Oecologia https://doi.org/10.1007/s00442-018-4130-6
COMMUNITY ECOLOGY – ORIGINAL RESEARCH
Amino acid composition reveals functional diversity of zooplankton in tropical lakes related to geography, taxonomy and productivity Nelson J. Aranguren‑Riaño1 · Cástor Guisande2 · Jonathan B. Shurin3 · Natalie T. Jones3 · Aldo Barreiro4 · Santiago R. Duque5 Received: 13 September 2017 / Accepted: 24 March 2018 © Springer-Verlag GmbH Germany, part of Springer Nature 2018
Abstract Variation in resource use among species determines their potential for competition and co-existence, as well as their impact on ecosystem processes. Planktonic crustaceans consume a range of micro-organisms that vary among habitats and species, but these differences in resource consumption are difficult to characterize due to the small size of the organisms. Consumers acquire amino acids from their diet, and the composition of tissues reflects both the use of different resources and their assimilation in proteins. We examined the amino acid composition of common crustacean zooplankton from 14 tropical lakes in Colombia in three regions (the Amazon floodplain, the eastern range of the Andes, and the Caribbean coast). Amino acid composition varied significantly among taxonomic groups and the three regions. Functional richness in amino acid space was greatest in the Amazon, the most productive region, and tended to be positively related to lake trophic status, suggesting the niche breadth of the community could increase with ecosystem productivity. Functional evenness increased with lake trophic status, indicating that species were more regularly distributed within community-wide niche space in more productive lakes. These results show that zooplankton resource use in tropical lakes varies with both habitat and taxonomy, and that lake productivity may affect community functional diversity and the distribution of species within niche space. Keywords Niche breadth · Crustaceans · Plankton · Productivity · Tropical lakes
Introduction Communicated by Ulrich Sommer. Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00442-018-4130-6) contains supplementary material, which is available to authorized users. * Jonathan B. Shurin
[email protected] 1
Unidad de Ecología en Sistemas Acuáticos UDESA, Universidad Pedagógica y Tecnológica de Colombia, Avenida Central del Norte 39‑115, Tunja, Boyacá, Colombia
2
Departamento de Ecología y Biología Animal, Universidad de Vigo, 36310 Vigo, Spain
3
Section of Ecology, Behavior and Evolution, University of California San Diego, 9500 Gilman Dr. #0116, La Jolla, CA 92093, USA
4
Centro Interdisciplinar de Investigação Marinha e Ambiental (CIIMAR), Avenida General Norton de Matos s/n, Matosinhos, Portugal
5
Grupo de Limnología Amazónica, Universidad Nacional de Colombia, Sede Amazonia, Leticia, Colombia
The concept of the ecological niche remains central to our understanding of co-existence, diversity and the role of species in ecosystems (Chase and Leibold 2003; Winemiller et al. 2015). However, the quantification of niches in terms of the use of alternative resources by different species and their impact on the environment remains challenging. Functional variation in the niches of different species can determine their impact on ecosystem processes, and functional diversity within a community in terms of the breadth of niches and functional traits represented among species may be an important determinant of ecosystem functioning (Grime 1998; Petchey and Gaston 2006; Cadotte et al. 2011). Functional diversity within a community includes a number of components, including the volume of niche space represented by co-occurring taxa, the degree of dispersion among species and the evenness of their distribution within functional trait space (Villeger et al. 2008). The biological, environmental and geographic determinants of community
13
Vol.:(0123456789)
Oecologia
functional diversity in zooplankton, and its significance for ecosystem functioning, are generally unknown. The functional traits of zooplankton that determine their partitioning of resources and the environment can be described in several ways (Litchman et al. 2013; Hebert et al. 2016). Barnett et al. (2007) identified a list of feeding and life history traits for classifying freshwater crustaceans and categorized a diverse group of common North American species based on literature reports of these traits. However, categorizing species based on published descriptions ignores the potential for intraspecific variation within and among populations in trait assignments. Isotopic signatures and characterizations based on fatty acids or other biochemical markers are powerful tools to quantitatively characterize resource use among species, but may not capture other niche axes such as habitat use (Kling et al. 1992; Dalsgaard et al. 2003; Matthews and Mazumder 2006). Different chemical markers characterize distinct aspects of the dietary niche of zooplankton and may offer complementary information for quantifying resource use and community-wide functional diversity. One powerful biomarker for characterizing zooplankton trophic niches is through the composition of the amino acids in their tissues. Guisande et al. (2003) showed that amino acid signatures successfully discriminated among zooplankton species in a wide range of Spanish lakes. Amino acids are acquired through the diet and selectively incorporated into tissues as proteins. Amino acid composition of the food is important for copepod reproduction, as higher reproductive success has been observed when the amino acid composition of copepods and food are more similar (Guisande et al. 1999). The amino acid composition of tissues therefore reflects both the sources in the diet and synthesis of different proteins. However, without detailed knowledge of the amino acid