local environment rather than past climate determines ......carmen elisa sáinz-cantero university...

Seediscussions,stats,andauthorprofilesforthispublicationat:https://www.researchgate.net/publication/319549189

Localenvironmentratherthanpastclimatedeterminescommunitycompositionofmountainstreammacroinvertebrat....

ArticleinMolecularEcology·September2017

DOI:10.1111/mec.14346

CITATIONS

0

READS

260

15authors,including:

Someoftheauthorsofthispublicationarealsoworkingontheserelatedprojects:

BiodiversitygenomicsViewproject

EffectsofHydrologicalAlterationonRiverFunctioningandServiceProvisioning:Implicationsfor

IntegratedCatchmentManagementViewproject

CescMúrria

UniversityofBarcelona

29PUBLICATIONS441CITATIONS

SEEPROFILE

CarmenZamora-Muñoz

UniversityofGranada

94PUBLICATIONS1,412CITATIONS

SEEPROFILE

JavierAlba-Tercedor

UniversityofGranada

135PUBLICATIONS1,926CITATIONS

SEEPROFILE

CarmenElisaSáinz-Cantero

UniversityofGranada

26PUBLICATIONS335CITATIONS

SEEPROFILE

AllcontentfollowingthispagewasuploadedbyAlfriedVogleron16November2017.

Theuserhasrequestedenhancementofthedownloadedfile.

Acc

epte

d A

rtic

le

This article has been accepted for publication and undergone full peer review but has not

been through the copyediting, typesetting, pagination and proofreading process, which may

lead to differences between this version and the Version of Record. Please cite this article as

doi: 10.1111/mec.14346

This article is protected by copyright. All rights reserved.

DR. CESC MURRIA (Orcid ID : 0000-0003-4319-0804)

PROF. ALFRIED VOGLER (Orcid ID : 0000-0002-2462-3718)

Article type : Original Article

LOCAL ENVIRONMENT RATHER THAN PAST CLIMATE DETERMINES COMMUNITY COMPOSITION OF

MOUNTAIN STREAM MACROINVERTEBRATES ACROSS EUROPE

Cesc Múrria1,2,3,*; Núria Bonada1; Mark Vellend3; Carmen Zamora-Muñoz4; Javier Alba-Tercedor4;

Carmen Elisa Sainz-Cantero4; Josefina Garrido5; Raul Acosta1; Majida El Alami6; Jose Barquín7; Tomáš

Derka8; Mario Álvarez-Cabria7; Marta Sáinz-Bariain4; Ana F. Filipe9,10 & Alfried P. Vogler2,11

1 Grup de Recerca Freshwater Ecology and Management (FEM) and Institut de Recerca de la

Biodiversitat (IRBio), Departament de Biologia Evolutiva, Ecologia i Ciències Ambientals, Facultat de

Biologia. Universitat de Barcelona, Avinguda Diagonal 643, 08028-Barcelona, Catalonia, Spain; 2

Department of Life Sciences, Natural History Museum, Cromwell Road, London SW7 5BD, UK; 3

Département de Biologie, Université de Sherbrooke, 2500, boul. de l'Université, Sherbrooke, QC J1K

2R1 Canada; 4 Departamento de Zoología, Facultad de Ciencias, Universidad de Granada, C/ Severo

Ochoa s/n, 18071-Granada, Spain; 5 Departamento de Ecología y Biología Animal, Facultad de

Biología. Universidad de Vigo, Campus Lagoas-Marcosende, 36310, Vigo, Spain; 6 Department of

Biology, University Abdelmalek Essâadi, 2121 Tétouan, Morocco; 7 Environmetal Hydraulics Institute,

Universidad de Cantabria, C/ Isabel Torres 15, Parque Científico y Tecnológico de Cantabria, 39011,

Santander, Spain; 8 Department of Ecology, Faculty of Natural Sciences, Comenius University, B-2

Mlynská dolina, Ilkovičova 6, SK-842 15, Bratislava, Slovakia; 9 CIBIO/InBIO, Centro de Investigação

em Biodiversidade e Recursos Genéticos da Universidade do Porto, Campus Agrário de Vairão, R.

Acc

epte

d A

rtic

le


Padre Armando Quintas, 4485-661 Vairão, Portugal; 10 CEABN/InBIO, Centro de Ecologia Aplicada,

Instituto Superior de Agronomia, Universidade de Lisboa, Tapada da Ajuda, 1349-017 Lisboa,

Portugal; 11 Department of Life Sciences, Imperial College London, Silwood Park Campus, Ascot,

Berkshire SL5 7PY, UK.

* Corresponding author: Cesc Múrria, Departament de Biologia Evolutiva, Ecologia i Ciències

Ambientals, Facultat de Biologia, Universitat de Barcelona, Avinguda Diagonal, 643, 08028

Barcelona, Catalonia. E-mail: [email protected]

Key words: α-diversity, β-diversity, latitudinal gradient, community DNA barcoding, multi-

hierarchical patterns, stream ecology, SGDC

Running title: Community assembly of mountain stream macroinvertebrates

Abstract

Community assembly is determined by a combination of historical events and contemporary

processes that are difficult to disentangle, but eco-evolutionary mechanisms may be uncovered by

the joint analysis of species and genetic diversity across multiple sites. Mountain streams across

Europe harbour highly diverse macroinvertebrate communities whose composition and turnover

(replacement of taxa) among sites and regions remain poorly known. We studied whole-community

biodiversity within and among six mountain regions along a latitudinal transect from Morocco to

Scandinavia at three levels of taxonomic hierarchy: genus, species and haplotypes. Using DNA

barcoding of four insect families (>3100 individuals, 118 species) across 62 streams, we found that

measures of local and regional diversity and intraregional turnover generally declined slightly

towards northern latitudes. However, at all hierarchical levels we found complete (haplotype) or

Acc

epte

d A

rtic

le


high (species, genus) turnover among regions (and even among sites within regions), which counters

the expectations of Pleistocene postglacial northward expansion from southern refugia. Species

distributions were mostly correlated with environmental conditions, suggesting a strong role of

lineage- or species-specific traits in determining local and latitudinal community composition,

lineage diversification and phylogenetic community structure (e.g., loss of Coleoptera, but not

Ephemeroptera, at northern sites). High intraspecific genetic structure within regions, even in

northernmost sites, reflects species-specific dispersal and demographic histories and indicates

postglacial migration from geographically scattered refugia, rather than from only southern areas.

Overall, patterns were not strongly concordant across hierarchical levels, but consistent with the

overriding influence of environmental factors determining community composition at the species

and genus levels.

Introduction

Biodiversity encompasses multiple levels of biological organization, from populations to ecosystems.

Although recent work has begun to integrate these various facets of biodiversity, identifying the

mechanisms of diversification and community assembly at large spatiotemporal scales remains a

major research challenge (Ricklefs 2004; Brooker et al. 2009; Vellend 2016). For example, at the

intraspecific level, comparative phylogeographic studies have focused largely on the genetic

structure within one or a few species, linking population demographic factors and vicariance to

palaeoclimatic and biogeographical history (Avise 2009; Marske et al. 2013). At the interspecific

level, molecular phylogenetic studies can reveal the drivers of species diversification and their role in

community assembly from variation in functional and ecological traits (Losos et al. 1998; Gillespie

2004). In parallel, community ecologists have studied turnover in local species composition in

attempts to identify contemporary processes of community assembly (e.g., environmental filtering,

dispersion distance, species interactions), while often ignoring the evolutionary history of the

interacting lineages (Qian & Ricklefs 2007; De Cáceres et al. 2012; Myers et al. 2013). Although

Acc

epte

d A

rtic

le


ecologists have recently incorporated phylogenetic distances between species into analyses of

community assembly, these approaches have most often been used as surrogates of ecological

differentiation across communities (Webb et al. 2002; Cavender-Bares et al. 2009), rather than as a

means to study communities through time. Hence, studies of community assembly have rarely been

conducted at multiple hierarchical levels from intraspecific genetic differentiation to deeper lineage

divergence, i.e. along a conceptual time-continuum in analogy to stratigraphic records (Rull 2014).

If the factors causing intraspecific genetic diversification and geographic structure (Leffler et

al. 2012, Romiguier et al. 2014) act over long time-scales, the initial differentiation progressively

leads to species divergence and to deeper phylogenetic divergence of lineages, resulting in

evolutionary divergences across all levels (Papadopoulou et al. 2009; Diniz-Filho et al. 2011; Múrria

et al. 2015). With the increasing scale of DNA sequencing studies, the uniform generation of

mitochondrial ‘DNA-barcodes’ (based on the sequencing of the cytochrome oxidase C subunit I,

cox1) allows quantification of a broad range of genetic differences within and across species

simultaneously for entire communities (Baselga et al. 2013). Patterns of variation from community-

wide DNA barcodes reflect the action of similar evolutionary processes at multiple hierarchical

levels, leading to: the divergence of populations and intraspecific genetic structure (Avise 2009), the

formation of species evident from genetic clusters (Hebert et al. 2003; Papadopoulou et al. 2008;

Tänzler et al. 2012), and the phylogenetic divergence of species and higher clades (Papadopoulou et

al. 2009; Múrria et al. 2015). For example, a study of insects across tropical highland rivers revealed

high inter- and intraspecific genetic structure and divergence among communities, which coincided

across distantly related lineages, suggesting long-term limited dispersal across river sites as the main

driver of community composition (Múrria et al. 2015). Hence, DNA sequencing of entire species

assemblages is a promising approach for exploring the ecological and evolutionary mechanisms of

community assembly.

