Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Waterbirds in a floodplain: influence of spatial and environmental factors through time

Waterbirds in a floodplain: influence of spatial and environmental factors through time Revista Brasileira de Ornitologia, 24(4), 314–322 ARTICLE December 2016 Waterbirds in a floodplain: influence of spatial and environmental factors through time 1,4 1 2 3 Bia de Arruda Almeida , Camila Barbosa Silva , Márcio Rodrigo Gimenes and Luiz dos Anjos Núcleo de Pesquisas em Limnologia, Ictiologia e Aquicultura (Nupelia), Programa de Pós Graduação em Ecologia de Ambientes Aquáticos Continentais, Universidade Estadual de Maringá (UEM), Av. Colombo, 5790, Maringá, PR, CEP 87020-900, Brazil. Universidade Estadual de Mato Grosso do Sul (UEMS), Unidade Universitária de Ivinhema, Av. Brasil, 679, CEP 79740-000, Ivinhema, MS, Brazil. Universidade Estadual de Londrina, Centro de Ciências Biológicas, Departamento de Biologia Animal e Vegetal, Universidade Estadual de Londrina (UEL), Rodovia Celso Garcia Cid, km 380, s/n, CEP 86057-970, Londrina, PR, Brazil. Corresponding author: bialmeida182@hotmail.com Received on 29 May 2015. Accepted on 29 December 2016. ABSTRACT: Wetlands are rapidly being lost and fragmented around the world, making it imperative to seek an understanding of the drivers of their diversity. Among aquatic assemblages, birds constitute a conspicuous group that provides many ecosystem services. Here, we use a metacommunity approach to understand the influence of spatial ( distance among patches) and environmental factors (local characteristics) in wading bird (Pelecaniformes and Ciconiiformes) assemblages in a river-floodplain system t hrough time. We tested the hypothesis that, due to the small scale of our study, spatial factors have no determinant role in waterbird assemblages, and, due to the annual occurrence of flood pulses, the importance of environmental factors varies through time, accor ding to the hydrological cycle. We tested this hypothesis using Partial Redundancy Analysis (pRDA). We used abundance data for the birds in 20 lagoons, sampled quarterly during two years. The spatial factors did not explain variation in community structure in any sampled month, whereas environmental factors explained variation in the assemblages only in two months. Due to high waterbird mobility, the non-significance of the spatial factor is expected a mong lagoons in the same floodplain. Environmental factors are important in determining the community structure only in two sampled months, evidencing that their importance varies through time in the floodplain, partially agreeing with our hypothesis. The non-conformity between the influence of environmental factors on assemblages and the hydrological cycle may be due to human impacts caused by the operation of upstream reservoirs, which alter the natural flood events, and caused a long drought period previous to this study. A multiscale approach is fundamental to the understanding on how anthropogenic impacts on wetlands affect waterbir d assemblages. Thus, this study contributes to the understanding of how seasonality, environmental conditions of lakes, and a local spatial scale act in structuring waterbird assemblages. KEY-WORDS: Ciconiiformes, metacommunity, Paraná River, Pelecaniformes, variance partitioning. INTRODUCTION 2005). However, few studies have considered space as a structuring factor for local waterbird communities (Pagel Wetlands are mosaic-like environments, which generally et al. 2014, Sebastián-González & Green 2014). Here, comprise patches of different types, sizes and distances we used metacommunity theory to better understand the from other patches (Whited et al. 2000). Additionally, distribution of waterbirds in their patchy environment. Metacommunity theory is a theoretical and wetlands have suffered continuous loss and fragmentation, which has negatively affected t he ecosystem services they mechanistic framework that serves to explain the provide and biodiversity they harbor, including waterbirds interdependence of local and regional processes in (Ma et al. 2009). Waterbirds (e.g. Pelecaniformes, structuring communities (Logue et al. 2011). As viewed Ciconiiformes, Charadriiformes, Anseriformes) are through the lens of this concept, two general types of forces affect community structure: local ( biotic interactions highly dispersive birds, moving between favorable areas in search of resources and are dependent on wetlands and environmental conditions) and regional (spatial during most or all of their life cycle (Haig et al. 1998). dynamics, linked to the dispersal of organisms). These The inherent patchiness and rapid loss of wetlands processes interact to produce a local species assemblage around the world makes it important to integrate spatial (Leibold et al. 2004). Depending on the degree of influence of local and components into analyses of waterbird diversity and distribution (Paracuellos & Tellería 2004, Luis et al. regional processes in structuring the local communities, Revista Brasileira de Ornitologia, 24(4), 2016 Waterbirds in a floodplain: influence of spatial and environmental factors through time Bia de Arruda Almeida, Camila Barbosa Silva, Márcio Rodrigo Gimenes and Luiz dos Anjos four perspectives were defined concerning the organization (Junk et al. 1989). Flood pulses act by connecting and of metacommunities: neutral model, patch dynamics, homogenizing the patches of aquatic environments in a species sorting and mass effects. The neutral model floodplain and are a major driver of biodiversity in these assumes environmental homogeneity and species with ecosystems (Junk et al. 1989, Thomaz et al. 2007). similar competitive capacities, so that their distribution Pelecaniformes and Ciconiiformes are medium-to- is influenced only by dispersive capacity. Patch dynamics large waders that occur in most floodplains, and many considers patches also to be homogeneous, but species aspects of their ecology are affected by the hydrologic differ among them, in what constitutes a trade-off regime (Kushlan 1986, Dimalexis & Pyrovetsi 1997, between colonization and competition capacities. Russell et al. 2002). The density and vulnerability of Species sorting assumes environmental heterogeneity prey vary seasonally and spatially (Fernandes et al. 2009, associated with differences among species and with a Gimenes & Anjos 2011), and depend on the frequency, dispersive capacity that is sufficient for the species to duration and magnitude of the flooding (Petry et al. reach different patches, but which does not influence 2003). Because wading birds forage in shallow waters, their distribution. Finally, mass effects considers that prey availability is highest when water is shallower during there is a high dispersive capacity in a heterogeneous dry seasons (Gawlik 2002). The seasonal variation in the environment, which brings to communities affected by environment therefore makes it necessary to consider source-sink relations, with a greater influence of dispersal the influence of the hydrological cycle on the local than environmental heterogeneity in the structure of communities of waders. communities (Leibold et al. 2004). However, the complex The goal of this study was to assess the relative dynamics of communities suggests that these models are importance of regional (spatial) and local (environmental) not discrete or mutually exclusive (Logue et al. 2011). factors in structuring the wading bird communities Winegardner et al. (2012) reviewed the terminology used in the upper Paraná River floodplain throughout the by Leibold and proposed that the mechanisms underlying hydrological cycle. Due to the small scale of our study these paradigms interact. Thus, they reorganized t he (a single floodplain spanning c. 50 km ) and to the models according to the relative importance of dispersal, vagility and habitat selection capacity of birds (tending suggesting that metacommunities can be classified as to an ideal free distribution, see Fretwell & Lucas-Jr. follows: neutral, species-sorting with limited dispersal 1970), we expect that dispersal limitation and mass (which would be equivalent to patch dynamics, sensu effects will not play important roles in structuring wader Leibold et al. 2004), species-sorting with efficient communities in this floodplain. At the same time, due to dispersal (equivalent to species sorting), or species-sorting the cyclic alterations of the water level, we expect that the with high dispersal (equivalent to mass effects). importance of environmental variables is not equivalent The metacommunity approach contributes to a throughout the year. Therefore, we hypothesize that better understanding of how spatial dynamics and local the wader metacommunity of the upper Paraná River interactions influence community ecology (Logue et floodplain is shaped by species sorting with efficient al. 2011). Studying metacommunities of waterbirds dispersal (Winegardner et al. 2012), and that the role at different scales is necessary to understand how of environmental characteristics in structuring the local local communities are structured within and among communities of waders changes seasonally. Thus, we wetlands. Considering the importance of waterbirds as predict that the regional component will not impact suppliers of ecological services such as dispersal of seeds metacommunity structure and that the local component and eggs, ecosystem engineering, population control (environmental conditions) will be more important and scavenging, understanding their metacommunity during drier periods, when there is more heterogeneity dynamics is essential for wetland management (Green among the water bodies in the floodplain. & Elmberg 2013). Knowledge on spatial dynamics for waterbird assemblages enables one to answer questions METHODS such as how distances between wetland patches influence assemblages. As an example of how the metacommunity Study area approach can be used to better explore the mechanisms The Paraná River stretches for 4695 km from t he that shape communities of waterbirds, we evaluated the local and regional drivers of assemblages of wading birds Brazilian