composition of alternative food sources, it is impossible to link consumer composition with specific dietary components. Boechat and Adrian (2005) showed that two ciliate species fed the same algal food showed distinct amino acid profiles, indicating that metabolism and retention of amino acids are species specific. Amino acid composition is therefore determined not only by resource use, but also by differences among species and environments in how resources are acquired, metabolized and incorporated. The features of lake habitats that determine functional diversity of zooplankton remain poorly understood. Barnett and Beisner (2007) found that while taxonomic diversity of zooplankton was unimodally related to lake trophic status, functional diversity in terms of feeding and size-related traits declined linearly with productivity. They suggested that the decline in functional diversity with productivity was related to lower spatial heterogeneity in phytoplankton resource distribution throughout the water column. As primary productivity becomes concentrated in the surface waters and light becomes
13
limiting at high nutrient supply, the potential for niche partitioning among zooplankton along a depth gradient is lost. Ecosystem productivity may also be correlated with other aspects of the resource supply such as overall resource quality or the imbalance or heterogeneity among phytoplankton and microbial taxa that make up the seston (Cardinale et al. 2009). Other environmental factors such as resource diversity in addition to productivity may also play large roles in consumer functional diversity (Connell 1978). For instance, Aranguren-Riaño et al. (2011) found that species richness of crustacean zooplankton in Colombian lakes was unrelated to chlorophyll concentration, but declined with increasing environmental variability and increased with the pigment richness of the phytoplankton community. In temperate lakes, zooplankton taxonomic richness is strongly related to the means and variability of dissolved organic carbon concentration, pH, temperature and total phosphorus (Shurin et al. 2010). The relationship between taxonomic and functional diversity, as well as the causes of geographic and environmental variation in the functional trait diversity of freshwater zooplankton, are largely unknown. In this study, we used amino acid profiles of 27 common freshwater zooplankton taxa in 14 Colombian lakes in three regions (the Amazonian floodplain, the Andes and the Caribbean coastal plain) to identify the environmental and geographic factors related to community-wide functional diversity. Four to 11 common species from each lake were collected and the concentrations of 15 amino acids in bulk samples of their tissues were quantified. We examined variation among species, orders and regions to determine how different levels of taxonomic identity and geography shaped amino acid composition. Next, we quantified functional diversity in terms of functional richness, dispersion and evenness (Villeger et al. 2008; Laliberte and Legendre 2010) among species within each lake and tested for variation among regions and in relation to measured limnological factors related to lake trophic status. Productivity of lakes is a strong predictor of both functional and species diversity of zooplankton in temperate lakes (Dodson et al. 2000; Barnett and Beisner 2007; Vogt et al. 2013; Nevalainen and Luoto 2017); however, a previous survey of these same lakes found that indicators of productivity were poor predictors of crustacean species richness (Aranguren-Riaño et al. 2011). Our goal was to understand the geographical and taxonomic determinants of resource use among freshwater zooplankton and the geographic and environmental control of communitywide functional diversity in terms of amino acid profiles.
Materials and methods Study area and zooplankton collection We sampled 14 lakes between March 2005 and February 2007 distributed among three major biogeographic provinces
Oecologia
of Colombia defined by Hernández et al. (1992) as shown in Fig. 1. The number of sampling points increased with the size of the lakes to capture the diversity of zooplankton found in each lake (Table 1). Zooplankton samples were collected by drawing 45 μm mesh nets through the water column in the area of open water from a small boat at the approximate center of each lake. A fraction of each sample was preserved by freezing for separation of species and amino acid analysis. Amino acid compositions were analyzed for all crustacean species that were abundant enough to provide sufficient material for analysis. The taxonomic identification of species was carried out following the procedures described in Aranguren-Riaño et al. (2011). A sample for quantifying species abundances was collected using a 10 L Schindler Patalas trap in the subsurface waters at half the depth of the photic zone. The samples for amino acid analysis tended to contain more species than the quantitative samples because the tow spanned the entire water column.
SiO2 by Method No. G-177-96. N H4 was analyzed using the Bran Luebbe modification of Koroleff (1969).
Limnological sampling
Associations among the different aspects of the physical and chemical environment of the lakes were analyzed by principal components analysis (PCA). For the analysis of amino acids by species, a matrix was prepared of the proportional representation of each of 15 amino acids for each species analyzed in each lake as in Guisande et al. (2002). Variation in amino acid signatures among orders, regions and species within lakes was analyzed by linear discrimination analysis (LDA). The fit of LDA ordinations to the data was assessed by calculating the proportion of predicted classifications that were correctly assigned by “leave one out” cross validation (Guisande et al. 2011).