Acc

epte

d A

rtic

le


Among widely studied macroecological patterns, the latitudinal gradient of species diversity

has been recognised in many groups of plants and animals, showing a reduction of both local (α) and

regional (γ) diversity towards the poles (Gaston & Blackburn 2000; Hawkins et al. 2003; Field et al.

2009; Heino 2011; Svenning et al. 2011). Some studies also find a latitudinal decrease of genetic

diversity at the intraspecific level (Bernatchez & Wilson 1998; Martin & McKay 2004; Adams & Hadly

2013; Miraldo et al. 2016). In Europe, latitudinal decreases of species diversity have been attributed

to contemporary environmental factors such as decreases of energy and productivity at northern

latitude (Shah et al. 2015), or to historical factors such as distribution shifts associated with climatic

changes during the Pleistocene (Hewitt 2000; Svenning & Skov 2007; Hof et al. 2008; Stewart et al.

2010; Hortal et al. 2011). Similarly, the latitudinal decrease of diversity at the intraspecific level has

been attributed to slower rates of evolution at lower temperature (Adams & Hadly 2013; Oppold et

al. 2016; Miraldo et al. 2016) or to genetic drift during repeated fragmentation and northward post-

glacial range expansion in the Pleistocene (Pfenninger & Posada 2002).

In contrast to patterns of decreasing α- and γ-diversity with latitude, patterns of β-diversity

(i.e., site-to-site variation in composition) are more variable and the underlying processes have been

actively debated. For instance, decreasing β-diversity among local communities of native vascular

plants with latitude in North America is associated with lower climate differentiation at northerly

latitudes and post-glacial dispersal limitation (Qian & Ricklefs 2007). Similarly, species richness in

terrestrial Coleoptera across Europe exhibited significant species richness decline in the north,

presumably due to cold-intolerance, whereas β-diversity among diverse southern regions was driven

by turnover of many specialist species with restricted distributions (Baselga 2010; Hortal et al.,

2011). Shifting mechanisms with latitude also explain β-diversity in tree communities as the

predominant result of dispersal limitation in tropical forests but environmental filtering in temperate

regions (Condit et al. 2002; De Cáceres et al. 2012; Myers et al. 2013).

Acc

epte

d A

rtic

le


In this study, we asked whether the dynamics of post-glacial recolonization (i.e. past climate)

or contemporary environmental conditions were more important in determining diversity patterns

in communities of mountain freshwater macroinvertebrates across Europe. Despite their known

great species diversity, mountain freshwater ecosystems have to date received surprisingly little

attention with regard to genetic and species diversity at the continental scale. We studied diversity

patterns at two geographic scales: among stream reaches within a single mountain region, and

across these regions along a south-north transect from Morocco to Scandinavia. These regions have

been affected differently by past climate and geological factors and differ in contemporary

ecological conditions. Specifically, the biotas at high latitudes occupy recently glaciated areas and

have formed in a comparatively short time. In addition, local environmental factors such as litter

inputs, seasonality in flow regimes or frequency of discharge disturbance differ between the

Mediterranean and Scandinavian regions (Bonada & Resh 2013). Water bodies in these regions

therefore differ greatly in spatio-temporal habitat connectivity and stability (Ribera et al. 2003; Hof

et al. 2008), as well as the amount and quality of available energy associated to primary production

dependent on latitudinal difference in insolation (Bonada & Resh 2013).

If biodiversity patterns are determined mainly by biogeographical history of postglacial

south-to-north range expansions (Hewitt 2000), latitudinal α-, β- and γ-diversity would decrease at

high latitudes. Southern communities are expected to accumulate high diversity and strong

intraregional structure due to long-term population and species persistence in comparatively stable

habitats (Ribera & Vogler 2004; Hof et al. 2008), whereas in postglacial northern regions the

communities should be recolonized by dispersive species and therefore these regions are expected

to show shallow local diversification with little spatial structure at both the species (Ribera et al.

2003) and genetic levels (Bernatchez & Wilson 1998). Under the “southern richness and northern

purity” paradigm (Hewitt 2000) that has been confirmed for many terrestrial groups, we predict a

general trend of locally diverse and spatially highly structured communities in southern Europe (high

α- and β-diversity), and less diverse northern communities which constitute a nested subset of those

Acc

epte

d A

rtic

le


present in southern refugia (Hof et al. 2008; Baselga 2010; Hortal et al. 2011). At phylogenetic scales,

diversity of northern communities is reduced equally to include the most dispersive (Hewitt 2000) or

warm-adapted clades (Hortal et al. 2011), which may constitute lineages of related species that

exhibit traits affecting dispersal propensity or thermal tolerance, and thus northern communities

may be phylogenetically clustered (Hortal et al., 2011).

Alternatively, if contemporary environmental conditions override biogeographical-historical

signatures, community composition should be strongly correlated with environmental conditions

within and across regions (e.g., climate, streamflow, instream physical habitat heterogeneity,

particulate organic matter) and life-history traits of species that respond to these conditions. Given

that not all environmental factors vary in parallel across latitude, we can expect that they exert

selection on species traits, such as cold tolerance, and determine diversity distribution. Therefore,

according to this hypothesis, α-, β- and γ-diversity should not necessarily decrease with latitude. The

environmental hypothesis thus predicts that: (1) there is high turnover of species diversity across

regions in accordance with climatic differences, but only limited thinning towards northern latitudes

(Svenninig et al. 2011); (2) turnover of species distribution and community composition within

narrowly defined regions is correlated with the degree of local environmental heterogeneity

(Astorga et al. 2014); (3) phylogenetic community structuring reflects differences in species traits

across lineages, e.g., a latitudinal decrease of phylogenetic-β-diversity for warm adapted species

(Hortal et al. 2011); and (4) intraspecific diversity and structure reflect species-specific dispersal and

demographic histories (determined by for instance the frequency of floods and droughts or local

barriers to gene flow; Múrria et al. 2013; Paz-Vinas & Blanchet 2015), which is independent of

latitude, unless these parameters differ with latitude.

We investigated the patterns of community diversity at three hierarchical levels: the

intraspecific genetic level, by analysing variation of cox1 haplotypes within DNA-delimited species;

the species level, by delimiting putative species defined by cox1 clusters; and the genus level, by

identifying all major groups present at a site based on morphology. We then assessed patterns of

Acc

epte

d A

rtic

le


diversity at these levels and at various spatial scales, and tested for correlations of diversity metrics

across organizational levels. Only a historical scenario driven primarily by the role of dispersal would

predict such correlation along the latitudinal gradient.

Materials and methods

Sampling sites, macroinvertebrates, and environmental variables

In total, 62 streams were sampled across six mountain regions covering three climatic zones (Fig. 1):

Jämtland in Sweden (Subarctic climate); Carpathians in Slovakia; Jura in France; Picos de Europa in

northern Spain (Temperate climate); Betic in southern Spain; and Rif in north Morocco

(Mediterranean climate) (Appendix S1). Intraregional spatial arrangement of sites was as similar as

possible between regions: all sites were pristine, permanent (continuous flow regime) first or second

order headwater streams which were located in calcareous zones and spatially independent (not

flow-connected). Calcareous soils were scarce in North Africa (Rif massif), where fewer sites were

sampled at shorter inter-site distance (Table 1). Similar limitations in Temperate and Subarctic

regions (European Soil Bureau Network European Commission, 2005) also imposed a spatial

sampling gap between the Carpathians and Jämtland and limited the choice of elevation in Jämtland.

The elevation of sampling sites was highest in the Mediterranean Rif and Betic regions, intermediate

at the Temperate regions, and lowest in Jämtland (Table 1). In each stream, reaches of 150 meters

(hereafter, a “site”) were sampled twice in spring and fall of 2008, and both time points were pooled

for the current analysis.

Macroinvertebrates were sampled using both qualitative and quantitative methods in order

to assess occurrence and abundance at the genus level. The qualitative sampling took place in all

available habitats using a 250 μm mesh kick net for at least 150 minutes (including kick sampling

together with material sorting and identification) until no new taxa (visually identifiable) were

captured in two consecutive kick samples. Three quantitative samples were collected at randomly

selected spots within the site using a cylindrical Hess Sampler (Ø=30.5 cm) with 250 μm mesh size.