central plateau southward to the Plate River (Pelecaniformes and Ciconiiformes) in a Neotropical between Argentina and Uruguay. In Brazil, the Paraná floodplain. F loodplains are heterogeneous environments River is affected by dams along most of its length, and with high biological diversity and are influenced by only a stretch of 230 km between the Porto Primavera and Itaipu reservoirs remains as a floodplain (22°40'S to cyclical variation in river discharge, or flood pulses Revista Brasileira de Ornitologia, 24(4), 2016 Waterbirds in a floodplain: influence of spatial and environmental factors through time Bia de Arruda Almeida, Camila Barbosa Silva, Márcio Rodrigo Gimenes and Luiz dos Anjos 22°52'S and 53°12'W to 53°38'W). The dams installed water levels, while smaller pulses (< 0.5 m) occur weekly along the hydrographic basin have modified t he natural in the dry season, caused by the operation of upstream pattern of magnitude, duration, periodicity and discharge reservoirs. frequency of the Paraná River, increasing the control on The study area, located between the mouths of the discharge regime, eliminating the highest water level the Paranapanema and Ivinhema Rivers, encompasses values and causing variability between night and day environments with very different characteristics. At this discharges (Thomaz et al. 2004). Thus, the seasonal water point, the Paraná River has an extensive alluvial plain level fluctuation in the upper Paraná River floodplain is on its west margin. This floodplain consists of a mosaic not continuously sinusoidal, as uninterrupted periods of of habitats, including rivers, streams, channels, and falling and rising water are rarely observed (Agostinho et marginal lagoons with different degrees of hydrological al. 2000). connectivity. Two types of lagoons are used in this The climate of the area is classified as Tropical study: permanently connected lagoons, which have an Subtropical, with an annual average temperature of 22°C above-ground connection with the river all year long, (average of 26°C in summer and 19°C in winter). The and isolated lagoons, which are disconnected from the pluvial regime is marked by a wet summer, with monthly river during most of the year and have an above-ground average precipitation greater than 125 mm, and a dry connection only during flooding events. winter, with averages under 80 mm. The high-water period in the Paraná River usually occurs from November/ Data collection December to April/May and is characterized by an increase in the water level averaging 2.5 m and reaching The birds were sampled in 2002 and 2003 (both years 7.5 m in years of extreme flood events, with a lmost no on February, May, August and November) in connected variation observed in years in which the characteristic (permanently connected to rivers or canals, n = 10) flood period does not occur. It is quite common for two and isolated perennial lagoons (not connected to rivers, or three annual flood pulses to be observed during high except during the flood, n = 10) (Figure 1). The areas FIGURE 1. Map of the study area. The numbers indicate the sampling units. Connected lagoons: 1, 3, 6, 8, 10, 12, 15, 17, 19, 20; isolated lagoons: 2, 4, 5, 7, 9, 11, 13, 14, 16, 18. Revista Brasileira de Ornitologia, 24(4), 2016 Waterbirds in a floodplain: influence of spatial and environmental factors through time Bia de Arruda Almeida, Camila Barbosa Silva, Márcio Rodrigo Gimenes and Luiz dos Anjos of the sampled lagoons varied from 0.43 to 89.8 ha, variables; and the residue, or the variation not explained and their perimeters from 251.4 to 7151.2 m. These by any of the above components (Borcard et al. 1992) months were chosen in order to sample all the phases of (Figure 2). We considered the adjusted R values as the the hydrological cycle. In each month, two visits lasting results of the variance partitioning procedure (Peres-Neto for 1 h were conducted in each sampling unit, always et al. 2006). The significance of each component (P < beginning 1 h after sunrise. In each sampling, a circular 0.05) was tested by Monte Carlo randomization tests transect was performed around the lagoon, through the (9999 runs) (Legendre & Legendre 1998). whole perimeter, either by boat in a constant speed of 30 -1 km.h (connected lagoons) or on foot (isolated lagoons), with identification and registration of each individual sighted (Bibby et al. 1992). The entire lagoon was fully visible along transects. Birds observed within a range of up to 5 m from the water were included, but individuals in flight were not recor ded, unless they were observed leaving or arriving at the lagoons. In all sampled lagoons only the margins of the water bodies are shallow enough to allow the presence of waders. Birds were identified to species according to Sigrist (2009). To characterize the environmental component, environmental variables considered important for the distribution of species of Pelecaniformes and Ciconiiformes and possibly related to the vulnerability of prey capture were selected. The recor ded environmental FIGURE 2. Venn diagram representing the decomposition of the variables were: perimeter (in m), average depth (in m), community structure into: E = environmental variation or local factor, S = spatial variation or regional factor, E|S = variation explained by type of dominant vegetation in the surrounding area environmental variables independent of spatial variables, or pure (forest, grasslands, macrophyte or Polygonum), and type environmental component, S|E = variation explained by spatial of lagoon (connected or isolated). Measurements of water variables independent of environmental variables, or pure spatial bodies were performed through aerial images. All datasets component, ES = variation explained together by environmental and spatial variables; and residue, variation not explained by any of were obtained from the PELD Technical Report, site 6 the previous components. (Souza et al. 2001, Souza-Filho & Stevaux 2002). Data analyses To construct the spatial matrix, a Euclidean distance matrix was generated between the sampled sites based To determine the relative contributions of local on their geographic coordinates, measured from the (environmental characteristics – E) and regional factors center of each lagoon. From the distance matrix, Moran (spatial determinants, or distances between patches (MEM) eigenvector maps were created to quantify the – S) to the community structure of Pelecaniformes spatial arrangement of the sampling points (Dray et al. and Ciconiiformes, we used a variance partitioning 2006). The eigenvectors (spatial variables) allow one to technique: partial redundancy analysis (pRDA) (Borcard represent the spatial relationships among the sampling et al. 1992, Cottenie 2005). This method of analysis can units at different spatial scales. S mall eigenvalues indicate be characterized as a multiple regression with multiple the absence of spatial autocorrelation and, therefore, are response variables. In this case, we used a dependent not suitable for defining spatial structures. We included matrix (species abundance) and two explanatory matrices all eigenvectors with a Moran's I coefficient greater (spatial and environmental variables) (Legendre & than 0.1 in the spatial predictor matrix (Nabout et al. Legendre 1998, Beisner et al. 2006). The variation in 2009). Variance partitioning was carried out using total community structure was decomposed into the following abundance data (number of birds recorded per lagoon), components: E+S - the total variation explained by the which were previously transformed into Hellinger values. analyses; E - environmental variation; S - spatial variation; The Hellinger distance is a measure recommended for E|S - the variation explained by the environmental the clustering or ordination of species abundance data variables, independent of spatial variables, or the pure (Legendre & Gallagher 2001). We performed eight environmental component; S|E - the variation explained pRDAs, one for each sampled month, in both years. All by spatial variables, independent of environmental analyses were conducted in R software (R Development variables, or the pure spatial component; ES - the Core Team 2012) using the packages vegan (Oksanen et explained variation shared by environmental and spatial al. 2013) and PCNM (Legendre et al. 2013). Revista Brasileira de Ornitologia, 24(4), 2016 Waterbirds in a floodplain: influence of spatial and environmental factors through time Bia de Arruda Almeida, Camila Barbosa Silva, Márcio Rodrigo Gimenes and Luiz dos Anjos RESULTS period of 2002. In fact, the water levels reached as high as 3.5 m in November 2003, the level that begins to provide connections between the river and isolated lagoons. The The water l evel of the Paraná River was higher between January and April in both the 2002 and 2003 annual year 2001, which preceded our sampling activities, was monitoring periods. However, it is noticeable that the water markedly dry, showing minimum water level values close levels varied more in the last half of 2003 than in the same to 1 m and maximum values not reaching 4 m (Figure 3). FIGURE 3. Daily fluviometric levels for years 2000–2005. Black arrows represent the sampling months. The line indicates the level at which a connection begins to be established between the environments. A total of 2028 individuals belonging to 13 species ibis, Syrigma sibilatrix, Mycteria americana, Ciconia of the orders Pelecaniformes and Ciconiiformes were maguari, Jabiru mycteria, Mesembrinibis cayennensis and recorded (Table 1). Six species were common (Ardea Theristicus caudatus) were less abundant, occurring in only cocoi, Butorides striata, Egretta thula, Ardea alba, Tigrisoma a few months and sites, mostly in connected lagoons, and lineatum and Nycticorax nycticorax), occurring in almost present in a maximum of six sites each month. The peak all sampling months (only in one month, one of the of bird abundance occurred in November 2002. February species was absent), with presence in a maximum of 22 2002 and February and August 2003, in turn, were the sites each month. Other species (Platalea ajaja, Bubulcus months