sfo

ail

Laboratrio de Conservao e Manejo de Vida Sivestre, Universidade Federal do Rio Grande do Sul, RS, Brazilc Systems Ecology Group, Leibniz Center for Tropical Marine Ecology, Bremen, Germanyd Escola Nacional de Sade Pblica/FIOCRUZ, Rio de Janee Jacobs University, Bremen, Germany

to rainfall and other environmental factors (Kjerfve, 1994). Theseecosystems occupy 13% of coastal areas in the world (Barnes,1980) and support a range of highly valued ecological services(Anthony et al., 2009). In addition, they are among the most bio-logically productive environments of our planet and harbor a richand unique biodiversity of species (Esteves et al., 2008; Whiteldet al., 2008). The high environmental heterogeneity of coastal

04; Tavares andoons are amongmultipleand Ch

2010), climate change (Anthony et al., 2009; ClauseClausen, 2014) and different forms of anthropogenic pertur(McKinney et al., 2010). For example, they are vulnerable tocial sandbar opening and drainage canals building, which couldpromotes arid conditions and salinization (Esteves et al., 2008).Previous studies have documented the sensitivity of waterbirdsto habitat alterations (Alexander and Hepp, 2014; Ma et al.,2010) and have highlighted the importance of management strate-gies, especially under the current context of rapid environmentalchange (IPCC, 2014) and biodiversity loss (Monastersky, 2014).

Corresponding author.E-mail address: wetlandbirdsbrazil@gmail.com (D.C. Tavares).

Biological Conservation 186 (2015) 1221

Contents lists availab

o

.e lTropical coastal lagoons are shallow aquatic ecosystems locatedat the boundary between terrestrial and marine environments.They are intermittently connected to the sea by restricted inletsand exhibit marked variations in their hydrological features due

Baidu et al., 1998; Paracuellos and Tellera, 20Siciliano, 2013a). However, tropical coastal lagthe most threatened ecosystems on Earth due tobances, including habitat alterations (Bullerihttp://dx.doi.org/10.1016/j.biocon.2015.02.0270006-3207/Crown Copyright 2015 Published by Elsevier Ltd. All rights reserved.distur-apman,n andbationsarti-Wetlands porting high avian abundance and biodiversity in tropical coastal lagoons. Management strategies shouldtherefore address a specic number of structural and hydrochemical attributes of coastal lagoons andshould carefully consider the effects of land use practices.

Crown Copyright 2015 Published by Elsevier Ltd. All rights reserved.

1. Introduction lagoons, in both temporal and spatial scales, provides habitats foraquatic bird species with different ecological needs (Ntiamoa-a r t i c l e i n f o

Article history:Received 23 December 2014Received in revised form 12 February 2015Accepted 20 February 2015

Keywords:Aquatic birdsConservationGeneralized Linear Mixed ModelsHuman-induced changesManagementTropical biologyWater depthWaterfowliro, RJ, Brazil

a b s t r a c t

Tropical coastal lagoons are among the most threatened ecosystems on Earth due to multiple factors. Theloss and degradation of coastal lagoons can adversely affect waterbirds, which depend crucially on wet-land habitats. This study investigates the effects of a number of environmental variables on waterbirdhabitat use. Specically, we assessed the abundances of six waterbird guilds in relation to hydrochemical,structural, and anthropogenic factors in 351 survey blocks of 22 tropical coastal lagoons of southeasternBrazil. Patterns of habitat use were analyzed using generalized mixed models. We found that water depthwas the most important variable inuencing the waterbird assemblage. Increasing water depth reducedthe abundance of small and large wading birds, dabbling ducks, and diving birds. Extreme values of salin-ity negatively affected diving birds and large wading birds. Vegetation height inuenced shorebirds nega-tively, while it inuenced vegetation gleaners positively. Anthropogenic factors played an important rolein structuring the habitat of small wading birds, dabbling ducks, diving birds and shing birds. Our studyindicates that the various guilds showed specic responses to different habitat variables. We concludethat the maintenance of water depths in the range of 020 cm, the preservation of a mosaic of vegetationheights, and the reduction of livestock grazing pressure are among the most important aspects for sup-a Laboratrio de Cincias Ambientais, Universidade Estadual do Norte Fluminense, Campos dos Goytacazes, RJ, BrazilbEnvironmental and anthropogenic factorsof tropical coastal lagoons: Implications

Davi Castro Tavares a,, Demtrio Luis Guadagnin b, JAgostino Merico c,e

Biological C

journal homepage: wwwtructuring waterbird habitatsr management

son Fulgencio de Moura c, Salvatore Siciliano d,

le at ScienceDirect

nservation

sevier .com/locate /b iocon

by increased abundance of resources (Takekawa et al., 2006).High vegetation can have negative effects for wading birds and

Congulls by impairing prey detectability and site selection (Butlerand Gillings, 2004; Tavares et al., 2013), but it constitutes a suit-able habitat for other groups such rallids and ciconids(Jedlikowski et al., 2014; Tavares and Siciliano, 2013b). Finally, aclear positive relationship has been observed between wetlandsize, habitat heterogeneity and waterbird species richness andabundance (Elmberg et al., 1994; Rosselli and Stiles, 2012;Sebastin-Gonzlez and Green, 2014).

Here we investigated the effects that a number of importantenvironmental variables have on waterbird habitat use in tropicalcoastal lagoons. Since waterbird abundance represents a recog-nized metric for habitat use (Baschuk et al., 2012; Lunardi et al.,2012; Paracuellos and Tellera, 2004), we assessed the effects ofhydrochemical, structural and anthropogenic factors on the abun-dance of different guilds. Specically, we tested the expectationthat different guilds should show specic responses to variationsin water depth, vegetation height, lagoon size, water salinity, waterpH, grazing pressure, and distance from human settlements(Cumming et al., 2012; Guadagnin and Maltchik, 2007; Prez-Crespo et al., 2013; Sebastin-Gonzlez and Green, 2014).Knowledge regarding the use of habitats by waterbirds is particu-larly poor in tropical and subtropical coastal systems, which harborcomplex arrays of habitats and unique bird communities (Gianucaet al., 2014; Lunardi et al., 2012; Tavares and Siciliano, 2014a).Furthermore, we provided an evaluation of suitable habitat condi-tions, which may nd useful applications for managementpurposes.