Acc

epte

d A

rtic

le


Qualitative and quantitative samples were sorted in the laboratory and macroinvertebrates were

identified to genus level (Tachet et al. 2010). Diptera, Gordiaceae and Oligochaeta were not

identified because of taxonomic complexity, and were not used in further analyses. Following

standard methods in plant ecology (cover-abundance index used by Braun-Blanquet; Poore 1955),

the overall abundance of each genus was assigned to one of four rank-abundance categories

(approximating equal bins of log-transformed abundance) after combining the qualitative and

quantitative datasets collected in spring and autumn: 1 for 1-3 individuals, 2 for 4-10 individuals, 3

for 11-100 individuals, and 4 for >100 individuals. The abundant and widely distributed genera Baetis

(Ephemeroptera: Baetidae) and Hydropsyche (Trichoptera: Hydropsychidae), and all members of the

coleopteran families Elmidae (7 genera) and Hydraenidae (3 genera) were selected for molecular

analyses and morphologically identified at the species level (Appendix S2). These four lineages

differed in species diversity and life-history traits such as dispersal capabilities or body size. For

instance, based on a combination of four biological traits (body size, number of generations per year,

dissemination strategy and adult life span), Elmidae, Hydraenidae and Hydropsyche were inferred to

exhibit strong flying propensity, whereas it was weak in Baetis (Sarremejane et al. 2017), which is

likely associated with the short adult life span of Ephemeroptera.

A total of 21 environmental variables related to stream habitat characteristics (e.g., channel

stability, substrate heterogeneity) and water physico-chemistry (e.g., pH, temperature, dissolved

organic matter) were measured in each site during the autumn collecting campaign (Appendix S3).

The number of variables was reduced to eliminate the redundancy and collinearity among

environmental variables, but capturing the key elements of environmental variation without

overfitting models. For pairs of highly correlated variables (Pearson r>0.8) we retained the one with

a clearer ecological interpretation (Dormann et al. 2013), yielding a total of 18 variables (maximum

and minimum stream width and total suspended solids were removed). Moreover, a principal

component analysis (PCA) was performed on these 18 variables and the first two orthogonal PCA

axes (i.e. perfectly uncorrelated; Dormann et al. 2013) were used in the variation partitioning

Acc

epte

d A

rtic

le


analyses. Geographical distances between sites associated with likely movement routes were

calculated as least-cost path distances using elevation as the cost layer (i.e., layer of impeding or

facilitating movement). These measures of landscape connectivity aim to account for the dispersal

pathway of winged adults (flying at low altitude) or for drifting larvae. Nevertheless, how widely and

how often freshwater invertebrates disperse passively or actively remains unknown but differs

greatly among taxa (Sarremejane et al. 2017). Geographical distances were calculated using the

Pathmatrix 1.1 Arcview 3.2 extension tool (Ray 2005). Digital elevation maps data used as a cost

layer were obtained from SRTM NASA (90x90 m resolution; Jarvis et al. 2008), except for Jämtland,

where data was obtained from MDT EEA (1x1 km resolution; EEA 2010).

DNA sequencing and molecular species delimitation

The cox1 gene was sequenced for up to 25 individuals per site for each recognisable species of the

12 focal genera. DNA was extracted from thorax or abdominal segments using WizardSV 96

extraction plates (Promega, Southampton, UK). A 658 bp fragment was amplified for Baetis species

using a combination of the standard oligonucleotide primers LCO1490 and HCO2198 (Folmer et al.

1994). For Hydropsyche, a 644 bp fragment was amplified using a combination of primers Dave and

Inger (Lehrian et al. 2009). An 822 bp fragment of cox1 was amplified for Elmidae and Hydraenidae

with primers C1-J-2183 (Jerry; Simon et al. 1994) and a modified TL2-N-3014 (S-PatR; Timmermans

et al. 2010). Amplification products were purified using Millipore Multiscreen 96-well plates

(Millipore, Billerica, MA, USA) and sequenced in both directions using the ABI BigDye Terminator kits

and ABI PRISM3700 DNA Analyzer (Applied Biosystems, Foster City, CA, USA). Sequences for all

unique haplotypes were submitted to GenBank under accession no. KY656166-KY656253 and

MF783180-MF783882.

Putative molecular species were delimited using the Generalized Mixed Yule Coalescent

(GMYC) model (Fujisawa & Barraclough 2013) after all identical cox1 sequences were collapsed into

unique haplotypes. The GMYC model optimizes the point of change in branching rates at the species-

Acc

epte

d A

rtic

le


population boundary to divide interspecific and intraspecific variation. GMYC was shown to be

consistent with other methods (e.g., ABGD, P ID (Liberal), PTP) (Hamilton et al, 2014; Tang et al,

2014). A GMYC model was run separately for each lineage. The GMYC analysis was conducted on a

Maximum Likelihood phylogenetic tree obtained with RAxML 7.0.4 (Stamatakis 2006) under a GTR +

Γ substitution model after adding one out-group per lineage (GenBank accession numbers:

JQ663124; DQ156052; DQ266502; JQ687925). The tree was made ultrametric using penalized

likelihood as implemented in r8s v. 1.7 (Sanderson 2003) with the optimal smoothing parameter

selected by cross-validation. The GMYC analysis was conducted using the R package (R Development

Core Team 2013) Splits with the ‘single threshold’ option (Fujisawa & Barraclough 2013) and the

resulting clusters were used for the analysis of the species level.

Estimates of diversity at multiple organizational levels

The same analytical framework for estimating diversity was used at the three hierarchical levels:

genus, species, and haplotypes. α-diversity was defined at the scale of the local site (reach) and γ-

diversity was defined at the regional scale, summing up all variants within a region. In order to assess

whether intraregional composition was distinct across regions, we conducted a correspondence

analysis (CA) using the site-by-taxon data tables to assess distinctiveness across sites and to test if

northern communities were a subset of southern communities. Differences across regions were

assessed using a Monte-Carlo procedure, which yielded a simulated probability obtained after 1000

permutations (Dolédec & Chessel 1989). Next, compositional differences (β-diversity) in pairs of

regions were quantified following Baselga (2010). This approach partitioned total β-diversity (βsor)

into nestedness (i.e., are northern taxa a subset of biota in southern taxa?; βnes) and turnover (i.e.,

replacement of taxa across regions; βsim), which are additive factors of βsor (Baselga 2010). For this

analysis, we used the presence/absence community matrix within a region, summing up all sites, and

the function “beta.pair” was run in the R package betapart (Baselga & Orme 2012).

Acc

epte

d A

rtic

le


β-diversity within each region was estimated using two indices, for which the observed and

expected values were estimated separately to account for the statistical dependence of β-diversity

on γ-diversity (Kraft et al. 2011). First, the Hellinger distance was used to measure compositional

dissimilarities between all pairs of sites, and its total variance within a region was the estimate of β-

diversity, which is the average dissimilarity between sampling units (De Cáceres et al. 2012;

Legendre & De Cáceres 2013). Since abundance at the genus level was coded as rank-abundance, for

this analysis the abundances of species and haplotypes were also transformed to rank categories.

The null model involved 1000 randomizations of taxon occurrence in a region fixing γ-diversity to the

total richness in the poorest region to remove effects of variation in γ-diversity among regions but

keeping the original relative abundance of each taxon in that region. Second, the Whittaker’s Index

of -diversity (β=1−γ/ᾱ) was estimated (Whittaker 1960). The expected values for these analyses

were obtained using a null model that randomly shuffles individuals among sites while preserving γ-

diversity and the total number of individuals per site for 1000 iterations (Kraft et al. 2011; Myers et

al. 2013). This was possible only at the species and haplotype levels. If differences between the

observed and (mean) expected dissimilarity divided by the standard deviations of expected values

were not significantly different from zero, local communities were considered a largely random draw

from the regional (γ) pool. In contrast, large deviations would indicate that certain factors (e.g.,

biogeography, dispersal limitations, ecological conditions) might be responsible for more aggregated

(positive) or dispersed (negative) species distributions than expected from stochastic sampling alone

(Kraft et al. 2011; Myers et al. 2013). To disentangle the processes contributing to the β-partitioning,

the total variation of the Hellinger distance among sites within each region was partitioned into four

components: pure environment, pure spatial structure (obtained from Principal Components of

Neighbor Matrices, PCNM; Borcard et al. 2004, Griffith & Peres-Neto 2006), spatially structured

environment, and the unexplained variation (Peres-Neto et al. 2006).