with the lowest abundances (Figure 4). TABLE 1. Species registered in the study area for all the sampled months and according to sampling sites (lagoons). Lagoons are numbered according to the map (Fig. 1). Classification of the species is accor ding to the American Ornithologists' Union (AOU). Connected Isolated Lagoons No. 16 2813 12 710 11 17 6114 18 20 15 935419 Ciconiiformes Ciconiidae Ciconia maguari xxx Jabiru mycteria x xxx Mycteria americana xxx x Pelecaniformes Ardeidae Ardea alba xxxxxxxxxxxxx xxxxxx Ardea cocoi xxxxxxxxxxxxxxxxxxxx Bubulcus ibis xx Butorides striata xxxxxxxxxxxx xxxx x Egretta thula xxxxxxxxxxxx xxxx Nycticorax nycticorax xx x xxxxx x x Syrigma sibilatrix x Tigrisoma lineatum xxxxxxxxxxxxx xxxxx Threskiornithidae Mesembrinibis cayennensisxx Platalea ajaja x xxx Revista Brasileira de Ornitologia, 24(4), 2016 Waterbirds in a floodplain: influence of spatial and environmental factors through time Bia de Arruda Almeida, Camila Barbosa Silva, Márcio Rodrigo Gimenes and Luiz dos Anjos FIGURE 4. Number of individuals of the species of Ciconiiformes and Pelecaniformes recorded in the perennial lagoons of the upper Paraná River floodplain. The letters F, M, A and N stand for the months February, May, August and Nove mber, respectively. The results of the RDA variance partitioning factor E explained the structure of the communities differed between months. In 2002, the environmental in February (R = 0.14; P = 0.0215) and November factor - E - explained the structure of the bird (R = 0.13; P = 0.0468), and the spatial factor S was 2 2 communities in February (R = 0.15; P = 0.0430), May also important in February (R = 0.13; P = 0.0060). 2 2 (R = 0.14; P = 0.0149) and November (R = 0.28; P The pure environmental factor - E|S - did not explain = 0.0001), whereas the pure environmental factor E|S the variation of community structure observed in any explained the variability of the communities of birds month of 2003, and the pure spatial factor S|E showed only in February (R = 0.22; P = 0.0257) and November no importance for the communities in any month of (R = 0.25; P = 0.0011). In 2003, the environmental this study (Table 2). TABLE 2. Variance partitioning of the community structure of Ciconiiformes and Pelecaniformes in different months during the two years of sampling in the 20 lagoons of the upper Paraná River floodplain. E = environmental variation, S = spatial variation, E|S = pure environmental component, S|E = pure spatial component, ES = shared variation explained by environmental and spatial variables; residue, or unexplained variation; significance level at P < 0.05, A dj. R = adjusted coefficient of determination. Significant values are underlined. FebruaryMay August November 2 2 2 2 Adj. R P Adj. R P Adj. R P Adj. R P E 0.15 0.0430 0.14 0.0149 0.03 0.2951 0.28 <0.0001 S 0.06 0.7305 0.01 0.3256 0.01 0.3322 0.05 0.0958 E|S 0.22 0.0257 0.10 0.0748 0.00 0.7105 0.25 0.0011 S|E 0.02 0.3161 0.00 0.6917 0.00 0.8977 0.02 0.2471 ES 0.00 0.04 0.07 0.04 Residue 0.82 0.89 1.00 0.69 E 0.14 0.0215 0.06 0.1681 0.00 0.4450 0.13 0.0468 S 0.13 0.0060 0.04 0.1824 0.00 0.5351 0.09 0.0521 E|S 0.04 0.2288 0.03 0.3188 0.00 0.6269 0.06 0.1864 S|E 0.04 0.1753 0.01 0.3996 0.00 0.7578 0.02 0.2976 ES 0.10 0.03 0.03 0.07 Residue 0.83 0.93 1.00 0.85 Revista Brasileira de Ornitologia, 24(4), 2016 Waterbirds in a floodplain: influence of spatial and environmental factors through time Bia de Arruda Almeida, Camila Barbosa Silva, Márcio Rodrigo Gimenes and Luiz dos Anjos DISCUSSION As we hypothesized, the pure spatial factor S|E did not play an important role in shaping the local communities Although we have used the metacommunity concept of waders. Wading birds are highly vagile, which allows to analyze local communities of waders in a single them to keep up with the seasonally fluctuating mosaic floodplain, we should e mphasize that this approach of suitable habitat (Fretwell & Lucas-Jr. 1968, Haig et can be a useful tool, especially to determine the al. 1998). For this reason, distances between sites may pattern of waterbird communities at larger scales, not be important in shaping communities at small e.g. among different wetlands. Given that human spatial scales. Meynard & Quinn (2008) suggested that activities often change the spatial structure within and for bird communities studied on a scale of 1000 km or among wetlands, a multiscale approach is essential for less, environmental factors will predominate, whereas understanding how these changes may affect waterbir d dispersion will become more important for the structure communities. Therefore, our approach s hould be of local communities at larger scales. Thus, it is expected viewed as a first step in understanding the importance that space does not play an important role for wader of the metacommunity concept in structuring local communities within patches of a single floodplain (see communities of waterbirds. Figure 5). FIGURE 5. Schematic representation of how Local (environmental - LF) and Regional (spatial – RF) factors may influence waterbir d assemblages within and among wetlands. Larger distances between sites may imply a more important role of the regional factor in structuring communities. The significance of the pure environmental are affected by dams, one of the main downstream component E|S in February and November 2002 impacts of the reservoirs is to change the natural water shows that these waterbird communities are influenced level fluctuations. This change affects the water exchange by local factors and also that the importance of the between the river main channel and the floodplain, and, purely environmental factors varies according to time in extreme years, even the absence of floods is observed. in the floodplain, partly corroborating our hypothesis. This alteration in the hydrological regime directly impacts Although we hypothesized that there would be temporal wading birds whose life cycles depend on water fluctuation variation in the importance of E|S, we also predicted (Agostinho et al. 2004). During the two sampled years that the pure environmental factor would be important of this study, the water regime was not consistent. The in most months, being more important during drought. first year (2002) was typical, with floods occurring from This prediction was not confirmed by data. The pure January to April and lower water levels characterizing environmental component was significant only in two all other months. The second year of sampling (2003) months, February and November 2002, the first of which also had a typical flood period in its first months, but falls within the beginning of the flood period. the expected dry period had a higher water level than the The contrasting results found in 2002 and 2003 same period of 2002, with several flood pulses reaching may be the outcome of human impacts in the Paraná the level at which the river starts to overflow. A dditionally, River floodplain. As the Paraná River and its tributaries 2002 was preceded by a dry year, in which the supposed Revista Brasileira de Ornitologia, 24(4), 2016 Waterbirds in a floodplain: influence of spatial and environmental factors through time Bia de Arruda Almeida, Camila Barbosa Silva, Márcio Rodrigo Gimenes and Luiz dos Anjos flood period was characterized by a water level closer to better information on the influence of the hydrological the levels observed in the last months of 2003 (Figure 3). cycle for these organisms. The variation in the importance of the environmental This study is an example of the use of the factor to the local communities of waders is linked to the metacommunity concept as an important tool to better flood regime of the floodplain. The pure environmental understand the drivers of waterbird diversity in space and component was important in shaping the community time. Knowing the factors driving waterbird distribution in February 2002, the beginning of the flood, and in in space, as well as the variation in the importance of these November 2002, the end of the dry period. This result factors seasonally is essential to understand waterbird might be explained by the differences in water fluctuation community dynamics. Here, we showed that space is among 2001, 2002 and 2003. Wading birds seek optimal not a strong driver of wading bird assemblages at small foraging habitats given their morphological and behavioral scales, considering the area of a floodplain. Furthermore, restrictions. They then select habitat based on local we showed that there is seasonal variation in the conditions. To reflect this, the environmental variables importance of the environmental variables used in this selected for this study are related to the availability of prey study for this group of waterbirds. Indeed, environmental to waders, the majority of which are piscivores (Bancroft variables other than those used in this study could be et al. 2002, Gawlik 2002). Previous studies performed important for these birds in different phases of the in the same floodplain showe d that differences among flood regime. These additional variables, such as water the density of fish in different lagoons increase during transparency and aquatic vegetation, could be the focus the year, becoming more apparent at the end of the dry of future investigations. This approach can be explored period (Fernandes et al. 2009, Gimenes & Anjos 2011). to recognize important factors for the maintenance of Additionally, according to Thomaz et al. (2007), a time waterbird communities in recognition of the continuous lag is expected after the flood period until t he effects of loss, alteration and fragmentation of wetlands around the environmental heterogeneity become apparent. Thus, we world. suggest that there is a tendency for the environmental factor to become more important in shaping communities at the end of the dry period (represented by November in ACKNOWLEDGEMENTS our samples). As we see, 2003 was an atypical year, in which higher water levels caused an interruption in the We thank the Center of Research in Limnology, Ichthyology and Aquaculture of the State University dynamics of the dry period and thus altered the role of the environment in