Predicting productivity in tropical reservoirs: The roles of phytoplankton taxonomic and functional diversity

Predicting productivity in tropical reservoirs: The roles of phytoplankton taxonomic and functional diversity

Ecological Indicators 48 (2014) 428–435 Contents lists available at ScienceDirect Ecological Indicators journal homepage: www.elsevier.com/locate/ec...

752KB Sizes 0 Downloads 32 Views

Ecological Indicators 48 (2014) 428–435

Contents lists available at ScienceDirect

Ecological Indicators journal homepage: www.elsevier.com/locate/ecolind

Predicting productivity in tropical reservoirs: The roles of phytoplankton taxonomic and functional diversity Ana M.C. Santos a,b,c, * ,1, Fernanda M. Carneiro d,1, Marcus V. Cianciaruso c a Centro de Biologia Ambiental and Ce3C – Centre for Ecology, Evolution and Environmental Changes, Faculdade de Ciências da Universidade de Lisboa, Edifício C2, Campo Grande, 1749-016 Lisboa, Portugal b Department of Biogeography & Global Change, Museo Nacional de Ciencias Naturales (CSIC), C/José Gutiérrez Abascal 2, 28006 Madrid, Spain c Departamento de Ecologia, Instituto de Ciências Biológicas, Universidade Federal de Goiás, 74001-970 Goiânia, GO, Brazil d Núcleo de Educação Ambiental e Pesquisa em Biologia – NEAP-Bio Universidade Estadual de Goiás (UEG), Unidade Universitária de Iporá (UnU-Iporá), Bairro jardim Novo Horizonte 2, CEP 76200-000, Iporá, GO, Brazil

A R T I C L E I N F O

A B S T R A C T

Article history: Received 22 May 2014 Received in revised form 20 August 2014 Accepted 25 August 2014

Primary productivity is intimately linked with biodiversity and ecosystem functioning. Much of what is known today about such relationship has been based on the manipulation of species richness. Other facets of biodiversity, such as functional diversity, have been neglected within this framework, particularly in freshwater systems. We assess the adequacy of different diversity measures, from species richness and evenness, to functional groups richness and functional diversity indices, to predict primary productivity in 19 tropical reservoirs of central Brazil, built to generate hydroelectric energy. We applied linear mixed models (and model selection based on the Akaike’s information criterion) to achieve our goal, using chlorophyll-a concentration as a surrogate for primary productivity. A total of 412 species were collected in this study. Overall we found a positive relation between productivity and diversity, with functional evenness representing the only exception. The most parsimonious models never included functional group classifications, with at least one continuous measure of functional diversity being present in many models. The best model included only species richness and explained 24.1% of variability in productivity. We therefore advise the use of species richness as an indicator of productivity in tropical freshwater environments. However, since the productivity–diversity relationship is known to be scale dependent, we recommend the use of continuous measures of functional diversity in future biodiversity and ecosystem functioning studies, in order to be certain that all functional differences between communities are being accounted for. ã 2014 Elsevier Ltd. All rights reserved.

Keywords: Biodiversity Chlorophyll-a Ecosystem functioning Functional groups Linear mixed models Species richness

1. Introduction Unraveling the relationship between biodiversity and ecosystem functioning remains a primary focus of ecological research (Tilman et al., 1997, 2012; Mittelbach et al., 2001; Hooper et al., 2005). This topic has received much attention due to the widespread impacts of human activities on natural ecosystems

(e.g., Hooper and Vitousek, 1997; Isbell et al., 2013). One of the most recurrent topics in this research area is the study of primary productivity drivers, particularly biodiversity (e.g., Tilman et al., 1996; Corcoran et al., 2012; Isbell et al., 2013). Primary productivity, i.e., the intrinsic rate of increase in biomass in an ecosystem (Bellinger and Sigee, 2010), is usually used as a common proxy for ecosystem functioning because it is directly related to

Abbreviations: BEF, biodiversity and ecosystem functioning; FD, functional diversity; S, species richness; Simp, Simpson index; FGRich_R, number of functional groups defined based on Reynolds et al. (2002) classification; FGRich_K, number of functional groups defined based on Kruk et al. (2010) classification; Simp_R, evenness of the functional groups defined based on Reynolds et al. (2002) classification; Simp_K, evenness of the functional groups defined based on Kruk et al. (2010) classification; FR, functional richness (convex hull volume, Villéger et al., 2008); MFD, unweighted mean functional distance; FEve, functional evenness (Villéger et al., 2008); MFDDens, mean functional distance weighted by species density; LMM, linear mixed models. * Corresponding author. Present address: Department of Biogeography & Global Change, Museo Nacional de Ciencias Naturales (CSIC), C/José Gutiérrez Abascal 2, 28006 Madrid, Spain. Tel.: +34 914111328 (ext. 1212). E-mail addresses: [email protected], [email protected] (A.M.C. Santos). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.ecolind.2014.08.033 1470-160X/ ã 2014 Elsevier Ltd. All rights reserved.