Simultaneously with the biological sampling, environmental variables were measured in situ, including temperature and pH (Boeco ad 110 pH meter), Secchi disk transparency, depth, conductivity (Gerate Schott CG-858) and dissolved oxygen (YSI oximeter 55). Dissolved nutrient analyses were performed in samples filtered through 0.45 µm cellulose acetate filters. Nutrient concentrations were measured in a Bran Luebbe (Buffalo Grove, IL, USA) Model III nutrient auto-analyzer. NO2 + NO3 were measured by Bran Luebbe Method No. G-172-96, P O4 by Method No. G-175-96 and
Table 1 Locations of the lakes sampled in the survey
Analysis of amino acids To evaluate the specific composition of amino acids, a sample containing between four and ten adult specimens of each of the most common pelagic crustacean species present in the sample were separated (Table 2), washed in distilled water and preserved in 0.1 N hydrochloric acid to dissolve tissues. Analysis of amino acids was performed by high-performance liquid chromatography (HPLC) using a Waters Alliance system, a fluorescence detector 474 and a 15 × 3.9 column Nova-Pak C18 (vanWandelen and Cohen 1997). Quantification of the concentrations of 15 essential amino acids was performed using the Thermo Scientific Pierce Amino Acid Standard H, a quantitative mixture of 18 amino acids supplied at 2.5 μMol/mL each in 0.1 N HCl.
Data analysis
Region
Lake
Coordinates
Altitude (m)
No. samMap number pling points
Andes
Tota Fúquene Iguaque Guatavita Ayapel Momil Purisima San Sebastian Yahuarcaca I Yahuarcaca II Yahuarcaca III Yahuarcaca IV Tarapoto El Correo
5°33′40″N, 72°53′52″W 5°27′38″N, 73°45′22″W 5°41′26″N, 72°26′17″W 4°58′89″N, 73°04′49″W 8°19′42″N, 75°06′11″W 9°14′10″N, 75°40′51″W 9°13′16″N, 75°44′27″W 9°12′39″N, 75°48′14″W 4°11′51S, 69°57′15″W 4°11′32″S, 69°57′26″W 4°11′13″S, 69°57′40″W 4°10′58″S, 69°58′00″W 3°48′34″S, 60°25′50″W 3°46′30″S, 79°22′50″W
3015 2543 3600 3000 18 15 15 16 90 90 90 90 97 97
5 4 1 1 5 2 2 2 1 1 1 1 2 2
Caribbean
Amazonia
1 2 3 4 5 6 7 8 9 10 11 12 13 14
13
Oecologia Table 2 List of species of planktonic crustaceans for which amino acid data were analyzed Group
Species
Code in Fig. S2
Cladocera Cladocera Cladocera Cladocera Cladocera Cladocera Cladocera Cladocera Cladocera Cladocera Cladocera Cladocera Cladocera Cyclopoida Cyclopoida Cyclopoida Cyclopoida Cyclopoida Cyclopoida Cyclopoida Cyclopoida Cyclopoida Calanoida Calanoida Calanoida Calanoida Calanoida
Bosmina freyi Bosminopsis deitersi Ceriodaphnia cornuta Ceriodaphnia pulchella Daphnia ambigua Daphnia pulex Diaphanosoma birgei Diaphanosoma brachyurum group Diaphanosoma brevireme Diaphanosoma polyspina Diaphanosoma spinulosum Moina micrura Moina minuta Ergasilus sp. Mesocyclops meridianus Mesocyclops venezolanus Metacyclops mendocinus Microcyclops anceps Thermocyclops crassus Thermocyclops decipiens Thermocyclops tenuis Tropocyclops prasinus Boeckella gracilis Colombodiaptomus brandorffi Dactylodiaptomus pearsei Notodiaptomus linus Notodiaptomus maracaibensis
B A C naa D E F G H I J K L M N O P Q S R T naa U naa V W X
a
na this species only occurred once and was excluded from ordination analyses
We used three measures to characterize the range and distribution of species in amino acid profiles among co-existing species within a lake. Functional richness (FRic) is the amount of functional space filled by species in a community, in this case measured as the volume of the convex hull containing all of the species (Villeger et al. 2008). Functional dispersion (FDis) is the mean distance of the position of species from the centroid of all species and is another measure of functional trait diversity (Laliberte and Legendre 2010). Functional evenness (FEve) is the regularity of the spacing of species within the multivariate trait space (the amino acid concentrations) (Villeger et al. 2008). Because we did not have abundance data for all species in our data set, and because we were not able to measure amino acid profiles for rare species that presented low quantities of biomass for analysis, we did not weight our measures of functional diversity by abundance. The species for which we analyzed amino acid profiles represent the dominant taxa
13
and therefore the bulk of the zooplankton biomass in each lake. In three lakes (Tarapoto, Guatavita and Iguaque), only two species were present in sufficient abundance for amino acid analysis; therefore, these lakes were excluded from functional diversity analysis. The three functional diversity metrics were calculated by averaging the proportion of each amino acid by species over multiple samples for each lake, so that they reflected variation among species but not among replicate samples of the same species from the same lake. All functional diversity metrics were calculated using the FD package in R (Laliberte et al. 2014). Patterns in FRic, FDis and FEve were visualized using principal coordinates analysis (PCoA) with convex hulls to show the distribution of species within lakes in amino acid signatures. Because the number of species sampled varied among lakes and FRic can be confounded by species richness (Naeem 2002), we also performed a rarefaction procedure to confirm that lake amino acid signatures were not driven by species richness alone. To accomplish this, we re-calculated functional richness based on 50 randomly drawn combinations of four species from each lake, the minimum number of species found in a site, then tested for differences among the three regions using linear mixed models. We found that values of FRic based on all species identified and rarefied to four species were positively correlated (Supplementary material). We present raw values of FRic; however, rarefied values produced similar patterns with respect to lake trophic status (Fig. S1).