Acc

epte

d A

rtic

le


Phylogenetic community structure was assessed based on ultrametric trees, by measuring

the mean pairwise phylogenetic distances among species (ComDist index) occurring in two different

communities (i.e., phylogenetic-β-diversity) within each region (Kembel et al. 2010). The mean value

of phylogenetic-β-diversity for each region was used for testing a trend with latitude. At the

intraspecific level, α-diversity was estimated as the local nucleotide diversity π (the average number

of nucleotide differences per site between two sequences; Nei 1987) for each species with at least

five individuals sequenced per site. Intraregional population genetic structure (β-diversity) was

estimated only for species with at least five individuals sequenced per site and located in at least

three sites within a region using pairwise ΦST values, as calculated in Arlequin v.3.1 (Excoffier et al.

2005). Finally, in order to test for Species-Genetic Diversity Correlations (SGDC; Vellend 2003), α-

diversity at the genus level and the mean nucleotide diversity π of species present at a site were

correlated using Pearson’s correlation coefficient.

The correlation between each estimate of diversity and latitude was tested using

Spearman’s rank-order correlation. These statistical analyses were performed with the ade4 (Dray &

Dufour 2007), stats (R Development Core Team 2013), ape (Paradis et al. 2004), and vegan (Oksanen

et al. 2015) packages of the R software (R Development Core Team 2013).

Results

A total of 18 orders, 87 families and 197 genera of freshwater macroinvertebrates were

morphologically identified (Appendix S4). Twenty out of 197 genera were distributed across the

entire latitudinal gradient, which included genera used in molecular analyses and others, such as

Dugesia (Tricladida), Isoperla, Leuctra (Plecoptera) or Rhyacophila and Wormaldia (Trichoptera). The

Rif had the most unique genera (13), followed by Picos (11), Jämtland (10) and Betic (9). The number

of genera restricted to the Mediterranean zone (15) was higher than in the Temperate (3) or

Subarctic (10) zones.

Acc

epte

d A

rtic

le


DNA sequences were obtained for 3113 individuals representing 791 unique cox1 haplotypes

and 118 species. The Maximum Likelihood trees were consistent with our species level

morphological identification (Figs. S1-S4). Baetis showed the highest richness (1084 sequences, 298

unique haplotypes, and 39 species, Appendix S5), followed by Hydraenidae (666 sequences, 206

unique haplotypes and 33 species, Appendix S6), Hydropsyche (508 sequences, 88 unique

haplotypes, and 25 species, Appendix S7) and Elmidae (855 sequences, 199 unique haplotypes and

21 species, Appendix S8). The intraregional number of species was similar at all latitudes for Baetis

(8-11 species) and Hydropsyche (4-7 species). In contrast, intraregional species richness for Elmidae

and Hydraenidae decreased from 10-11 species in Betic to two species per lineage in Jämtland. None

of the species was found across the entire gradient, and the most widely distributed species were

located either in southern (from Rif to Picos) or northern (from Jura to Jämtland) areas. Most

haplotypes were only found in a single region (Fig. 2, mono-coloured bars), and Hydraenidae and

Elmidae were the lineages with the highest number of haplotypes shared across regions (Fig. 2b and

2c, multi-coloured bars).

The α-diversity (at local reaches) ranged from 14 to 66 genera, 2 to 23 species, and 3 to 52

haplotypes (Fig. 3a). The highest estimates of α-diversity at all three levels were found at Picos, Betic

and Carpathians, whereas Jämtland harboured the lowest values. α-diversity significantly decreased

with latitude at the genus (Rho=-0.37, P

Acc

epte

d A

rtic

le


significant (P=0.001) at all three levels, indicating overall high turnover across regions. Genus

composition overlapped across Temperate sites (Picos, Jura, and Carpathians), where intraregional

variability was the lowest, while the Mediterranean and Subarctic zones were located at opposite

sides of the ordination axes. Species composition in the Temperate zone and Jämtland strongly

overlapped and showed low intraregional variability, whereas Rif and Betic were very distinct. At the

haplotype level, Picos and Jura showed the largest intraregional variation, while turnover among

regions was almost complete, and only Carpathians and Jämtland shared any haplotypes.

Compositional turnover among regions varied among hierarchical taxonomical levels. For genera,

42.1% were widely distributed, 29.5% were limited to one climatic zone and 28.4% were located at

only one region, whereas the percentages of species and haplotypes limited to a single region were

85.6% and 99.97%, respectively (Appendixes S4-S8). Overall β-diversity (βsor) between regions

increased significantly with geographical distance at the genus and species levels, whereas

haplotypes showed almost a complete turnover among all pairs of regions, regardless of their

physical proximity (Fig. 4d). The turnover component of β-diversity (βsim) between regions (as

opposed to nestedness, βnes) was the main contributor to overall -diversity across regions: 85.76%,

93.75%, and 99.10% at the genus, species and haplotype levels, respectively.

The degree of intraregional β-diversity (using Hellinger distances) was largely decoupled

from latitude, apart from a marginally significant negative relationship at the genus level (genus:

Rho=-0.82, P=0.06; species: Rho=-0.43, P=0.42; haplotypes: Rho=-0.37, P=0.5; Fig. 3c). Similar trends

were found using Whittaker’s Index of -diversity (species: Rho=-0.6, P=0.24; haplotypes: Rho=-0.08,

P=0.91; Fig. 3d). After controlling for differences in γ-diversity, null-model expectations and

observed patterns were coincident for Hellinger distances (Fig. 3c), whereas observed estimates of

β-diversity using Whittaker’s Index were greater than the values in the null models (Fig. 3d).

Deviations of Whittaker’s Index were larger at species than haplotype levels, indicating a greater

level of aggregated distribution for species than haplotypes.

Acc

epte

d A

rtic

le


Environmental and spatial contributions to β-diversity varied among organizational levels

and regions, but the unexplained (residual) variance never increased with latitude. For example, at

the genus level, the contribution of pure environment and spatial distance was higher in northern

regions, particularly in the Carpathians (Fig. 5a). At the species level, the pure environment

component was important (>15%) in all regions except Betic, and pure distance contributed the

most at Rif and Jämtland (~20%; Fig. 5b). The combined influence of environment and space was

lowest at the haplotype level, and the majority of variation was unexplained (Fig. 5c).

Mean phylogenetic-β-diversity (Fig. 6) among communities within a region clearly decreased

from southern to northern latitudes for Hydraenidae (Rho=-1, P=0.002) and Elmidae (Rho=-0.94,

P=0.01), while it was uncorrelated for Baetis (Rho=0.08, P=0.91) and Hydropsyche (Rho=-0.31,

P=0.56). Nucleotide diversity π ranged from 0 to 0.018 and decreased significantly albeit weakly with

latitude (R2=-0.03, P =0.004; Fig. 7a). The genus Baetis showed the highest values of nucleotide

diversity. Across the entire gradient, 22 out of the 37 species in the analysis showed significant

intraregional genetic structure ΦST (Fig. 7b). The magnitude of ΦST was not significantly related to

latitude, and both significant and non-significant ΦST values were found in all regions. All species in

Jämtland showed ΦST>0.1. The ΦST was particularly high for Baetis (maximum ΦST=0.62) at Picos and

Jura. Genus richness and average intraspecific nucleotide diversity π calculated for all species with

multiple individuals in a local site were positively correlated (SGDC: N=62, R2=0.072, P=0.035),

showing the highest values in Betic and Picos, and lowest correlation in Jura and Jämtland (Fig. 7c).

Discussion

Support for environmental filtering from distribution patterns at multiple organisational levels

In contrast to studies of terrestrial arthropods along the latitudinal gradient in Europe that

supported the postglacial recolonization hypothesis (Hawkins 2010; Hortal et al. 2011; Entling et al.

2012), distributions of freshwater organisms examined here only partly confirmed this model.

Diversity within sites (α-diversity) and regions (γ-diversity) decreased with latitude (Fig. 3a-b), as

Acc

epte

d A

rtic

le


predicted under the postglacial south-to-north range expansion scenario (Hewitt 2000), but the

decrease was mostly limited to the genus level and weak at the other taxonomical levels, as

previously observed at the global scale for freshwater taxa (Heino 2011). In addition, we found

numerous species confined to either the Mediterranean, Temperate or Subarctic zones, including 10

genera located exclusively in the Subarctic. These observations indicate the influence of climatic

differences and local factors in determining species distributions among regions (Heino 2011;

Astorga et al. 2014), rather than the dispersal from southern refugia. We also found high levels of

intraregional turnover (β-diversity), even in northern regions (Fig. 3c-d), which would not be

expected if community assembly occurred through region-wide postglacial colonization of a limited

number of expanding, southern taxa. Instead, our findings support a scenario driven by

environmental conditions, which is corroborated by the following observations (see Introduction):

(1) inter-regional distinctiveness in community composition across regions was high (Fig. 4a-c), with

β-diversity showing only a very small nestedness component (Fig. 4d); (2) the environmental effect

on community composition at the genus and species levels was as strong or stronger at northern

sites (Fig. 5a-b), where the northward migration scenario might have predicted more stochastic (or

unexplained) variation from the random long-distance dispersal of lineages; (3) latitudinal trends of

phylogenetic community structure varied among lineages (Fig. 6) probably reflecting species- or

lineage-specific life-history traits determining the geographic distributions; and (4) nucleotide

diversity π showed only a weak tendency to decrease with latitude (Fig. 7a). In addition, intraspecific

genetic differentiation among populations (ΦST) within a region was fairly high at all latitudes (Fig.