A.M.C. Santos et al. / Ecological Indicators 48 (2014) 428–435

how resources are utilized in natural communities (Tilman, 1999). Indeed, many studies in the field of biodiversity and ecosystem functioning (BEF) are based on the assumption that diversity, particularly species richness, controls biomass production (e.g., Declerck et al., 2007; Cardinale et al., 2009; Korhonen et al., 2011). Many studies relating primary productivity and biodiversity indicate a positive relationship between these two variables, at least for plant groups (Tilman et al., 1996; Van Ruijven and Berendse, 2005; Ptacnik et al., 2008; Zimmerman and Cardinale, 2014). However, this relationship is not universal, and in some cases it can either be hump-shaped (e.g., Declerck et al., 2007; Mittelbach et al., 2001; but see Whittaker, 2010), negative or even non-significant (e.g., Waide et al., 1999; Schmidtke et al., 2010; Adler et al., 2011). Most of our current knowledge on BEF has come from terrestrial ecosystems, particularly grasslands (Tilman et al., 1997; Loreau et al., 2002), raising the question of whether existing results can be extended to other ecosystems. Indeed, only a small number of studies have considered other organisms such as the phytoplankton (e.g., Ptacnik et al., 2008; Korhonen et al., 2011; Corcoran and Boeing, 2012) and few have taken into consideration further facets of phytoplankton diversity apart from species richness (e.g., Griffin et al., 2009). Focusing BEF research mainly in only one type of ecosystems is indeed very limited, especially considering that most primary production on earth occurs in aquatic environments (Falkowski et al., 1998), where a high diversity can be encountered (Hutchinson, 1961). Undeniably, the unique features of aquatic ecosystems may offer insights that help understand the role of biodiversity in different ecosystem processes (Giller et al., 2004; Hortal et al., 2014). Indeed, some of the currently known hump-shaped relationships between species diversity and productivity come from studies in lacustrine systems (e.g., Dodson et al., 2000). Traditionally, many of the advances made in the BEF agenda have been based on the manipulation of species richness (e.g., Tilman, 1999; Mittelbach et al., 2001; Corcoran et al., 2012). However, many ecosystem level processes are affected by the functional attributes of the coexisting species and not by their identity (Hooper et al., 2005; Naeem and Wright, 2003). Therefore, one important limitation of this approach is that it wrongly assumes that all species contribute equally to biodiversity (Hooper et al., 2005; Magurran, 2004), ignoring the fact that species have different traits and ecological roles (Tilman et al., 1997; Díaz and Cabido, 2001; Petchey et al., 2004). Thus, the last two decades have seen a growing interest in understanding the relationship between species richness, functional diversity and the functioning and maintenance of community processes (e.g., Díaz and Cabido, 2001; Naeem and Wright, 2003; Cianciaruso, 2011). Functional diversity (FD) can be defined as “the value and its range, for the species present in an ecosystem, of those traits that influence one or more aspects of the functioning of an ecosystem” (Tilman, 2001). In practical terms, FD is a representation of how species are distributed in an n-dimensional space defined by functional traits (Petchey and Gaston, 2006). Because FD links species and individuals with functions they perform on the ecosystems, it constitutes a better candidate measure than species richness to explain community and ecosystem processes (Díaz and Cabido, 2001; Hooper et al., 2005). Classically, functional diversity has been measured as the number of functional groups present in an assemblage, i.e., functional group richness (FGR; e.g., Tilman et al., 1997; Díaz and Cabido, 2001; Tilman, 2001; Naeem and Wright, 2003). Functional groups are usually defined as sets of species that show similar responses to the environment or have similar effects on ecosystem processes (Tilman, 2001), therefore being a simplified alternative to the taxonomic approach (Padisák et al., 2009). Such groups can be

429

defined by experts using a priori knowledge on the species’ biological traits related to their ecological role (e.g., Reynolds et al., 2002), or by using multivariate analyses, like hierarchical classification (Legendre and Legendre, 1998), to identify clusters of species with similar traits (e.g., Weithoff, 2003; Kruk et al., 2010). However, as pointed out by Petchey and Gaston (2006), such approximation has several drawbacks. First, it is based on arbitrary decisions regarding which differences among organisms are functionally significant (Petchey and Gaston, 2006). Second, the number of functional groups can be greatly influenced by species richness (Petchey and Gaston, 2002). Finally, by using functional groups, one has to follow two assumptions that are rarely true: (i) all species within a particular group are functionally similar (i.e., are completely redundant); and (ii) species from different groups are equally different (i.e., are complementary). Several alternative continuous measures have been proposed for measuring FD. These have the advantage of not having so many limitations and do not require making as many assumptions and decisions as with the FGR approach (Petchey and Gaston, 2006; but see Villéger et al., 2008; Pavoine and Bonsal, 2011). Nowadays there is an increasing range of continuous trait-based diversity indices (see review in Petchey and Gaston, 2006; Pavoine and Bonsal, 2011) that focus on three components of FD: (i) functional richness – “the amount of space filled by species in the community”; (ii) functional evenness – the equitability of abundance distribution in filled niche space; and (iii) functional divergence – “the degree to which abundance distribution in niche space maximizes divergence in functional characters within the community” (Mason et al., 2005). In this work, we evaluated the relationship between productivity and different diversity measures in freshwater reservoirs of central Brazil, particularly focusing on phytoplankton species. Phytoplankton communities are known to be responsible for a large amount of the global primary production, largely participating in the carbon cycle (Falkowski et al., 1998). Also, they can be related not only with productivity but also with other environmental variables like available nutrients, water characteristics and the surrounding landscape (e.g., Carpenter, 2005; Nabout et al., 2006). They are indeed the ideal candidates for such type of studies as they have well defined traits that determine their ecological niche (Litchman and Klausmeier, 2008). Typically, patterns of diversity in freshwater systems and their relationship with productivity and the environment have been addressed through species diversity (e.g., Dodson et al., 2000; Ptacnik et al., 2008; Korhonen et al., 2011) and functional groups (e.g., Kruk et al., 2002; Hoyer et al., 2009). Despite the potential advantages of using continuous measures of FD, and the fact that species richness and FGR are often an inadequate surrogate for productivity, these have rarely been used on studies related to phytoplankton (Hortal et al., 2014; but see Griffin et al., 2009; Longhi and Beisner, 2010; Vogt et al., 2010). Also, few attempts have been made to understand the interplay of distinct measures of biodiversity and functional aspects of biodiversity (Petchey and Gaston, 2002). Here we aim to reverse this trend by evaluating which measure(s) of phytoplankton taxonomic and functional diversity, either based or not on density data (measured using functional groups or continuous indices), are the most appropriate for predicting productivity in tropical reservoirs of central Brazil. Although some previous studies have focused on the identification of surrogates for predicting phytoplankton’s richness, community composition and response to environmental variability (e.g., Carneiro et al., 2010, 2013; Gallego et al., 2012; Hu et al., 2013), as far as we know, this paper represents one of the first attempts for testing the performance of different diversity measures as predictors of productivity (see Vogt et al., 2010).

430

A.M.C. Santos et al. / Ecological Indicators 48 (2014) 428–435

2. Materials and methods 2.1. Survey data We collected the data in 19 reservoirs that were built to generate hydroelectric energy. These are located in Central Brazil (Goiás State, Fig. 1; see Appendix A for a more detailed description of the studied sites), in a region where two of the largest Brazilian basins are located: Araguaia-Tocantins and Paraná. Samples were gathered once during the dry season (July 2009), with the number of sampling points varying according to reservoir size (from two to six) in order horizontal heterogeneity, totalling 66 samples. Sampling consisted of collecting subsurface phytoplankton samples from a constant depth (ca. 40 cm depth) in 100 mL amber bottles and fixing them with an acetic Lugol solution (Vollenweider, 1974; Bicudo and Menezes, 2006). We estimated phytoplankton relative density (individual mL1) with an inverted Zeiss microscope at 1000x following Utermöhl (1958), while also noting settling units (cells, colonies, and filaments) in random fields (Uhelinger, 1964). We counted the samples in random fields and the minimum number of fields per sedimentation camera (that varied from 5 to 10 mL) followed the stabilization curve of the species number, which was obtained based on new species added to each counted field. In general, phytoplankton samples were identified to the species level using specific literature for each