Results Limnological conditions The major axis of variation in physical and chemical conditions among lakes and regions was associated with temperature and lake trophic status (productivity). The first principal axis (PC1) explained 36% of the variation and was strongly correlated with temperature, chlorophyll, nutrients, lake depth and Secchi depth. Cold, deep, high elevation Andean lakes fell on the low end of the productivity gradient and Amazonian at the high end, with Caribbean lakes intermediate. The second PC axis explained 24% of the variation and was most correlated with pH, conductivity and dissolved oxygen concentration. Caribbean lakes showed high values for PC2 relative to the Amazon and Andes (Fig. 2).
Zooplankton amino acid composition We measured the composition of 15 amino acids for 28 species of planktonic crustaceans representing 67% of the identified species (Table 1; Aranguren-Riaño et al. 2011). The remaining species were present in too low abundance to
Oecologia Fig. 1 A map of the study lakes. Lake names are listed in Table 1
Fig. 2 Biplot of principal component analysis on physical and chemical variables of the systems by region: Amazon (black circles, n = 6), Andes (white squares, n = 4), Caribbean (gray diamonds, n = 4). The measured variables are depth at the sampling point and Secchi depth (m), temperature (°C), conductivity (µS cm−1), dissolved oxygen (µg L−1 and percent saturation), pH, chlorophyll-a (µg L−1), NO3, NH4, PO4 and SiO2 (all in mg/L)
provide sufficient material for amino acid analysis. Amino acid profiles differed substantially among broad taxonomic groups (Cladocerans, Cyclopoid and Calanoid Copepods), as well as the three study regions. Figure 3 shows the results of linear discriminant analysis based on the proportional representation of each amino acid among taxonomic groups (Fig. 3a) and regions (Fig. 3b). Amino acid composition successfully discriminated between regions and taxonomic groups. The discriminant analysis correctly predicted the region of origin of 70.3% of the samples and the taxonomic affiliation of 65.9% of the samples by leave-one-out cross validation (Guisande et al. 2011). Individual species showed distinct amino acid profiles in different lakes. Figure 4 shows linear discriminant analyses of four common species that were found in multiple lakes. In each case, the lake of origin significantly affected the amino acid composition, as illustrated by the different positions of the samples from different lakes. The three regions varied significantly in FEve (ANOVA, P = 0.02), while FRic varied marginally (ANOVA, P = 0.06) and FDis did not differ among regions (P = 0.42, Fig. 5).
13
Oecologia
Fig. 3 Biplots of linear discriminant analysis of amino acid composition of zooplankton taxa by a region and b order. Ellipses represent correlations between the two axes
The Amazon showed the highest FRic and FEve, while the Andes had the lowest FEve. Neither FRic nor FDis was significantly related to our measure of lake trophic status, the first principal component of the physical–chemical variables shown in Fig. 2. FRic showed a marginally significant positive association with trophic status when the least productive lake (Lago Tota) was excluded from the analysis (Fig. 5, P = 0.07, R2= 0.34). By contrast, FEve increased significantly with lake trophic status (Fig. 5, P = 0.02, R2= 0.66), indicating that species are more clustered in their amino acid composition in colder, deeper, less productive lakes. Lakes differed considerably in their functional diversity (Fig. 6). The functional richness of lakes in the Amazon was higher on average compared to the other two regions, as indicated by their larger convex hulls (Fig. 6), with Lake Yahuarcaca 3 having the highest functional richness overall. In contrast, the lakes with the smallest convex hulls were Fúquene in the Andes, Momil in the Caribbean and Correo in the Amazon.