2. Materials and methods

2.1. Study site

This study was conducted in a chain of 22 tropical coastallagoons, located along the coastline of northern Rio de Janeiro,southeastern Brazil (221104400S, 412600030W, Fig. 1A). TheRestinga de Jurubatiba National Park includes eleven of the sur-veyed coastal lagoons (Caris et al., 2013). Almost all lagoons arecharacterized by shallow waters, average depth = 0.82 cm, andare relatively small in size, average size = 0.94 km2 (Calimanet al., 2010). When considered together, they form a heterogeneousecosystem with contrasting hydrochemical and structural attri-butes both in the spatial and temporal dimensions (Macedo-Soares et al., 2010; Tavares and Siciliano, 2013a). For example,two lagoons separated by a 300 m sandbar can have, respectively,However, understanding the effects of habitat alterations onwaterbirds is a difcult task that requires rst an insight intohow environmental factors inuence habitat use. Most environ-mental and anthropogenic perturbations have contrasting effectson different waterbird guilds. For example, alteration in spatialattributes may cause the loss of certain species due to area and iso-lation effects (Guadagnin and Maltchik, 2007), but may favor otherspecies that use human-created habitats or resources (Prez-Garca et al., 2014). Also, hydrological and hydrochemical alter-ations can have opposite consequences to waterbirds. High waterlevels, for example, can make it difcult for small birds to land, restand nd food, while this same condition may provide advantagesto diving birds (Navedo et al., 2012). High salinities may negativelyaffect growth and survival of some waterbird guilds due to prob-lems associated with osmoregulation and feather damage(Rubega and Oring, 2004), although other guilds may be favoured

D.C. Tavares et al. / Biologicalsalinity values of 15 ppt and 130 ppt during a severe dry period.This heterogeneity allowed us to account for a wide range of habi-tat variables. Note, however, that the orientation of the lagoonsparallel to the coastline makes such variables relatively uniformwithin each lagoon (Di Dario et al., 2013; Kjerfve, 1994). These sys-tems operate like intermittently open estuaries, remaining most ofthe time closed off from the sea, especially during low rainfallregimes (Whiteld et al., 2008). Such condition together with theirsmall sizes, makes the lagoons particularly vulnerable to anthro-pogenic inuences (Ortega-Cisneros et al., 2014). Two principalkinds of management activities are erratically conducted by localauthorities and farmers in the study site (Caris et al., 2013): (1)sandbar opening to mitigate overow and (2) drainage canalsbuilding to claim land for agriculture and pasture. Such activitiescause broad shifts in the hydrochemical, structural, and spatialcharacteristics of the lagoons (Esteves and Suzuki, 2008; Santoset al., 2006).

2.2. Waterbird survey

We carried out point counts along the coastal lagoons, everymonth, between May 2012 and April 2013. Counts were performedevery 1 km in the center of survey blocks. We considered 34 surveyblocks distributed along a transect oriented linearly with thebeachline (Bibby et al., 2000, Fig. 1B). Each block covered a semi-circular visual eld with 300 m xed radius and the countinglasted no more than 5 min in order to reduce the probability of birdrecount (Gregory et al., 2004). The data were collected always bythe same ornithologist by means of spotting scope (2060 80),binoculars (8 42), and rangender. The small size of the surveyblocks and the low vegetation cover (average height = 0.13 m)reduced the survey error. Rails and Bitterns were not considereddue to their low detectability with the adopted survey method(Conway and Gibs, 2005).

2.3. Waterbirds guilds

Waterbirds were grouped into guilds reecting species foraginghabits and morphology, according to Tavares and Siciliano (2014b)and following the guild concept proposed by Blondel (2003). Weidentied six guilds: diving birds (grebes), dabbling ducks (belong-ing to the genera Dendrocygna and Anas), large wading birds (her-ons, egrets and storks), vegetation gleaners (jacanas andgallinules), shing birds (gulls and terns) and small wading birds also called shorebirds (mostly sandpipers and yellowlegs). Thespecies observed within each guild and basic statistics are reportedin Appendix A.

2.4. Habitat variables

We collected data on seven habitat variables at the end of eachwaterbird survey and for each sampling block. These variableswere classied into three groups: (1) structural, i.e. vegetationheight, lagoon size and water depth; (2) hydrochemical, i.e. watersalinity and pH; (3) anthropogenic, i.e. livestock grazing pressureand distance from human settlement.

Vegetation height was estimated at each survey block usinggraduated sticks in three random locations. Mean values were con-sidered for each survey block. The southern cattail (Typha domin-gensis Pers.) is the dominant plant species with height above 1 m.The effect of lagoon size on bird abundance was investigated usingLandSat imagery (available online at http://landsatlook.usgs.gov).Lagoon size was measured only for water covered surface, yieldinga metric corrected by water availability (Sebastin-Gonzlez andGreen, 2014). Water depth was estimated as the mean between

servation 186 (2015) 1221 13minimum water depth and maximum water depth, measured bymeans of xed graduated sticks at, respectively, 1 m and 300 mfrom the lagoon shore.

. Cirling

ConWater salinity and pH were measured in three points spaced atleast 50 m apart from each other using portable equipment, whichwere regularly calibrated during the survey. Mean values wereconsidered for both variables for the statistical analyses.

The livestock grazing pressure was estimated as the number of

Fig. 1. (A) Location of the study site, on northern Rio de Janeiro, Southeastern Brazilfalling outside lagoons boundaries over all period of study. (B) Diagram of the samp

14 D.C. Tavares et al. / Biologicaldomestic animals grazing in a survey block. These counts followedthe same protocol employed for the waterbirds, and were madeduring the morning (after the day of bird counts) when farmersusually put the cattle out to pasture. The distance from human set-tlement was estimated as the shortest distance between the centerof each lagoon and the nearest inhabited building or farming activ-ity (Longoni et al., 2011). This metric was obtained from theDigitalGlobe high resolution imagery produced in January 2012and offered by Google Earth. The distance from human settle-ment may inuence guild abundances because of its effect on birdsite permanency (Glover et al., 2011; Gyimesi et al., 2012; Petersand Otis, 2007).

After one year of surveys, we collected data from 420 points, but69 were excluded due to: (i) stick removal by unknown factors; (ii)shrinking of water surface; (iii) unfeasible lagoon size estimationdue to intense cloud cover on LandSat imagery. Therefore a totalof 351 points were analyzed. In addition, since the group of smallwading birds included nearctic migrants, we considered only sam-ples between September and April when these birds visit theBrazilian coastal environments (Rodrigues, 2000; Tavares andSiciliano, 2013a), so that only 224 points were analyzed for thisguild. Basic statistics of investigated habitat features are shownin Appendix B.

2.5. Statistical analyses

Our core statistical analyses were preceded by a data explo-ration phase during which we checked for outliers, zero ination,collinearity, and other possible problems pointed out by Zuuret al. (2010). The analysis of collinearity between variables neverexceeded a Variance Ination Factor (VIF) of 1.98. Typically, a prob-lematic collinearity is assumed to occur when the VIF is above 3(Zuur et al., 2009). In addition, we applied Principal ComponentAnalysis (PCA) on habitat variables in order to elucidate the covari-ance structure in the dataset.

To single out the habitat variables that exert the most importantinuence on the abundance of the six guilds, we used Generalized

cles indicate the location of survey points, while squares indicate points ignored fordesign used to investigate waterbird habitat use in a chain of 22 coastal lagoons.

servation 186 (2015) 1221Linear Mixed Models, GLMMs (Bolker et al., 2009). This approachallowed us to account for: (i) non-normal data; (ii) models contain-ing multiplicative terms; (iii) overdispersion caused by excess zer-oes in the data set; and (iv) temporal and spatial correlations,known as pseudoreplication (Crawley, 2007; Zuur et al., 2009). Weused random-intercept models with environmental variables setas xed effects and survey blocks nested within lagoons consideredas random effects, thus accounting for spatial pseudoreplication(Millar and Anderson, 2004). Months were also included as a sepa-rated random effect, correcting temporal correlationswhen the sur-veyed lagoons exhibited some temporal coherence (Caliman et al.,2010). A set of at least 10 models were constructed step-by-stepreducing from a full model that included all guilds (Bolker et al.,2009). Candidate models were tted by maximum likelihoodestimation using the Laplacian approximation, with a zero-inatednegative binomial family as the best error distribution accordingto diagnostics and distribution of residuals.