taxonomic group (e.g., Bicudo and Menezes, 2006; Bourrelly, 1966, 1968, 1970; González, 1996; Komárek and Anagnostidis, 1998, 2005; Wehr et al., 2003). Water samples were collected in a similar manner at subsurface, and were subjected to subsequent chemical analyses. Chlorophyll-a was measured spectrophotometrically using a Whatman GF/C filter and acetone for extraction (Golterman et al., 1978). Since productivity and chlorophyll-a usually have a positive relationship, we have followed the common practice of using chlorophyll-a concentration as a surrogate of biomass (e.g., Huot et al., 2007; Ptacnik et al., 2008; Cardinale et al., 2009; Giordani et al., 2009; Bellinger and Sigee, 2010; Søndergaard et al., 2011), and therefore a proxy of phytoplankton productivity. 2.2. Measuring biodiversity For each sample we measured species richness (S) and the Simpson index (D) in its form 1-D (herein referred to as Simp), which takes into account species identity and evenness (Magurran, 2004). Species were assigned to two different functional groups classifications: (i) the updated proposal of Reynolds et al. (2002) classification made by Padisák et al. (2009) (herein mentioned as Reynolds’ classification for brevity) and (ii) Kruk et al. (2010). The main difference between these two classifications is that while the first one greatly relies on expert knowledge by incorporating

Fig. 1. Location of the sampling points in the Paraná (R1–R10) and Araguaia-Tocantins (R11–R19) basins, State of Goiás, Brazil.

A.M.C. Santos et al. / Ecological Indicators 48 (2014) 428–435

information on the ecologies of the species, the second is based solely on morphological aspects, being therefore simpler. For each sample we accounted for both the number of functional groups present (FGRich_R and FGRich_K, for the classifications based on Reynolds et al., 2002 and Kruk et al., 2010; respectively) and the evenness within such groups measured with the Simpson index (see details above; Simp_R and Simp_K, for the classifications based on Reynolds et al., 2002 and Kruk et al., 2010; respectively). Continuous measures of functional diversity were calculated using eight morphological, physiological and behavioral traits that were either easily measured or obtained from the literature. Functional traits can be defined as “any trait that impacts fitness indirectly via its effects on growth, reproduction and survival” (Violle et al., 2007). Traits can be classified as effect traits, i.e., traits that reflect the impact of species on ecosystem functioning, or as response traits, i.e., traits that influence the species ability to respond to environmental changes (Lavorel and Garnier, 2002). Therefore, traits usually correspond to morphological, physiological or behavioral features that are expressed by each individual and that are thought to be associated with the organism response to the environment or to how it affects ecosystem properties (Díaz et al., 2013). We selected traits that reflect the major ecological axes that define ecological niches of phytoplankton and affect species fitness (Litchman and Klausmeier, 2008). Such traits are related to different ecological functions, from reproduction, resource acquisition (light and nutrients) and predator/pathogen avoidance (see Table 1; Litchman and Klausmeier, 2008; also see Weithoff, 2003), ultimately being linked with phytoplankton productivity. Thus, for each species, we calculated the maximum linear dimension. Almost 30 individuals of each species were measured to this end. Individuals were then classified according to their form (unicellular, coenobium, colonial or filamentous). The presence of toxins, aerotopes, flagella, mucilage, siliceous exoskeletal structures and heterocysts were also noted for each individual as a binary trait, either by making use of literature sources or by direct observations of the individuals collected (Table 1). With the abovementioned traits we calculated a distance matrix using Gower distance (Pavoine et al., 2009) that was used to calculate different indices reflecting the multiple facets of functional diversity (Mason et al., 2005): functional richness, functional evenness, and functional divergence was obtained. Functional richness was measured through FR (convex hull volume; Villéger et al., 2008), while functional evenness was assessed through FEve (Villéger et al., 2008) and functional divergence using unweighted MFD (mean functional distance; this metric was adapted from MPD, which was originally used in community phylogenetic studies and that corresponds to the mean pairwise distance in the communities; Webb, 2000) and MFD weighted by species density (MFDDens). All indices were calculated in R (R development Core Team, 2012), using the functions dbFD in the package FD (Laliberté and Legendre, 2010), and mpd in the package picante (Kembel et al., 2010). To summarise, in total we considered 10 different diversity measures: 5 non-density-based indices (NDB indices herein; S, FR,

431

FGRich_R, FGRich_K, MFD) and 5 density-based indices (DB indices herein; Simp, Simp_R, Simp_K, FEve and MFDDens). 2.3. General data analyses The best diversity predictors of productivity (measured using chlorophyll-a concentration as a proxy) were selected through linear mixed models and model selection based on the Akaike’s information criterion (Burnham and Anderson, 2002). In order to normalize model residuals, chlorophyll-a concentration (Chlor_a) was log10 transformed (log(Chlor_a + 1)). The constant (+1) was added to allow the log transformation of chlorophyll-a concentrations recorded as 0. From here onwards, the abbreviation Chlor_a corresponds to the transformed values of chlorophyll-a concentration. All continuous predictors were standardized to mean = 0 and standard deviation = 1. An assumption of linear models is that the data points are all independent. However, as samples within a particular reservoir are not independent from each other it is necessary to use statistical methods that help overcome this problem. Linear mixed models (LMM) are ideal candidates for this; like linear models, they allow determining the effect of explanatory variables over a response variable, but they also include information on the structure of the data by considering two different types of explanatory variables: fixed (i.e., the variables for which we want to estimate the slope and/or intercept; in our case all the explanatory variables) and random effects (i.e., the grouping within the data that is used to drawn random samples from a population; in our case the reservoir and the basin). LMMs were applied separately to NDB and DB indices using a top-down approach for model selection (Zuur et al., 2009; Bunnefeld and Phillimore, 2012). In a first step, the best random effect structures (considering all the fixed effects) were selected by identifying the model with the lowest small-samples corrected Akaike’s information criterion (AICc; Burnham and Anderson, 2002). If the difference between a model’s AICc and the lowest AICc, the DAICc, was lower than 2, it was then assumed that particular model to be part of the set of best models. Akaike weights derived from the AICc (AICc-w) give the probability that a particular model is the best model, considering the data and candidate models available (Burnham and Anderson, 2002). Models were fit with the lme function of the lme4 package (ver. 0.999375-39) in R (R Development Core Team, 2012), using restricted maximum likelihood. We ran models with basin, reservoirs and reservoirs within basins (basin: reservoirs) as random effects, identifying the most parsimonious set of random effects. In a second step, it was necessary to identify the most parsimonious set of fixed effects, which was achieved using maximum likelihood methods. The dredge function in the MuMIn package in R was used to run a complete set of models with all possible combinations of the fixed effects and to identify the set of best models according to the criterion of DAICc < 2 (Burnham and