structuring waterbird communities. In of Maringa (Nupelia/UEM) for logistical support the same way, the significance of the pure environmental and PROEX/CAPES for financial support. Sebastião factor in February 2002 may be a consequence of the dry Rodrigues, Leandro Rodrigues and Alfredo da Silva year that preceded it. Specifically, as the flood period was helped with the fieldwork. We appreciate valuable comments and suggestions by Nadson Ressye Simões da still at its beginning in February, it is possible that the effects of the flood pulse were delayed by the long 2001 Silva. Fabrício Oda and Gabriel Rosa helped with figures. drought. The lack of signal from our explanatory variables may be in part due to limitations of our study. The lack REFERENCES of importance of the spatial factor may be associated with the small scale of the floodplain. As wading birds are Agostinho, A. A.; Thomaz, S. M.; M inte-Vera, C. V. & Winemiller, K. O. 2000. Biodiversity in the high Paraná River floodplain, vagile organisms, the size of the floodplain may be too p. 89–118. In: Gopal, B.; Junk, W. J. & Davis, J. A. (eds.). small for distances between sites to influence the structure Biodiversity in wetlands: assessment, function and conservation. of the community. At the same time, environmental Leiden: Backhuys Publishers (Wetlands Biodiversity). characteristics other than those considered in this study Agostinho, A. A.; Thomaz, S. M. & Gomes, L. C. 2004. Threats for are known to influence waterbir d assemblages, such as biodiversity in the floodplain of the upper Paraná River: effects of hydrological regulation by dams. Ecohydrology and Hydrobiolog y, water transparency, productivity, aquatic vegetation and 4: 255–256. marginal vegetation (Bancroft et al. 2002, Guadagnin Bancroft, G. T.; Gawlik, D. E. & Rutchey K. 2002. Distribution & Maltchik 2007, Cintra 2015). The addition of such of wading birds relative to vegetation and water depths in the variables could increase the importance of environmental northern Everglades of Florida, USA. Waterbirds, 25: 265–277. Beisner, B. E.; Peres-Neto, P. R.; Lindström, E. S.; Barnett, A. factors or change its seasonal variation. Furthermore, & Longhi, M. L. 2006. The role of environmental and spatial sampling along only two years could not be enough to processes in structuring lake communities from bacteria to fish. determine a seasonal variation pattern of the importance Ecology, 87: 2985–2991. of environmental or spatial factors for wading bird Bibby, C. J.; Burgess, N. D. & Hill, D. A. 1992. Bird census community structure. More years of study could bring techniques. London: Academic Press. Revista Brasileira de Ornitologia, 24(4), 2016 Waterbirds in a floodplain: influence of spatial and environmental factors through time Bia de Arruda Almeida, Camila Barbosa Silva, Márcio Rodrigo Gimenes and Luiz dos Anjos Borcard, D.; Legendre, P. & Drapeau, P. 1992. Partialling out the Meynard, C. N. & Quinn, J. F. 2008. Bird metacommunities in spatial component of ecological variation. Ecology, 73: 1045– temperate South American forest: vegetation structure, area, and 1055. climate effects. Eco logy, 89: 981–990. Cintra, R. 2015. Spatial distribution and composition of waterbirds Nabout, J. C.; Siqueira, T.; Bini, L. M. & Nogueira, I. S. 2009. No in relation to limnological conditions in the Amazon Basin. evidence for environmental and spatial processes in structuring Hydrobiologia, 747: 235–252. phytoplankton communities. Acta Oecologica, 35: 720–726. Cottenie, K. 2005. Integrating environmental and spatial processes in Oksanen, J.; Blanchet, F. G.; Kindt, R.; Legendre, P.; Minchin, ecological community dynamics. Ecology Letters, 8: 1175–1182. P. R.; O'Hara, R. B.; Simpson, G. L.; Solymos, P.; Stevens, Dimalexis, A. & Pyrovetsi, M. 1997. Effect of water level fluctuations M. H. H. & Wagner, H. 2013. vegan: community ecology on wading bird foraging habitat use at an irrigation reservoir, Lake package. R package version 2.0-8. http://CRAN.R-project.org/ Kerkini, Greece. Colonial Waterbirds, 20: 244–252. package=vegan (access on 28 January 2015). Dray, S.; Legendre, P. & Peres-Neto, P. R. 2006. Spatial modeling: Pagel, J.; Martínez-Abraín, A.; Gómez, J. A.; Jiménez, J. & Oro, a comprehensive framework for principal coordinate analysis of D. 2014. A long-term macroecological analysis of the recovery neighbor matrices (PCNM). Ecological Modelling, 196: 483–493. of a waterbird metacommunity after site protection. PLoS ONE, Fernandes, R.; Gomes, L. C.; Pelicice, F. M. & Agostinho, A. A. 9: e105202. 2009. Temporal organization of fish assemblages in floodplain Paracuellos, M. & Tellería, J. L. 2004. Factors affecting the lagoons: the role of hydrological connectivity. Environmental distribution of a waterbird community: the role of habitat Biology of Fishes, 85: 99–108. configuration and bir d abundance. Waterbirds, 27: 446–453. Fretwell, S. D. & Lucas-Jr., H. L. 1970. On territorial behavior Peres-Neto, P. R.; Legendre, L.; Dray, S. & Borcard, D. 2006. and other factors influencing habitat distribution in birds. I. Variation partitioning of species data matrices: estimation and Theoretical development. Ac ta Biotheoretica, 19: 16–36. comparison of fractions. Ecology, 87: 2614–2625. Gawlik, D. E. 2002. The effects of prey availability on the numerical Petry, A. C.; Agostinho, A. A. & Gomes, L. C. 2003. Fish assemblages response of wading birds. Ecological Monographs, 72: 329–346. of tropical floodplain lagoons: exploring the role of connectivity Gimenes, M. R. & Anjos, L. 2011. Quantitative analysis of foraging in a dry year. Neotropical Ichthyology, 1: 111–119. habitat use by Ciconiiformes in the upper Paraná River floodplain, R Development Core Team. 2012. R: a language and environment Brazil. Brazilian Archives of Biology and Technology, 54: 415–427. for statistical computing. Vienna: R Foundation for Statistical Green, A. J. & Elmberg, J. 2013. Ecosystem services provided by Computing. waterbirds. Biological Reviews, 89: 105–122. Russell, G. J.; Bass, O. L. & Pimm, S. L. 2002. The effect of Guadagnin, D. L. & Maltchik, L. 2007. Habitat and landscape hydrological patterns and breeding-season flooding on the factors associated with Neotropical waterbird occurrence and numbers and distribution of wading birds in Everglades National richness in wetland fragments. Biodiversity and Conservation, 16: Park. Animal Conservation, 5: 185–199. 1231–1244. Sebastián-González, E. & Green, A. J. 2014. Habitat use by Haig, S. M.; Mehlman, D. W. & Oring, L. W. 1998. Avian waterbirds in relation to pond size, water depth, and isolation: movements and wetland connectivity in landscape conservation. lessons from a restoration in southern Spain. Restoration Ecology, Conservation Biology, 12: 749–758. 22: 311–318. nd Junk, W. J.; Bayley, P. B. & Sparks, R. E. 1989. The flood pulse Sigrist, T. 2009. Guia de campo Avis Brasilis - avifauna brasileira, 2 concept in river-floodplain systems. Canadian Special Publication edn. São Paulo: Avis Brasilis. of Fisheries and Aquatic Sciences, 106: 110–127. Souza, M. C.; Kita, K. K.; Romagnolo, M. B.; Soares, J. J.; Kushlan, J. A. 1986. Responses of wading birds to seasonally Tomazini, V. & Albuquerque, E. C. 2001. Mata ciliar, p. 231– fluctuating water levels: strategies and t heir limits. Colonial 240. In: Agostinho, A. A.; Nakatani, K.; Antônio, R. R.; Zavadski, Waterbirds, 9: 155–162. F. C.; Thomaz, S. M.; Júlio-Jr., H. F.; Luz, K. D. G. & Pereira, J. Legendre, P.; Borcard, D.; Blanchet, F. G. & Dray, S. 2013. PCNM: L. L. (eds.). A planície de inundação do alto Rio Paraná. Maringá: MEM spatial eigenfunction and principal coordinate analyses. PELD/CNPq Technical Report. R package version 2.1-2/r109. http://R-Forge.R-project.org/ Souza-Filho, E. E. & Stevaux, J. C. 2002. Locais de amostragem, p. projects/sedar/ (access on 28 January 2015). 7–13. In: Agostinho, A. A.; Thomaz, S. M.; Nakatani, K.; Júlio-Jr., Legendre, P. & Gallagher, E. D. 2001. Ecologically meaningful H. F.; Antônio, R. R.; Luz, K. D. G.; Zavadski, F. C. & Pereira, transformations for ordination of species data. Oecologia, 129: J. L. L. (eds.). A planície alagável do alto rio Paraná: estrutura e 271–280. processos ambientais. Maringá: PELD/CNPq Technical Report. Legendre, P. & Legendre, L. 1998. Numerical ecolog y. Amsterdam: Thomaz, S. M.; A gostinho, A. A. & Hahn, N. S. 2004. The upper Elsevier. Paraná River and its floodplain: physical aspects, ecology and Leibold, M. A.; Holyoak, M.; Mouquet, N.; Amarasekare, P.; conservation. Leiden: Backhuys Publishers. Chase, J. M.; Hoopes, M. F.; Holt, R. D.; Shurin, J. B.; Thomaz, S. M.; B ini, L. M. & Bozelli, R. L. 2007. Floods increase Law, R.; Tilman, D.; Loreau, M. & Gonzalez, A. 2004. The similarity among aquatic habitats in river-floodplain systems. metacommunity concept: a framework for multi-scale community Hydrobiologia, 579: 1–13. ecology. Ecology Letters, 7: 601–613. Whited, D.; Galatowitsch, S.; Tester, J. R.; Schik, K.; Lehtinen, Logue, J. B.; Mouquet, N.; Peter, H. & Hillebrand, H. 2011. R. & Husveth, J. 2000. The importance of local and regional Empirical approaches to metacommunities: a review and factors in predicting effective conservation planning strategies for comparison with theory. Trends in Ecology and Evolution, 26: wetland bird communities in agricultural and urban landscapes. 482–491. Landscape and Urban Planning, 49: 49–65. Guadagnin, D. L.; Peter, Â. S.; Perello, L. F. C. & Maltchik, L. Winegardner, A. K.; Jones, B. K.; Ng, I. S. Y.; Siqueira, T. & 2005. Spatial and temporal patterns of waterbird assemblages in Cottenie, K. 2012. The terminology of metacommunity ecology. fragmented wetlands of southern Brazil. Waterbirds, 28: 261–272. Trends in Ecology and Evolution, 27: 253–254. Ma, Z.; Cai, Y.; Li, B. & Chen, J. 2009. Managing wetland habitats for waterbirds: an international perspective. Wetlands, 30: 15–27. Associate Editor: Caio G. Machado. Revista Brasileira de Ornitologia, 24(4), 2016 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Ornithology Research Springer Journals

Waterbirds in a floodplain: influence of spatial and environmental factors through time

Download PDF
9 pages

Loading next page...