Discussion The composition of amino acids in tissues reflects variation in the niches and resource use among zooplankton species, lake habitat conditions and geographic regions of Colombia, and can be used to estimate functional diversity
13
or community-wide niche breadth within local lake communities. Amino acid profiles successfully discriminated among the three broad biogeographic regions of Colombia (the Amazon, Andes and Caribbean), among the three most common orders of crustacean zooplankton found in tropical lakes (Cladocerans, Cyclopoid and Calanoid copepods), among species inhabiting the same lake and among populations of the same species found in different lakes. In addition, functional richness, the volume of multivariate space in amino acid profiles found among species within a lake, was greater in the Amazon, the most productive region, compared with the other two regions, although zooplankton functional richness was not significantly correlated with lake trophic status across lakes. Functional evenness increased along the gradient in productivity, indicating that species in less productive lakes tend to be more tightly clustered in amino acid space. Our results show that amino acid profiles provide a useful indicator of trophic niches of tropical lake zooplankton living under variable conditions and the environmental control of functional diversity in communities. Zooplankton of the Andes, Amazon and Caribbean regions of Colombia showed distinct amino acid compositions, as composition successfully predicted the region of origin of 76% of the crustacean species found in the lakes we sampled. The lakes in the three regions also differed substantially in physical–chemical environment, with the most eutrophic conditions found in the Amazon, lowest in the
Oecologia
Andes and intermediate in the Caribbean. In addition, Caribbean lakes tended to show elevated pH and conductivity. The three regions differ substantially in geology, with Amazonian lakes being seasonally connected to the river through water level cycles, Caribbean lakes are situated in alluvial terraces and marine deposits (Thouret 1981), while Andean lakes are of tectonic and glacial origin (Donato 2001). Such broad environmental gradients are likely to strongly affect communities of micro-organisms including phytoplankton and bacteria that are the main food resources of crustacean zooplankton. The patterns in amino acid composition indicate that these shifts alter the diets or assimilation of microbial resources by meso-zooplankton. The next most important indicator of amino acid composition was taxonomic order, with cladocerans, cyclopoid and calanoid copepods showing distinct profiles across all lakes. These three groups are known to have distinct diets, with cladocerans often consuming more herbivorous diets, and smaller cells on average (Cyr and Curtis 1999; Sommer et al. 2001; Sommer and Sommer 2006). Cyclopoids and calanoids often show ontogenetic omnivory as they develop from herbivory as juveniles to predation on microzooplankton such as flagellates or rotifers as adults (Kerfoot and Kirk 1991; Barnett et al. 2007). Whether ontogenetic diet shifts are reflected in amino acid composition is unknown. The distinct compositions of the three groups indicate that variation among them in either their uptake of microbial resources or assimilation of amino acids in proteins is consistent across a wide range of environments in tropical lakes. Finally, amino acid composition also revealed variation among species within lakes, as well as among populations of the same species inhabiting different lake environments. Partitioning of microbial resources among co-existing species is strongly suggested by the distinct positions in amino acid space of different species occupying the same water body (Guisande 2006). Alternatively, species-specific amino acid profiles may represent distinct metabolic pathways and incorporation of different amino acids into the synthesis of proteins. Amino acid profiles may therefore indicate differences in assimilation of resources rather than in acquisition of prey items. This result is consistent with a large body of literature showing distinct trophic niches among crustacean zooplankton (DeMott 1986; Gliwicz 1990; Kerfoot and Kirk 1991; Cyr and Curtis 1999; Sommer et al. 2001). In addition, species found across multiple lakes often showed different amino acid profiles in different habitats, indicating that the spectrum of available resources affected the proportions of different amino acids found in the tissues of consumers. We found regional variation in different components of functional diversity in relation to lake productivity. Functional richness was greater in the Amazon, the most productive of our three regions, than either the Caribbean or Andes, which are more oligotrophic. However, there was