Table 1 Traits used to measure phytoplankton functional diversity and their relation to different ecological functions. Trait

Scale

Ecological funcion Reproduction

Resource aquisition

Avoidance

Maximum linear dimension Body form Toxins Aerotopes Flagella Mucilage Siliceous exoskeletal structures Heterocysts

Continuous Categorical (4 categories) Binary (presence/abscense) Binary (presence/abscense) Binary (presence/abscense) Binary (presence/abscense) Binary (presence/abscense) Binary (presence/abscense)

x x

x x

x x x x x x x x

x x x x x

432

A.M.C. Santos et al. / Ecological Indicators 48 (2014) 428–435

the model that included density-based indices) (Appendix E). Reservoir was then included as a random factor in all subsequent models used to select the best predictor(s) of chlorophyll-a. When considering only the relationship between non-densitybased indices and chlorophyll-a, the model with the lowest AICc included only species richness (S) as a predictor of the variability in chlorophyll-a concentration, with FGRich_K not being included in the best set of models (i.e., those with DAICc < 2) (Table 2). The averaged model, which included S, FR, FGRich_R and MFD, explained 25.3% of variability in chlorophyll-a concentration, and included S as the most important predictor. Regarding the density-based indices, the best subset of models (i.e., when DAICc < 2) always included functional evenness (FEve) as a predictor (presenting a negative relationship with chlorophylla), with Simp and Simp_K never being included in any of these models (Table 3). In fact, the model with the lowest AICc included the variables FEve and MFDDens and explained 21% of variability in chlorophyll-a. The averaged model included Simp_R, FEve and MFDDens and explained 21% of data variation, with FEve being the most important predictor for this model (Table 3). The most parsimonious models obtained from the combination of the previously selected density- and non-density-based indices are presented in Table 4. Species richness was identified in the most parsimonious model (that explained 24.1% of variability) as the best predictor of chlorophyll-a. The averaged model included S and MFDDens and explained 24.4% of variability in chlorophyll-a; in this case, species richness was the most important predictor (Table 4).

Anderson, 2002). We then performed a model averaging that returned the estimated coefficients and the relative importance of the predictor variables (calculated as the sum of the Akaike weights of all the models in which a particular predictor appears; Burnham and Anderson, 2002) for each subset of indices. Finally, the NDB and DB indices present in the most parsimonious model were subject altogether to the same steps, which allowed the selection of a final model based on both types of indices. The biodiversity indices included in this final model were considered to be the most adequate predictors of chorophyll-a, and therefore the best indices for studying the relationship between biodiversity and ecosystem productivity. One of the difficulties of working with LMMs is obtaining comparable R2 values with the same meaning as in simple or multiple linear regression (Zuur et al., 2009). Following Patiño et al. (2013), we used a R2 measure that compares the deviance of the LMM with the deviance of a linear intercept-only model (Kvalseth, 1985): 2  ^ 1S YY R2 ¼ 2 YY These R2 values were used as indicators of the proportion of the total variation in chlorophyll-a among samples that is explained by selected LMMs. 3. Results A total of 412 species were collected in this study, with species richness varying across all samples from 4 to 65 (Appendix B). These species were distributed among 26 functional groups defined based on Reynolds et al. (2002), and all seven groups delimited by Kruk et al. (2010). After verifying the relationship between chlorophyll-a and each measure of diversity individually we identified an obvious outlier: sample R16P3 (taken from reservoir São Domingos – R16) presented excessively low levels of species evenness (see Appendix C). Subsequent results were obtained without such sample. Overall, there was a positive relationship between diversity measures and chlorophyll-a concentration (Fig. 2(a); Appendix D), with the exception of functional evenness (measured with FEve) that presented a negative relationship (Fig. 2(b)). The best random effect structure (i.e., the one with the lowest AICc value) included reservoir as random effect for both the models (i.e., for the model that included non-density-based indices and for 0.8

In accordance with other studies focusing on phytoplankton (e.g., Ptacnik et al., 2008; Behl et al., 2011; Vogt et al., 2010), in general terms we found a positive linear relationship between diversity and productivity. For example, species richness showed a positive relationship with productivity, a trend that has been previously reported for this group (e.g., Ptacnik et al., 2008; Korhonen et al., 2011). The only exception to this positive trend was FEve, a continuous measure of functional diversity (FD) that quantifies “the regularity with which the functional space is filled by species, weighted by their abundances” (Villéger et al., 2008). A priori, one could expect that higher functional evenness would result in higher productivity, and even resilience, as species would use resources in a complementary manner (Mouillot et al., 2005). However, here we found that productivity was higher when 0.8

a)

0.7

Chlorophyll-a concentration

4. Discussion

0.6

0.6

0.5

0.5

0.4

0.4

0.3

0.3

0.2

0.2

0.1

0.1

0.0

0.0

-0.1

0

b)

0.7

10

20

30

40

50

60

70

-0.1 0.1

0.2

0.3

0.4

0.5

0.6

0.7

FEve

Species richness 2

Fig. 2. Relationship between chlorophyll-a concentration and (a) species richness (N = 65; R = 0.241; p < 0.001) and (b) functional evenness (FEve; N = 65; R2 = 0.117; 0.01 > p > 0.001).

A.M.C. Santos et al. / Ecological Indicators 48 (2014) 428–435

433

Table 2 Coefficients for the fixed non-density-based diversity predictors of chlorophyll-a concentration included in the most parsimonious (with DAICc < 2) and for the averaged models that considered reservoir as a random effect. The intercept, the number of parameters in the model (k), the AICc, AICc difference (DAICc) and Akaike weights derived from the AICc (AICc-W) are given for each model. Predictors (S, FR, FGRich_R, FGRich_K and FD) are as explained in Section 2. Values in parentheses correspond to the contribution (i.e., importance) of each variable calculated over the best set of models. The proportion of the total variation in chlorophyll-a concentration (R2) explained by the model is indicated. Model

Intercept

S

1 2 3 4 5 Averaged model

0.182 0.182 0.182 0.182 0.182 0.182

0.072 0.066 0.095 0.100 0.06 0.077 (1)

FR

FGRich_R

FGRich_K

MFD 0.020

0.027 0.043 0.016 0.002 (0.13)

0.028

0.010 (0.31)