 
/lp/springer-journals/waterbirds-in-a-floodplain-influence-of-spatial-and-environmental-yobxHyDmx0

References (40)

Publisher
Springer Journals
Copyright
Copyright © Sociedade Brasileira de Ornitologia 2016
eISSN
2178-7875
DOI
10.1007/bf03544360
Publisher site
See Article on Publisher Site

Abstract

Revista Brasileira de Ornitologia, 24(4), 314–322 ARTICLE December 2016 Waterbirds in a floodplain: influence of spatial and environmental factors through time 1,4 1 2 3 Bia de Arruda Almeida , Camila Barbosa Silva , Márcio Rodrigo Gimenes and Luiz dos Anjos Núcleo de Pesquisas em Limnologia, Ictiologia e Aquicultura (Nupelia), Programa de Pós Graduação em Ecologia de Ambientes Aquáticos Continentais, Universidade Estadual de Maringá (UEM), Av. Colombo, 5790, Maringá, PR, CEP 87020-900, Brazil. Universidade Estadual de Mato Grosso do Sul (UEMS), Unidade Universitária de Ivinhema, Av. Brasil, 679, CEP 79740-000, Ivinhema, MS, Brazil. Universidade Estadual de Londrina, Centro de Ciências Biológicas, Departamento de Biologia Animal e Vegetal, Universidade Estadual de Londrina (UEL), Rodovia Celso Garcia Cid, km 380, s/n, CEP 86057-970, Londrina, PR, Brazil. Corresponding author: bialmeida182@hotmail.com Received on 29 May 2015. Accepted on 29 December 2016. ABSTRACT: Wetlands are rapidly being lost and fragmented around the world, making it imperative to seek an understanding of the drivers of their diversity. Among aquatic assemblages, birds constitute a conspicuous group that provides many ecosystem services. Here, we use a metacommunity approach to understand the influence of spatial ( distance among patches) and environmental factors (local characteristics) in wading bird (Pelecaniformes and Ciconiiformes) assemblages in a river-floodplain system t hrough time. We tested the hypothesis that, due to the small scale of our study, spatial factors have no determinant role in waterbird assemblages, and, due to the annual occurrence of flood pulses, the importance of environmental factors varies through time, accor ding to the hydrological cycle. We tested this hypothesis using Partial Redundancy Analysis (pRDA). We used abundance data for the birds in 20 lagoons, sampled quarterly during two years. The spatial factors did not explain variation in community structure in any sampled month, whereas environmental factors explained variation in the assemblages only in two months. Due to high waterbird mobility, the non-significance of the spatial factor is expected a mong lagoons in the same floodplain. Environmental factors are important in determining the community structure only in two sampled months, evidencing that their importance varies through time in the floodplain, partially agreeing with our hypothesis. The non-conformity between the influence of environmental factors on assemblages and the hydrological cycle may be due to human impacts caused by the operation of upstream reservoirs, which alter the natural flood events, and caused a long drought period previous to this study. A multiscale approach is fundamental to the understanding on how anthropogenic impacts on wetlands affect waterbir d assemblages. Thus, this study contributes to the understanding of how seasonality, environmental conditions of lakes, and a local spatial scale act in structuring waterbird assemblages. KEY-WORDS: Ciconiiformes, metacommunity, Paraná River, Pelecaniformes, variance partitioning. INTRODUCTION 2005). However, few studies have considered space as a structuring factor for local waterbird communities (Pagel Wetlands are mosaic-like environments, which generally et al. 2014, Sebastián-González & Green 2014). Here, comprise patches of different types, sizes and distances we used metacommunity theory to better understand the from other patches (Whited et al. 2000). Additionally, distribution of waterbirds in their patchy environment. Metacommunity theory is a theoretical and wetlands have suffered continuous loss and fragmentation, which has negatively affected t he ecosystem services they mechanistic framework that serves to explain the provide and biodiversity they harbor, including waterbirds interdependence of local and regional processes in (Ma et al. 2009). Waterbirds (e.g. Pelecaniformes, structuring communities (Logue et al. 2011). As viewed Ciconiiformes, Charadriiformes, Anseriformes) are through the lens of this concept, two general types of forces affect community structure: local ( biotic interactions highly dispersive birds, moving between favorable areas in search of resources and are dependent on wetlands and environmental conditions) and regional (spatial during most or all of their life cycle (Haig et al. 1998). dynamics, linked to the dispersal of organisms). These The inherent patchiness and rapid loss of wetlands processes interact to produce a local species assemblage around the world makes it important to integrate spatial (Leibold et al. 2004). Depending on the degree of influence of local and components into analyses of waterbird diversity and distribution (Paracuellos & Tellería 2004, Luis et al. regional processes in structuring the local communities, Revista Brasileira de Ornitologia, 24(4), 2016 Waterbirds in a floodplain: influence of spatial and environmental factors through time Bia de Arruda Almeida, Camila Barbosa Silva, Márcio Rodrigo Gimenes and Luiz dos Anjos four perspectives were defined concerning the organization (Junk et al. 1989). Flood pulses act by connecting and of metacommunities: neutral model, patch dynamics, homogenizing the patches of aquatic environments in a species sorting and mass effects. The neutral model floodplain and are a major driver of biodiversity in these assumes environmental homogeneity and species with ecosystems (Junk et al. 1989, Thomaz et al. 2007). similar competitive capacities, so that their distribution Pelecaniformes and Ciconiiformes are medium-to- is influenced only by dispersive capacity. Patch dynamics large waders that occur in most floodplains, and many considers patches also to be homogeneous, but species aspects of their ecology are affected by the hydrologic differ among them, in what constitutes a trade-off regime (Kushlan 1986, Dimalexis & Pyrovetsi 1997, between colonization and competition capacities. Russell et al. 2002). The density and vulnerability of Species sorting assumes environmental heterogeneity prey vary seasonally and spatially (Fernandes et al. 2009, associated with differences among species and with a Gimenes & Anjos 2011), and depend on the frequency, dispersive capacity that is sufficient for the species to duration and magnitude of the flooding (Petry et al. reach different patches, but which does not influence 2003). Because wading birds forage in shallow waters, their distribution. Finally, mass effects considers that prey availability is highest when water is shallower during there is a high dispersive capacity in a heterogeneous dry seasons (Gawlik 2002). The seasonal variation in the environment, which brings to communities affected by environment therefore makes it necessary to consider source-sink relations, with a greater influence of dispersal the influence of the hydrological cycle on the local than environmental heterogeneity in the structure of communities of waders. communities (Leibold et al. 2004). However, the complex The goal of this study was to assess the relative dynamics of communities suggests that these models are importance of regional (spatial) and local (environmental) not discrete or mutually exclusive (Logue et al. 2011). factors in structuring the wading bird communities Winegardner et al. (2012) reviewed the terminology used in the upper Paraná River floodplain throughout the by Leibold and proposed that the mechanisms underlying hydrological cycle. Due to the small scale of our study these paradigms interact. Thus, they reorganized t he (a single floodplain spanning c. 50 km ) and to the models according to the relative importance of dispersal, vagility and habitat selection capacity of birds (tending suggesting that metacommunities can be classified as to an ideal free distribution, see Fretwell & Lucas-Jr. follows: neutral, species-sorting with limited dispersal 1970), we expect that dispersal limitation and mass (which would be equivalent to patch dynamics, sensu effects will not play important roles in structuring wader Leibold et al. 2004), species-sorting with efficient communities in this floodplain. At the same time, due to dispersal (equivalent to species sorting), or species-sorting the cyclic alterations of the water level, we expect that the with high dispersal (equivalent to mass effects). importance of environmental variables is not equivalent The metacommunity approach contributes to a throughout the year. Therefore, we hypothesize that better understanding of how spatial dynamics and local the wader metacommunity of the upper Paraná River interactions influence community ecology (Logue et floodplain is shaped by species sorting with efficient al. 2011). Studying metacommunities of waterbirds dispersal (Winegardner et al. 2012), and that the role at different scales is necessary to understand how of environmental characteristics in structuring the local local communities are structured within and among communities of waders changes seasonally. Thus, we wetlands. Considering the importance of waterbirds as predict that the regional component will not impact suppliers of ecological services such as dispersal of seeds metacommunity structure and that the local component and eggs, ecosystem engineering, population control (environmental conditions) will be more important and scavenging, understanding their metacommunity during drier periods, when there is more heterogeneity dynamics is essential for wetland management (Green among the water bodies in the floodplain. & Elmberg 2013). Knowledge on spatial dynamics for waterbird assemblages enables one to answer questions METHODS such as how distances between wetland patches influence assemblages. As an example of how the metacommunity Study area approach can be used to better explore the mechanisms The Paraná River stretches for 4695 km from t he that shape communities of waterbirds, we evaluated the local and