Models were ranked according to the Akaikes InformationCriterion, AIC (Burnham and Anderson, 2002); the preferred modelbeing the one with the lowest AIC score. Since for each guild weconstructed a relatively large number of models, small differencesin AIC scores prevented us to select one best single model out of aset of models showing very similar performances. Thus, weadopted a model averaging with shrinkage (a so called zeromethod) using a cut-off of 2 AIC, because estimates from modelswith poor weights tend to be spurious (Grueber et al., 2011).Note that the use of the zero method is particularly recommendedfor studies like the present, which aim is to detect the factors thathave the strongest effects on some dependent variables (Nakagawaand Frackleton, 2010).

To investigate the inuence that habitat variables had on theentire waterbird assemblage, we measured the importance of each

The most suitable habitat conditions for each signicant vari-able were assessed graphically by incorporating smoothing func-tions into the Generalized Additive Mixed Models, GAMMs (Zuuret al., 2009). In practice, such models are GLMMs with smoothingfunctions linked to the predictive variables, thus allowing for aexible and effective way to t linear and non-linear responses(Gilman et al., 2014; Wood, 2006).

The statistical analyses were performed with R (version 3.0.2)using the packages glmmADMB for model t; MuMIn forestimating the importance of each variable; bbmle for calculatingthe AIC values; and gamm4 for producing the response curves (RCore Team, 2013). In addition, we assessed collinearity betweenexploratory variables by means of the function corvif, providedby Zuur et al. (2009).

3. Results

The rst and the second principal components explain, respec-tively, 34% and 22% of the variation in the data (Fig. 2). Accordingto these two principal components, salinity is negatively correlatedwith water depth, water pH, and vegetation height. Lagoon size ispositively correlated with distance from human settlements and isnegatively correlated with livestock grazing pressure.

Table 1 shows the best Generalized Linear Mixed Models for

Fig. 2. Ordination diagram of Principal Components Analysis showing the relation-ship of the observed environmental variables. Dots indicate observations. Thedirections of the arrows indicate positive or negative correlations with eachprincipal component. Abbreviations indicate water salinity (ws), livestock grazingpressure (lgp), water pH (wp), water depth (wd), vegetation height (vh), distancefrom human settlements (dhs), and lagoon size (ls).

D.C. Tavares et al. / Biological Conservation 186 (2015) 1221 15habitat variable as the sum of the Akaike weights in a set of modelsrandomly generated from the full model (Burnham and Anderson,2002; Nummi et al., 2013). However, sometimes high Akaikeweights can be associated with variables with low statistical sig-nicance (Burnham and Anderson, 2002). To minimize this effectwe calculated a global importance index for each variable by sum-ming the importance values of a given variable on the six guildsand by multiplying this number by the sum of the signicant stan-dardized parameter estimates of the variable on the guilds.Modular values of parameter estimates were used to avoid can-cellations between negative and positive estimates. Therefore, thisglobal importance index incorporates Akaike weights, variableestimates (slope strengths) and their signicances.Table 1Ranking of the best Generalized Linear Mixed Models tted with negative bidifferent habitat variables on 22 tropical coastal lagoons on southeastern B

Best Generalized Linear Mixed Models for guild abundances

Dabbling ducksDH lagoon size + livestock grazing pressureWater depth + livestock grazing pressure + DH lagoon sizeLivestock grazing pressure + water depth + DH + salinity + lagoon size

Fishing birdsDH + water depth + livestock grazing pressureDH + vegetation height + livestock grazing pressure + pH

Vegetation gleanersVegetation height + salinity + livestock grazing pressure + DHVegetation height salinity + livestock grazing pressure + DHLarge wading birdsWater depth salinity + lagoon size + DHWater depth salinity + DHWater depth salinity + livestock grazing pressure + lagoon size + DHWater depth salinity + lagoon sizeDiving birdsSalinity water depth + livestock grazing pressure + lagoon sizeSmall wading birdsWater depth + vegetation height + DH + livestock grazing pressureWater depth + distance from human settlement + livestock grazing presWater depth + vegetation height + livestock grazing pressure + DH + pHWater depth + distance from human settlement + vegetation height

K = number of parameters estimates + intercept; DAIC = difference in Akweights; DH = distance from human settlement.waterbird guild abundance with respect to environmental vari-ables. The negative binomial models revealed that the varioushabitat variables have a diverse effect on different waterbird guilds(Table 2). In Fig. 3 we summarize the relationships between habitatvariables and waterbird guilds. Water depth affects the abundanceof small and large wading birds negatively, while it has a non-lin-ear effect on dabbling ducks and diving birds. Salinity has a non-linear inuence on diving and large wading birds. Vegetationheight inuences vegetation gleaners positively and small wadingbirds negatively. Livestock grazing pressure affects both dabblingducks and diving birds negatively. Distance from human settle-ment affects shing birds negatively and small wading birds posi-tively. Interestingly, water pH and lagoon size have no signicanteffect on any guilds.

nomial errors for predicting waterbird guild abundances as functions ofrazil.

K DAIC wi

5 0 0.266 0.4 0.216 1.2 0.14

4 0 0.415 1.9 0.16

5 0 0.536 1.4 0.26

6 0 0.315 0.4 0.257 0.8 0.215 1.6 0.14

6 0 0.52

5 0 0.26sure 4 0.4 0.21

6 0.5 0.204 1 0.16aikes Information Criterion score between ranked models; wi = AIC

d gules

estimates.

e

ConModel-averaged coefcients Estimat

Dabbling ducksLivestock grazing pressure 0.19Water depth 0.08Distance from human settlement 0.49Lagoon size 1.83Salinity 0.01Distance from human settlement lagoon Size 0.53Diving birdsWater depth 0.2Livestock grazing pressure 0.11Salinity 0.06Salinity water depth 0.01Lagoon size 1.12

Vegetation gleanersVegetation height 4.27Salinity 0.05Livestock grazing pressure 0.07Distance from human settlement 0.5Vegetation height salinity 0.03Small wading birdsTable 2Parameters of the Generalized Linear Mixed Models for waterbirtropical coastal lagoons on southeastern Brazil. Signicant variab

16 D.C. Tavares et al. / BiologicalRegarding the global importance of each variable on the wholewaterbird assemblage, water depth exhibits by far the greatesteffect, followed in order of importance by salinity, livestock grazingpressure, vegetation height, and distance from human settlement(Fig. 3). These results are consistent with those obtained byestimating the effect of habitat variables on single guilds.

Fig. 4 shows, for each guild, response curves as functions of themost signicant variables. The appropriate water depth range formost of the affected guilds appears to be 020 cm. Diving birdsand large wading birds show greater abundances under inter-mediate values of salinity, between 20 and 40 ppt, and show a sev-ere decline when salinity exceeded 100 ppt. The abundance ofvegetation gleaners becomes stable in vegetation above one meter.Distance from human settlement above about 2.5 km affects sh-ing birds negatively. In contrast, the presence of humans has anegative effect on small wading birds when at distances below2.5 km, although this impact is less severe than the ones exertedby other variables (Fig. 4).

4. Discussion

We investigated the effects of a comprehensive set of environ-mental variables on waterbird habitat use in a chain of 22 tropicalcoastal lagoons of southeastern Brazil.