0.009 (0.36)

k

AICc

DAICc

AICc-W

R2

4 5 5 6 5 –

84.6 83.5 82.9 82.9 82.6 –

0 1.02 1.62 1.7 1.97 –

0.206 0.124 0.092 0.088 0.077 –

0.24 0.244 0.244 0.257 0.241 0.253

Table 3 Coefficients for the fixed density-based diversity predictors of chlorophyll-a concentration included in the most parsimonious (with DAICc < 2) and for the averaged models that considered reservoir as a random effect. The intercept, the number of parameters in the model (k), the AICc, AICc difference (DAICc) and Akaike weights derived from the AICc (AICc-W) are given for each model. Predictors (Simp, Simp_R, Simp_K, FEve and MFDDens) are as explained in Section 2. Values in parentheses correspond to the contribution (i.e., importance) of each variable calculated over the best set of models. The proportion of the total variation in chlorophyll-a concentration (R2) explained by the model is indicated. Model

Intercept

1 2 3 Averaged model

0.181 0.181 0.181 0.181

Simp

Simp_R



0.04 0.019 0.017 (0.52)

Simp_K

FEve

MFDDens

k

AICc

DAICc

AICc-W

R2

0.045



0.041 0.035 0.04 0.039 (1)

5 5 6 –

78 77.2 76.3 –

0 0.81 1.72 –

0.165 0.011 0.07 –

0.2 0.191 0.212 0.21

0.03 0.028 (0.683)

Table 4 Coefficients for the best diversity predictors of chlorophyll-a concentration included in the non-density and density-based most parsimonious models (with DAICc < 2) that considered reservoir as a random effect. The intercept, the number of parameters in the model (k), the AICc, AICc difference (DAICc) and Akaike weights derived from the AICc (AICc-W) are given for each model. Predictors (S, FEve and MFDDens) are as explained in Section 2. Values in parentheses correspond to the contribution (i.e., importance) of each variable calculated over the best set of models. The proportion of the total variation in chlorophyll-a concentration (R2) explained by the model is indicated. Model

Intercept

S

1 2 Averaged model

0.182 0.182 0.182

0.079 0.065 0.070 (1)

FEve

MFDDens

k

AICc

DAICc

AICc-W

R2



0.019 0.007 (0.35)

4 5 –

84.6 83.3 –

0 1.23 –

0.481 0.260 –

0.241 0.241 0.244

functional distances among species were less regular, which might indicate that in these communities resources are being used more efficiently by species with particular traits (i.e., there is low functional regularity). Functional group richness is one of the most common measures of FD (Díaz and Cabido, 2001; Naeem and Wright, 2003), being also one of the most common measures of diversity used in phytoplankton related studies (e.g., Nabout et al., 2006; Hoyer et al., 2009; Behl et al., 2011). Paradoxically, it is also one of the diversity measures that requires the largest number of decisions and assumptions (Petchey and Gaston, 2006). Surprisingly, none of the functional groups classifications evaluated here were included in the most parsimonious models (nor as an important predictor in the averaged models), which could indicate that these measures are in fact less reliable surrogates of phytoplankton productivity than species richness and/or continuous functional diversity measures like FEve or MFDDens. Continuous measures of FD have the additional advantage of allowing us to get closer to understanding the processes shaping community assembly (Pavoine and Bonsall, 2011). Our knowledge about lakes and phytoplankton communities would certainly benefit from further developments on this line of research (Hortal et al., 2014; see e.g., Vogt et al., 2013). It is important to note that our results do not invalidate previous findings based on functional group classifications; we agree that these classifications can help simplify the complexity inherent to biological systems, while also reducing

time and effort devoted in ecological studies. However, as such classifications have many disadvantages, we believe that the use of functional groups for understanding BEF presents more limitations than benefits, and its use as surrogates of productivity should be cautiously evaluated for each particular study system. Species richness and MFDDens were the only two variables included in the final averaged model, which indicates that these are the most important predictors of productivity. The importance of FD measures as predictors of productivity may be explained by the fact that one of the mechanisms behind the BEP relationship might be niche complementarity, where species morphological and physiological differences enhance the use of different resources, consequently increasing overall productivity in species rich systems (Hooper et al., 1997; Loreau et al., 2002; Griffin et al., 2009). Indeed, functional divergence can lead to higher niche differentiation as the most abundant species have disparate traits and therefore do not tend to compete. However, and contrary to our expectations, the importance of species richness as an indicator of phytoplankton productivity overrides the importance of any of the FD measures considered. In agreement with our results, Griffin et al. (2009) and Vogt et al. (2010) also found that the identity of species were better predictors of the total magnitude of productivity than continuous measures of FD. Such findings might be related to a sampling effect, as highly diverse communities are more likely to include highly productive species (Loreau et al., 2002). Some studies have also suggested that

434

A.M.C. Santos et al. / Ecological Indicators 48 (2014) 428–435

ecosystem functioning might be related to species richness, at least on small spatio-temporal scales (Cardinale et al., 2004). All field, laboratory and statistical methodologies present limitations and advantages. Available methods for determining phytoplankton biomass are not an exception, and one can argue that chlorophyll-a concentration may not be the best proxy of phytoplankton biomass. Indeed, when using such approach it is necessary to make some assumptions: (i) the precise relationship between chlorophyll-a concentration and biomass usually is unknown for the particular conditions of the study system; and (ii) pigments of photosynthetic bacteria and of zooplankton that contain ingested algae could be altering total chlorophyll-a concentration, but such effect is usually thought to be minimal (Bellinger and Sigee, 2010). Alternative methods would be using biovolumes obtained from microscopic counts, or measuring dry weight; however obtaining such data can be time and resource consuming, and such information may not always be available, such as in our study. Still, we argue that our approach is valid, as many studies indicate that chlorophyll-a concentration can be used as a surrogate for biomass (Vörös and Padisák, 1991; Huot et al., 2007), and indeed have been widely used as so in many cases (e.g., Declerck et al., 2007; Ptacnik et al., 2008; Cardinale et al., 2009; Søndergaard et al., 2011). Finally, the statistical approach we have used based on mixed modelling has several advantages that could benefit BEF studies in the field of limnology. The main advantage is that allows using a larger sample size; this is possible because this method allows samples within a reservoir to be considered individually without the need of merging them and having only a single point representing each reservoir. This way it is possible to also take into account the variability that occurs within reservoirs and to deal with unbalanced sampling designs (e.g., different number of samples per reservoir), while avoiding the problems associated with rarefaction or extrapolation techniques (e.g., Gotelli and Colwell, 2001). We suggest that this approach could be used as standard methodology in future limnological studies with a nested sampling design. In summary, we found a significant relationship between diversity and chlorophyll-a concentration, which indicates that productivity is linked with taxonomic and functional diversity. Our findings indicate that species richness is the most parsimonious predictor of phytoplankton productivity, at least for small spatial and temporal scales. Measuring functional traits to obtain continuous FD measures involve considerably more time and resources than simply using species richness as a diversity measure. However, the fact that species richness was here identified as the best predictor of productivity could mislead researchers into discarding the importance of FD measures; we argue that FD indices might provide important insights in the BEF agenda and therefore should not be neglected in such studies. Moreover, since the diversity–productivity relationship might change across geographical and temporal scales (e.g., Cardinale et al., 2004; Korhonen et al., 2011) we advise researchers to verify which can potentially be the best diversity measure for their particular goal and scale at which their study is being conducted, concentrating particularly on continuous measures of FD. This way one increases the chances of accounting for all functional differences between communities. Acknowledgements We want to thank Luis Mauricio Bini, Paulo De Marco Júnior, Joaquín Hortal, Jorge Lobo, and Alberto Jiménez-Valverde for their help and suggestions. This research was partially funded by the Brazilian CNPq (projects 567794/2008-3 and 562756/2008-6). A. M.C.S. was supported by a Brazilian CNPq junior postdoctoral fellowship (159763/2010-0) and a Portuguese Fundação para a