regional drivers of assemblages of wading birds Brazilian central plateau southward to the Plate River (Pelecaniformes and Ciconiiformes) in a Neotropical between Argentina and Uruguay. In Brazil, the Paraná floodplain. F loodplains are heterogeneous environments River is affected by dams along most of its length, and with high biological diversity and are influenced by only a stretch of 230 km between the Porto Primavera and Itaipu reservoirs remains as a floodplain (22°40'S to cyclical variation in river discharge, or flood pulses Revista Brasileira de Ornitologia, 24(4), 2016 Waterbirds in a floodplain: influence of spatial and environmental factors through time Bia de Arruda Almeida, Camila Barbosa Silva, Márcio Rodrigo Gimenes and Luiz dos Anjos 22°52'S and 53°12'W to 53°38'W). The dams installed water levels, while smaller pulses (< 0.5 m) occur weekly along the hydrographic basin have modified t he natural in the dry season, caused by the operation of upstream pattern of magnitude, duration, periodicity and discharge reservoirs. frequency of the Paraná River, increasing the control on The study area, located between the mouths of the discharge regime, eliminating the highest water level the Paranapanema and Ivinhema Rivers, encompasses values and causing variability between night and day environments with very different characteristics. At this discharges (Thomaz et al. 2004). Thus, the seasonal water point, the Paraná River has an extensive alluvial plain level fluctuation in the upper Paraná River floodplain is on its west margin. This floodplain consists of a mosaic not continuously sinusoidal, as uninterrupted periods of of habitats, including rivers, streams, channels, and falling and rising water are rarely observed (Agostinho et marginal lagoons with different degrees of hydrological al. 2000). connectivity. Two types of lagoons are used in this The climate of the area is classified as Tropical study: permanently connected lagoons, which have an Subtropical, with an annual average temperature of 22°C above-ground connection with the river all year long, (average of 26°C in summer and 19°C in winter). The and isolated lagoons, which are disconnected from the pluvial regime is marked by a wet summer, with monthly river during most of the year and have an above-ground average precipitation greater than 125 mm, and a dry connection only during flooding events. winter, with averages under 80 mm. The high-water period in the Paraná River usually occurs from November/ Data collection December to April/May and is characterized by an increase in the water level averaging 2.5 m and reaching The birds were sampled in 2002 and 2003 (both years 7.5 m in years of extreme flood events, with a lmost no on February, May, August and November) in connected variation observed in years in which the characteristic (permanently connected to rivers or canals, n = 10) flood period does not occur. It is quite common for two and isolated perennial lagoons (not connected to rivers, or three annual flood pulses to be observed during high except during the flood, n = 10) (Figure 1). The areas FIGURE 1. Map of the study area. The numbers indicate the sampling units. Connected lagoons: 1, 3, 6, 8, 10, 12, 15, 17, 19, 20; isolated lagoons: 2, 4, 5, 7, 9, 11, 13, 14, 16, 18. Revista Brasileira de Ornitologia, 24(4), 2016 Waterbirds in a floodplain: influence of spatial and environmental factors through time Bia de Arruda Almeida, Camila Barbosa Silva, Márcio Rodrigo Gimenes and Luiz dos Anjos of the sampled lagoons varied from 0.43 to 89.8 ha, variables; and the residue, or the variation not explained and their perimeters from 251.4 to 7151.2 m. These by any of the above components (Borcard et al. 1992) months were chosen in order to sample all the phases of (Figure 2). We considered the adjusted R values as the the hydrological cycle. In each month, two visits lasting results of the variance partitioning procedure (Peres-Neto for 1 h were conducted in each sampling unit, always et al. 2006). The significance of each component (P < beginning 1 h after sunrise. In each sampling, a circular 0.05) was tested by Monte Carlo randomization tests transect was performed around the lagoon, through the (9999 runs) (Legendre & Legendre 1998). whole perimeter, either by boat in a constant speed of 30 -1 km.h (connected lagoons) or on foot (isolated lagoons), with identification and registration of each individual sighted (Bibby et al. 1992). The entire lagoon was fully visible along transects. Birds observed within a range of up to 5 m from the water were included, but individuals in flight were not recor ded, unless they were observed leaving or arriving at the lagoons. In all sampled lagoons only the margins of the water bodies are shallow enough to allow the presence of waders. Birds were identified to species according to Sigrist (2009). To characterize the environmental component, environmental variables considered important for the distribution of species of Pelecaniformes and Ciconiiformes and possibly related to the vulnerability of prey capture were selected. The recor ded environmental FIGURE 2. Venn diagram representing the decomposition of the variables were: perimeter (in m), average depth (in m), community structure into: E = environmental variation or local factor, S = spatial variation or regional factor, E|S = variation explained by type of dominant vegetation in the surrounding area environmental variables independent of spatial variables, or pure (forest, grasslands, macrophyte or Polygonum), and type environmental component, S|E = variation explained by spatial of lagoon (connected or isolated). Measurements of water variables independent of environmental variables, or pure spatial bodies were performed through aerial images. All datasets component, ES = variation explained together by environmental and spatial variables; and residue, variation not explained by any of were obtained from the PELD Technical Report, site 6 the previous components. (Souza et al. 2001, Souza-Filho & Stevaux 2002). Data analyses To construct the spatial matrix, a Euclidean distance matrix was generated between the sampled sites based To determine the relative contributions of local on their geographic coordinates, measured from the (environmental characteristics – E) and regional factors center of each lagoon. From the distance matrix, Moran (spatial determinants, or distances between patches (MEM) eigenvector maps were created to quantify the – S) to the community structure of Pelecaniformes spatial arrangement of the sampling points (Dray et al. and Ciconiiformes, we used a variance partitioning 2006). The eigenvectors (spatial variables) allow one to technique: partial redundancy analysis (pRDA) (Borcard represent the spatial relationships among the sampling et al. 1992, Cottenie 2005). This method of analysis can units at different spatial scales. S mall eigenvalues indicate be characterized as a multiple regression with multiple the absence of spatial autocorrelation and, therefore, are response variables. In this case, we used a dependent not suitable for defining spatial structures. We included matrix (species abundance) and two explanatory matrices all eigenvectors with a Moran's I coefficient greater (spatial and environmental variables) (Legendre & than 0.1 in the spatial predictor matrix (Nabout et al. Legendre 1998, Beisner et al. 2006). The variation in 2009). Variance partitioning was carried out using total community structure was decomposed into the following abundance data (number of birds recorded per lagoon), components: E+S - the total variation explained by the which were previously transformed into Hellinger values. analyses; E - environmental variation; S - spatial variation; The Hellinger distance is a measure recommended for E|S - the variation explained by the environmental the clustering or ordination of species abundance data variables, independent of spatial variables, or the pure (Legendre & Gallagher 2001). We performed eight environmental component; S|E - the variation explained pRDAs, one for each sampled month, in both years. All by spatial variables, independent of environmental analyses were conducted in R software (R Development variables, or the pure spatial component; ES - the Core Team 2012) using the packages vegan (Oksanen et explained variation shared by environmental and spatial al. 2013) and PCNM (Legendre et al. 2013). Revista Brasileira de Ornitologia, 24(4), 2016 Waterbirds in a floodplain: influence of spatial and environmental factors through time Bia de Arruda Almeida, Camila Barbosa Silva, Márcio Rodrigo Gimenes and Luiz dos Anjos RESULTS period of 2002. In fact, the water levels reached as high as 3.5 m in November 2003, the level that begins to provide connections between the river and isolated lagoons. The The water l evel of the Paraná River was higher between January and April in both the 2002 and 2003 annual year 2001, which preceded our sampling activities, was monitoring periods. However, it is noticeable that the water markedly dry, showing minimum water level values close levels varied more in the last half of 2003 than in the same to 1 m and maximum values not reaching 4 m (Figure 3). FIGURE 3. Daily fluviometric levels for years 2000–2005. Black arrows represent the sampling months. The line indicates the level at which a connection begins to be established between the environments. A total of 2028 individuals belonging to 13 species ibis, Syrigma sibilatrix, Mycteria americana, Ciconia of the orders Pelecaniformes and Ciconiiformes were maguari, Jabiru mycteria, Mesembrinibis cayennensis and recorded (Table 1). Six species were common (Ardea Theristicus caudatus) were less abundant, occurring in only cocoi, Butorides striata, Egretta thula, Ardea alba, Tigrisoma a few months and sites, mostly in connected lagoons, and lineatum and Nycticorax nycticorax), occurring in almost present in a maximum of six sites each month. The peak all sampling months (only in one month, one of the of bird abundance occurred in November 2002. February species was absent), with presence in a maximum of 22 2002 and February and August 2003, in turn, were the sites each month. Other