Vegetation height 1.28Distance from human settlement 0.3Water depth 0.29Livestock grazing pressure 0.02pH 0.06Vegetation height grazing pressure 0.24Fishing birdsDistance from human settlement 0.91Water depth 0.07Livestock grazing pressure 0.3Vegetation height 1.37pH 0.44

Large wading birdsWater depth 0.13Salinity 0.04Salinity water depth 0.01Distance from human settlement 0.26Lagoon size 0.43Livestock grazing pressure 0.02ild abundances as functions of different habitat variables on 22(P < 0.05) are marked in bold and ordered following parameter

CI lower CI upper Z P

0.31 0.07 3.08

blesd adterp

Contolerant of saline conditions thanks to their salt glands (Gutirrez

Fig. 3. Relationship between habitat variables and waterbird guilds. Signicant variaindicate positive effects, and black lines indicate non-linear effects. Size of circles anHorizontal black bars inside guild boxes represent relative mean abundances. (For inweb version of this article.)

D.C. Tavares et al. / Biologicalet al., 2012). The pattern of a slightly greater abundance at inter-mediate levels of salinity (2040 ppt) was also observed in otherstudies focusing on different waterbird species, and it appears tobe related to prey density, especially sh and macroinvertebrates(Takekawa et al., 2006; Warnock et al., 2002). Moreover, by impos-ing constraints on vegetation types and abundances, salinity cancreate habitat gradients in coastal environments (Glass andWatts, 2009; Prado et al., 2014). In the region we studied, tall veg-etation wane under salinities above 10 (Esteves and Suzuki, 2008).

Vegetation height was especially important for vegetationgleaners and small wading birds. Gleaners are closely related withmarshes. Such environments are suitable to breeding, feeding, andpredatory avoidance (Forman and Brain, 2004; Shirley et al., 2003).Whereas vegetation gleaners benets from high vegetation, smallwading birds do not. Douglas and Pearce-Higgins (2014) found astrong negative correlation between vegetation height and breed-ing density of golden plovers and attributed this to a greater preyaccessibility in shorter vegetation. In addition, vegetation may actas a barrier reducing the capability of small wading birds to detectpredators (Brindock and Colwell, 2011). Thus, these birds selectopen habitats in order to reduce predation risk (Pomeroy, 2006).In addition, shorebirds tend to increase vigilance under reducedvisibility, with negative consequences on their feeding efciency(Metcalfe, 1984) and thus on habitat occupancy.

No much attention has been devoted so far to the effects ofanthropogenic activities on waterbird habitat use in tropicalregions. We showed that human-related factors are important forstructuring the habitat of four guilds. At increasing distance fromhuman settlement, small wading birds became slightly more abun-dant. Cestari (2008) found no difference in small wading birdsabundance between areas with contrasting levels of human distur-bance in the Southeastern Brazilian coast. However, recent evi-dence suggests that human activities may induce stress in thesebirds (Borneman et al., 2014; Pagel et al., 2014), with negativeconsequences on site occupancy due to correlations between

are in grey boxes. Red lines indicate negative effects to respective guilds, blue linesjacent numbers indicate the importance of each variable in affecting the six guilds.retation of the references to colour in this gure legend, the reader is referred to the

servation 186 (2015) 1221 17environmental features and space (Gianuca et al., 2013). Humanactivity may force small wading birds to abandon habitats afictedby chronic disturbances (Kirby et al., 1993). We found quite puz-zling that shing bird abundance exhibited an inverse relationshipwith distance from human settlement. This pattern could probablybe linked to the great availability of small-sized livebearers in shal-low and brackish lagoons closely situated to villages and farmingactivities. In these locations predatory sh abundances, potentialcompetitors of shing birds, may be reduced by shing activities(pers. comm. A.C. Petry). In addition, it has been suggested thatwaste and organic inputs increase the abundance of shing birds(Marateo et al., 2013). A previous study conducted in the northernpart of Rio de Janeiro showed that some shing birds (i.e. grey-headed gulls) are more abundant in coastal environments withhigh organic waste, typically close to human settlements(Tavares et al., 2013). Therefore, we suggest that higher food avail-ability in proximity of human activities may be the cause for thenegative relationship we observed between abundance of shingbirds and distance from human settlements.

Livestock grazing pressure had a negative effect on dabblingducks and diving birds. Most species of ducks and diving birdsare resident and build nests on the ground (Tavares and Siciliano,2014a), making them vulnerable to livestock trampling (Burgeret al., 2004). In contrast, some diving birds (e.g. neotropical cor-morants) build nests on trees (del Hoyo et al., 1992), and so theyare not affected by trampling. Moreover, in the presence of cattle,birds may nd themselves spending more time and energy withdefensive and avoidance behaviors (Otieno et al., 2011), with nega-tive consequences on site occupancy. Livestock grazing has also anindirect effect on waterbirds due to removal of vegetation.Richmond et al. (2012), for example, found that the site occupancyof a vegetation gleaner (the californian black rail) was reduced ingrazed marshes of Sierra Nevada. Davies et al. (2010) reported thatlivestock grazing affected waterbirds directly by degrading feeding

Con18 D.C. Tavares et al. / Biologicaland nesting habitats. Other guilds inhabiting marshes during thebreeding season, such as shing birds (the grey-headed gull) andlarge wading birds (the maguari stork), showed similar responsesin our earlier observations (Tavares et al., 2013; Tavares andSiciliano, 2013b). Some studies reported that small wading birdsbeneted by areas where grazing kept vegetation low (Alfaroet al., 2008; Colwell and Dodd, 1995; Isacch and Cardoni, 2011).

Fig. 4. Responses and 95% condence bands (shaded areas delimited by dashed lines) of wsplines obtained with generalized additive mixed models. Circles indicate partial residuservation 186 (2015) 1221However, none of the guilds we observed showed increased abun-dance under higher grazing pressure.

Our nding that lagoon size plays aminor role inwaterbird habi-tat use is somewhat unusual (Riffel et al., 2001; Rosselli and Stiles,2012; Sebastin-Gonzlez and Green, 2014), although a few earlierstudies have obtained similar results. For example, Guadagninet al. (2009) found that waterbird abundance decreased with

aterbird abundances to signicant habitat variables. Such responses are smoothingals.

garded anthropogenic factors. But our work showed that distance

assessment of habitat use by focusing on habitat quality, for exam-

Superior-CAPES. Jailson Fulgencio de Moura is supported by a fel-

Conincreasing area of fragmentedwetlands in South Brazil. This ndingwas attributed to the lowabundanceof the larger species (e.g. divingbirds) that could benet by increased water levels, which presum-ably covariatewithwetland area. Such larger species normally needlarger areas for breeding and foraging, while species with smallerbody size benet from resources within short reach (Prez-Garcaet al., 2014). Although lagoon size was part of our observations, wecould not detect any signicant relationship between this variableand guild abundance (Fig. 3). A possible explanation for this resultcould be the presence of a size threshold above which size effectsare less pronounced. The determination of such a threshold, how-ever, is not a simple task because of the contrasting inuences thatthis variable can have on different guilds. Scheffer and van Geest(2006), for example, suggested that small habitat size couldpromotespecies richness because low sh biomass and high abundance ofsubmerged vegetation provide a better environment for small birds(e.g. dabbling ducks, vegetation gleaners, and small wading birds).These conditions, conversely, are unsuitable for larger species suchas shing birds and large wading birds. In addition, lagoon size cansometimes hide the effects of other variables (e.g. water depth,salinity, pH, etc.) via the pure area effect (Guadagnin andMaltchik, 2007;Marini et al., 2010) thus resulting relevant to habitatuse. With the Generalized LinearMixedModels approach, however,we could extract the effects of each single variable and we couldtherefore show that size plays, in fact, a negligible role in these tropi-cal coastal lagoons.