Ciência e a Tecnologia grant (SFRH/BPD/70709/2010) co-funded by the European Social Fund POPH-QREN program. F.M.C. was supported by CNPq doctoral fellowship (562756/2008-6). M.V.C. has a productivity grant from CNPq (306843/2012-9). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ecolind.2014.08.033. References Adler, P.B., Seabloom, E.W., Borer, E.T., Hillebrand, H., Hautier, Y., Hector, A., Harpole, W.S., O’Halloran, L.R., Grace, J.B., Anderson, T.M., Bakker, J.D., Biederman, L.A., Brown, C.S., Buckley, Y.M., Calabrese, L.B., Chu, C.-J., Cleland, E.E., Collins, S.L., Cottingham, K.L., Crawley, M.J., Damschen, E.I., Davies, K.F., Decrappeo, N.M., Fay, P.A., Firn, J., Frater, P., Gasarch, E.I., Gruner, D.S., Hagenah, N., Hille Ris Lambers, J., Humphries, H., Jin, V.L., Kay, A.D., Kirkman, K.P., Klein, J.A., Knops, J. M.H., La Pierre, K.J., Lambrinos, J.G., Li, W., Macdougall, A.S., Mcculley, R.L., Melbourne, B.A., Mitchell, C.E., Moore, J.L., Morgan, J.W., Mortensen, B., Orrock, J. L., Prober, S.M., Pyke, D.A., Risch, A.C., Schuetz, M., Smith, M.D., Stevens, C.J., Sullivan, L.L., Wang, G., Wragg, P.D., Wright, J.P., Yang, L.H., 2011. Productivity is a poor predictor of plant species richness. Science 333, 1750–1753. Behl, S., Donval, A., Stibor, H., 2011. The relative importance of species diversity and functional group diversity on carbon uptake in phytoplankton communities. Limnol. Oceanogr. 56, 683–694. Bellinger, E.G., Sigee, D., 2010. Freshwater Algae: Identification and Use Bioindicators. Wiley-Blackwell, Chichester, West Sussex. Bicudo, C.E.M., Menezes, M., 2006. Gêneros de Algas de Águas Continentais do Brasil. RiMa, São Carlos. Bourrelly, P., 1968. Les Algues d’ Eau Douce. Tome II. Les Algues Jaunes et Brunes, Chrysophycées, Phéophycés, Xanthophycées et Diatomées, N. Boubée et Cie Paris. Bourrelly, P., 1970. Les Algues d’. Tome III. Les Algues Bleues et Rouges, Les Eugléniens, Peridiniens et Cryptomonadines, N. Boubée et Cie Paris. Bourrelly, P., 1966. Les Algues d’. Tome I. Les Algues Vertes., N. Boubée et Cie Paris. Bunnefeld, N., Phillimore, A.B., 2012. Island, archipelago and taxon effects: mixed models as a means of dealing with the imperfect design of nature’s experiments. Ecography 35, 15–22. Burnham, K.P., Anderson, D.R., 2002. Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach. Springer-Verlag, New York. Cardinale, B.J., Hillebrand, H., Harpole, W.S., Gross, K., Ptacnik, R., 2009. Separating the influence of resource availability from resource imbalance on productivity– diversity relationships. Ecol. Lett. 12, 475–487. Cardinale, B.J., Ives, A.R., Inchausti, P., 2004. Effects of species diversity on the primary productivity of ecosystems: extending our spatial and temporal scales of inference. Oikos 104, 437–450. Carneiro, F.M., Bini, L.M., Rodrigues, L.C., 2010. Influence of taxonomic and numerical resolution on the analysis of temporal changes in phytoplankton communities. Ecol. Indic. 10, 249–255. Carneiro, F.M., Nabout, J.C., Vieira, L.C.G., Lodi, S., Bini, L.M., 2013. Higher taxa predict plankton beta-diversity patterns across an eutrophication gradient. Nat. Conserv. 11, 43–47. Carpenter, S.R., 2005. Eutrophication of aquatic ecosystems: bistability and soil phosphorus. Proc. Natl. Acad. Sci. 102, 10002–10005. Cianciaruso, M.V., 2011. Beyond taxonomical space: large-scale ecology meets functional and phylogenetic diversity. Front. Biogeogr. 3, 87–90. Corcoran, A.A., Boeing, W.J., 2012. Biodiversity increases the productivity and stability of phytoplankton communities. PLoS One 7, e49397. Declerck, F., Vanderstukken, M., Pals, A., Muylaert, K., De Meester, L., 2007. Plankton biodiversity along a gradient of productivity and its mediation by macrophytes. Ecology 88, 2199–2210. Díaz, S., Cabido, M., 2001. Vive la différence: plant functional diversity matters to ecosystem processes. Trends Ecol. Evol. 16, 646–655. Díaz, S., Purvis, A., Cornelissen, J.H.C., Mace, G.M., Donoghue, M.J., Ewers, R.M., Jordano, P., Pearse, W.D., 2013. Functional traits, the phylogeny of function, and ecosystem service vulnerability. Ecol. Evol. 3, 2958–2975. Dodson, S.I., Arnott, S.E., Cottingham, K.L., 2000. The relationship in lake communities between primary productivity and species richness. Ecology 81, 2662–2679. Falkowski, P.G., Barber, R.T., Smetacek, V., 1998. Biogeochemical controls and feedbacks on ocean primary production. Science 281, 200–206. Gallego, I., Davidson, T.A., Jeppesen, E., Pérez-Martínez, C., Sánchez-Castillo, P., Juan, M., Fuentes-Rodríguez, F., León, D., Peñalver, P., Toja, J., Casas, J., 2012. Taxonomic or ecological approaches? Searching for phytoplankton surrogates in the determination of richness and assemblage composition in ponds. Ecol. Indic. 18, 575–585. Giller, P.S., Hillebrand, H., Berninger, U.-G., Gessner, M.O., Hawkins, S., Inchausti, P., Inglis, C., Leslie, H., Malmqvist, B., Monaghan, M.T., Morin, P.J., O'Mullan, G., 2004. Biodiversity effects on ecosystem functioning: emerging issues and their experimental test in aquatic environments. Oikos 104, 423–436.

