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: Primary productivity is intimately linked with biodiversity and ecosystem functioning. Much of what is
Received 22 May 2014 known today about such relationship has been based on the manipulation of species richness. Other
Received in revised form 20 August 2014 facets of biodiversity, such as functional diversity, have been neglected within this framework,
Accepted 25 August 2014
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
Keywords: productivity in 19 tropical reservoirs of central Brazil, built to generate hydroelectric energy. We applied
Biodiversity
linear mixed models (and model selection based on the Akaike’s information criterion) to achieve our
Chlorophyll-a
Ecosystem functioning
goal, using chlorophyll-a concentration as a surrogate for primary productivity. A total of 412 species
Functional groups were collected in this study. Overall we found a positive relation between productivity and diversity, with
Linear mixed models functional evenness representing the only exception. The most parsimonious models never included
Species richness 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.

1. Introduction
Unraveling the relationship between biodiversity and ecosys-
tem 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: ana.margarida.c.santos@gmail.com, amcsantos@mncn.csic.es (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 consider-
ation 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 produc-
tivity 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 participat-
ing in the carbon cycle (Falkowski et al., 1998). Also, they can be
related not only with productivity but also with other environ-
mental 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 (Vollen-
weider, 1974; Bicudo and Menezes, 2006). We estimated phyto-
plankton relative density (individual mL
1) 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
Fig. 1. Location of the sampling points in the Paraná (R1–R10) and Araguaia-Tocantins (R11–R19) basins, State of Goiás, Brazil.
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 subsur-
face, 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
 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, physiologi-
cal 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 acqui-
sition (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 (unicellu-
lar, 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,
Table 1
Traits used to measure phytoplankton functional diversity and their relation to different ecological functions.
Trait Scale
Maximum linear dimension Continuous
Body form Categorical (4 categories)
Toxins Binary (presence/abscense)
Aerotopes Binary (presence/abscense)
Flagella Binary (presence/abscense)
Mucilage Binary (presence/abscense)
Siliceous exoskeletal structures Binary (presence/abscense)
Heterocysts Binary (presence/abscense)
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 concen-
trations 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
Ecological funcion
Reproduction Resource aquisition Avoidance
x x x
x x x
x
x x
x x
x x
x x
x x
432
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
1
S Y
Y ^ most important predictor for this model (Table 3).
R2 ¼
2
Y
Y
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
0.7
a)
0.6
Chlorophyll-a concentration
0.5
0.4
0.3
0.2
0.1
0.0
-0.1
0 10 20 30
Species richness
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
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-density-
based 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 chlorophyll-
a), 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
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).
4. Discussion
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
0.7
b)
0.6
0.5
0.4
0.3
0.2
0.1
0.0
-0.1
40 50 60 70 0.1 0.2 0.3 0.4 0.5 0.6 0.7
FEve
2
 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 FR FGRich_R FGRich_K MFD k AICc DAICc AICc-W R2

1 0.182 0.072 4
84.6 0 0.206 0.24
2 0.182 0.066 0.020 5
83.5 1.02 0.124 0.244
3 0.182 0.095
0.027 5
82.9 1.62 0.092 0.244
4 0.182 0.100
0.043 0.028 6
82.9 1.7 0.088 0.257
5 0.182 0.06 0.016 5
82.6 1.97 0.077 0.241
Averaged model 0.182 0.077 (1) 0.002 (0.13)
0.010 (0.31) – 0.009 (0.36) – – – – 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 Simp Simp_R Simp_K FEve MFDDens k AICc DAICc AICc-W R2
1 0.181
0.041 0.045 5
78 0 0.165 0.2
2 0.181 0.04
0.035 5
77.2 0.81 0.011 0.191
3 0.181 0.019
0.04 0.03 6
76.3 1.72 0.07 0.212
Averaged model 0.181 – 0.017 (0.52) –
0.039 (1) 0.028 (0.683) – – – – 0.21

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 FEve MFDDens k AICc DAICc AICc-W R2

1 0.182 0.079 4
84.6 0 0.481 0.241
2 0.182 0.065 0.019 5
83.3 1.23 0.260 0.241
Averaged model 0.182 0.070 (1) – 0.007 (0.35) – – – – 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 classifica-
tions; 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.eco-
lind.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 produc-
tivity. 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 phyto-
plankton. 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 phytoplankton-
methodik. 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 meta-
analysis 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 experi-
ments? 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.