7b), contrary to expectations of reduced diversity and greater homogeneity in recently established

northern populations (Hewitt 2000).

Correlations of species diversity and phylogenetic turnover with environment

Studying patterns of diversity at the three hierarchical levels simultaneously provided key insights

into the eco-evolutionary mechanisms that shaped community composition of mountain freshwater

macroinvertebrates. At the species level, we found that Mediterranean species are different from

Acc

epte

d A

rtic

le


those in the Temperate zones, and only a few species (e.g., Baetis rhodani, Hydraena gracilis, Elmis

aena) are shared between the Temperate and Subarctic zones, while the latter also harbours several

exclusive species (Baetis fuscatus, Baetis subalpinus, Hydraena britteni, Hydropsyche silfvenii).

Phylogenetic-β-diversity of beetles significantly decreased with latitude because Hydraenidae and

Elmidae are strongly diversified in the Mediterranean zones, but not in the Subarctic zone (Fig. 2b-c,

6a-b). Limited cold tolerance has likely reduced the northward expansion and diversification of

freshwater beetles, in agreement with findings for terrestrial beetles (Baselga 2010; Hortal et al.

2011). In contrast, northern communities of Baetis (Fig. 6c) were composed both of species

belonging to the widely distributed rhodani group and of distantly-related, cold-tolerant species

belonging to the niger, fuscatus and subalpinus groups. Low-latitude Baetis assemblages were

composed of closely related and locally diverged species within the rhodani and maurus groups.

Similarly, the latitudinal turnover of Hydropsyche across regions (Fig. 6d) affected species drawn

from the entire phylogeny (Fig. 2d), without obvious region-specific factors structuring species

replacement. Overall, patterns of phylogenetic-β-diversity thus appear to reflect lineage-specific life-

history traits in determining distributions and accumulation of species in different biogeographic

regions, rather than the canonical south-to-north postglacial dispersion from the Mediterranean

area (Taberlet et al. 1998; Provan & Bennett 2008; Stewart et al. 2010; Heino 2011). Since species-

specific ecological preferences presumably arose long before the contemporary climatic gradients

were established, it is more plausible that northern mountain stream species expanded from a

number of different glacial refugia, presumably in parts of Central and Southeastern Europe that

remained ice-free (Pauls et al. 2006; Stewart et al. 2010; Theissinger et al. 2013), which would

explain the high degree of differentiation between contemporary southern and northern

communities. The strong latitudinal structuring of species ranges and phylogenetic lineages,

together with intraregional environmental effects on species composition (in all regions except Betic,

Fig. 5b), indicates the role of environmental filtering.

Acc

epte

d A

rtic

le


Intraspecific genetic structure of communities

The distribution of haplotypes differed from the species and genus level patterns: intraregional β-

diversity of haplotypes was largely independent of environmental parameters. Cox1 sequence

variation is presumably neutral with regard to environment but instead is expected to track the

demographic history of species (Craft et al. 2010; Baselga et al. 2015). Therefore, intraspecific

genetic differentiation may be ‘unexplained’ by any of our measured variables (Fig. 5c), but this does

not contradict the scenario of community assembly due to environmental filtering. We also find that

haplotype β-diversity was also largely unrelated to geographical distance, contrary to expectation

under a scenario of limited dispersal and gradual accumulation of genetic divergence (Baselga et al.

2013; Fujisawa et al. 2015), but consistent with large-scale migration from various source areas and

species distributions determined by the local environment. Within regions, topography and physical

habitats are generally highly heterogeneous across mountain streams (Gooderham et al. 2007), thus

leading to ecological differentiation of communities as we found here. In addition to environmental

heterogeneity, the mountain settings and dendritic connectivity present physical barriers to

dispersal among headwater streams, and thus can contribute to community and population

differentiation (Finn et al. 2011; Múrria et al. 2013; Paz-Vinas & Blanchet 2015). The strong genetic

structure (ΦST, Fig. 7b) in all regions (including the Subarctic Jämtland region) suggests limited

dispersal among streams, which exacerbates the constraints imposed by the environment.

Genetic structure found in freshwater phylogeographic studies at a regional scale has been

commonly associated with Pleistocene post-glacial population expansion from periglacial refugia

(e.g., Pauls et al. 2006; Múrria & Hughes 2008; Benke et al. 2011) and high isolation across

headwater streams and among high elevation mountain ranges (Bálint et al. 2011; Finn et al. 2013;

Pauls et al. 2013). The current analysis confirms the earlier findings of limited dispersion, extending

the scope of the analysis to multiple species and at the continental scale, and now also revealing

isolated biogeographical climatic zones and high intraspecific genetic structure unrelated to latitude.

The strongest genetic structure was found in the karstic landscapes of Picos and Jura where several

Acc

epte

d A

rtic

le


streams emerge from springs. These habitats are stable and typically characterized by unique fauna

and limited gene flow across reaches (Barquín & Death 2006; Sei et al. 2009). In contrast, the few

examples from existing studies of genetically unstructured species (from various families along all

latitudes) might possibly be linked to local habitat characteristics that promote homogenization,

such as frequent disturbance by drought or floods (Abellan et al. 2009; Múrria et al. 2010; Phillipsen

et al. 2015).

Diversity correlations across organisational levels

Patterns of local diversity at multiple hierarchical (species and haplotypes) levels might be coupled

under various scenarios of stochastic dispersal and community persistence (Vellend 2003; Vellend et

al. 2014). We found a weak correlation of genus diversity and average nucleotide diversity of

individual species within a community, which is largely driven by the latitudinal decrease in diversity

at both levels (Fig. 3a, 7c). Although assembly mechanisms appear to vary across latitudes and

lineages (Fig. 5), certain factors (e.g., isolation among regions, local ecological conditions, and drift)

might affect genus and nucleotide diversity in similar ways. In previous studies, no significant SGDC

was found across latitude for plant or tree species, likely due to different factors influencing diversity

at the community (e.g., habitat diversity, environmental filtering) and population levels (e.g., glacial

and post-glacial history, habitat instability) (Puşcaş et al. 2008; Taberlet et al. 2012; Wei & Jiang

2012). Commonly, estimates of intraspecific genetic diversity were based either on only one species

or on data from several widely-distributed species analysed separately, thereby capturing most

often species-specific genetic trajectories. In contrast, community DNA barcoding allows estimation

of intraspecific genetic diversity by summing genetic diversity for all species within a community,

which provides a more comprehensive genetic characterization of entire communities. For instance,

positive SGDCs were explained by long-term habitat stability and strong isolation that promoted

divergence of both communities and populations of tropical highland freshwater insects (Múrria et

al. 2015), or dispersal limitation that have influenced species and genetic diversity of tenebrionid

Acc

epte

d A

rtic

le


beetles similarly on the Aegean Archipelago (Papadopoulou et al. 2011). Although we found a

significant SGDC here (Fig. 7c), overall patterns of variation were not strongly concordant across

hierarchical levels, pointing to an important role of species-specific selection, rather than a

predominant role of neutral processes that are thought to be generally responsible for producing

positive SGDCs (Vellend 2003; Baselga et al. 2013).

Conclusions

We found that the α- and γ-diversity of macroinvertebrates of mountain freshwater communities at

the genus level decreased with latitude – a pattern well-established for terrestrial arthropods (e.g.,

Baselga 2010; Hortal et al. 2011). However, the high level of turnover among regions and among

sites within regions at multiple hierarchical levels suggests an overriding influence of environmental

factors determining the large-scale community composition of freshwater macroinvertebrates

(Heino, 2011; Astorga et al. 2014). The novelty of our approach is in accounting for an evolutionary

time-continuum, integrating diversity from populations through deeper phylogenetic levels and

examining the entire communities, to disentangle the mechanisms underlying macroecological

patterns. We found pronounced differences in community composition across European highland

streams, each with their own evolutionary history and local differentiation resulting in high levels of

endemism. This has practical relevance for biological assessments of water ecological quality.

Because freshwater biodiversity is declining at faster rates than terrestrial biodiversity (Sala et al.

2000) due to losses of habitat quality and connectivity (e.g., through pollution, river regulation,

water over-extraction; Woodward et al., 2010; Markovic et al., 2015), understanding the

mechanisms of origin and maintenance of freshwater diversity is crucial for management and

conservation. To date, freshwater biomonitoring has been assessed commonly at the family

taxonomic level for many groups, but there is an increasing application of DNA sequencing

technologies that allow high-resolution assessments at the genus or even species level (Leese et al.