no significant correlation between lake trophic status and FRic across all lakes. The greater functional richness of Amazonian zooplankton may be due to other aspects of the environment that differentiate these lakes. For instance, Amazonian floodplain lakes contain high concentrations of dissolved and particulate organic compounds of terrestrial origin as a result of seasonal cycles of flooding (Petry et al. 2003; Bozelli et al. 2015). Zooplankton in these lakes may acquire resources from the detrital food chain, linked through detritivorous bacteria and protozoan grazers that can be consumed by meso-zooplankton such as those in our survey (Forsberg et al. 1993). A reliance on the microbial resources sustained by greater detrital inputs rather than in situ primary productivity may therefore explain the elevated functional richness of Amazonian zooplankton. The greater FRic in the most productive region is the opposite of the patterns observed by Barnett and Beisner (2007) in Canadian lakes from estimates of functional diversity based on published feeding traits. However, they agree with a paleolimnological showing an increase in cladoceran functional diversity with productivity in two Finnish lakes until lakes became highly eutrophic, at which point further increases in trophic status were associated with declines in zooplankton functional diversity (Nevalainen and Luoto 2017). However, lake trophic status per se was not a significant predictor of either FRic or FDis, indicating either that the niche breadth of zooplankton communities does not consistently vary with lake productivity or that our survey lacked sufficient statistical power to detect a real trend. The range in productivity among the lakes we sampled (chlorophyll-a, 0.8–11.2 µg/L) was largely overlapping, but smaller than that covered by Barnett and Beisner (0.9–24.2 µg/L). The apparent contrast among studies of lake zooplankton functional diversity in relation to productivity may result from differences in functional traits used to estimate FD, or the range of productivities included in surveys. Nevertheless, our estimate of FRic is based on traits measured in situ for each species, and therefore reflects habitat-specific differences among populations in resource use and assimilation. Intraspecific variation in amino acid composition among lakes indicates that the trophic niches of species are significantly constrained by their environment. The only aspect of interspecific diversity in amino acid composition to show a strong relationship with the trophic status of the environment was functional evenness (FEve), which measures the variability in distances among species in amino acid composition. FEve is greatest when species are regularly distributed within the trophic niche space of the community, and declines as species become more clumped. This pattern indicates that while the overall variability in amino acid composition among species within a community is unrelated or only weakly related to ecosystem productivity, species are more tightly grouped within that space in
13
Oecologia
13
Oecologia ◂Fig. 4 a–d Linear discriminant analysis (LDA) of four common spe-
less productive lakes with a few outliers occupying distinct niches. The processes generating this pattern are unknown but open to several possible interpretations. One is that the
evenness of the resource base increases with productivity. Another is that resource partitioning due to competition becomes more intense as productivity increases. Our data provide no means of discriminating among these possibilities, but suggest that further mechanistic studies are needed to elucidate the effects of productivity on niche overlap or partitioning among zooplankton of tropical lakes. Amino acids provide one perspective on the functional traits of zooplankton related to their use of resources and incorporation into their tissues. Other classifications of zooplankton functional traits have been proposed based on
Fig. 5 Three components of functional diversity (richness, dispersion and evenness) for each lake plotted against PC1 of the physical and chemical conditions, which corresponds closely to indicators of lake productivity including chlorophyll and nutrient concentrations
(Fig. 1). The dashed line on the top-right panel is for the regression for all lakes excluding Lago Tota, the largest and most oligotrophic lake in the data set. Right panels show box plots of each diversity descriptor by region
cies illustrating differences in amino acid composition in different lakes. Each point is a sample consisting of multiple individuals and the text on the graphs indicates the name of the lake from which the sample was collected, centered on the mean position of the replicate samples. e–h LDA of co-occurring species within each of four lakes. Each panel represents the lake indicated in the title, and points with the same symbol represent replicate samples of the same species
13
Oecologia
Fig. 6 Principal coordinates analysis (PCoA) of the amino acid profiles of zooplankton with each lake highlighted in color by region. The colored points indicate samples of different species from the same lake enclosed within a convex hull to indicate the degree of dispersion among species found in the same lake. Functional rich-
ness (FRic) is measured by the area enclosed within the convex hull. For instance, the lakes with the smallest convex hulls (Fúquene in the Andes, Momil in the Caribbean, Correo in the Amazon) have the lowest FRic, while Yahuarcaca 3 in the Amazon has the highest