Water pH had a negligible effect on the various guilds. Thisresult appears consistent with previous literature (Halse et al.,1993; Takekawa et al., 2006), although to date only a very fewstudies have assessed the role of pH on aquatic organisms(Holland et al., 2012).

Counterworking the natural mechanisms operating on tropicalcoastal lagoons, human-induced environmental changes canexacerbate the problems affecting such ecosystems. In a future cli-mate scenario, the southeastern Brazilian coast will experienceincreased frequency of extreme precipitation events and increasedtemperatures (Alexander et al., 2006; Marengo et al., 2009). Sincerainfall regimes are the main driver of water depth in these lagoons(Caliman et al., 2010), an increase in mean water level is to beexpected under these climatic conditions. Our results show quiteclearly that such a situation will most likely disrupt suitable habi-tats for large and small wading birds, diving birds, and dabblingducks. Habitat displacement has very negative consequences forwaterbirds because they exhibit high site delity and spatial mem-ory (Buchanan et al., 2012; Rogers et al., 2006). This consequencecan be more severe for migrating waders, because they are subjectto extreme energetic constraints during long distance movements(Newton, 2007). Articial sandbar opening and drainage canalsinduces radical changes in the hydrochemical characteristics oftropical coastal lagoons, including shifts from low to high salinityregimes (Santangelo et al., 2007; Suzuki et al., 2002). These salinityshifts will then have an impact on the abundances of diving birdsand large wading birds, as we showed.

Our ndings have a number of important implications for themanagement of tropical coastal lagoons. Management should focuson maintaining water depths within the range 020 cm, whichappears to be critical for supporting high waterbird abundances.Farmers and local public authorities could regulate water levelsby means of articial canals and barriers. This operation could beeasily implemented and at limited costs if combined with the cur-rent practice of creating and managing drinking water pools forlivestock. Appropriate water levels could also help to ensure a suit-able salinity range of 2040 ppt. A control on vegetation height and

D.C. Tavares et al. / Biologicallivestock grazing should be also considered in order to provide amosaic of vegetation heights (Colwell and Dodd, 1995), whichcould sustain high waterbird diversity.lowship of the von Humboldt foundation and CAPES with grantnumber BEX 0128/14-7.

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.biocon.2015.02.027.

References

Alexander, B.W., Hepp, G.R., 2014. Estimating effects of habitat characteristics onabundances of three species of secretive marsh birds in central orida.Waterbirds 37, 274285.

Alexander, L.V., Zhang, X., Peterson, T.C., Caesar, J., Gleason, B., Klein Tank, A.M.G.,Haylock, M., Collins, D., Trewin, B., Rahimzadeh, F., Tagipour, A., Rupa Kumar, K.,Revadekar, J., Grifths, G., Vincent, L., Stephenson, D.B., Burn, J., Aguilar, E.,Brunet, M., Taylor, M., New, M., Zhai, P., Rusticucci, M., Vazquez-Aguirre, J.L.,2006. Global observed changes in daily climate extremes of temperature andprecipitation. J. Geophys. Res.: Atmosph. 111, 122.

Alfaro, M., Azpiroz, A., Rabau, T., Abreu, M., 2008. Distribution, relative abundance,and habitat use of four species of neotropical shorebirds in Uruguay. Ornitolog.Neotropic. 19, 461472.ple by including measures of breeding success, survival rates, bodysize, and energy intake (Johnson, 2007; Ma et al., 2010).

Acknowledgments

We gratefully acknowledge the local authorities of Quissamand the Instituto Brasileiro do Meio Ambiente e dos RecursosNaturais Renovveis (IBAMA, Brazilian Institute for theEnvironment and Renewable Natural Resources) for the logisticsupport provided; Dr. M. Chame and Dr. H. Rajo for valuable sug-gestions on the study design; Profs. L.R. Monteiro, M. Suzuki, C.Ruiz, G. Kristosch and A. Petry for helpful comments on the study;Dr. Acevedo-Trejos for assistance with statistical analysis; Dr. M.Rubega for pointing out some relevant literature; D. Awabdi, E.Zart, L. Valadares, M. Almeida, M. Gonalves and M. Sucunza fortheir enduring help in eldworks; P. Fbregas for hosting DaviCastro Tavares during the last eld surveys. Finally, we thanktwo anonymous reviewers for their constructive and insightfulcomments. This study is part of the wetland birds monitoring pro-gram in the northern coast of the Rio de Janeiro State. SalvatoreSiciliano is supported by Conselho Nacional de DesenvolvimentoCientco e Tecnolgico-CNPq. Davi Castro Tavares is supportedby Coordenao de Aperfeioamento de Pessoal de Nvelfrom human settlement and livestock grazing pressure are impor-tant for structuring the habitat of these organisms. Although theabundance of waterbird is a well-recognized indication of habitatuse, this ecological parameter does not necessarily reect therequirements of different species. Future studies may extend ourIn summary, we found compelling evidence that various guildsshow specic responses to different habitat variables, with waterdepth being a major factor affecting the waterbird assemblage oftropical coastal lagoons of Southeastern Brazil. Quite unexpectedlylagoon size had no impact on waterbirds. This result could be dueto a size threshold effect or to the fact that with our statisticalapproach we were able to extract the contribution of each singlevariable (water depth, salinity, pH, etc.) thus eluding pure areaeffects that could have overweigh the impact of other variables.Classical studies on waterbird habitat structure (Holm andClausen, 2006; Ma et al., 2010; Taft et al., 2002) have often disre-

servation 186 (2015) 1221 19Anthony, A., Atwood, J., August, P.V., Byron, C., Cobb, S., Foster, C., Fry, C., Gold, A.,Hagos, K., Heffner, L., Kellogg, D.Q., Lellis-Dibble, K., Opaluch, J.J., Oviatt, C.A.,Pfeiffer-Herbert, A., Rohr, N., Smith, L., Smuthe, T., Swift, J., Vinhateior, N., 2009.

ConCoastal lagoons and climate change: ecological and social ramications in theU.S. Atlantic and Gulf Coast Ecosystems. Ecol. Soc. 14, 18.

Barnes, R.S.K., 1980. Coastal lagoons. Cambridge University Press, Cambridge, UK.Baschuk, M.S., Koper, N., Wrubleski, D.A., Goldsborough, G., 2012. Effects of water

depth, cover and food resources on habitat use of marsh birds and waterfowl inboreal wetlands of Manitoba, Canada. Waterbirds 35, 4455.

Bibby, C., Martin, J., Marsden, S., 2000. Expedition eld techniques: bird surveys.BirdLife International, Cambridge.

Blondel, J., 2003. Guilds or functional groups: does it matter? Oikos 100, 223231.Bolker, B.M., Brooks, M.E., Clark, C.J., Geange, S.W., Poulsen, J.R., Stevens, M.H.,

White, J.S., 2009. Generalized linear mixed models: a practical guide for ecologyand evolution. Trends Ecol. Evol. 24, 127135.

Borneman, T.E., Rose, E.T., Simons, T.R., 2014. Minimal changes in heart rate ofincubating American Oystercatchers (Haematopus palliatus) in response tohuman activity. The Condor 116, 493503.