species (Platalea ajaja, Bubulcus months with the lowest abundances (Figure 4). TABLE 1. Species registered in the study area for all the sampled months and according to sampling sites (lagoons). Lagoons are numbered according to the map (Fig. 1). Classification of the species is accor ding to the American Ornithologists' Union (AOU). Connected Isolated Lagoons No. 16 2813 12 710 11 17 6114 18 20 15 935419 Ciconiiformes Ciconiidae Ciconia maguari xxx Jabiru mycteria x xxx Mycteria americana xxx x Pelecaniformes Ardeidae Ardea alba xxxxxxxxxxxxx xxxxxx Ardea cocoi xxxxxxxxxxxxxxxxxxxx Bubulcus ibis xx Butorides striata xxxxxxxxxxxx xxxx x Egretta thula xxxxxxxxxxxx xxxx Nycticorax nycticorax xx x xxxxx x x Syrigma sibilatrix x Tigrisoma lineatum xxxxxxxxxxxxx xxxxx Threskiornithidae Mesembrinibis cayennensisxx Platalea ajaja x xxx Revista Brasileira de Ornitologia, 24(4), 2016 Waterbirds in a floodplain: influence of spatial and environmental factors through time Bia de Arruda Almeida, Camila Barbosa Silva, Márcio Rodrigo Gimenes and Luiz dos Anjos FIGURE 4. Number of individuals of the species of Ciconiiformes and Pelecaniformes recorded in the perennial lagoons of the upper Paraná River floodplain. The letters F, M, A and N stand for the months February, May, August and Nove mber, respectively. The results of the RDA variance partitioning factor E explained the structure of the communities differed between months. In 2002, the environmental in February (R = 0.14; P = 0.0215) and November factor - E - explained the structure of the bird (R = 0.13; P = 0.0468), and the spatial factor S was 2 2 communities in February (R = 0.15; P = 0.0430), May also important in February (R = 0.13; P = 0.0060). 2 2 (R = 0.14; P = 0.0149) and November (R = 0.28; P The pure environmental factor - E|S - did not explain = 0.0001), whereas the pure environmental factor E|S the variation of community structure observed in any explained the variability of the communities of birds month of 2003, and the pure spatial factor S|E showed only in February (R = 0.22; P = 0.0257) and November no importance for the communities in any month of (R = 0.25; P = 0.0011). In 2003, the environmental this study (Table 2). TABLE 2. Variance partitioning of the community structure of Ciconiiformes and Pelecaniformes in different months during the two years of sampling in the 20 lagoons of the upper Paraná River floodplain. E = environmental variation, S = spatial variation, E|S = pure environmental component, S|E = pure spatial component, ES = shared variation explained by environmental and spatial variables; residue, or unexplained variation; significance level at P < 0.05, A dj. R = adjusted coefficient of determination. Significant values are underlined. FebruaryMay August November 2 2 2 2 Adj. R P Adj. R P Adj. R P Adj. R P E 0.15 0.0430 0.14 0.0149 0.03 0.2951 0.28 <0.0001 S 0.06 0.7305 0.01 0.3256 0.01 0.3322 0.05 0.0958 E|S 0.22 0.0257 0.10 0.0748 0.00 0.7105 0.25 0.0011 S|E 0.02 0.3161 0.00 0.6917 0.00 0.8977 0.02 0.2471 ES 0.00 0.04 0.07 0.04 Residue 0.82 0.89 1.00 0.69 E 0.14 0.0215 0.06 0.1681 0.00 0.4450 0.13 0.0468 S 0.13 0.0060 0.04 0.1824 0.00 0.5351 0.09 0.0521 E|S 0.04 0.2288 0.03 0.3188 0.00 0.6269 0.06 0.1864 S|E 0.04 0.1753 0.01 0.3996 0.00 0.7578 0.02 0.2976 ES 0.10 0.03 0.03 0.07 Residue 0.83 0.93 1.00 0.85 Revista Brasileira de Ornitologia, 24(4), 2016 Waterbirds in a floodplain: influence of spatial and environmental factors through time Bia de Arruda Almeida, Camila Barbosa Silva, Márcio Rodrigo Gimenes and Luiz dos Anjos DISCUSSION As we hypothesized, the pure spatial factor S|E did not play an important role in shaping the local communities Although we have used the metacommunity concept of waders. Wading birds are highly vagile, which allows to analyze local communities of waders in a single them to keep up with the seasonally fluctuating mosaic floodplain, we should e mphasize that this approach of suitable habitat (Fretwell & Lucas-Jr. 1968, Haig et can be a useful tool, especially to determine the al. 1998). For this reason, distances between sites may pattern of waterbird communities at larger scales, not be important in shaping communities at small e.g. among different wetlands. Given that human spatial scales. Meynard & Quinn (2008) suggested that activities often change the spatial structure within and for bird communities studied on a scale of 1000 km or among wetlands, a multiscale approach is essential for less, environmental factors will predominate, whereas understanding how these changes may affect waterbir d dispersion will become more important for the structure communities. Therefore, our approach s hould be of local communities at larger scales. Thus, it is expected viewed as a first step in understanding the importance that space does not play an important role for wader of the metacommunity concept in structuring local communities within patches of a single floodplain (see communities of waterbirds. Figure 5). FIGURE 5. Schematic representation of how Local (environmental - LF) and Regional (spatial – RF) factors may influence waterbir d assemblages within and among wetlands. Larger distances between sites may imply a more important role of the regional factor in structuring communities. The significance of the pure environmental are affected by dams, one of the main downstream component E|S in February and November 2002 impacts of the reservoirs is to change the natural water shows that these waterbird communities are influenced level fluctuations. This change affects the water exchange by local factors and also that the importance of the between the river main channel and the floodplain, and, purely environmental factors varies according to time in extreme years, even the absence of floods is observed. in the floodplain, partly corroborating our hypothesis. This alteration in the hydrological regime directly impacts Although we hypothesized that there would be temporal wading birds whose life cycles depend on water fluctuation variation in the importance of E|S, we also predicted (Agostinho et al. 2004). During the two sampled years that the pure environmental factor would be important of this study, the water regime was not consistent. The in most months, being more important during drought. first year (2002) was typical, with floods occurring from This prediction was not confirmed by data. The pure January to April and lower water levels characterizing environmental component was significant only in two all other months. The second year of sampling (2003) months, February and November 2002, the first of which also had a typical flood period in its first months, but falls within the beginning of the flood period. the expected dry period had a higher water level than the The contrasting results found in 2002 and 2003 same period of 2002, with several flood pulses reaching may be the outcome of human impacts in the Paraná the level at which the river starts to overflow. A dditionally, River floodplain. As the Paraná River and its tributaries 2002 was preceded by a dry year, in which the supposed Revista Brasileira de Ornitologia, 24(4), 2016 Waterbirds in a floodplain: influence of spatial and environmental factors through time Bia de Arruda Almeida, Camila Barbosa Silva, Márcio Rodrigo Gimenes and Luiz dos Anjos flood period was characterized by a water level closer to better information on the influence of the hydrological the levels observed in the last months of 2003 (Figure 3). cycle for these organisms. The variation in the importance of the environmental This study is an example of the use of the factor to the local communities of waders is linked to the metacommunity concept as an important tool to better flood regime of the floodplain. The pure environmental understand the drivers of waterbird diversity in space and component was important in shaping the community time. Knowing the factors driving waterbird distribution in February 2002, the beginning of the flood, and in in space, as well as the variation in the importance of these November 2002, the end of the dry period. This result factors seasonally is essential to understand waterbird might be explained by the differences in water fluctuation community dynamics. Here, we showed that space is among 2001, 2002 and 2003. Wading birds seek optimal not a strong driver of wading bird assemblages at small foraging habitats given their morphological and behavioral scales, considering the area of a floodplain. Furthermore, restrictions. They then select habitat based on local we showed that there is seasonal variation in the conditions. To reflect this, the environmental variables importance of the environmental variables used in this selected for this study are related to the availability of prey study for this group of waterbirds. Indeed, environmental to waders, the majority of which are piscivores (Bancroft variables other than those used in this study could be et al. 2002, Gawlik 2002). Previous studies performed important for these birds in different phases of the in the same floodplain showe d that differences among flood regime. These additional variables, such as water the density of fish in different lagoons increase during transparency and aquatic vegetation, could be the focus the year, becoming more apparent at the end of the dry of future investigations. This approach can be explored period (Fernandes et al. 2009, Gimenes & Anjos 2011). to recognize important factors for the maintenance of Additionally, according to Thomaz et al. (2007), a time waterbird communities in recognition of the continuous lag is expected after the flood period until t he effects of loss, alteration and fragmentation of wetlands around the environmental heterogeneity become apparent. Thus, we world. suggest that there is a tendency for the environmental factor to become more important in shaping communities at the end of the dry period (represented by November in ACKNOWLEDGEMENTS our samples). As we see, 2003 was an atypical year, in which higher water levels caused an interruption in the We thank the Center of Research in Limnology, Ichthyology and Aquaculture of the State University dynamics of the dry period and thus altered the