Acc

epte

d A

rtic

le


2016). Our findings highlight that future routine application in biomonitoring will require a region-

specific approach, taking into account local and regional habitat differentiation to establish

reference conditions and the ecological status of each water body.

Acknowledgements

We are grateful to Kvetoslava Bottová, Kamilia Hajji and Sanae Errochdi for their assistance in the

field. This project was funded by the Spanish Ministry of Education and Science/FEDER (Fondo

Europeo de Desarrollo Regional) (CGL2007-60163/BOS). CM was supported by a Beatriu de Pinós

postdoctoral fellowship (BP-DGR-2011) from Agència de Gestió d'Ajuts Universitaris i de Recerca,

Catalunya. AFF was supported by the FRESHING Project funded by FCT and COMPETE (PTDC/AAG-

MAA/2261/2014 – POCI-01-0145-FEDER-356 016824).

References

Abellan P, Millan A, Ribera I (2009) Parallel habitat-driven differences in the phylogeographical structure of two independent lineages of Mediterranean saline water beetles. Molecular Ecology, 18, 3885-3902.

Adams RI, Hadly EA (2013) Genetic diversity within vertebrate species is greater at lower latitudes. Evolutionary Ecology, 27, 133-143.

Avise JC (2009) Phylogeography: retrospect and prospect. Journal of Biogeography, 36, 3-15. Bálint M, Domisch S, Engelhardt CHM, Haase P, Lehrian S, Sauer J, Theissinger K, Pauls SU, Nowak C

(2011) Cryptic biodiversity loss linked to global climate change. Nature Climate Change, 1, 313-318.

Barquín J, Death RG (2006) Spatial patterns of macroinvertebrate diversity in New Zealand springbrooks and rhithral streams. Journal of the North American Benthological Society, 25, 768-786.

Baselga A (2010) Partitioning the turnover and nestedness components of beta diversity. Global Ecology and Biogeography, 19, 134-143.

Baselga A, Fujisawa T, Crampton-Platt A, Bergsten J, Foster PG, Monaghan MT, Vogler AP (2013) Whole-community DNA barcoding reveals a spatio-temporal continuum of biodiversity at species and genetic levels. Nature Communications, 4, 1892.

Baselga A, Gómez-Rodríguez C, Vogler AP (2015) Multi-hierarchical macroecology at species and genetic levels to discern neutral and non-neutral processes. Global Ecology and Biogeography, 24, 873-882.

Baselga A, Orme CDL (2012) betapart: an R package for the study of beta diversity. Methods in Ecology and Evolution, 3, 808-812.

Acc

epte

d A

rtic

le


Benke M, Brändle M, Albrecht C, Wilke T (2011) Patterns of freshwater biodiversity in Europe: lessons from the spring snail genus Bythinella. Journal of Biogeography, 38, 2021-2032.

Bernatchez L, Wilson CC (1998) Comparative phylogeography of Nearctic and Palearctic fishes. Molecular Ecology, 7, 431-452.

Bonada N, Resh VH (2013) Mediterranean-climate streams and rivers: geographically separated but ecological comparable freshwater systems. Hydrobiologia, 719, 1-29.

Borcard D, Legendre P, Avois-Jacquet C, Tuomisto H (2004) Dissecting the spatial structure of ecological data at multiple scales. Ecology, 85, 1826–1832.

Brooker RW, Callaway RM, Cavieres LA, Kikvidze Z, Lortie CJ, Michalet R, Pugnaire FI, Valiente-Banuet A, Whitham TG (2009) Don’t diss integration: a comment on Ricklefs’s disintegrating communities. The American Naturalist, 174, 919-927.

Cavender-Bares J, Kozak KH, Fine PVA, Kembel SW (2009) The merging of community ecology and phylogenetic biology. Ecology Letters, 12, 693-715.

Condit R, Pitman N, Leigh Jr EG et al. (2002) Beta-diversity in tropical forest trees. Science, 295, 666-669.

Craft KJ, Pauls SU, Darrow K, Miller SE, Hebert PDN, Helgen LE, Novotny V, Weiblen GD (2010). Population genetics of ecological communities with DNA barcodes: an example from Guinea Lepidoptera. Proceedings of the National Academy of Sciences of the United States of America, 107, 5041-5046.

De Cáceres M, Legendre P, Valencia R et al. (2012) The variation of tree beta diversity across a global network of forest plots. Global Ecology and Biogeography, 21, 1191-1202.

Diniz-Filho, JAF, Bini LM (2011) Geographical patterns in biodiversity: towards an integration of concepts and methods from genes to species diversity. Natureza & Conservação, 9, 179-187.

Dolédec S, Chessel D (1989) Rythmes saisonniers et composantes stationnelles en milieu aquatique. II. Prise en compte et élimination d’effets dans un tableau faunistique. Acta Oecologica, Oecologia Generalis, 10, 207-232.

Dormann CF, Elith J, Bacher S et al. (2013) Collinearity: a review of methods to deal with it and a simulation study evaluating their performance. Ecography, 36, 27-46.

Dray S, Dufour AB (2007) The ade4 package: implementing the duality diagram for ecologists. Journal of Statistical Software, 22, 1-20.

EEA (2010) EU-DEM - Digital Elevation Model over Europe derived from GTOPO30 (cell size 1000m). GMES RDA project of the European Environmental Agency.

Entling MH, Schweiger O, Bacher S, Espadaler X, Hickler T, Kumschick S, Woodcock BA,Nentwig W (2012) Species richness-environment relationships of European arthropods at two spatial grains: habitats and countries. Plos One, 7, e45875.

European Soil Bureau Network European Commission (2005) Soil Atlas of Europe, Office for Official Publications of the European Communities, Luxembourg.

Excoffier L, Laval G, Schneider S (2005) Arlequin ver. 3.0: an integrated software package for population genetics data analysis. Evolutionary Bioinformatics Online, 1, 47-50.

Field R, Hawkins BA, Cornell HVet al. (2009). Spatial species-richness gradients across scales: a meta-analysis. Journal of Biogeography, 36, 132-147.

Finn DS, Bonada N, Múrria C, Hughes JM (2011) Small but mighty: headwaters are vital to stream network biodiversity at two levels of organization Journal of North American Benthological Society, 30, 963-980.

Finn DS, Khamis K, Milner AM (2013) Loss of small glaciers will diminish beta diversity in Pyrenean streams at two levels of biological organization. Global Ecology and Biogeography, 22, 40-51.

Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R (1994) DNA primers for amplification of mitochondrial cytochrome C oxidase subunit I from diverse metazoan invertebrates. Molecular Marine Biology and Biotechnology, 3, 294-299.

Acc

epte

d A

rtic

le


Fujisawa T, Barraclough TG (2013) Delimiting species using single-locus data and the Generalized Mixed Yule Coalescent approach: A revised method and evaluation on simulated data sets. Systematic Biology, 62, 707-724.

Fujisawa T, Vogler AP, Barraclough TG (2015) Ecology has contrasting effects on genetic variation within species versus rates of molecular evolution across species in water beetles. Proceedings of the Royal Society B, 282, 20142476.

Gaston KJ, Blackburn JH (2000) Patterns and process in macroecology. Blackwell Science, Oxford. Gillespie RG (2004) Community assembly through adaptive radiation in Hawaiian spiders. Science,

303, 356-359. Gooderham JP, Barmuta LA, Davies PE (2007) Upstream heterogeneous zones: small stream systems

structured by a lack of competence? Journal of the North American Benthological Society, 26, 365-374.

Griffith DA, Peres-Neto PR (2006) Spatial modelling in ecology: the flexibility of eigenfunction spatial analyses. Ecology, 87, 2603–2613.

Hawkins BA (2010) Multiregional comparison of the ecological and phylogenetic structure of butterfly species richness gradients. Journal of Biogeography, 37, 647-656.

Hawkins BA, Porter EE, Diniz-Filho JAF (2003) Productivity and history as predictors of the latitudinal diversity gradient of terrestrial birds. Ecology, 84, 1608-1623.

Hebert PDN, Cywinska A, Ball SL, deWaard JR (2003) Biological identifications through DNA barcodes. Proceedings of the Royal Society B, 270, 313-321.

Heino J (2011) A macroecological perspective of diversity patterns in the freshwater realm. Freshwater Biology, 56, 1703-1722.

Hewitt GM (2000) The genetic legacy of the Quaternary ice ages. Nature, 68, 87-112. Hof C, Brändle M, Brandl R (2008) Latitudinal variation of diversity in European freshwater animals is

not concordant across habitat types. Global Ecology and Biogeography, 17, 539-546. Hortal J, Diniz-Filho JAF, Bini LM et al. (2011) Ice age climate, evolutionary constraints and diversity

patterns of European dung beetles. Ecology Letters, 14, 741-748. Jarvis A., H.I. R., AN, EG (2008) Void-filled SRTM (NASA Shuttle Radar Topographic Mission) 90m

Digital Elevation Data (DEM) for the globe. Version 4. Kembel SW, Cowan PD, Helmus MR, Cornwell WK, Morlon H, Ackerly DD, Blomberg SP, Webb CO

(2010) Picante: R tools for integrating phylogenies and ecology. Bioinformatics, 26, 1463-1464.

Kraft NJ, Comita LS, Chase JM et al., (2011) Disentangling the drivers of β diversity along latitudinal and elevational gradients. Science, 333, 1755-1758.

Leese F, Altermatt F, Bouchez A et al., (2016) DNAqua-Net: Developing new genetic tools for bioassessment and monitoring of aquatic ecosystems in Europe. Research Ideas and Outcomes, 2, e11321.

Leffler EM, Bullaughey K, Matute DR, Meyer WK, Ségurel L, Venkat A, Andolfatto P, Przeworski, M (2012) Revisiting an old riddle: what determines genetic diversity levels within species? PLoS Biology, 10, e1001388.

Legendre P, De Cáceres M (2013) Beta diversity as the variance of community data: dissimilarity coefficients and partitioning. Ecology Letters, 16, 951-963.

Lehrian S, Pauls SU, Haase P (2009) Contrasting patterns of population structure in the montane caddisflies Hydropyshe tenuis and Drusus discolor in the Central European highlands. Freshwater Biology, 54, 283-295.

Losos JB, Jackman TR, Larson A, Queiroz de K, Rodríguez-Schettino L (1998) Contingency and determinism in replicated adaptive radiations of island lizards. Science, 279, 2115-2118.

Markovic D, Carrizo S, Freyhof J, Cid N, Lengyel S, Scholz M, Kasperdius H, Darwall W (2014) Europe’s freshwater biodiversity under climate change: Distribution shifts and conservation needs. Diversity and Distributions, 20, 1097–1107.

Acc

epte

d A

rtic

le


Marske KA, Rahbek C, Nogués-Bravo D (2013) Phylogeography: spanning the ecology‐evolution continuum. Ecography, 36, 1169-1181.

Martin PR, McKay JK (2004) Latitudinal variation in genetic divergence of populations and the potential for future speciation. Evolution, 58, 938-945.

Miraldo A, Li S, Borregaard MK et al., (2016) An Anthropocene map of genetic diversity. Science, 353, 1532-1535.

Múrria C, Bonada N, Arnedo MA, Prat N, Vogler AP (2013) Higher β- and γ-diversity at species and genetic levels in headwaters than in mid-order streams in Hydropsyche (Trichoptera). Freshwater Biology, 58, 2226-2236.

Múrria C, Bonada N, Ribera C, Prat N (2010) Homage to the virgin of Ecology, or why an aquatic insect unadapted to desiccation may maintain population in very small, temporary Mediterranean streams? Hydrobiologia, 653, 179-190.

Múrria C, Hughes JM (2008) Cyclic habitat displacements during Pleistocene glaciations have induced independent evolution of Tasimia palpata populations (Trichoptera: Tasimiidae) in isolated subtropical rain forest patches. Journal of Biogeography, 35, 1727-1737.

Múrria C, Rugenski AT, Whiles MR, Vogler AP (2015) Long-term isolation and endemicity of Neotropical aquatic insects limit the community responses to recent amphibian decline. Diversity and Distributions, 21, 938-949.

Myers JA, Chase JM, Jiménez I, Jørgensen PM, Araujo-Murakami A, Paniagua‐Zambrana N, Seidel R (2013) Beta-diversity in temperate and tropical forests reflects dissimilar mechanisms of community assembly. Ecology Letters, 16, 151-157.

Nei M (1987) Molecular Evolutionary Genetics. Columbia University Press, New York. Oksanen J, Blanchet FG, Kindt R et al., 2013 Package ‘vegan’. Community ecology package, version,

2(9). Oppold AM, Pedrosa JA, Bálint M, Diogo JB, Ilkova J, Pestana JL, Pfenninger M (2016) Support for the

evolutionary speed hypothesis from intraspecific population genetic data in the non-biting midge Chironomus riparius. Proceedings of the Royal Society B, 283, 20152413.

Papadopoulou A, Bergsten J, Fujisawa T, Monaghan MT, Barraclough TG, Vogler AP (2008) Speciation and DNA barcodes: testing the effects of dispersal on the formation of discrete sequence clusters. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 363, 2987-2996.

Papadopoulou A, Anastasiou I, Keskins B, Vogler AP (2009) Comparative phylogeography of tenebrionid beetles in the Aegean archipelago: the effects of dispersal ability and habitat preference. Molecular Ecology, 18, 2503-2517.

Papadopoulou A, Anastasiou I, Spagopoulou F, Stalimerou M, Terzopoulou S, Legakis A, Vogler AP (2011) Testing the Species-Genetic Diversity Correlations in the Aegean Archipelago: toward a haplotype-based macroecology? The American Naturalist, 178, 241-255.

Paradis E, Claude J, Strimmer K (2004) APE: analyses of phylogenetics and evolution in R language. Bioinformatics, 20, 289-290.

Pauls SU, Lumbsch HT, Haase P (2006) Phylogeography of the montane caddisfly Drusus discolor: evidence for multiple refugia and periglacial survival. Molecular Ecology, 15, 2153-2169.

Pauls SU, Nowak C, Balint M, Pfenninger M (2013) The impact of global climate change on genetic diversity within populations and species. Molecular Ecology, 22, 925-946.

Paz-Vinas I, Blanchet S (2015) Dendritic connectivity shapes spatial patterns of genetic diversity: a simulation-based study. Journal of Evolutionary Biology, 28, 986-994.

Peres-Neto PR, Legendre P, Dray S, Borcard D (2006) Variation partitioning of species data matrices estimation and comparison of fractions. Ecology, 87, 2614-2625.

Pfenninger M, Posada D (2002) Phylogeographic history of the land snail Candidula unifasciata (Helicellinae, Stylommatophora): fragmentation, corridor migration, and secondary contact. Evolution, 56, 1776-1788.

Acc

epte

d A

rtic

le


Phillipsen IC, Kirk EH, Bogan MT, Mims MC, Olden JD, Lytle DA (2015) Dispersal ability and habitat requirements determine landscape‐level genetic patterns in desert aquatic insects. Molecular Ecology, 24, 54-69.

Provan J, Bennett KD (2008) Phylogeographic insights into cryptic glacial refugia. Trends in Ecology and Evolution, 23, 564-571.

Puşcaş M, Taberlet P, Choler P (2008) No positive correlation between species and genetic diversity in European alpine grasslands dominated by Carex curvula. Diversity and Distributions, 14, 852-861.

Qian H, Ricklefs RE (2007) A latitudinal gradient in large-scale beta diversity for vascular plants in North America. Ecology Letters, 10, 737-744.

R Development Core Team (2013). R: a language and environment for statistical computing. In. R Foundation for Statistical Computing Vienna, Austria. Available at: http://www.R-project.org/.

Ray N (2005) Pathmatrix: a geographical information system tool to compute effective distances among samples. Molecular Ecology Notes, 5, 177-180.

Ribera I, Foster GN, Vogler AP (2003) Does habitat use explain large scale species richness patterns of aquatic beetles in Europe? Ecography, 26, 145-152.

Ribera I, Vogler AP (2004) Speciation of Iberian diving beetles in Pleistocene refugia (Coleoptera, Dystiscidae). Molecular Ecology, 13, 179-193.

Ricklefs RE (2004) A comprehensive framework for global patterns in biodiversity. Ecology Letters, 7, 1-15.

Romiguier J, Gayral P, Ballenghien M et al. (2014) Comparative population genomics in animals uncovers the determinants of genetic diversity. Nature, 515, 261-263.

Rull V (2014) Time continuum and true long-term ecology: from theory to practice. Frontiers in Ecology and Evolution, 2, 75.

Sanderson MJ (2003) r8s: Inferring absolute rates of molecular evolution and divergence times in the absence of a molecular clock. Bioinformatics, 19, 301-302.

Sei M, Lang BK, Berg DJ (2009) Genetic and community similarities are correlated in endemic-rich springs of the northern Chihuahuan Desert. Global Ecology and Biogeography, 18, 192-201.

Sarremejane R, Mykrä H, Bonada N, Aroviita J, Muotka T (2017) Habitat connectivity and dispersal ability drive the assembly mechanisms of macroinvertebrate communities in river networks. Freshwater Biology, 62, 1073-1082.

Shah DN, Tonkin JD, Haase P, Jähnig SC (2015) Latitudinal patterns and large-scale environmental determinants of stream insect richness across Europe. Limnologica-Ecology and Management of Inland Waters, 55, 33-43.

Simon C, Frati F, Beckenbach A, Crespi B, Liu H, Flook P (1994) Evolution, weighting and phylogenetic utility of mitochondrial gene sequences and compilation of conserved polymerase chain reactions primers. Annals of the Entomological Society of America, 87, 651-701.

Stamatakis A (2006) RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics, 22, 2688-2690.

Stewart JR, Lister AM, Barnes I, Dalén L (2010) Refugia revisited: individualistic responses of species in space and time. Proceedings of the Royal Society of London B: Biological Sciences, 277, 661-671.

Svenning J-C, Fløjgaard C, Baselga A (2011) Climate, history and neutrality as drivers of mammal beta diversity in Europe: insights from mutilscale deconstruction. Journal of Animal Ecology, 80, 393-402.

Svenning JC, Skov F (2007) Could the tree diversity pattern in Europe be generated by postglacial dispersal limitation? Ecology Letters, 10, 453-460.

Taberlet P, Fumagalli L, Wust-Saucy AG, Cosson JF (1998) Comparative phylogeography and postglacial colonization routes in Europe. Molecular Ecology, 7, 453-464.

Acc

epte

d A

rtic

le


Taberlet P, Zimmermann NE, Englisch T et al., (2012) Genetic diversity in widespread species is not congruent with species richness in alpine plant communities. Ecology Letters, 15, 1439-1448.

Tachet H, Richoux P, Bournaud M, Usseglio-Polaterra P (2010) Invertébrés d'eau douce. CNRS éditions, Paris.

Tänzler R, Sagata K, Surbakti S, Balke M, Riedel A (2012) DNA barcoding for community ecology - how to tackle a hyperdiverse, mostly undescribed Melanesian fauna. Plos ONE, 7, e28832.

Theissinger K, Bálint M, Feldheim KA, Haase P, Johannesen J, Laube I, Pauls SU (2013) Glacial survival and post‐glacial recolonization of an arctic–alpine freshwater insect (Arcynopteryx dichroa, Plecoptera, Perlodidae) in Europe. Journal of Biogeography, 40, 236-248.

Timmermans MJTN, Dodsworth S, Culverwell CL, Bocak L, Ahrens D, Pons J, Vogler AP (2010) Why barcode? High-throughput multiplex sequencing of mitochondrial genomes for molecular systematics. Nucleic Acids Research, 38, 1-14.

Vellend M (2003) Island biogeography of genes and species. The American Naturalist, 162, 358-365. Vellend M, Lajoie G, Bourret A, Múrria C, Kembel SW, Garant D (2014) Drawing ecological inferences

from coincident patterns of population- and community-level biodiversity. Molecular Ecology, 23, 2890-2901.

Vellend M (2016) The theory of ecological communities. Princeton University Press, Princeton, US. Webb CO, Ackerly DD, McPeek MA. Donoghue MJ (2002) Phylogenies and community ecology.

Annual Review of Ecology and Systematics, 33, 475-505. Wei X, Jiang M (2012) Contrasting relationships between species diversity and genetic diversity in

natural and disturbed forest tree communities. New Phytologist, 193, 779-786. Whittaker RH (1960) Vegetation of the Siskiyou Mountains, Oregon and California. Ecological

monographs, 30, 279-338. Woodward G, Perkins DM, Brown LE (2010) Climate change and freshwater ecosystems: impacts

across multiple levels of organization. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 365, 2093-2106.

Author Contributions CM, NB, CZM and APV conceived the study. CM, NB, CZM, JG, MEA, JB, TD, MAC and MSBS collected the field data and samples. NB, CZM, JAT, CSC, JG, TD and RA did morphological examination. CM did the physicochemical analyses and the molecular work and APV supervised the molecular work. CM and APV conducted the phylogenetic and population genetic analyses, and CM and MV conducted the statistical analyses. AFF prepared the maps and calculated geographical distances. CM and APV wrote the first draft, and NB, MV, CZM and AFF improved successive versions. All authors read, edited and approved the final version of the article.

Acc

epte

d A

rtic

le


Table 1. Description of spatial extend and configuration of the sites for each region. n, number of sites in each region.

Region n Inter-site

range (km) Inter-site

mean (km) Altitude

range (m) Altitude

mean (m)

Rif 9 1.83-35.78 15.65 63-1362 887.44

Betic 10 4.98-165.75 71.75 124-1314 983.6

Picos 10 8.47-171.54 71.85 137-855 473.6

Jura 11 11.16-128.64 58.22 332-931 590.72

Carpathians 11 7.47-101.34 53.96 614-938 725.09

Jämtland 11 13.97-107.11 53.27 306-478 342.72 Data Accessibility DNA sequences: deposited in GenBank (accessions nos. KY656166-KY656253 and MF783180-MF783882) Supporting information Additional Supporting information may be found in the online version of this article: Appendix S1. Sampled localities Appendix S2. Bibliography used to identify all specimens used in the molecular analyses at the species level Appendix S3. Description of the 21 environmental variables measured Appendix S4. Genus composition and rank-abundance per site Appendix S5-S8. GMYC and cox1 haplotype composition per site for the four lineages: Baetis (Appendix S5), Hydraenidae (Appendix S6), Hydropsyche (Appendix S7), and Elmidae (Appendix S8) Figures S1-S4. Maximum likelihood phylogenetic trees of unique cox1 haplotypes for four lineages: Baetis (Fig. S1), Hydraenidae (Fig. S2), Elmidae (Fig. S3), and Hydropsyche (Fig. S4).

Acc

epte

d A

rtic

le


Figure 1. Location of the 62 sites in the six mountain regions in Europe and North Morocco. The

regions sampled are the mountain ranges of Rif, Betic, Picos, Jura, Carpathians, and Jämtland. In

each region, 9 to 11 sites were sampled.

Acc

epte

d A

rtic

le


Figure 2. Maximum likelihood phylogenetic trees of unique cox1 haplotypes for four lineages: (a)

Baetis, (b) Hydraenidae, (c) Elmidae, and (d) Hydropsyche (scale bars at the bottom of the trees

correspond to 0.2 nucleotide substitutions per site). Branches coloured in red correspond to

variation within the inferred GMYC entities. Each haplotype (terminal on trees) is coloured in

accordance with its mountain region. Note that haplotypes located in more than one site are

coloured with different colours. Names of genera are given for Elmidae and Hydraenidae.

Acc

epte

d A

rtic

le


Figure 3. Trends of the estimates of (a) α-diversity; (b) γ-diversity; (c) β-diversity using the Hellinger

distance; and (d) the Whittaker’s Index of -diversity along the latitude. In purple genus, in green

GMYC species and in orange cox1 haplotypes. In (c) and (d), circles: observed values, and crosses:

expected values. Only significant correlations (plain lines: P

Acc

epte

d A

rtic

le


Figure 4. Correspondence analyses (CA) of community composition across the 62 sites at three

levels: (a) genus, (b) species; and (c) cox1 haplotype. The ellipses envelop 95% of sites of each

mountain region (from left to right R-Rif, B-Betic, P-Picos, J-Jura, C-Carpathians, and S-Jämtland). (d)

Overall β-diversity (βsor in Baselga 2010) between regions along geographical distances at the genus

(purple, r2=0.67, P

Acc

epte

d A

rtic

le


Figure 5. Variation partitioning of the contribution to β-diversity of environment and space across

regions at the (a) genus; (b) species; and (c) haplotype levels. Regions are ordered by increasing

latitude; acronyms for mountain regions: R-Rif, B-Betic, P-Picos; J-Jura, C-Carpathians, and S-

Jämtland.

Acc

epte

d A

rtic

le


Figure 6. Variation of pairwise phylogenetic-β-diversity (measured using the ComDist index) per each

region. Mean value of phylogenetic-β-diversity for a region is correlated with latitude: (a) Elmidae

(Rho=-0.94, P=0.01), (b) Hydraenidae (Rho=-1, P=0.002), (c) Hydropsyche (Rho=-0.31, P=0.56), and

(d) Baetis (Rho=0.08, P=0.91). Acronyms for mountain regions: R-Rif, B-Betic, P-Picos; J-Jura, C-

Carpathians, and S-Jämtland.

Acc

epte

d A

rtic

le


Figure 7. Variation of intraspecific genetic diversity with latitude. (a) Nucleotide diversity for each

site and species with ≥ five individuals sequenced. (b) Genetic structure (φST) of species distributed in

more than three sites with at least 5 individuals sequenced per site; empty circles show non-

significant φST values. (c) Correlation between genus richness and average nucleotide diversity π per

site across the 62 sites (R2=0.072, P=0.035). Acronyms for mountain regions: R-Rif, B-Betic, P-Picos; J-

Jura, C-Carpathians, and S-Jämtland.

View publication statsView publication stats

local environment rather than past climate determines ......carmen elisa sáinz-cantero university...

Documents