Brindock, K.M., Colwell, M.A., 2011. Habitat selection by the western snowy ploversduring the nonbreeding season. J. Wildl. Manag. 75, 786793.

Buchanan, J.B., Lyons, J.E., Salzer, L.J., Carmona, R., Arce, N., Wiles, G.J., Brady, K.,Hayes, G.E., Desimone, S.M., Schirato, G., Michaelis, W., 2012. Among-year sitedelity of red knots during migration in Washington. J. Field Ornithol. 83, 282289.

Bulleri, F., Chapman, M.G., 2010. The introduction of coastal infrastructure as adriver of change in marine environments. J. Appl. Ecol. 47, 2635.

Burger, J., Jeitner, C., Niles, L.J., 2004. Effect of human activities on shorebirds:successful adaptative management. Environ. Conserv. 31, 283288.

Burnham, K.P., Anderson, D.R., 2002. Model Selection and Multimodel Inference.Springer, New York.

Butler, S.J., Gillings, S., 2004. Quantifying the effects of habitat structure on preydetectability and accessibility to farmland birds. Ibis 146, 123130.

Caliman, A., Carneiro, L.S., Santangelo, J.M., Guariento, R.D., Pires, A.P.F., Suhett, A.L.,Quesado, L.B., Scoeld, V., Fonte, E.S., Lopes, P.M., Sanches, L.F., Azevedo, F.D.,Marinho, C.C., Bozelli, R.L., Esteves, F.A., Farjalla, V.F., 2010. Temporal coherenceamong tropical coastal lagoons: a search for patterns and mechanisms. Brazil. J.Biol. 70, 803814.

Caris, E.A.P., Kurtz, B.C., Cruz, C.B.M., Scarano, F.R., 2013. Vegetation cover and landuse of protected coastal area and its surroundings, southeast Brazil. Rodrigusia64, 747755.

Cestari, C., 2008. O uso de praias arenosas com diferentes concentraes humanaspor espcies de aves limcolas (Charadriidae e Scolopacidae) nerticas nosudeste do Brasil. Biota Neotropica 8, 8388.

Clausen, K.K., Clausen, P., 2014. Forecasting future drowning of coastal waterbirdhabitats reveals a major conservation concern. Biol. Conserv. 171, 177185.

Colwell, M.A., Dodd, S.L., 1995. Waterbird communities and habitat relationships incoastal pastures of Northern California. Conserv. Biol. 9, 827834.

Conway, C.J., Gibs, J.P., 2005. Effectiveness of call-broadcast surveys for monitoringmarsh birds. Auk 122, 2635.

Crawley, M.J., 2007. The R Book. John Wiley & Sons, UK.Cumming, G.S., Paxton, M., King, J., Beuster, H., 2012. Foraging guild membership

explains variation in waterbird responses to the hydrological regime of an arid-region ood-pulse river in Namibia. Freshw. Biol. 57, 12021213.

Davies, K.F., Melbourne, B.A., James, C.D., Cunningham, R.B., 2010. Using traits ofspecies to understand responses to land use change: birds and livestock grazingin the Australian Arid zone. Biol. Conserv. 143, 7885.

Del Hoyo, J., Elliot, A., Sargatal, J., 1992. Handbook of the birds of the world. vol. 1.Ostrich to Ducks. Lynx Edicions, Barcelona, Spain.

Di Dario, F., Petry, A.C., Pereira, M.M.S., Mincarone, M.M., Agostinho, L.S., Camara,E.M., Caramaschi, E.P., Britto, M.R., 2013. An update on the sh composition(Teleostei) of the coastal lagoons of the Restinga de Jurubatiba National Parkand the Imboassica Lagoon, northern Rio de Janeiro State. Acta Limnol. Bras. 25,257278.

Douglas, D.J.T., Pearce-Higgins, J.W., 2014. Relative importance of prey abundanceand habitat structure as drivers of shorebird breeding success and abundance.Anim. Conserv. 17, 535543.

Elmberg, J., Nummi, P., Pysa, H., Sjberg, K., 1994. Relationships between speciesnumber, lake size and resourse diversity in assemblages of breeding waterfowl.J. Biogeogr. 21, 7584.

Esteves, B.E., Suzuki, M.S., 2008. Efeito da salinidade sobre as plantas. OecologiaAustralis 12, 662679.

Esteves, F.A., Caliman, A., Santangelo, J.M., Guariento, R.D., Farjalla, V.F., Bozelli, R.L.,2008. Neotropical coastal lagoons: an appraisal of their biodiversity,functioning, threats and conservation management. Brazil. J. Biol. 68, 967981.

Euliss, N.H., Jarvis, R.L., Gilmer, D.S., 1989. Carbonate deposition on tail feathers ofRuddy Ducks using evaporation ponds. Condor 99, 803806.

Forman, D.W., Brain, P.F., 2004. Reproductive strategies used by moorhens (Gallinulachloropus) colonizing an articial wetland habitat in south Wales. J. Nat. History38, 389401.

Gianuca, A.T., Bastazini, V.A.G., Dias, R.A., Hernndez, M.I.M., 2013. Independent andshared effects of environmental features and space driving avian communitybeta diversity across a coastal gradient in southern Brazil. Austral Ecol. 38, 864873.

Gianuca, A.T., Dias, R.A., Debastiani, V.J., Duarte, L.D.S., 2014. Habitat lteringinuences the phylogenetic structure of avian communities across a coastalgradient in southern Brazil. Austral Ecol. 39, 2938.

20 D.C. Tavares et al. / BiologicalGilman, E., Chaloupla, M., Wiedoff, B., Willson, J., 2014. Mitigating seabird bycatchduring Hauling by pelagic longline vessels. PLoS ONE 9, e84499.Glass, K.A., Watts, B.D., 2009. Osprey diet composition and quality in high- and low-salinity areas of lower Chesapeake Bay. J. Raptor Res. 43, 2736.

Glover, H.K., Weston, M.A., Maguire, G.S., Miller, K.K., Christie, B.A., 2011. Towardsecologically meaningful and socially acceptable buffers: response distances ofshorebirds in Victoria, Australia, to human disturbance. Landsc. Urban Plann.103, 326334.

Gregory, R.D., Gibbons, D.W., Donald, P.F., 2004. Bird census and survey techniques.In: Sutherland, W.J., Newton, I., Green, R.E. (Eds.), Bird ecology andconservation. Oxford University Press, Oxford, pp. 1755.

Grueber, C.E., Nakagawa, S., Laws, R.J., Jamison, I.G., 2011. Multimodel inference inecology and evolution: challenges and solutions. J. Evol. Biol. 24, 699711.

Guadagnin, D.L., Maltchik, L., 2007. Habitat and landscape factors associated withneotropical waterbird occurrence and richness in wetland fragments. Biodivers.Conserv. 16, 12311244.

Guadagnin, D.L., Maltchik, L., Fonseca, C.R., 2009. Species-area relationship ofneotropical waterbird assemblages in remnant wetlands: looking at themechanisms. Divers. Distrib. 15, 319327.

Gutirrez, J.S., Dietz, M.W., Masero, J.A., Gill Jr, R.E., Dekinga, A., Battley, P., Snchez-Guzmn, J.M., Piersma, T., 2012. Functional ecology of saltglands in shorebirds:exible responses to variable environmental conditions. Funct. Ecol. 26, 236244.