qualitative definitions of trophic position (e.g., herbivore, omnivore, predator), feeding traits (e.g., filter feeder, suspension feeder), or quantitative traits like body size, tissue stoichiometry, nutrient excretion rates or grazing rates (Barnett and Beisner 2007; Barnett et al. 2007; Litchman et al. 2013; Nevalainen and Luoto 2017). These have been applied to assessments of functional diversity in natural communities by applying literature values to field populations. This approach is constrained in its application by the considerable intraspecific variation in many of the measured traits, as revealed by our amino acid signatures. Other measures of dietary or habitat niche that can be measured in situ include fatty acids (Dalsgaard et al. 2003) or stable isotopes (Kling et al. 1992; Matthews and Mazumder 2006) These have been applied to measuring variation among species in resource use, but not community-wide niche breadth or functional diversity as in our survey. A more comprehensive assessment of functional diversity based on metrics that incorporate intraspecific variation in resource use is needed to understand the environmental drivers of community-wide niche breadth. Caveats While patterns of amino acid composition successfully discriminated among regions, broad taxonomic groups, co-occurring species and different populations, the mechanistic interpretation of the amino acid signature in terms of resource use or ecosystem impact remains uncertain for several reasons. First, amino acids reflect both resource use and incorporation into proteins. That is, zooplankton may consume certain microbes and either excrete or metabolize their amino acids, leaving no footprint in the amino acids profile of the consumer’s tissues. Amino acid profiles therefore represent aspects of both the
response niche or resource use of zooplankton in relation to their environment, and the effect niche, or assimilation vs. metabolism and excretion of different amino acids (Leibold 1995; Guisande et al. 2003; Hebert et al. 2017). Second, we have no means of linking concentrations of specific amino acids to the resource species in which they originated. Further work is needed to determine how phytoplankton and microbial taxa differ in amino acid composition and how this translates into the composition of their consumers. Finally, while we present data on the composition of zooplankton tissues, we lack data on their relative abundances within their communities. Measuring functional diversity optimally includes weighting by abundance since rare species contribute less to the functional niche space of the community than abundant ones. All of the species represented in our data set are dominant community members, since measuring amino acid composition requires a large amount of tissue; therefore, rare species were excluded. Our estimates of FRic, FDis and FEve are all based on equal weighting of each species measured, which is equivalent to an assumption of equal abundance. The species in our data set certainly varied in abundance, but all were the most common members of their respective communities and therefore contributed the most to community-wide resource use among zooplankton. The amino acid composition of tissues successfully discriminated among regions, orders, species within lakes and multiple lake populations of the same species in zooplankton found in tropical lakes throughout Colombia. Amino acid profiles therefore characterized aspects of the niche, including both the acquisition of resources and
13
Oecologia
their incorporation in the tissues, as determined by both taxonomy and the environment. In addition, although functional diversity was greater in the Amazon than the Andes or Caribbean, only functional evenness increased significantly with lake trophic status. This pattern suggests that ecosystem productivity had a stronger effect on the spacing of co-existing species within community-wide niche space than the breadth of feeding niches within a community. Further work is needed to understand the implications of amino acid profiles for zooplankton diets and physiology, but our data indicate that amino acids discriminate among species, orders, habitats and regions, and can also be used to characterize functional diversity and its relationship to the environment in tropical lake zooplankton communities. Acknowledgements NA was supported by a graduate fellowship from Colciencias 1892–2006. JS was supported by a Fulbright Colombia Fellowship, JS and NTJ were supported by NSF DEB 1457737. Author contribution statement NJA, CG, AB and SD collected field samples and analyzed amino acids in the laboratory, NJA, JBS and NTJ analyzed the data, and all authors wrote and edited the manuscript.
References Aranguren-Riaño N, Guisande C, Ospina R (2011) Factors controlling crustacean zooplankton species richness in Neotropical lakes. J Plankton Res 33:1295–1303 Barnett A, Beisner BE (2007) Zooplankton biodiversity and lake trophic state: explanations invoking resource abundance and distribution. Ecology 88:1675–1686 Barnett AJ, Finlay K, Beisner BE (2007) Functional diversity of crustacean zooplankton communities: towards a trait-based classification. Freshw Biol 52:796–813 Boechat IG, Adrian R (2005) Biochemical composition of algivorous freshwater ciliates: you are not what you eat. FEMS Microbiol Ecol 53:393–400 Bozelli RL, Thomaz SM, Padial AA, Lopes PM, Bini LM (2015) Floods decrease zooplankton beta diversity and environmental heterogeneity in an Amazonian floodplain system. Hydrobiologia 753:233–241 Cadotte MW, Carscadden K, Mirotchnick N (2011) Beyond species: functional diversity and the maintenance of ecological processes and services. J Appl Ecol 48:1079–1087 Cardinale BJ, Hillebrand H, Harpole WS, Gross K, Ptacnik R (2009) Separating the influence of resource ‘availability’ from resource ‘imbalance’ on productivity–diversity relationships. Ecol Lett 12:475–487 Chase JM, Leibold MA (2003) Ecological niches: linking classical and contemporary approaches. University of Chicago Press, Chicago Connell JH (1978) Diversity in tropical rain forests and coral reefs— high diversity of trees and corals is maintained only in a nonequilibrium state. Science 199:1302–1310 Cyr H, Curtis JM (1999) Zooplankton community size structure and taxonomic composition affects size-selective grazing in natural communities. Oecologia 118:306–315