A.M.C. Santos et al. / Ecological Indicators 48 (2014) 428–435 Giordani, G., Zaldívar, J.M., Viaroli, P., 2009. Simple tools for assessing water quality and trophic status in transitional water ecosystems. Ecol. Indic. 9, 982–991. Golterman, H.L., Clymo, R.S., Ohmstad, M., 1978. Methods for Physical and Chemical Analysis of Freshwaters. Blackwell Scientific Publication, Oxford. González, A.C., 1996. Las Chlorococcales dulciacuícolas de Cuba. Bibliotheca Phycologica 99, 1–192. Gotelli, N.J., Colwell, R.K., 2001. Quantifying biodiversity: procedures and pitfalls in the measurement and comparison of species richness. Ecol. Lett. 4, 379–391. Griffin, J.N., Méndez, V., Johnson, A.F., Jenkins, S.R., Foggo, A., 2009. Functional diversity predicts overyielding effect of species combination on primary productivity. Oikos 118, 37–44. Hooper, D.U., Chapin III, F.S., Ewel, Hector, J.J., Inchausti, A., Lavorel, P., Lawton, S., Lodge, J.H., Loreau, D.M., Naeem, M., Schmid, S., Setälä, B., Symstad, H., Vandermeer, A.J., Wardle, J., 2005. Effects of biodiversity on ecosystem functioning: a consensus of current knowledge. Ecol. Monogr. 75, 3–35. Hooper, D.U., Vitousek, P.M., 1997. The effects of plant composition and diversity on ecosystem processes. Science 277, 1302–1305. Hortal, J., Nabout, J.C., Calatayud, J., Carneiro, F.M., Padial, A., Santos, A.M.C., Siqueira, T., Bokma, F., Bini, L.M., Ventura, M., 2014. Perspectives on the use of lakes and ponds as model systems for macroecological research. J. Limnol. 73, 46–60. Hoyer, A.B., Moreno-Ostos, E., Vidal, J., Blanco, J.M., Palomino-Torres, R.L., Basanta, A., Escot, C., Rueda, F.J., 2009. The influence of external perturbations on the functional composition of phytoplankton in a Mediterranean reservoir. Hydrobiologia 636, 49–64. Hu, R., Han, B., Naselli-Flores, L., 2013. Comparing biological classifications of freshwater phytoplankton: a case study from south china. Hydrobiologia 701, 219–233. Huot, Y., Babin, M., Bruyant, F., Grob, C., Twardowski, M.S., Claustre, H., 2007. Does chlorophyll-a provide the best index of phytoplankton biomass for primary productivity studies? Biogeosci. Discuss. 4, 707–745. Hutchinson, G.E., 1961. The paradox of the plankton. Am. Nat. 95, 137–145. Isbell, F., Reich, P.B., Tilman, D., Hobbie, S.E., Polasky, S., Binder, S., 2013. Nutrient enrichment, biodiversity loss, and consequent declines in ecosystem productivity. Proc. Natl. Acad. Sci. 110, 11911–11916. Kembel, S.W., Cowan, P.D., Helmus, M.R., Cornwell, W.K., Morlon, H., Ackerly, D.D., Blomberg, S.P., Webb, C.O., 2010. Picante: R tools for integrating phylogenies and ecology. Bioinformatics 26, 1463–1464. Komárek, J., Anagnostidis, K., 1998. Cyanoprokaryota I. Teil chroococcales. In: Ettl, H., Gartner, G., Heynig, H., Mollenhauer, D. (Eds.), Susswasserflora Von Mitteleuropa 19/1. G. Fischer, Jena, pp. 548. Komárek, J., Anagnostidis, K., 2005. Cyanoprokaryota II. Teil oscillatoriales. In: Büdel, B., Gärtner, G., Krienitz, L., Schagerl, M. (Eds.), Susswasserflora Von Mitteleuropa 19/2. Elsevier, Heidelberg, pp. 759. Korhonen, J.J., Wang, J., Soininen, J., 2011. Productivity–diversity relationships in lake plankton communities. PLoS One 6, e22041. Kruk, C., Huszar, V.L.M., Peeters, E.T.H.M., Bonilla, S., Costa, L., Lürling, M., Reynolds, C.S., Scheffer, M., 2010. A morphological classification capturing functional variation in phytoplankton. Freshw. Biol. 55, 614–627. Kruk, C., Mazzeo, N., Lacerot, G., Reynolds, C.S., 2002. Classification schemes for phytoplankton: a local validation of a functional approach to the analysis of species temporal replacement. J. Plankton Res. 24, 901–912. Kvalseth, T.O., 1985. Cautionary note about R2. Am. Stat. 39, 279–285. Laliberté, E., Legendre, P., 2010. A distance-based framework for measuring functional diversity from multiple traits. Ecology 91, 299–305. Lavorel, S., Garnier, E., 2002. Predicting changes in community composition and ecosystem functioning from plant traits: revisiting the Holy Grail. Funct. Ecol. 16, 545–556. Legendre, P., Legendre, L., 1998. Numerical Ecology. Elsevier, Amsterdam. Litchman, E., Klausmeier, C.A., 2008. Trait-based community ecology of phytoplankton. Annu. Rev. Ecol. Evol. Syst. 39, 615–639. Longhi, M.L., Beisner, B.E., 2010. Patterns in taxonomic and functional diversity of lake phytoplankton. Freshw. Biol. 55, 1349–1366. Loreau, M., Naeem, S., Inchausti, P., 2002. Biodiversity and ecosystem functioning. Synthesis and Perspectives. Oxford University Press, Oxford. Magurran, A.E., 2004. Measuring Biological Diversity. Blackwell, Oxford. Mason, N.W.H., Mouillot, D., Lee, W.G., Wilson, J.B., 2005. Functional richness: functional evenness and functional divergence: the primary components of functional diversity. Oikos 111, 112–118. Mittelbach, G.G., Steiner, C.F., Scheiner, S.M., Gross, K.L., Reynolds, H.L., Waide, R.B., Willig, M.R., Dodson, S.I., Gough, L., 2001. What is the observed relationship between species richness and productivity? Ecology 82, 2381–2396. Mouillot, D., Mason, W.H.N., Dumay, O., Wilson, J.B., 2005. Functional regularity: a neglected aspect of functional diversity. Oecologia 142, 353–359. Nabout, J.C., Nogueira, I.S., Oliveira, L.G., 2006. Phytoplankton community of foodplain lakes of the Araguaia River, Brazil, in the rainy and dry seasons. J. Plankton Res. 28, 181–193. Naeem, S., Wright, J.P., 2003. Disentangling biodiversity effects on ecosystem functioning: deriving solutions to a seemingly insurmountable problem. Ecol. Lett. 6, 567–579. Padisák, J., Crossetti, L.O., Naselli-Flores, L., 2009. Use and misuse in the application of the phytoplankton functional classification: a critical review with updates. Hydrobiologia 621, 1–19.