Gyimesi, A., Franken, M.S., Feige, N., Nolet, B.A., 2012. Human disturbance ofBewicks Swans is reected in giving-up net energy intake rate, but not ingiving-up food density. Ibis 154, 781790.

Halse, S.A., Williams, M.R., Jaensch, R.P., Lane, J.A.K., 1993. Wetland characteristicsand waterbird use of wetlands in south-western Australia. Wildl. Res. 20, 103125.

Hannam, K.M., Oring, L.W., Herzog, M.P., 2003. Impacts of salinity on growth andbehavior of American Avocet chicks. Waterbirds 26, 119125.

Holland, A., Duivenvoorden, L.J., Kinnear, S.H.W., 2012. Naturally acidic waterways:conceptual food webs for better management and understanding of ecologicalfunctioning. Aqua. Conserv.: Mar. Freshw. Ecosyst. 22, 836847.

Holm, T.E., Clausen, P., 2006. Effects of water level management on autumn stagingwaterbird and macrophyte diversity in three Danish coastal lagoons. Biodivers.Conserv. 15, 43994423.

IPCC, 2014. Climate Change 2014: mitigation of climate change. Contribution ofworking group III to the fth assessment report of the intergovernamentalpanel on climate change. In: Edenhofer, O., Pichs-Madruga, R., Sokona, Y.,Farahani, E., Kadner, S., Seyboth, K., Adler, A., Baum, I., Brunner, S., Eickemeier,P., Kriemann, B., Savolainen, J., Schlmer, S., Stechow, C.V., Zwickel, T., Minx, J.C.(Eds.). Cambridge University Press, Cambridge, United Kingdom and New York,USA.

Isacch, J.P., Cardoni, D.A., 2011. Different grazing strategies are necessary toconserve endangered grassland birds in short and tall salty grasslands of theooding pampas. The Condor 113, 724734.

Jedlikowski, J., Brambilla, M., Suska-Malawska, M., 2014. Fine-scale selection ofnesting habitat in Little Crake Porzana parva and Water Rail Rallus aquaticus insmall ponds. Bird Stud. 61, 171181.

Johnson, M.D., 2007. Measuring habitat quality: a review. The Condor 109, 489504.Kirby, J.S., Clee, C., Seager, V., 1993. Impact and extent of recreational disturbance to

roosts on the Dee estuary: some preliminary results. Wader Study Group Bullet.68, 5358.

Kjerfve, B., 1994. Coastal lagoon processes. In: Kjerfve, B. (Ed.), Coastal LagoonProcesses. Elsevier, New York, pp. 18.

Lantz, S.M., Gawlik, D.E., Cook, M.I., 2010. The effects of water depth and submergedaquatic vegetation on foraging habitat selection and foraging success of wadingbirds. Condor 112, 460469.

Lantz, S.M., Gawlik, D.E., Cook, M.I., 2011. The effects of water depth and emergentvegetation on foraging success and habitat selection of wading birds in theEverglades. Waterbirds 34, 439447.

Longoni, V., Rubolini, D., Ambrosini, R., Bogliani, G., 2011. Habitat preferences ofEurasian Bitterns Botaurus stellaris booming in riceelds: implications formanagement. Ibis 153, 695706.

Lunardi, V.O., Macedo, R.H., Granadeiro, J.P., Palmeirim, J.M., 2012. Migratory owsand foraging habitat selection by shorebirds along the northeastern coast ofBrazil: the case of Baa de Todos os Santos. Estuar. Coast. Shelf Sci. 96, 179187.

Ma, Z., Cai, Y., Li, B., Chen, J., 2010. Managing wetland habitats for waterbirds: aninternational perspective. Wetlands 30, 1527.

Macedo-Soares, P.H.M., Petry, A.C., Farjalla, V.F., Camaraschi, E.P., 2010.Hydrological connectivity in a coastal inland systems: lessons from aNeotropical sh metacommunity. Ecol. Freshw. 19, 718.

Marateo, G., Grilli, P., Bouzas, N., Jensen, R., Ferretti, V., Jurez, M., Soave, G., 2013.Habitat use by birds in landlls of northeastern Buenos Aires province,Argentina. Ecolog. Austral 23, 202208.

Marengo, J.A., Jones, R., Alves, L.M., Valverde, M.C., 2009. Future change oftemperature and precipitation extremes in South America as derived fromPRECIS regional climate modeling system. Int. J. Climatol. 9, 22412255.

Marini, L., Bommarco, R., Fontana, P., Battisti, A., 2010. Disentangling effects ofhabitat diversity and area on the orthopteran species with contrasting mobility.Biol. Conserv. 143, 21642171.

McKinney, R.A., Raposa, K.B., Kutcher, T.E., 2010. Use of urban marine habitats byforaging wading birds. Urban Ecosyst. 13, 191208.

Metcalfe, N.B., 1984. The effects of habitat on the vigilance of shorebirds: is visibility

servation 186 (2015) 1221important? Anim. Behav. 32, 981985.Millar, R.B., Anderson, M.J., 2004. Remedies for pseudoreplication. Fish. Res. 70,

397407.

Monastersky, R., 2014. Life a status report. Nature 516, 159161.Nakagawa, S., Frackleton, R.P., 2010. Model averaging, missing data and multiple

imputation: a case study for behavioural ecology. Behav. Ecol. Sociobiol. 65,103116.

Navedo, J.G., Sauma-Castillho, L., Fernndez, G., 2012. Foraging activity and capturerate of large Nearctic shorebirds wintering at a tropical coastal lagoon.Waterbirds 35, 301311.

Newton, I., 2007. The Migration Ecology of Birds. Academic Press/Elsevier, London.Ntiamoa-Baidu, Y., Piersma, T., Wiersma, P., Poot, M., Battley, P., Gordon, C., 1998.

Water depth selection, daily feeding routines and diets of waterbirds in coastallagoons in Ghana. Ibis 140, 89103.

Nummi, P., Paasivaara, A., Suhonen, S., Poysa, H., 2013. Wetland use by brood-rearing female ducks in a boreal forest landscape: the importance of food andhabitat. Ibis 155, 6879.

Ortega-Cisneros, K., Scharler, U.M., Whiteld, A.K., 2014. Inlet mouth phaseinuences density, variability and standing stocks of plankton assemblages intemporarily open/closed estuaries. Estuar. Coast. Shelf Sci. 136, 139148.

Otieno, N.E., Oyieke, H., Ogoma, M., Kochey, J., 2011. Bird density and distributionpatterns in relation to anthropogenic habitat modication around an eastafrican estuary. West. Ind. Ocean J. Mar. Sci. 10, 191200.

Pagel, J., Martnez-Abran, A., Gmez, J.A., Jimnez, J., Oro, D., 2014. A long-termmacroecological analysis of the recovery of a waterbird metacommunity after

Santangelo, J.M., Rocha, A.M., Bozelli, R.L., Carneiro, L.S., Esteves, F.A., 2007.Zooplankton responses to sandbar opening in a tropical eutrophic coastallagoon. Estuar. Coast. Shelf Sci. 71, 657668.

Santos, A.M., Amado, A.M., Minello, M., Farjalla, V.F., Esteves, F.A., 2006. Effects ofthe sand bar breaching on Typha domingensis (Pers.) in a tropical coastal lagoon.Hydrobiologia 556, 6168.

Scheffer, M., van Geest, G.J., 2006. Small habitat size and isolation can promotespecies richness: second-order effects on biodiversity in shallow lakes andponds. Oikos 112, 227231.

Sebastin-Gonzlez, E., Green, A.J., 2014. Habitat use by waterbirds in relation topond size, water depth and isolation: lessons from a restoration in SouthernSpain. Restor. Ecol. 22, 311318.

Shirley, A.M., Goldizen, A.W., Jones, D.N., Krebs, E.A., Putland, D.A., 2003. Effects ofhabitat characteristics and climate on the distribution and colouration of DuskyMoorhens (Gallinula tenebrosa) in south-east Queensland. Emu AustralOrnithol. 103, 8186.

Suzuki, M.S., Figueiredo, R.O., Castro, S.C., Silva, C.F., Pereira, E.A., Silva, J.A., Aragon,G.T., 2002. Sand bar opening in a coastal lagoon (Iquipari) in the northern regionof Rio de Janeiro state, hydrological and hydrochemical changes. Brazil. J. Biol.62, 5162.

Taft, O.W., Colwell, M.A., Isola, C.R., Safran, R.J., 2002. Waterbird responses toexperimental drawdown: implications for the multispecies management ofwetland mosaics. J. Appl. Ecol. 39, 9871001.

D.C. Tavares et al. / Biological Conservation 186 (2015) 1221 21site protection. PLoS ONE 9, e105202.Paracuellos, M., Tellera, J.L., 2004. Factors affecting the distribution of a waterbird

community: the role of habitat conguration and bird abundance. Waterbirds27, 446453.

Prez-Crespo, M., Fonseca, J., Pineda-Lpez, R., Palacios, E., Lara, C., 2013. Foragingguild structure and niche characteristics of waterbirds in an epicontinental lakein Mexico. Zool. Stud. 52, 117.

Prez-Garca, J.M., Sebastin-Gonzlez, E., Alexander, K., Snchez-Zapata, J., Botella,F., 2014. Effect of landscape conguration and habitat quality on thecommunity structure of waterbirds using a man-made habitat. Eur. J. Wildl.Res., 19

Peters, K.A., Otis, D.L., 2007. Shorebird roost-site selection at two temporal scales: ishuman disturbance a factor? J. Appl. Ecol. 44, 196209.

Pomeroy, A.C., 2006. Tradeoffs between food abundance and predation danger inspatial usage of a stopover site by western sandpipers, Calidris mauri. Oikos 112,629637.

Prado, P., Vergara, C., Caiola, N., Ibez, C., 2014. Inuence of salinity regime on thefood-web structure and feeding ecology of sh species from mediterraneancoastal lagoons. Estuar. Coast. Shelf Sci. 139, 110.

R Core Team, 2013. A Language and Environment for Statistical Computing. RFoundation for Statistical Computing, Vienna.

Richmond, O.M.W., Tecklin, J., Beissinger, S.R., 2012. Impact of cattle grazing on theoccupancy of a cryptic threatened rail. Ecol. Appl. 22, 16551664.

Riffel, S.K., Keas, B.E., Burton, T.M., 2001. Area and habitat relationships of birds ingreat lakes coastal wet meadows. Wetlands 21, 492507.

Rodrigues, A.A.F., 2000. Seasonal abundance of neartic shorebirds in the Gulf ofMaranho, Brazil. J. Field Ornithol. 71, 665675.

Rogers, D.I., Battley, P.F., Piersma, T., Van Gils, J.A., Rogers, K.G., 2006. High-tidehabitat choice: insights from modelling roost selection by shorebirds around atropical bay. Anim. Behav. 72, 563575.

Rosselli, L., Stiles, F.G., 2012. Wetland habitats of the Sabana de Bogot andeanhighland plateau and their birds. Aquat. Conserv.: Mar. Freshw. Ecosyst. 22,303317.

Rubega, M.A., Oring, L.W., 2004. Excretory organ growth and implications for salttolerance in hatchling American avocets Recurvirostra americana. J. Avian Biol.35, 1315.

Rubega, M.A., Robinson, J.A., 1996. Water salinization and shorebirds emergingissues. Int. Wader Stud. 9, 4554.Takekawa, J.Y., Miles, A.K., Schoellhamer, D.H., Athearn, N.D., Saiki, M.K., Duffy,W.D., Kleinschmidt, S., Shellenbarger, G., Jannusch, C.A., 2006. Trophic structureand avian communities across a salinity gradient in evaporation ponds of theSan Francisco Bay estuary. Hydrobiologia 567, 307327.

Tavares, D.C., Siciliano, S., 2013a. An inventory of wetland non-passerine birds alonga southeastern Brazilian coastal area. J. Threatened Taxa 5, 45864597.

Tavares, D.C., Siciliano, S., 2013b. Notes on records of Ciconia maguari (Gmelin,1789) (Aves, Ciconiidae) on northern Rio de Janeiro State, Southeast Brazil. Pan-Am. J. Aquat. Sci. 8, 352357.

Tavares, D.C., Siciliano, S., 2014a. The bird community in a threatened coastallagoon in southeastern Brazil. Open J. Ecol. 4, 98112.

Tavares, D.C., Siciliano, S., 2014b. Temporal variation in the abundance of waterbirdspecies in a coastal lagoon in the northern Rio de Janeiro state, Brazil. Biotemas27, 121132.

Tavares, D.C., Moura, J.F., Amorim, C.E., Siciliano, S., 2013. Distribution andabundance of the Grey-headed Gull Chroicocephalus cirrocephalus (Vieillot,1818) in Rio de Janeiro, Southeastern Brazilian coast. Brazil. J. Ornithol. 21, 168175.

Warnock, N., Page, G.W., Ruhlen, T.D., Nur, N., Takekawa, J.Y., Hanson, J.T., 2002.Management and Conservation of San Francisco Bay Salt Ponds: effects of pondsalinity, area, tide, and season on Pacic Flyway Waterbirds. Waterbirds 25(Special Publication 2), pp. 7992.

Whiteld, A.K., Adams, J.B., Bate, G.C., Bezuidenhout, K., Bornman, T.G., Cowley, P.D.,Froneman, P.W., Gama, P.T., James, N.C., Mackenzie, B., Riddin, T., Snow, G.C.,Strydom, N.A., Taljaard, S., Terorde, A.I., Theron, A.K., Turpie, J.K., Niekerk, L.V.,Vorwerk, P.D., Wooldridge, T.H., 2008. A multidisciplinary study of a small,temporarily open/closed South African estuary, with particular emphasis on theinuence of mouth state on the ecology of the system. Afr. J. Mar. Sci. 30, 453473.

Wood, S.N., 2006. Generalized Additive Models: An Introduction with R. Chapmanand Hall/CRC.

Zuur, A.F., Ieno, E.N., Walker, N.J., Saveliev, A.A., Smith, G.M., 2009. Mixed EffectsModels and Extensions in Ecology with R. Springer, New York.

Zuur, A.F., Ieno, E.N., Elphick, C.S., 2010. A protocol for data exploration to avoidcommon statistical problems. Methods Ecol. Evol. 1, 314.

Environmental and anthropogenic factors structuring waterbird habitats of tropical coastal lagoons: Implications for management1 Introduction2 Materials and methods2.1 Study site2.2 Waterbird survey2.3 Waterbirds guilds2.4 Habitat variables2.5 Statistical analyses

3 Results4 DiscussionAcknowledgmentsAppendix A Supplementary materialReferences