role of the environment in structuring waterbird communities. In of Maringa (Nupelia/UEM) for logistical support the same way, the significance of the pure environmental and PROEX/CAPES for financial support. Sebastião factor in February 2002 may be a consequence of the dry Rodrigues, Leandro Rodrigues and Alfredo da Silva year that preceded it. Specifically, as the flood period was helped with the fieldwork. We appreciate valuable comments and suggestions by Nadson Ressye Simões da still at its beginning in February, it is possible that the effects of the flood pulse were delayed by the long 2001 Silva. Fabrício Oda and Gabriel Rosa helped with figures. drought. The lack of signal from our explanatory variables may be in part due to limitations of our study. The lack REFERENCES of importance of the spatial factor may be associated with the small scale of the floodplain. As wading birds are Agostinho, A. A.; Thomaz, S. M.; M inte-Vera, C. V. & Winemiller, K. O. 2000. Biodiversity in the high Paraná River floodplain, vagile organisms, the size of the floodplain may be too p. 89–118. In: Gopal, B.; Junk, W. J. & Davis, J. A. (eds.). small for distances between sites to influence the structure Biodiversity in wetlands: assessment, function and conservation. of the community. At the same time, environmental Leiden: Backhuys Publishers (Wetlands Biodiversity). characteristics other than those considered in this study Agostinho, A. A.; Thomaz, S. M. & Gomes, L. C. 2004. Threats for are known to influence waterbir d assemblages, such as biodiversity in the floodplain of the upper Paraná River: effects of hydrological regulation by dams. Ecohydrology and Hydrobiolog y, water transparency, productivity, aquatic vegetation and 4: 255–256. marginal vegetation (Bancroft et al. 2002, Guadagnin Bancroft, G. T.; Gawlik, D. E. & Rutchey K. 2002. Distribution & Maltchik 2007, Cintra 2015). The addition of such of wading birds relative to vegetation and water depths in the variables could increase the importance of environmental northern Everglades of Florida, USA. Waterbirds, 25: 265–277. Beisner, B. E.; Peres-Neto, P. R.; Lindström, E. S.; Barnett, A. factors or change its seasonal variation. Furthermore, & Longhi, M. L. 2006. The role of environmental and spatial sampling along only two years could not be enough to processes in structuring lake communities from bacteria to fish. determine a seasonal variation pattern of the importance Ecology, 87: 2985–2991. of environmental or spatial factors for wading bird Bibby, C. J.; Burgess, N. D. & Hill, D. A. 1992. Bird census community structure. More years of study could bring techniques. London: Academic Press. Revista Brasileira de Ornitologia, 24(4), 2016 Waterbirds in a floodplain: influence of spatial and environmental factors through time Bia de Arruda Almeida, Camila Barbosa Silva, Márcio Rodrigo Gimenes and Luiz dos Anjos Borcard, D.; Legendre, P. & Drapeau, P. 1992. Partialling out the Meynard, C. N. & Quinn, J. F. 2008. Bird metacommunities in spatial component of ecological variation. Ecology, 73: 1045– temperate South American forest: vegetation structure, area, and 1055. climate effects. Eco logy, 89: 981–990. Cintra, R. 2015. Spatial distribution and composition of waterbirds Nabout, J. C.; Siqueira, T.; Bini, L. M. & Nogueira, I. S. 2009. No in relation to limnological conditions in the Amazon Basin. evidence for environmental and spatial processes in structuring Hydrobiologia, 747: 235–252. phytoplankton communities. Acta Oecologica, 35: 720–726. Cottenie, K. 2005. Integrating environmental and spatial processes in Oksanen, J.; Blanchet, F. G.; Kindt, R.; Legendre, P.; Minchin, ecological community dynamics. Ecology Letters, 8: 1175–1182. P. R.; O'Hara, R. B.; Simpson, G. L.; Solymos, P.; Stevens, Dimalexis, A. & Pyrovetsi, M. 1997. Effect of water level fluctuations M. H. H. & Wagner, H. 2013. vegan: community ecology on wading bird foraging habitat use at an irrigation reservoir, Lake package. R package version 2.0-8. http://CRAN.R-project.org/ Kerkini, Greece. Colonial Waterbirds, 20: 244–252. package=vegan (access on 28 January 2015). Dray, S.; Legendre, P. & Peres-Neto, P. R. 2006. Spatial modeling: Pagel, J.; Martínez-Abraín, A.; Gómez, J. A.; Jiménez, J. & Oro, a comprehensive framework for principal coordinate analysis of D. 2014. A long-term macroecological analysis of the recovery neighbor matrices (PCNM). Ecological Modelling, 196: 483–493. of a waterbird metacommunity after site protection. PLoS ONE, Fernandes, R.; Gomes, L. C.; Pelicice, F. M. & Agostinho, A. A. 9: e105202. 2009. Temporal organization of fish assemblages in floodplain Paracuellos, M. & Tellería, J. L. 2004. Factors affecting the lagoons: the role of hydrological connectivity. Environmental distribution of a waterbird community: the role of habitat Biology of Fishes, 85: 99–108. configuration and bir d abundance. Waterbirds, 27: 446–453. Fretwell, S. D. & Lucas-Jr., H. L. 1970. On territorial behavior Peres-Neto, P. R.; Legendre, L.; Dray, S. & Borcard, D. 2006. and other factors influencing habitat distribution in birds. I. Variation partitioning of species data matrices: estimation and Theoretical development. Ac ta Biotheoretica, 19: 16–36. comparison of fractions. Ecology, 87: 2614–2625. Gawlik, D. E. 2002. The effects of prey availability on the numerical Petry, A. C.; Agostinho, A. A. & Gomes, L. C. 2003. Fish assemblages response of wading birds. Ecological Monographs, 72: 329–346. of tropical floodplain lagoons: exploring the role of connectivity Gimenes, M. R. & Anjos, L. 2011. Quantitative analysis of foraging in a dry year. Neotropical Ichthyology, 1: 111–119. habitat use by Ciconiiformes in the upper Paraná River floodplain, R Development Core Team. 2012. R: a language and environment Brazil. Brazilian Archives of Biology and Technology, 54: 415–427. for statistical computing. Vienna: R Foundation for Statistical Green, A. J. & Elmberg, J. 2013. Ecosystem services provided by Computing. waterbirds. Biological Reviews, 89: 105–122. Russell, G. J.; Bass, O. L. & Pimm, S. L. 2002. The effect of Guadagnin, D. L. & Maltchik, L. 2007. Habitat and landscape hydrological patterns and breeding-season flooding on the factors associated with Neotropical waterbird occurrence and numbers and distribution of wading birds in Everglades National richness in wetland fragments. Biodiversity and Conservation, 16: Park. Animal Conservation, 5: 185–199. 1231–1244. Sebastián-González, E. & Green, A. J. 2014. Habitat use by Haig, S. M.; Mehlman, D. W. & Oring, L. W. 1998. Avian waterbirds in relation to pond size, water depth, and isolation: movements and wetland connectivity in landscape conservation. lessons from a restoration in southern Spain. Restoration Ecology, Conservation Biology, 12: 749–758. 22: 311–318. nd Junk, W. J.; Bayley, P. B. & Sparks, R. E. 1989. The flood pulse Sigrist, T. 2009. Guia de campo Avis Brasilis - avifauna brasileira, 2 concept in river-floodplain systems. Canadian Special Publication edn. São Paulo: Avis Brasilis. of Fisheries and Aquatic Sciences, 106: 110–127. Souza, M. C.; Kita, K. K.; Romagnolo, M. B.; Soares, J. J.; Kushlan, J. A. 1986. Responses of wading birds to seasonally Tomazini, V. & Albuquerque, E. C. 2001. Mata ciliar, p. 231– fluctuating water levels: strategies and t heir limits. Colonial 240. In: Agostinho, A. A.; Nakatani, K.; Antônio, R. R.; Zavadski, Waterbirds, 9: 155–162. F. C.; Thomaz, S. M.; Júlio-Jr., H. F.; Luz, K. D. G. & Pereira, J. Legendre, P.; Borcard, D.; Blanchet, F. G. & Dray, S. 2013. PCNM: L. L. (eds.). A planície de inundação do alto Rio Paraná. Maringá: MEM spatial eigenfunction and principal coordinate analyses. PELD/CNPq Technical Report. R package version 2.1-2/r109. http://R-Forge.R-project.org/ Souza-Filho, E. E. & Stevaux, J. C. 2002. Locais de amostragem, p. projects/sedar/ (access on 28 January 2015). 7–13. In: Agostinho, A. A.; Thomaz, S. M.; Nakatani, K.; Júlio-Jr., Legendre, P. & Gallagher, E. D. 2001. Ecologically meaningful H. F.; Antônio, R. R.; Luz, K. D. G.; Zavadski, F. C. & Pereira, transformations for ordination of species data. Oecologia, 129: J. L. L. (eds.). A planície alagável do alto rio Paraná: estrutura e 271–280. processos ambientais. Maringá: PELD/CNPq Technical Report. Legendre, P. & Legendre, L. 1998. Numerical ecolog y. Amsterdam: Thomaz, S. M.; A gostinho, A. A. & Hahn, N. S. 2004. The upper Elsevier. Paraná River and its floodplain: physical aspects, ecology and Leibold, M. A.; Holyoak, M.; Mouquet, N.; Amarasekare, P.; conservation. Leiden: Backhuys Publishers. Chase, J. M.; Hoopes, M. F.; Holt, R. D.; Shurin, J. B.; Thomaz, S. M.; B ini, L. M. & Bozelli, R. L. 2007. Floods increase Law, R.; Tilman, D.; Loreau, M. & Gonzalez, A. 2004. The similarity among aquatic habitats in river-floodplain systems. metacommunity concept: a framework for multi-scale community Hydrobiologia, 579: 1–13. ecology. Ecology Letters, 7: 601–613. Whited, D.; Galatowitsch, S.; Tester, J. R.; Schik, K.; Lehtinen, Logue, J. B.; Mouquet, N.; Peter, H. & Hillebrand, H. 2011. R. & Husveth, J. 2000. The importance of local and regional Empirical approaches to metacommunities: a review and factors in predicting effective conservation planning strategies for comparison with theory. Trends in Ecology and Evolution, 26: wetland bird communities in agricultural and urban landscapes. 482–491. Landscape and Urban Planning, 49: 49–65. Guadagnin, D. L.; Peter, Â. S.; Perello, L. F. C. & Maltchik, L. Winegardner, A. K.; Jones, B. K.; Ng, I. S. Y.; Siqueira, T. & 2005. Spatial and temporal patterns of waterbird assemblages in Cottenie, K. 2012. The terminology of metacommunity ecology. fragmented wetlands of southern Brazil. Waterbirds, 28: 261–272. Trends in Ecology and Evolution, 27: 253–254. Ma, Z.; Cai, Y.; Li, B. & Chen, J. 2009. Managing wetland habitats for waterbirds: an international perspective. Wetlands, 30: 15–27. Associate Editor: Caio G. Machado. Revista Brasileira de Ornitologia, 24(4), 2016

Journal

Ornithology Research – Springer Journals

Published: Dec 1, 2016

Keywords: Ciconiiformes; metacommunity; Paraná River; Pelecaniformes; variance partitioning

There are no references for this article.