435

Patiño, J., Guilhaumon, F., Whittaker, R.J., Triantis, K.A., Gradstein, S.R., Hedenäs, L., González-Mancebo, J.M., Vanderpoorten, A., 2013. Accounting for data heterogeneity in patterns of biodiversity: an application of linear mixed effect models to the oceanic island biogeography of spore-producing plants. Ecography 36, 904–913. Pavoine, S., Bonsall, M.B., 2011. Measuring biodiversity to explain community assembly: a unified approach. Biol. Rev. 86, 792–812. Pavoine, S., Vallet, J., Dufour, A.-B., Gachet, S., Daniel, H., 2009. On the challenge of treating various types of variables: application for improving the measurement of functional diversity. Oikos 118, 391–402. Petchey, O.L., Gaston, K.J., 2002. Functional diversity (FD), species richness and community composition. Ecol. Lett. 5, 402–411. Petchey, O.L., Gaston, K.J., 2006. Functional diversity: back to basics and looking forward. Ecol. Lett. 9, 741–758. Petchey, O.L., Hector, A., Gaston, K.J., 2004. How do different measures of functional diversity perform? Ecology 85, 847–857. Ptacnik, R., Solimini, A.G., Andersen, T., Tamminen, T., Brettum, P.L., Lepistö, L., Willén, E., Rekolainen, S., 2008. Diversity predicts stability and resource use efficiency in natural phytoplankton communities. Proc. Natl. Acad. Sci. 105, 5134–5138. R Development Core Team, 2012. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Reynolds, C.S., Huszar, V.L.M., Kruk, C., Naeselli-Flores, L., Melo, S., 2002. Towards a functional classification of the freshwater phytoplankton. J. Plankton Res. 24, 417–428. Schmidtke, A., Gaedke, U., Weithoff, G., 2010. A mechanistic basis for underyielding in phytoplankton communities. Ecology 91, 212–221. Søndergaard, M., Larsen, S.E., Jørgensen, T.B., Jeppesen, E., 2011. Using chlorophyll a and cyanobacteria in the ecological classification of lakes. Ecol. Indic. 11, 1403–1412. Tilman, D., 1999. The ecological consequences of changes in biodiversity: a search for general principles. Ecology 80, 1455–1474. Tilman, D., 2001. Functional diversity. In: Levins, S.A. (Ed.), Encyclopedia of Biodiversity. Academic Press, San Diego, CA, pp. 109–120. Tilman, D., Knops, J., Wedin, D., Reich, P., Ritchie, M., Siemann, E., 1997. The influence of functional diversity and composition on ecosystem processes. Science 277, 1300–1302. Tilman, D., Reich, P.B., Isbell, F., 2012. Biodiversity impacts ecosystem productivity as muchasresourcesdisturbance,orherbivory.Proc.Natl.Acad.Sci.109,10394–10397. Tilman, D., Wedin, D., Knops, J., 1996. Productivity and sustainability influenced by biodiversity in grassland ecosystems. Nature 379, 718–720. Uhelinger, V., 1964. Étude statistique des méthodes de dénobrement planctonique. Arch. Sci. 17, 121–233. Utermöhl, H., 1958. Zur vervollkomnung der quantitativen phytoplanktonmethodik. Mitteilungen Int. Vereinigung Theoretische Angew. Limnol. 9, 1–38. Van Ruijven, J., Berendse, F., 2005. Diversity–productivity relationships: initial effects, long-term patterns, and underlying mechanisms. Proc. Natl. Acad. Sci. 102, 695–700. Villéger, S., Mason, N., Mouillot, D., 2008. New multidimensional functional diversity indices for a multifaceted framework in functional ecology. Ecology 89, 2290–2301. Violle, C., Navas, M.L., Vile, D., Kazakou, E., Fortunel, C., Hummel, I., Garnier, E., 2007. Let the concept of trait be functional!. Oikos 116, 882–892. Vogt, R.J., Beisner, B.E., Prairie, Y.T., 2010. Functional diversity is positively associated with biomass for lake diatoms. Freshw. Biol. 55, 1636–1646. Vogt, R.J., Peres-Neto, P.R., Beisner, B.E., 2013. Using functional traits to investigate the determinants of crustacean zooplankton community structure. Oikos 122, 1700–1709. Vollenweider, R.A., 1974. A Manual on Methods for Measuring Primary Production in Aquatic Environments. Blackwell Scientific Publishers, Oxford. Vörös, L., Padisák, J., 1991. Phytoplankton biomass and chlorophyll-a in some shallow lakes in central Europe. Hydrobiologia 215, 111–119. Waide, R.B., Willig, M.R., Steiner, C.F., Mittelbach, G., Gough, L., Dodson, S.I., Juday, G. P., Parmenter, R., 1999. The relationship between productivity and species richness. Annu. Rev. Ecol. Syst. 30, 257–300. Webb, C.O., 2000. Exploring the phylogenetic structure of ecological communities: an example for rain forest trees. Am. Nat. 156, 145–155. Wehr, J.D., Sheath, R.G., 2003. Freshwater Algae of North America: Ecology and Classification. Academic Press, New York. Weithoff, G., 2003. The concepts of ‘plant functional types’ and ‘functional diversity’ in lake phytoplankton – a new understanding of phytoplankton ecology? Freshw. Biol. 48, 1669–1675. Whittaker, R.J., 2010. Meta-analyses and mega-mistakes: calling time on metaanalysis of the species richness—productivity relationship. Ecology 91, 2522–2533. Zimmerman, E.K., Cardinale, B.J., 2014. Is the relationship between algal diversity and biomass in North American lakes consistent with biodiversity experiments? Oikos 123, 267–278. Zuur, A.F., Ieno, E.N., Walker, N.J., Saveliev, A.A., Smith, G.M., 2009. Mixed Effects Models and Extensions in Ecology with R. Springer, New York.