Dalsgaard J, St John M, Kattner G, Muller-Navarra D, Hagen W (2003) Fatty acid trophic markers in the pelagic marine environment. Adv Mar Biol 46(46):225–340 DeMott WR (1986) The role of taste in food selection by fresh-water zooplankton. Oecologia 69:334–340 Dodson SI, Arnott SE, Cottingham KL (2000) The relationship in lake communities between primary productivity and species richness. Ecology 81:2662–2679 Donato JC (2001) Fitoplancton de los lagos andinos del norte de Sudamerica (Colombia): composicion y factores de distribucion. Academia Colombiana de Ciencias Exactas, Fisicas y Naturales Bogota, Colombia Forsberg BR, Araujolima C, Martinelli LA, Victoria RL, Bonassi JA (1993) Autotrophic carbon-sources for fish of the central Amazon. Ecology 74:643–652 Gliwicz ZM (1990) Food thresholds and body size in cladocerans. Nature 343:638–640 Grime JP (1998) Benefits of plant diversity to ecosystems: immediate, filter and founder effects. J Ecol 86:902–910 Guisande C (2006) Biochemical fingerprints in zooplankton. Limnetica 25:369–376 Guisande C, Maneiro I, Riveiro I (1999) Homeostasis in the essential amino acid composition of the marine copepod Euterpina acutifrons. Limnol Oceanogr 44:691–696 Guisande C, Maneiro I, Riveiro I, Barreiro A, Pazos Y (2002) Estimation of copepod trophic niche in the field using amino acids and marker pigments. Mar Ecol Prog Ser 239:147–156 Guisande C, Bartumeus F, Ventura M, Catalan J (2003) Role of food partitioning in structuring the zooplankton community in mountain lakes. Oecologia 136:627–634 Guisande C, Barreiro A, Vaamonde A (2011) Tratamiento de datos con R. Statistica y SPSS, Ediciones Díaz de Santos, Madrid Hebert MP, Beisner BE, Maranger R (2016) A meta-analysis of zooplankton functional traits influencing ecosystem function. Ecology 97:1069–1080 Hebert MP, Beisner BE, Maranger R (2017) Linking zooplankton communities to ecosystem functioning: toward an effect-trait framework. J Plankton Res 39:3–12 Hernández J, Hurtado A, Ortiz R, Walschburger T (1992) Unidades biogeográficas de Colombia. In: Halffter G (ed) La diversidad biológica de Iberoamérica I. Instituto de Ecología, A.C., Mexico, pp 105–151 Kerfoot WC, Kirk KL (1991) Degree of taste discrimination among suspension-feeding cladocerans and copepods—implications for detritivory and herbivory. Limnol Oceanogr 36:1107–1123 Kling GW, Fry B, Obrien WJ (1992) Stable isotopes and planktonic trophic structure in arctic lakes. Ecology 73:561–566 Koroleff F (1969) Direct determination of ammonia in natural waters as Indophenol Blue. In: ICES, information on techniques and methods for seawater analysis, an interlaboratory report, pp 19–22 Laliberte E, Legendre P (2010) A distance-based framework for measuring functional diversity from multiple traits. Ecology 91:299–305 Laliberte E, Legendre P, Shipley B (2014) FD: measuring functional diversity from multiple traits, and other tools for functional ecology, R package version 1.2 Leibold MA (1995) The niche concept revisited—mechanistic models and community context. Ecology 76:1371–1382 Litchman E, Ohman MD, Kiorboe T (2013) Trait-based approaches to zooplankton communities. J Plankton Res 35:473–484 Matthews B, Mazumder A (2006) Habitat specialization and the exploitation of allochthonous carbon by zooplankton. Ecology 87:2800–2812 Naeem S (2002) Disentangling the impacts of diversity on ecosystem functioning in combinatorial experiments. Ecology 83:2925–2935
13
Oecologia Nevalainen L, Luoto TP (2017) Relationship between cladoceran (Crustacea) functional diversity and lake trophic gradients. Funct Ecol 31:488–498 Petchey OL, Gaston KJ (2006) Functional diversity: back to basics and looking forward. Ecol Lett 9:741–758 Petry P, Bayley PB, Markle DF (2003) Relationships between fish assemblages, macrophytes and environmental gradients in the Amazon River floodplain. J Fish Biol 63:547–579 Shurin JB, Winder M, Adrian R, Keller W, Matthews B, Paterson AM, Paterson MJ, Pinel-Alloul B, Rusak JA, Yan ND (2010) Environmental stability and lake zooplankton diversity—contrasting effects of chemical and thermal variability. Ecol Lett 13:453–463 Sommer U, Sommer F (2006) Cladocerans versus copepods: the cause of contrasting top-down controls on freshwater and marine phytoplankton. Oecologia 147:183–194 Sommer U, Sommer F, Santer B, Jamieson C, Boersma M, Becker C, Hansen T (2001) Complementary impact of copepods and cladocerans on phytoplankton. Ecol Lett 4:545–550
13
Thouret J (1981) Una mapa geomorfoestructural de los Andes colombianos. Instituto Geografico “Agustin Codazzi” Bogota, Colombia vanWandelen C, Cohen SA (1997) Using quaternary high-performance liquid chromatography eluent systems for separating 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate-derivatized amino acid mixtures. J Chromatogr A 763:11–22 Villeger S, Mason NWH, Mouillot D (2008) New multidimensional functional diversity indices for a multifaceted framework in functional ecology. Ecology 89:2290–2301 Vogt RJ, Peres-Neto PR, Beisner BE (2013) Using functional traits to investigate the determinants of crustacean zooplankton community structure. Oikos 122:1700–1709 Winemiller KO, Fitzgerald DB, Bower LM, Pianka ER (2015) Functional traits, convergent evolution, and periodic tables of niches. Ecol Lett 18:737–751
