  H  # = log  S  ; and (d) the index of polydo-minance Hill  N  2  ¼  1 = ð  p 21 C  p 22 / C  p 2 n Þ , where  p 1  is theproportional abundance of the first species compared tothe total number of individuals in the  n  samples. Toexplore differences between mean values of richness,diversity, evenness, and dominance among stations theMann e Whitney  U   test was used (Zar, 1984). 3 D.A. Giberto et al./Estuarine, Coastal and Shelf Science 61 (2004) 1 e 13  3.2.2. Interactions between environmental variablesand benthic patterns Differences between the environmental characteristics(depth, salinity, temperature and frontal density) of thedefined zones were assessed using the Mann e Whitney  U  test (Zar, 1984). Spearman’s correlation coefficients wereused to examine the relationships between the biologicalvariables (diversity, dominant species, abundance andbiomass of the total macrobenthic fauna) and depth,bottom salinity, bottom temperature and frontal density.However, since it would not be expected that a singleenvironmental variable would provide the best explana-tion of biological patterns, the BIO-ENV procedure wasused to determine which set of variables (salinity, depth,temperature, frontal density, similarity calculated withthe Euclidean distance coefficient) best explains thebiological matrices (diversity and abundance e biomassdata, using Bray e Curtis similarity measure) (Clarke andAinsworth, 1993; Clarke and Warwick, 2001). Salinity,temperature and frontal density were transformed withthe log ð x C 1 Þ . In line with recommendations of  Clarkeand Warwick (2001), prior to this analysis a draftsmanplot (scatter plots between pairs of environmentalvariables) was used to assess the linearity of the dataand the inter-correlation between variables, in orderto remove from the analysis all subsets of variableswhich have mutual Spearman’s correlation coefficientsaveraging more than about 0.90. 4. Results 4.1. Composition of the macrobenthos We recorded a total of 134 species (most from marinewaters) (Table 1). Thirty-four species were recorded inmixohaline waters ( ! 30) and only seven ( Glypteuthriaagnesia , Idoteidae unident. 1 e 2, Nepthyidae unident.,Polychaeta unident.,  Pyrene isabellei   and  Travisia  sp.)were exclusively found at low salinity conditions(salinities  ! 15). The highest densities (station 25;1931 ind. km  2 ) and biomass (station 21; 2291.7 gkm  2 )were found in front of the Uruguayan coast; the lowestvalues were obtained at station 14 (19 ind. km  2 ) andstation 36 (23.32 gkm  2 ) (Fig. 1).Crustaceans were most diverse with 48 species,followed by polychaetes with 33 species and molluskswith 31 species. Mollusks, on the other hand, occurredin highest number and biomass (65.59% and 65.90% of the total respectively) throughout the study area (Table2). Fifteen species contributed  O 90% of the totalabundance and biomass, with the bivalve  Mactraisabelleana  (Mactridae) being most numerous, followedby the red shrimp  Artemesia longinaris  and the sanddollar  Encope emarginata  (Table 2). 4.2. Benthic assemblages and environmental conditions Classification and ordination analysis separatedsampling stations into two main groups, correspondingto inner (hereafter referred to as the ‘‘Area 1’’) and outer(hereafter ‘‘Area 2’’) stations. Based on this two plots,Area 1 included 16 sampling sites (13, 18, 20 e 31, 33 and34) and 37 species, while Area 2 included 25 sites (1 e 17,19, 32, 35 e 41) and 125 species. Fig. 2 shows the MDSordination of sites clustered at a similarity of 25%,according to (a) abundance and (b) biomass of species.The biomass plot separated Area 2 into two sub-groups,one corresponding to the Argentinean coast (character-ized mainly by the presence of   Encope emarginata ) andanother to the Uruguayan coast (characterized by manyspecies) (Fig. 2b).Area 1 included stations with salinity ranging from12.6 to 32.9, depths between 8 and 23 m, frontaldensities ranging from 0 to 8.5, temperatures between9.2 and 11.4 ( C, and with 85.71% of sites correspondingto muddy bottoms. Area 2 included stations withsalinity ranging between 30.2 and 33.8, depths between10 and 27, frontal densities ranging from 0 to 5,temperatures between 9 and 11.8 ( C, and with stationsin front of the Uruguayan coast characterized by theshell debris fraction (48% of sites) and mud e sandfraction (28%), while the Argentinean coast wascharacterized by the sand medium fraction (24%).The species that typified Area 1 were the bivalve Mactra isabelleana , the decapod  Artemesia longinaris and the gastropod  Buccinanops duartei   (SIMPERanalysis, Table 3). Area 2 was typified by  A. longinaris ,the bivalve  Corbula patagonica , and the serolid  Serolismarplatensis  (Table 3). The red shrimp  A. longinaris  wastypical in both regions, but with higher densities in Area2. Therefore, together with  M. isabelleana , they were themain discriminating species (Table 3). Classification andordination analysis among taxa separated species of Area 1 and Area 2. Dominant taxa (those species thatcontributed more than 10% of similarity at any site),both in abundance and biomass are displayed in Fig. 3.Area 1 had higher values of total invertebrates abun-dance (10,024 ind. km  2 vs. 3829  2 ) and bio-mass (8349.4 gkm  2 vs. 6257.6 g km  2 ) than Area 2.Mean abundance (2131.75 G 1614.65 ind. km  2 vs.534.39 G 572.8  2 ) and biomass (1849.24 G 2088.7 gkm  2 vs. 841.80 G 793.8 gkm  2 ) values dis-played highly significant (  p ! 0 : 001) and significant(  p ! 0 : 05) differences, respectively, always highest forArea 1. The lowest values of diversity and richness, withunevenly represented species, characterized Area 1(Table 4). Diversity reached the highest value in Area2 at station 5 (Hill  N  1  ¼  22 : 7).A two-way ANOVA based on abundance andbiomass data (see Fig. 4) showed that the meanpercentage per site was significantly different among 4  D.A. Giberto et al./Estuarine, Coastal and Shelf Science 61 (2004) 1 e 13  areas and taxonomic composition, both for abundance( F   ¼  5 : 478,  p  ¼  0 : 02, and  F   ¼  67 : 597,  p ! 0 : 001) andbiomass ( F   ¼  8 : 524,  p  ¼  0 : 004, and  F   ¼  46 : 17,  p ! 0 : 001), with also a significant interaction term area ! taxonomic composition for abundance ( F   ¼  8 : 571,  p  ¼  0 : 0002) and biomass ( F   ¼  20 : 23,  p ! 0 : 0001). Re-sults of the Tukey test showed that the mollusks weredominants in terms of biomass in Area 1 (  p ! 0 : 001), Fig. 2. MDS ordination of stations using data of (a) abundance and (b) biomass of species. Groups (Area 1= C , Area 2= 6 ) are stations delimitedfrom classification analysis (group average sorting of the Bray e Curtis similarity measure) at a similarity level of 25%.5 D.A. Giberto et al./Estuarine, Coastal and Shelf Science 61 (2004) 1 e 13
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks