Functional diversity of fish communities in two tropical ...

MPB-07938; No of Pages 11 Marine Pollution Bulletin xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Marine Pollution Bulletin journal homepage: www.elsevier.com/locate/marpolbul

Functional diversity of fish communities in two tropical estuaries subjected to anthropogenic disturbance M. Dolbeth a,b,⁎, A.L. Vendel c, A. Pessanha a, J. Patrício d a

Postgraduate Program in Ecology and Conservation, Paraiba State University, 58429-500 Campina Grande, Brazil Biology Department & Centre for Environmental and Marine Studies (CESAM), University of Aveiro, Campus Universitário de Santiago, 3810-193 Aveiro, Portugal Centre for Applied Biological and Social Sciences, Paraiba State University, Campus V, Horacio Trajano Street, 58070-450 João Pessoa, Paraiba, Brazil d MARE - Marine and Environmental Sciences Centre, Faculty of Sciences and Technology, University of Coimbra, Portugal b c

a r t i c l e

i n f o

Article history: Received 20 April 2016 Received in revised form 2 August 2016 Accepted 3 August 2016 Available online xxxx Keywords: Brazil Estuary Functional complementarity Functional identity Functional redundancy Morphological traits

a b s t r a c t The functional diversity of fish communities was studied along the salinity gradient of two estuaries in Northeast Brazil subjected to different anthropogenic pressures, to gain a better understanding of the response of fish communities to disturbance. We evaluated functional complementarity indices, redundancy and analysed functional composition through functional groups based on combinations of different traits. The fish communities in both estuaries share similar functions performed by few functional groups. The upstream areas had generally lower taxonomic, functional diversity and lower redundancy, suggesting greater vulnerability to impacts caused by human activities. Biomass was slightly more evenly distributed among functional groups in the less disturbed estuary, but total biomass and redundancy were lower in comparison to the urbanized estuary. The present findings lend strength to the notion that the less disturbed estuary may be more susceptible to anthropogenic impacts, underscoring the need for more effective conservation measures directed at this estuary. © 2016 Elsevier Ltd. All rights reserved.

1. Introduction Coastal and estuarine ecosystems are often home to a variety of human activities and constitute an important source of wealth for populations around the world (Costanza et al., 2014). In recent decades, anthropogenic stress has intensified in tropical estuarine ecosystems due to urban and industry effluents, agriculture, aquiculture, overfishing and poaching activities (Barletta et al., 2010; Blaber, 2013; Diegues, 1999). Undisturbed or nearly pristine tropical systems are now rare (Blaber, 2013). Efforts have been made for the conservation of such aquatic ecosystems locally, nationally or internationally through management plans as well as the establishment of Ramsar sites and world heritage sites. However, many tropical regions continue to suffer from human activities that lead to environmental degradation stemming from inadequate management, unenforced laws, poverty and over-population (Barletta et al., 2010; Diegues, 1999). Northeast Brazil exemplifies the conflicting uses of estuarine areas. Human activities flourish in estuaries in the region, which is home to sugarcane production, intensive shrimp aquaculture (97% of the national production) and fishing activities, together with high degrees of urbanization and the dumping of untreated solid waste and effluents ⁎ Corresponding author at: Biology Department & CESAM, University of Aveiro, Campus Universitário de Santiago, 3810-193 Aveiro, Portugal. E-mail address: [email protected] (M. Dolbeth).

(Barletta et al., 2010; Lacerda, 2006; Sá et al., 2013). Such activities have been leading to the general impoverishment of aquatic communities. Despite protection actions from both governmental and non-governmental organizations in recent years (Diegues, 1999), environmental degradation continues and has the potential to intensify in the near future. The largest estuaries of the state of Paraiba (Northeast Brazil) are the Paraiba and Mamanguape estuaries, which have distinct intensity profiles with regard to anthropogenic pressures. The Paraiba estuary is a highly impacted system, with anthropogenic pressures from several sources, while the Mamanguape estuary has been declared a conservation unit by the IUCN (category V), despite also having some anthropogenic pressure on the system. Human activities in both estuaries have considerable economic importance in the region for different endusers (Alves et al., 2005; Sá et al., 2013). Thus, any management program must take into account the multiple and often conflicting uses of the estuary without compromising its ecological quality and overall functioning. For such, one needs to gain an understanding of the responses of biological communities to anthropogenic disturbances in order to gather information on ecological quality and the capacity of a system to recover from such disturbances. As species do not contribute equally to the functioning of an ecosystem, it is important to understand the way these organisms use the system and how they cope with environmental changes (Stuart-Smith et al., 2013). The evidence suggests that the determination of functional

http://dx.doi.org/10.1016/j.marpolbul.2016.08.011 0025-326X/© 2016 Elsevier Ltd. All rights reserved.

Please cite this article as: Dolbeth, M., et al., Functional diversity of fish communities in two tropical estuaries subjected to anthropogenic disturbance, Marine Pollution Bulletin (2016), http://dx.doi.org/10.1016/j.marpolbul.2016.08.011

2

M. Dolbeth et al. / Marine Pollution Bulletin xxx (2016) xxx–xxx

diversity provides a way to understand the responses of a community to disturbances in a more integrated approach (Mouillot et al., 2013) and to gain a better understanding of the functioning of an ecosystem (Dolbeth et al., 2015; Strong et al., 2015). Functional diversity is generally described as the functional component of biodiversity that is measured by the traits of species (Violle et al., 2007). Traits reflect the mechanisms underlying species-environment relationships and may provide considerable insight into the responses of a community (e.g., Mouillot et al. 2013). However, functional diversity encompasses different components, which may all be important to revealing the impacts of disturbances, and several methods have been proposed to quantify this aspect (Carmona et al., 2016; van der Linden et al., 2016; Villéger et al., 2008). The aim of the present study was to characterize and explore functional composition and diversity in the Mamanguape and Paraiba estuaries, assuming that the two ecosystems experience distinct degrees of anthropogenic pressure. However, since different facets of functional diversity are important to revealing the impact of disturbances (Mouillot et al., 2013; Villéger et al., 2008, 2010), two different approaches were employed to evaluate functional diversity: 1) based on a functional complementarity effect (i.e., evaluating mean trait dissimilarity using different functional indices and evaluating functional redundancy); and 2) based on a functional identity effect. To study functional identity and understand how it could help in explaining the community response to disturbances, the concept of functional groups was employed rather than analysing each trait separately. Recent studies have shown that a particular combination of traits may reflect stability in the presence of environmental disturbance better than each trait alone (Dolbeth et al., 2015; Verberk et al., 2013; Winemiller et al., 2015). The goal was to broaden knowledge on how tropical systems function and understand what characteristics allow fish communities to recover better from environmental disturbances. Ultimately, we want to test whether the analysis of the functional diversity of fish communities is

a useful tool for the assessment of the effectiveness of environmental protection practices. 2. Materials and methods 2.1. Study area This study was conducted in two tropical estuaries located on the coast of Northeast Brazil (Fig. 1): the Paraiba estuary (3012 ha) and the Mamanguape estuary (690 ha). According to the Köppen-Geiger classification, the climate in the two estuaries is “As”, i.e., equatorial with a dry summer (Alvares et al., 2013). In both estuaries, the rainy season extends from February to August, with the greatest rainfall occurring in June and the lowest in November. The Paraiba estuary has a wetter climate (1717 mm/year) than the Mamanguape estuary (1392 mm/year) (data from 1999 to 2014; CPTEC/INPE 2015). The Paraiba River valley drains the driest region of Brazil (the Borborema plateau). Most rainfall is retained in reservoirs, except during wetter years. During the study period, freshwater entering the Paraiba estuary originated on the humid coastal plains (Executive Agency for the Management of Waters in the State of Paraiba [AESA]; accessed in August 2015). The watersheds that drain directly into the Paraiba River have approximately 38,472 ha. The Mamanguape River valley drains areas that are less dry, including a humid range, leading to frequent water spill over onto the coastal plains (AESA 2015). The watersheds that drain directly into the Mamanguape River have approximately 25,055 ha. Moreover, the Mamanguape estuary has a reef line 8.5 km in length running parallel to the shoreline that creates a protected region at the mouth of the estuary. Both estuaries have mangroves that grow around the main channel and intertidal creeks, along with remnants of the Atlantic rainforest (Campos et al., 2015). Both systems are subjected to different intensities of anthropogenic pressures. The Paraiba estuary is situated in a metropolitan area with

b)

a)

Fig. 1. Location of estuaries in Northeast Brazil, sampling sites and main land uses.



approximately one million inhabitants as well as sugarcane plantations along the riverbanks and intensive shrimp aquaculture activities that occupy part of the remaining mangrove area. These activities have been leading to the overall environmental degradation of the estuary, with visible signs of anthropogenic impact. Such conditions are aggravated by the lack of surveillance and law enforcement for specific activities as well as the increasing trend in human population growth. In contrast, Brazilian governmental agencies have recognized the ecological relevance of the Mamanguape estuary due to the importance of its coastal habitats and as an area for the conservation of the marine manatee, which is a protected species. This estuary has been declared a conservation unit for sustainable use since 1993 (IUCN, category V) and has extensive, well-preserved mangrove areas, providing food sources and nursery areas for different species (Alves et al., 2005; Claudino et al., 2015; Pessanha et al., 2015; Xavier et al., 2012). However, shrimp aquaculture activities and sugarcane plantations also occupy the areas surrounding the estuarine region, thereby contributing towards eutrophication and agrochemical pollution (Silvestre et al., 2011; Lacerda et al., 2011; Nóbrega et al., 2014), despite the longstanding environmental protection legislation. Moreover, local fishermen have detected a decrease in fish production in recent years as a consequence of such activities (Alves et al., 2005). 2.2. Sampling and laboratory procedures Fish assemblages were sampled using a beach seine (10 × 1.5 m; mesh size: 8 mm) along the salinity gradient of the two estuaries (15 sampling sites in the Paraiba estuary and 12 sampling sites in the Mamanguape estuary, Fig. 1). At each sampling site, three 30-m hauls were collected during low tide in the dry (November 2013) and wet (July 2014) seasons. Physicochemical parameters were measured in situ (surface water temperature, salinity, pH and turbidity with multiparameter probe and water transparency with a Secchi disc) and water samples were collected for the analysis of nutrient content and the determination of chlorophyll a to infer the possible effects of organic enrichment from agriculture and aquaculture activities. Concentrations of ammonium (NH3-N, μg/L), nitrite + nitrate (NOx-N, μg/L) and total phosphorous (P, μg/L) were measured and chlorophyll a (Chl a, μg/L) was determined in the laboratory, as described in Alves et al. (2016). All fish were identified, counted, weighed (g in wet mass) and morphological measurements were determined on a sub-sample of individuals: total length, body depth, body width, mouth width, mouth depth, pectoral fin length, pectoral fin surface, caudal fin depth, minimum depth of caudal peduncle, distance between the centre of the eye to the bottom of the head, head depth along the vertical axis of the eye. 2.3. Selection of traits With the above-mentioned morphological measurements, eight effect traits were defined to quantify food acquisition and locomotion (Villéger et al., 2010). Effect traits are those that produce an effect on ecosystem processes (Violle et al., 2007). Body mass, body transversal shape, relative head length, oral gape shape, eye position, aspect ratio of the pectoral fin, relative peduncle length, caudal peduncle compression were evaluated (equations used to define effect traits in Table 1S, supplementary information). All traits were initially checked for collinearity after the inspection of correlations and the variation inflation factor (VIF N 3) (Zuur et al., 2009). As the fish communities in the tropical systems studied were mainly composed of juveniles and sub-adults, ontogenic changes were not considered. 2.4. Data analyses Environmental data were explored using PERMANOVA based on a Euclidean distance matrix to test differences between estuaries for each variable. Analyses were performed taking into account the

3

estuarine gradient, classifying sites as upstream, middle and downstream sections. Three factors were considered: “estuary”, with two fixed levels (Paraiba and Mamanguape); “season”, with two fixed levels (wet and dry); and “estuarine section”, with three levels (upstream, middle and downstream, nested in “estuary” and “season”). Two different approaches were employed to evaluate the functional composition of the fish communities in each estuary. First, we evaluated functional complementarity by computing different facets of functional diversity with dissimilarity indices, as suggested by Mason et al. (2005) and Villéger et al. (2008): functional richness (FRic), which is not weighted by species abundance/biomass; and functional evenness (FEve) and functional divergence (FDiv), both of which are weighted by species biomass. FRic was standardized to constrain values between 0 and 1 (Laliberté et al., 2015). These indices were measured as proposed by Villéger et al. (2008) and Laliberté et al. (2015). For samples with less than three species, both FEve and FDiv were forced to zero to guide the interpretation of the results; however, these indices cannot be computed in such scenarios. Redundancy was computed as the difference between Simpson's diversity index and Rao's quadratic entropy (RaoQ) (de Bello et al., 2007). The SR'-FRED index proposed by van der Linden et al. (2016) incorporates a measurement of a reference species number (SR') and redundancy computed as RaoQ divided by Simpson's index [SR'-FRED = SR' − (1 − (RaoQ/Simpson))]. According to the authors, SR'-FRED can be used to discriminate disturbance levels within the same estuary. RaoQ is a measurement of the trait dissimilarity between individuals (de Bello et al., 2016). Functional diversity was then evaluated through functional identity analysis by defining a priori life history groups and analysing their biomass variation patterns through space and time. We used life-history groups because the adaptive value of a trait is often context dependent (Verberk et al., 2013). Therefore, a specific combination of traits may be more informative with regard to the response of the community to environmental disturbance than a single trait alone (Dolbeth et al., 2015; Murray et al., 2014; Verberk et al., 2013). Functional groups were defined taking into account the ecomorphologic groups proposed by Pessanha et al. (2015) complemented with principal coordinate (PCO) analysis on a Bray-Curtis similarity matrix for the traits matrix. Boxplots of the trait measurements from each functional group were also created. As we were interested in understanding how environmental background affects the functional composition of the community, we further explored the relationships between all functional diversity descriptors with the environmental data: 1) exploratory scatterplots and Spearman's correlations were used for functional indices; 2) relationships with environmental data were explored for the functional groups using distance-based linear models (DISTLM) on a Bray-Curtis similarity matrix for the biomass of fish functional groups. Prior to the DISTLM, all environmental variables were checked for collinearity using a Draftsman plot. As none was collinear, all variables were tested. The Best model selection procedure was used for the DISTLM based on the AICc selection criteria (Anderson et al., 2008). Analyses were performed using the FD library (i.e., FRic, FEve, FDiv) (Laliberté and Legendre, 2010, Laliberté et al., 2015) and the Rao function (Simpson and RaoQ) (de Bello et al., 2010) implemented with the aid of the R statistical software (R core team, 2016), and using PRIMER v6 and PERMANOVA + routines (for PERMANOVA, PCO, DISTLM and the Draftsman plot) (Anderson et al., 2008). 3. Results and discussion 3.1. Environmental background Salinity varied along the gradient, as expected for an estuarine system (Fig. 2a). In upstream areas of the Mamanguape estuary, salinity increased and depth was low in the dry season (N13) due to the lack of freshwater input in this semi-arid region. In contrast, oligohaline values (b 5; Fig. 2a) were found in the wet season. Details on variation trends


4


a) Salinity Mamanguape

Salinity

40

Upstream

Middle

Paraiba

40

Downstream

35

Dry

35

30

Wet

30

25

25

20

20

15

15

10

10

5

5

0

Upstream

Middle

Downstream

Dry Wet

0 1

2

3

4

5

6

7

8

9

10

11

12

1

2

3

4

5

6

7

8

9 10 11 12 13 14 15

b) Nutrients Mamanguape Upstream

Middle

2800

Middle

Downstream

2800

2100

2100

1400

1400

700

700 0 1

2

4200

3

4

5

6

7

Upstream

8

9

Middle

10 11

1

12

2

4200

Downstream

3500

3500

2800

2800

µg/L

µg/L

Upstream

3500

3500

0

Wet Season

Paraiba

4200

Downstream

µg/L

µg/L

Dry Season

4200

2100

2100

1400

1400

700

700

3

4

5

6

7

Upstream

8 Middle

9 10 11 12 13 14 15 Downstream

0

0 1

2

3

4

5

6 7 Sites

8

9

10 11

Nutrients:

1

12 NH4-N

2

NOx-N

3

4

5

6

7

8 9 10 11 12 13 14 15 Sites

PO4-P

Fig. 2. Variations in a) salinity and b) nutrient concentrations along the salinity gradient of Mamanguape and Paraiba estuaries in dry and wet seasons, with indication of upper-middledownstream estuarine gradient on grey scale.

regarding turbidity, transparency, temperature and pH are found in Alves et al. (2016). Concentrations of all nutrients (total P, NH3-N and NOx-N) were much higher in the Paraiba estuary than the Mamanguape estuary (P b 0.05; Fig. 2b), reflecting the higher degree of anthropogenic disturbance in the urban estuary. These results were more evident in the wet season, although season was not a significant factor. Moreover, significant differences were found along the salinity gradient regarding all nutrients in the Paraiba estuary (P b 0.01), with higher total nutrient concentrations in the upstream and middle sections of the estuary (Fig. 2b). Besides the large urban area surrounding the estuary with the discharge of untreated effluents, the middle section of the Paraiba estuary has shrimp aquaculture activities that discharge directly into the estuary as well as through surrounding tributaries (Fig. 1), which may be one of the causes for the high nutrient concentrations in this area. Effluents from shrimp aquaculture are saline and typically enriched with ammonia, whereas sugarcane and sewage effluents are freshwater and may have higher proportions of nitrates, depending on the temperature (Jones et al., 2001; Martinelli and Filoso, 2008). Thus, the extremely high ammonia found in the Paraiba estuary is probably due to aquaculture effluents, which may be unreported, since Brazilian water quality standards for aquaculture effluents limits ammonia to ≤500 μg/L (Sá et al., 2013) and may also be due to untreated domestic

sewage effluents. In the Mamanguape estuary, nutrient concentrations were slightly higher in the dry season, although season was not a significant factor. NOx-N was generally higher in downstream areas and total phosphorous was higher in upstream areas (Fig. 2b), but both nutrients were considerably lower than those recorded for the highly urbanized Paraiba estuary. Despite the difference in the nutrients measured in the water column between estuaries, which may vary depending on the assimilative capacity of the environment (Primavera, 2006), a previous study has shown that both systems have enriched δ15N in the biota, which is a sign of anthropogenic impact (Dolbeth et al., submitted). Chlorophyll-a was highest in the upstream oligohaline areas of the Paraiba estuary (36.8 ± 17.5 μg/L; P b 0.01 for the estuarine gradient) particularly in the dry season, although no statistically significant differences were found between seasons (P N 0.05). In the Mamanguape estuary, chlorophyll-a concentrations were generally lower than 3 μg/L in both seasons as well as along the estuarine gradient. 3.2. Mean trait dissimilarity among species We analysed the evolution of the different functional indices along the estuarine gradient of each estuary. However, as fish communities are highly dynamic, the results along the gradient should not be interpreted in a static way.



FRic is described as the amount of functional space occupied by a community (Mason et al., 2005; Schleuter et al., 2010). Both estuaries had generally lower FRic values in the less saline upstream areas (Fig. 3), suggesting a narrower range of functions in these areas. This means that resources may be explored in similar way by existing species or that similar resources exist in such areas. Salinity was considerably higher in the upstream areas of the Mamanguape estuary in the dry season compared to the wet season (Fig. 2). However, FRic did not differ substantially between seasons. Moving towards the middle sections (sites 7 to 9), FRic began to increase, particularly in the Mamanguape estuary, which was a tendency maintained in both seasons. In the Paraiba estuary, FRic was highest in the downstream section (sites 10, 12 and 13), where salinity was N25 (Fig. 3), suggesting a wider range of functional possibilities (see functional identity groups, [Section 3.4]). At sites closer to the sea, however, FRic decreased again in both estuaries and both seasons. Comparing estuaries, FRic was generally higher in the Paraiba estuary in the dry season. In the wet season, the Paraiba estuary had the highest peak FRic, but mean FRic was higher in the Mamanguape estuary. A similar trend in the variation in FRic and number of species was found in both systems (Fig. 3, Fig. 1S), as expected, since FRic is positively correlated with species diversity (Schleuter et al., 2010). The relationship between FRic and environmental parameters was not clear in either estuary (Fig. 2S). As FRic does not take into account abundance/biomass values, it is important to consider the variation in biomass-weighted indices, i.e., how the functional space is occupied (Schleuter et al., 2010). FEve maintained a similar range of values in the Mamanguape estuary, except for sites 2 and 5 in the wet season (Fig. 3) due to the low number of species (b3; Fig. 1S), which limits the quantification of FEve (value forced to zero to guide interpretation). FEve describes how species biomass is

distributed in a functional space. A high FEve is usually obtained with regular distribution and a low FEve is obtained in the presence of separate trait clouds (Schleuter et al., 2010). This constancy around intermediate values in the Mamanguape estuary suggests more than one dominant group with particular functions, as the distribution of functions within the existing biomass is not completely even. Moreover, FDiv was generally higher than 0.5, except at sites with few species, where FDiv could not be evaluated (forced to zero to guide interpretation). FDiv reveals the proportion of the species biomass with extreme traits as well as the degree of resource differentiation (Mason et al., 2005; Mouillot et al., 2013). The FDiv results also suggest more than a single function/way to use the resources in the system, since high values indicate a high degree of niche differentiation (Mason et al., 2005). However, some sites had a lower FDiv, suggesting competition for resources. Similarly to FRic, FEve and FDiv did not differ considerably along the gradient between seasons, despite the differences in salinity. No correlations with particular environmental variables were evident with regard to either FEve or FDiv (Fig. 2S). For the Paraiba estuary, a decreasing variation trend in FEve was found, particularly in the wet season, while FDiv increased towards downstream areas and remained at high levels at sites closer to the sea (Fig. 3). FEve and FDiv could not be computed at sites 7 and 9 in the wet season, (forced to zero; Fig. 3) due to the low number of species (Fig. 1S). The FEve and FDiv results suggest that functions may be more regularly distributed among individuals (except at sites 7 and 9) and that resource differentiation is also high, meaning that species could have different ways of using the available resources, particularly in downstream areas. As no correlations were found between functional diversity indices and environmental parameters in the Paraiba estuary, relationships between trait dissimilarity and the parameters measured to characterize the system could not be ascertained.

Functional Richness Functional Evenness

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

Functional Divergence

Mamanguape 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

Upstream

1

2

3

4

Middle

5

6

7

Upstream

1

2

3

4

5

6

7

Upstream

1

2

3

4

8

8

9

6

Sites

7

8

10

11

9

10

11

10

11

12

Dry season

Upstream

1

2

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

12

Downstream

9

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

12

Downstream

Middle

5

Paraiba Downstream

Middle

5

3

4

Middle

5

6

7

Upstream

1

2

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

3

4

2

3

4

9

Middle

5

6

7

Upstream

1

8

Downstream

8

Downstream

9

Middle

5

6

7

8

10 11 12 13 14 15

10 11 12 13 14 15

Downstream

9

10 11 12 13 14 15

Sites

Wet season

Fig. 3. Functional diversity multi-metric indices weighted by biomass (except functional richness) variation along the salinity gradient of Mamanguape and Paraiba estuaries in dry and wet seasons (indices for communities with less than three species forced to zero – dashed lines, but index not computed for such scenarios).


6


1 and −1, respectively representing limited and maximum impact (van der Linden et al., 2016). Most sites had SR'-FRED lower than 0, except the middle section of the Mamanguape estuary and the most downstream section of the Paraiba estuary (Fig. 4). The low SR'-FRED suggests that both systems are subjected to sources of disturbance along the salinity gradients, with upstream areas generally exhibiting worse ecological conditions regarding fish communities. The middle section of the Mamanguape estuary and the downstream section of the Paraiba estuary exhibited better ecological quality along the gradient with regard to fish communities. As SR' was defined for each estuary (i.e., 90th percentile of measured species richness in each estuary) (van der Linden et al., 2016), the interpretation of sites more prone to human impacts should be restricted to each estuary separately. SR'-FRED tended to decrease in the Paraiba estuary with the increase in nutrient concentrations (Fig. 2SB), which, based on the definition proposed by van der Linden et al. (2016), was an expected result.

3.3. Functional redundancy Functional redundancy is generally described as the amount of different taxa that exhibit similar functions (Guillemot et al., 2011). As such, this aspect is important to the assessment of the impact of disturbances in a system, as it can ensure against the loss of ecosystem functioning following a decline in species diversity, thereby conferring stability (i.e., resistance and resilience) to the system (Guillemot et al., 2011; Micheli and Halpern, 2005; Mouillot et al., 2014). Different ways to evaluate functional redundancy have been proposed (Carmona et al., 2016; van der Linden et al., 2016). For the present study, we computed redundancy as the difference between Simpson's index and RaoQ (de Bello et al., 2007), that is, redundancy is zero (Simpson = RaoQ) when all species in the community are functionally different, whereas redundancy is maximum and equal to Simpson's index when species have the same function (RaoQ = 0). On average, redundancy in the Mamanguape estuary varied close to 0.4 in the dry season, whereas an increasing tendency from upstream to downstream sectors was found in the wet season (Fig. 4). The low degree of redundancy implies that functions may be lost when diversity declines (Micheli and Halpern, 2005), which is a plausible scenario, especially in upstream areas of the Mamanguape estuary (Fig. 1S). In the Paraiba estuary, redundancy was slightly higher than that in the Mamanguape estuary at several sites (Fig. 4), which may allow generally greater resilience for the Paraiba. However, the lowest values were found in the middle sections in the dry season (Fig. 4), when species diversity was extremely low (Fig. 1S), the concentrations of nutrients were high (Fig. 2) and water residence time was potentially high due to the limited freshwater input. This result shows that the middle section may be extremely vulnerable to disturbance in the dry season. However, the increase in redundancy in this section during the wet season (Fig. 4) suggests that the system has a certain capacity for resilience, i.e., the system seems to be capable of restoring lost functions in the following season. As estuarine fish communities are highly dynamic, we can only presume that adjacent sites, which have fish from all functional groups, are important to maintaining such resilience, as refuges for fish that return when conditions are more favourable. SR'-FRED was also evaluated based on the idea that the importance of redundancy with regard to the impact of disturbances depends on species diversity (van der Linden et al., 2016). SR'-FRED varies between

3.4. Patterns of functional identity groups To address the issue of the possible meaning of the present results in terms of strategies for coping with environmental constraints caused by natural and human-induced stress in each system, one must bear in mind that the value of a trait is context-dependent and that a particular combination of traits may reflect stability in the presence of environmental disturbances better than a single trait alone (Dolbeth et al., 2015; Verberk et al., 2013; Winemiller et al., 2015). We first sought to define functional groups based on the clustering of traits that could reflect locomotion and/or feeding strategies to cope with environmental changes along each estuary over time. We adapted the definition of ecomorphological groups proposed by Pessanha et al. (2015) together with a PCO analysis of the trait measurements (explaining 72% of variability) (Fig. 5A). Three main functional groups were found, which are hereafter designated as 1) flatfish, 2) cylindrical and 3) laterally compressed forms. Moreover, extreme forms were included, which were not clearly associated with the three groups (Fig. 5A). Fig. 5B displays boxplots of the measurements of the traits in each functional group. Table 2S (supplementary material) lists the species that compose each group and the description of the ecological relevance of each functional group.

Mamanguape Redundancy (Simpson-Rao)

1

Upstream

Middle

Paraiba Downstream

0.8

0.8

0.6

0.6

0.4

0.4

0.2

0.2

0

Upstream

Middle

Downstream

0 1

SR’-FRed

1

2

1 0.8 0.6 0.4 0.2 0 -0.2 -0.4 -0.6 -0.8 -1

3

4

5

6

7

Upstream

1

2

3

4

8

9

Middle

5

6 7 Sites

8

10

11

12

Downstream

9

10

11

12

Dry season

1

2

1 0.8 0.6 0.4 0.2 0 -0.2 -0.4 -0.6 -0.8 -1

3

4

5

6

7

Upstream

1

2

3

4

8

9

Middle

5

6

7

8

9

10 11

12 13 14 15 Downstream

10 11 12 13 14 15

Sites

Wet season

Fig. 4. Functional redundancy and SR'-FRED variation along the salinity gradient of Mamanguape and Paraiba estuaries in dry and wet seasons (red line for value 0: limit between impacted systems [maximum impact at −1] and limited-impacted systems [no impact at 1]). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)



A

B


7

8


Flatfish species occurred mainly in saline waters in the middle and downstream areas, which were subjected to some hydrodynamics, despite the reduced mobility associated to the group (Table 2S). This spatial distribution is associated to the salinity requirements of the species, as all the species found were marine organisms with benthivorous habits (Piet, 1998) (Table 2S), feeding essentially on benthic invertebrates (Pessanha et al., 2015). These species were more abundant in the Paraiba estuary (Fig. 6) and also occurred in areas with a high abundance of benthic polychaetes (Nóbrega et al., 2014). The cylindrical forms are essentially demersal species living in the water layer just above the sediment. Within this group, we separated fish with greater maneuverability and the ability to catch larger prey (orange group, Fig. 5B - Table 2S). The morphological characteristics of this group seem to be advantageous for species that feed on benthic organisms (Karpouzi and Stergiou, 2003; Villéger et al., 2008) (Table 2S) and in fact, the diet of these fishes may include benthic invertebrates as preferential food items (e.g., Campos et al., 2015; Pessanha et al., 2015) or even other fish (e.g., Strongylura spp.). These cylindrical forms were usually more abundant in the wet season in both estuaries, particularly in the middle sections (Fig. 6). The laterally compressed forms comprehended a wide range of body mass values (Fig. 5B). This group is essentially pelagic, with good swimming ability and a wide array of potential preys and feeding possibilities (Table 2S). These characteristics may be highly beneficial when facing the impact of disturbances by allowing better use of the limited food sources and the possibility of escaping from toxic conditions at specific sites and times. This group contained some of the most abundant species in both estuaries, such as Rhinosardinia bahiensis and Atherinella brasiliensis (Table 1S). A sub-division was also considered in this group, essentially separating fish with greater mobility and highly piscivorous (darker blue, Fig. 5B, Table 2S). The extreme forms of this group were highly compressed fishes (body transversal shape N5), such as Chaetodipterus faber and Selene vomer (Table 1S), the forms of which may be advantageous for vertical stability in the water column (Table 2S). Similarly to other extreme forms, the biomass of these species was low in comparison to the other functional groups (Fig. 6). Regarding the variation patterns of the functional identity groups in the dry season, the vast majority of species belonged to the laterally compressed forms in the Paraiba estuary, while the biomass seemed more evenly distributed among the different groups in the Mamanguape estuary (Fig. 6). In the latter estuary, functional composition changed with the season, with lower biomass and less evenly distributed groups in the upstream section and a higher percentage of cylindrical fish in the middle section. This result could be related to differences in the intrusion of salinity in the Mamanguape estuary between seasons. In the Paraiba estuary, the biomass of laterally compressed fish remained high in the wet season, with a comparatively more even distribution among the other groups, except at site 12, where laterally compressed forms were clearly dominant (Fig. 6) due to the occurrence of schools of R. bahiensis. Total biomass levels were considerably higher in the Paraiba estuary than the Mamanguape estuary, except at stations 8 to 10 in the dry season. These latter stations are strongly influenced by shrimp aquaculture effluents, typically resulting in high ammonia concentrations (Jones et al., 2001), which may cause an abnormal decline in fish biomass at these sites. As the sampling scheme was the same for both seasons, the extremely high ammonia values probably resulted from direct effluent discharges at the time of data collection. The impact on the fish community may be harsher in dry season due to the limited water exchange in the estuary. The few remaining fish at these stations were mainly laterally compressed species, which lends support to the notion that this functional group may be better adapted to coping with anthropogenic disturbances.

For the biomass of functional groups, we investigated relationships with the environmental variables using DISTLM. In both estuaries, as no environmental variable was collinear (b 0.75 correlation), all were used in the DISTLM. For the Mamanguape estuary, the parsimonious model explained only 22% of total variability, with salinity and pH demonstrating to be significant variables. Samples were clustered mostly by season, with higher pH and salinity associated with the dry season, probably as a result of the higher salinity incursion during the dry season. Samples from the downstream areas in the wet season also demonstrated a tendency to cluster associated with the high salinity, as expected. For the Paraiba estuary, only 7.3% of the total variability was explained, with log total P as the single significant variable, separating the upstream and middle sections with higher P concentrations from the downstream section. Since this dichotomy between upstream/middle and downstream areas reflects differences in the concentrations of all nutrients in the Paraiba estuary (Fig. 2), log P was probably a proxy for the nutrient effect in the estuary associated with anthropogenic activities (Sá et al., 2013; Nóbrega et al., 2014), with flatfish and the red category of cylindrical fish generally associated with downstream areas. However, the low percentage of explained variability in both estuaries suggests that important variables may not have been taken into account (e.g., habitat diversity, untreated effluent discharges, etc.) and the relationship between the variation in biomass and environmental background may not be linear. For instance, Xavier et al. (2012) found that the composition of fish communities in the Mamanguape estuary was correlated with the availability of microhabitats, which was not evaluated in the present study. Moreover, environmental-biological community relationships are often masked by the highly dynamic nature of the fish communities in estuarine systems (Elliott and Hemingway, 2002). Nevertheless, our focus was not to strictly model these relationships, but rather attempt to understand the impact of disturbances on fish functional organization. 3.5. Integrating knowledge to support management recommendations In the present study, functional diversity was studied taking into account both a complementarity approach and a compositional identity approach (i.e., dissimilarity among traits quantified through indices and functional group analyses, respectively). Both approaches are complementary to the study of functional diversity and provide different types of information (Dolbeth et al. 2015; Gagic et al. 2015). Particularly with regard to the impact of environmental disturbances: how does disturbance affect the functional structure of communities, as synthesised by different complementary indices (e.g., Mouillot et al., 2013; van der Linden et al., 2016)?; which functional groups/traits contribute to such changes (e.g., Baptista et al., 2015)? We were able to synthesise the functional space and identify potential vulnerable areas in each estuary with the indices approach, while identifying the ways fish used each estuary with the functional group approach. The present results suggest that the fish communities in both estuaries share similar functions, which result in few functional groups representing different ways of using the resources in the systems. This finding was expected when taking into account the harsh physicochemical conditions of an estuarine system, which impose strong habitat selection. Within functional groups, laterally compressed fish were generally dominant in the Paraiba estuary, followed by cylindrical forms and flatfish, with the latter occurring only in downstream areas. In the Mamanguape estuary, biomass was slightly more evenly distributed among the different groups, but total fish biomass was also considerably lower than that in the Paraiba estuary. This could result from the smaller estuarine area and the downstream reef line, which limits the entrance of new fish from adjacent oceanic waters, resulting in a

Fig. 5. Functional identity groups from fish communities in both estuaries: A) PCO with trait measurements, with indication of groups and species identity from the “extreme form” groups; B) boxplots with measurements of each trait for each functional identity group (width of box proportional to number of species in each group); species in each functional group and ecological relevance of morphological traits can be consulted in Table 2S. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)



Mamanguape Upstream

Middle

Downstream

1800

1800

1600

1600

1400

1400

1200

1200

g/m2

g/m2

Paraiba 2000

2000

Dry Season

9

1000

Upstream

Middle

Downstream

1000

800

800

600

600

400

400

200

200 0

0 1

2

3

4

5

6

7 Sites

8

9

10

11

1

12

2

3

4

5

6

7

8

9 Sites

10

11

12

13

14

15

14

15

5629 g/m2 Upstream

Middle

2000

Downstream

1800

1800

1600

1600

1400

1400

1200

1200

g/m2

g/m2

Wet Season

2000

1000

Middle

Downstream

5018 g/m2

1000

800

800

600

600

400

400

200

200

0

Upstream

0 1

2

3

4

5

6 7 Sites

8

Functional types:

9

10

11

Flatfish

12

1

Cylindrical

2

3

4

5

6

Lateral

7

8 9 Sites

10

11

12

13

Extreme

Fig. 6. Cumulative biomass from each functional type group along the salinity gradient of Mamanguape and Paraiba estuaries in dry and wet seasons; at site 12 in Paraiba estuary, total biomass of laterally compressed fish was 5018 g m−2 and total fish community biomass was 5629 g m−2.

lower percentage of marine species in the Mamanguape estuary compared to the Paraiba, particularly in downstream areas. Such a limitation may contribute to the lower number of species and functional richness found in downstream areas of the Mamanguape estuary, as these indices are highly correlated (Schleuter et al., 2010). This result may also result from monitoring only the main channel, since, according to Xavier et al. (2012), estuarine margins in downstream areas of the Mamanguape estuary generally have greater diversity and abundance than the channel. As fish communities are highly dynamic, the interpretation of the variation patterns along the estuarine gradient should not be static. However, the results demonstrate that the upstream areas of both estuaries may be more prone to the effects of disturbances on the communities, with consequent impacts on the functioning of the overall ecosystem due to generally lower species and functional diversity coupled with low redundancy. In these areas, laterally compressed fish were generally dominant. However, sampling occurred in the main channel and not in vegetated areas (mangrove and seagrass), where potential fish diversity was not taken into account (Blaber, 2007, 2013; Xavier et al., 2012). Indeed, connectivity among mangrove, seagrass and reef areas seems to be highly important for fish communities of the Mamanguape estuary (Xavier et al., 2012). Nonetheless, fishes would need to have characteristics that would allow them to move in restricted areas, such as stability in the water column and good maneuverability (Weihs, 2002). These characteristics are mainly associated with laterally compressed fish, which was the group dominant in upstream and downstream areas. The middle section of the Mamanguape estuary and the downstream section of the Paraiba estuary exhibited richer communities with regard to functional diversity. For the particular case of the Paraiba estuary, this may be related to the decrease in nutrient concentrations in the downstream section. Additionally, we found a correlation between the decrease in SR'-FRED and increase in nutrient concentration. Although this correlation does not imply causation, a decrease in SR'FRED has been considered to be a response to increased disturbance (van der Linden et al., 2016), which supports the former hypothesis.

This compositional redundancy index was the metric that best reflected the potential spatial impact of nutrient enrichment in the Paraiba estuary. Moreover, one would expect to find greater differences in functional composition and organization between the two estuaries taking into account the considerable differences with regard to human pressures, resulting in substantially greater nutrient input in the Paraiba estuary. The present results were based on two sampling occasions and did not consider the structural complexity of the habitats, which may be critiqued in the context of an environmental impact assessment. However, the results suggest that downstream areas of the Paraiba estuary may exhibit slightly better ecological conditions regarding the biomass levels of the fish community, functional diversity and greater redundancy, allowing this system to recover from disturbances. Despite its conservation status, the Mamanguape estuary cannot be regarded as a reference system for a sustainable conservation unit, as it also subjected to human pressures, which are manifested in the fish community. The downstream reef line may also impose additional challenges due to the increase in water residence time in this sector. As mentioned before, greater fish diversity has been associated to the margins and habitat complexity of downstream areas of this estuary (Xavier et al., 2012), which were not considered in the present study. However, based on the results presented herein, the Mamanguape estuary seems to be more vulnerable to negative impacts stemming from human activities. Specific management measurements for both estuaries require better knowledge on the habitats and their connectivity (Barletta et al., 2010), as the anthropogenic activities described here often translate into habitat loss (either the habitat itself or its capacity to function as nursery or feeding grounds) (Martinelli and Filoso, 2008; Primavera, 2006; Sá et al., 2013). Besides habitat protection and recovery (e.g., vegetated habitats), the need for the regulation and better surveillance of human practices in coastal areas, environmental risk assessment considering the entire basin, the improvement of practices to reduce pollution (e.g., effluent treatment or bioremediation) (Olivera and Brito, 2005) and the involvement of the local population seem essential to the sustainable use and conservation of the natural resources in both


10


systems (e.g., Barletta et al., 2010; Martinelli and Filoso, 2008; Primavera, 2006; Barletta et al., 2010; Sá et al., 2013). Particularly for the Mamanguape estuary, a management plan was approved in 2014, with the definition of several priority conservation areas (Management plan for APA and mangrove areas - ARIE of Mamanguape River, Chico Mendes Institute – ICMBio 2014, www.icmbio.gov.br/portal/ planosmanejo). Thus, efforts should be directed at the application and surveillance of the approved plan. The local community has also raised important issues, such as the need for better public services, including basic sanitation, which would improve domestic sewage discharge in the estuarine areas (APA, ARIE – ICMBio 2014), demonstrating the concern of local communities with regard to preserving the system. Our main conclusion is that both estuaries need more effective conservation measures, especially the protected Mamanguape estuary, which seems more vulnerable to the anthropogenic impact. Acknowledgments This study was funded by the Brazilian Coordination for the Improvement of Higher Education Personnel (CAPES; Project no. 173/ 2012): “What lessons can be learned from ecological functioning in the estuarine systems of the state of Paraíba? An analysis of the effect of natural and anthropogenic disturbances”, under the Brazilian program Science without Borders (Special Visiting Researcher). We also acknowledge the grant awarded by the Portuguese Foundation for Science and Technology (FCT) attributed to M. Dolbeth (SFRH/BPD/110441/ 2015). We would like to thank S. Vital for providing the figure of the study site. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.marpolbul.2016.08.011. References Alvares, C.A., Stape, J.L., Sentelhas, P.C., Gonçalves, M.J.L., Sparovek, G., 2013. Köppen's climate classification map for Brazil. Meteorol. Z. 22, 711–728. Alves, R.R.N., Nishida, A.K., Hernández, M.I.M., 2005. Environmental perception of gatherers of the crab “caranguejo-uçá” (Ucides cordatus, Decapoda, Brachyura) affecting their collection attitudes. J. Ethnobiol. Ethnomed. 1, 10. http://dx.doi.org/10.1186/ 1746-4269-1-10. Alves, V.E.N., Patrício, J., Dolbeth, M., Pessanha, A., Palma, A.R.T., Dantas, E.W., Vendel, A.L., 2016. Do different degrees of human impact affect the diet of Atherinella brasiliensis (Brazilian silverside) in two Brazilian estuaries? J. Fish Biol. 89, 1239–1257. Anderson, M.J., Gorley, R.N., Clarke, K.R., 2008. PERMANOVA+ for PRIMER: Guide to Software and Statistical Methods. Primer-e, Plymouth, UK. Baptista, J., Martinho, F.M.D., Nyitrai, D., Pardal, M.A., Dolbeth, M., 2015. Long-term functional changes in an estuarine fish assemblage. Mar. Pollut. Bull. 97, 125–134. http:// dx.doi.org/10.1016/j.marpolbul.2015.06.025. Barletta, M., Jaureguizar, A.J., Baigun, C., Fontoura, N.F., Agostinho, A.A., Almeida-Val, V.M.F., Val, A.L., Torres, R.A., Jimenes-Segura, L.F., Giarrizzo, T., Fabré, N.N., Batista, V.S., Lasso, C., Taphorn, D.C., Costa, M.F., Chaves, P.T., Vieira, J.P., Corrêa, M.F.M., 2010. Fish and aquatic habitat conservation in South America: a continental overview with emphasis on neotropical systems. J. Fish Biol. 76, 2118–2176. http://dx.doi.org/ 10.1111/j.1095-8649.2010.02684.x. Blaber, S.J.M., 2007. Mangroves and fishes: issues of diversity, dependence, and dogma. Bull. Mar. Sci. 80, 457–472. Blaber, S.J.M., 2013. Fishes and fisheries in tropical estuaries: the last 10 years. Estuar. Coast. Shelf Sci. 135, 57–65. http://dx.doi.org/10.1016/j.ecss.2012.11.002. Campos, D.M.d.A.R., Silva, A.F.D., Sales, N.D.S., Oliveira, R.E.M.C.C., Pessanha, A.L.M., 2015. Trophic relationships among fish assemblages in a mudflat within Brazilian marine protected area. Braz. J. Oceanogr. 63, 135–146. http://dx.doi.org/10.1590/S167987592015091306302. Carmona, C.P., De Bello, F., Mason, W.H.N., Lepš, J., 2016. Traits without borders: integrating functional diversity across scales. Trends Ecol. Evol. 1–13. http://dx.doi.org/10. 1016/j.tree.2016.02.003. Claudino, M.C., Pessanha, A.L.M., Araújo, F.G., Garcia, A.M., 2015. Trophic connectivity and basal food sources sustaining tropical aquatic consumers along a mangrove to ocean gradient. Estuar. Coast. Shelf Sci. 167, 45–55. http://dx.doi.org/10.1016/j.ecss.2015.07. 005. Costanza, R., de Groot, R., Sutton, P., van der Ploeg, S., Anderson, S.J., Kubiszewski, I., Farber, S., Turner, R.K., 2014. Changes in the global value of ecosystem services. Glob. Environ. Chang. 26, 152–158. http://dx.doi.org/10.1016/j.gloenvcha.2014.04. 002.

de Bello, F., Lepš, J., Lavorel, S., Moretti, M., 2007. Importance of species abundance for assessment of trait composition: an example based on pollinator communities. Community Ecol. 8, 163–170. de Bello, F., Lavergne, S., Meynard, C.N., Lepš, J., Thuiller, W., 2010. The partitioning of diversity: showing Theseus a way out of the labyrinth. J. Veg. Sci. 21, 992–1000. de Bello, F., Carmona, C.P., Lepš, J., Szava-Kovats, R., Pärtel, M., 2016. Functional diversity through the mean trait dissimilarity: resolving shortcomings with existing paradigms and algorithms. Oecologia 1–9 http://dx.doi.org/10.1007/s00442-016-3546-0. Diegues, A.C., 1999. Human populations and coastal wetlands: conservation and management in Brazil. Ocean Coast. Manag. 42, 187–210. Dolbeth, M., Dolédec, S., Pardal, M.A., 2015. Relationship between functional diversity and benthic secondary production in a disturbed estuary. Mar. Ecol. Prog. Ser. 539, 33–46. http://dx.doi.org/10.3354/meps11473. Dolbeth, M., Vendel, A.L., Baeta, A., Pessanha, A., Patrício, J., 2016. Exploring ecosystem functioning in two Brazilian estuaries with fish richness, traits and food webs: building knowledge or increasing confusion? Mar. Ecol. Prog. Ser. (Submitted). Elliott, M., Hemingway, K.L., 2002. Fishes in Estuaries. Blackwell Science, Oxford. Gagic, V., Bartomeus, I., Jonsson, T., Taylor, A., Winqvist, C., Fischer, C., Slade, E.M., SteffanDewenter, I., Emmerson, M., Potts, S.G., Tscharntke, T., Weisser, W., Bommarco, R., 2015. Functional identity and diversity of animals predict ecosystem functioning better than species-based indices. Proc. R. Soc. B 282, 20142620. http://dx.doi.org/10. 1073/pnas.0704716104. Guillemot, N., Kulbicki, M., Chabanet, P., Vigliola, L., 2011. Functional redundancy patterns reveal non-random assembly rules in a species-rich marine assemblage. PLoS ONE 6, e26735. http://dx.doi.org/10.1371/journal.pone.0026735.s003. Jones, A.B., O'Donohue, M.J., Udy, J., Dennison, W.C., 2001. Assessing ecological impacts of shrimp and sewage effluent: biological indicators with standard water quality analyses. Estuar. Coast. Shelf Sci. 52, 91–109. Karpouzi, V.S., Stergiou, K.I., 2003. The relationships between mouth size and shape and body length for 18 species of marine fishes and their trophic implications. J. Fish Biol. 62, 1353–1365. http://dx.doi.org/10.1046/j.1095-8649.2003.00118.x. Lacerda, L.D., 2006. Inputs of nitrogen and phosphorus to estuaries of northeastern Brazil from intensive shrimp farming. Braz. J. Aquat. Sci. Technol. 10, 13–27. Lacerda, L.D., Soares, T.M., Costa, B.G.B., Godoy, M.D.P., 2011. Mercury emission factors from intensive shrimp aquaculture and their relative importance to the Jaguaribe River estuary, NE Brazil. Bull. Environ. Contam. Toxicol. 87, 657–661. http://dx.doi. org/10.1007/s00128-011-0399-4. Laliberté, E., Legendre, P., 2010. A distance-based framework for measuring functional diversity from multiple traits. Ecology 91, 299–305. Laliberté, E., Legendre, P., Shipley, B., 2015. Package “FD” (R software). Martinelli, L.A., Filoso, S., 2008. Expansion of sugarcane ethanol production in Brazil: environmental and social challenges. Ecol. Appl. 18, 885–898. Mason, W.H.N., Mouillot, D., Lee, W., Wilson, J.B., 2005. Functional richness, functional evenness and functional divergence: the primary components of functional diversity. Oikos 111, 112–118. Micheli, F., Halpern, B.S., 2005. Low functional redundancy in coastal marine assemblages. Ecol. Lett. 8, 391–400. http://dx.doi.org/10.1111/j.1461-0248.2005.00731.x. Mouillot, D., Graham, N.A.J., Villéger, S., Mason, W.H.N., Bellwood, D.R., 2013. A functional approach reveals community responses to disturbances. Trends Ecol. Evol. 28, 167–177. http://dx.doi.org/10.1016/j.tree.2012.10.004. Mouillot, D., Villéger, S., Parravicini, V., Kulbicki, M., Arias-González, J.E., Bender, M., Chabanet, P., Floeter, S.R., Friedlander, A., Vigliola, L., Bellwood, D.R., 2014. Functional over-redundancy and high functional vulnerability in global fish faunas on tropical reefs. Proc. Natl. Acad. Sci. U. S. A. 111, 13757–13762. Murray, F., Douglas, A., Solan, M., 2014. Species that share traits do not necessarily form distinct and universally applicable functional effect groups. Mar. Ecol. Prog. Ser. 516, 23–34. http://dx.doi.org/10.3354/meps11020. Nóbrega, G.N., Otero, X.L., Macías, F., Ferreira, T.O., 2014. Phosphorus geochemistry in a Brazilian semiarid mangrove soil affected by shrimp farm effluents. Environ. Monit. Assess. 186, 5749–5762. http://dx.doi.org/10.1007/s10661-014-3817-3. Olivera, A., Brito, L.O., 2005. Treating shrimp farming effluent using the native oyster, Crassostrea rhizophorae, in Brazil. World Aquacult. 60–63. Pessanha, A.L.M., Araújo, F.G., Oliveira, R.E.M.C.C., Silva, A.F.D., Sales, N.S., 2015. Ecomorphology and resource use by dominant species of tropical estuarine juvenile fishes. Neotrop. Ichthyol. 13, 401–412. http://dx.doi.org/10.1590/1982-0224-20140080. Piet, G.J., 1998. Ecomorphology of a size-structured tropical freshwater fish community. Environ. Biol. Fish 51, 67–86. Primavera, J.H., 2006. Overcoming the impacts of aquaculture on the coastal zone. Ocean Coast. Manag. 49, 531–545. Sá, T.D., de Sousa, R.R., Rocha, Í.R.C.B., de Lima, G.C., Costa, F.H.F., 2013. Brackish shrimp farming in northeastern Brazil: the environmental and socio-economic impacts and sustainability. NR 04, 538–550. http://dx.doi.org/10.4236/nr.2013.48065. R Core Team, 2016. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria URL https://www.R-project.org/. Schleuter, D., Daufresne, M., Massol, F., 2010. A user's guide to functional diversity indices. Ecol. Monogr. 80 (3), 469–484. Silvestre, L.C., dos Santos Farias, D.L., Silva, J.D., 2011. Diagnóstico dos impactos ambientais advindo de atividades antrópicas na APA da barra do rio Mamanguape. Enciclopédia Biosfera. Centro Científico Conhecer, Goiânia. Strong, J.A., Andonegi, E., Bizsel, K.C., Danovaro, R., Elliott, M., Franco, A., Garces, E., Little, S., Mazik, K., Moncheva, S., Papadopoulou, N., Patrício, J., Queirós, A.M., Smith, C., Stefanova, K., Solaun, O., 2015. Marine biodiversity and ecosystem function relationships: the potential for practical monitoring applications. Estuar. Coast. Shelf Sci. 161, 46–64. http://dx.doi.org/10.1016/j.ecss.2015.04.008. Stuart-Smith, R.D., Bates, A.E., Lefcheck, J.S., Duffy, J.E., Baker, S.C., Thomson, R.J., StuartSmith, J.F., Hill, N.A., Kininmonth, S.J., Airoldi, L., Becerro, M.A., Campbell, S.J.,


M. Dolbeth et al. / Marine Pollution Bulletin xxx (2016) xxx–xxx Dawson, T.P., Navarrete, S.A., Soler, G.A., Strain, E.M.A., Willis, T.J., Edgar, G.J., 2013. Integrating abundance and functional traits reveals new global hotspots of fish diversity. Nature 501, 539–542. http://dx.doi.org/10.1038/nature12529. van der Linden, P.V.D., Marchini, A., Dolbeth, M., Patrício, J., Veríssimo, H., Marques, J.C., 2016. The performance of trait-based indices in an estuarine environment. Ecol. Indic. 61, 378–389. http://dx.doi.org/10.1016/j.ecolind.2015.09.039. Verberk, W.C.E.P., van Noordwijk, C.G.E., Hildrew, A.G., 2013. Delivering on a promise: integrating species traits to transform descriptive community ecology into a predictive science. Freshw. Sci. 32, 531–547. http://dx.doi.org/10.1899/12-092.1. Villéger, S., Mason, W.H.N., Mouillot, D., 2008. New multidimensional functional diversity indices for a multifaceted framework in functional ecology. Ecology 89, 2290–2301. Villéger, S., Ramos Miranda, J., Flores Hernandez, D., Mouillot, D., 2010. Contrasting changes in taxonomic vs. functional diversity of tropical fish communities after habitat degradation. Ecol. Appl. 20, 1512–1522.

11

Violle, C., Navas, M., Vile, D., Kazakou, E., Fortunel, C., Hummel, I., Garnier, E., 2007. Let the concept of trait be functional! Oikos 116, 882–892. Weihs, D., 2002. Stability versus maneuverability in aquatic locomotion. Integr. Comp. Biol. 42, 127–134. http://dx.doi.org/10.1093/icb/42.1.127. Winemiller, K.O., Fitzgerald, D.B., Bower, L.M., Pianka, E.R., 2015. Functional traits, convergent evolution, and periodic tables of niches. Ecol. Lett. 18, 737–751. http://dx.doi. org/10.1111/ele.12462. Xavier, J.H.d.A., Cordeiro, C.A.M.M., Tenório, G.D., Diniz, A.d.F., Paulo-Júnior, E.P.N., Rosa, R.S., Rosa, I.L., 2012. Fish assemblage of the Mamanguape Environmental Protection Area, NE Brazil: abundance, composition and microhabitat availability along the mangrove-reef gradient. Neotrop. Ichthyol. 10, 109–122. http://dx.doi.org/10.1590/ S1679-62252012000100011. Zuur, A.F., Ieno, E.N., Elphick, C.S., 2009. A protocol for data exploration to avoid common statistical problems. Methods Ecol. Evol. 1, 3–14.
