ArticlePDF Available

Abstract and Figures

Identifying barriers that govern parasite community assembly and parasite invasion risk is critical to understand how shifting host ranges impact disease emergence. We studied regional variation in the phylogenetic compositions of bird species and their blood parasites (Plasmodium and Haemoproteus spp.) to identify barriers that shape parasite community assembly. Australasia and Oceania. We used a data set of parasite infections from >10,000 host individuals sampled across 29 bioregions. Hierarchical models and matrix regressions were used to assess the relative influences of interspecies (host community connectivity and local phylogenetic distinctiveness), climate and geographic barriers on parasite local distinctiveness and composition. Parasites were more locally distinct (co-occurred with distantly related parasites) when infecting locally distinct hosts, but less distinct (co-occurred with closely related parasites) in areas with increased host diversity and community connectivity (a proxy for parasite dispersal potential). Turnover and the phylogenetic symmetry of parasite communities were jointly driven by host turnover, climate similarity and geographic distance. Interspecies barriers linked to host phylogeny and dispersal shape parasite assembly, perhaps by limiting parasite establishment or local diversification. Infecting hosts that co-occur with few related species decreases a parasite's likelihood of encountering related competitors, perhaps increasing invasion potential but decreasing diversification opportunity. While climate partially constrains parasite distributions, future host range expansions that spread distinct parasites and diminish barriers to host shifting will likely be key drivers of parasite invasions.
Content may be subject to copyright.
Diversity and Distributions. 2017;1–11.  
© 2017 John Wiley & Sons Ltd
DOI: 10.1111/ddi.12661
Climate, host phylogeny and the connectivity of host
communities govern regional parasite assembly
Nicholas J. Clark1| Sonya M. Clegg2| Katerina Sam3| William Goulding4,5|
Bonny Koane6| Konstans Wells7
1School of Veterinary Science, University of
Queensland, Gatton, Qld, Australia
2Edward Grey Institute of Field
Ornithology, Department of
Zoology, University of Oxford, Oxford, UK
3Biology Centre CAS, Faculty of
Science, Institute of Entomology and
University of South Bohemia, Branisovska,
Ceske Budejovice, Czech Republic
4The Landscape Ecology and Conservation
Group, School of Earth and Environmental
Science, University of Queensland, St Lucia,
Qld, Australia
5Biodiversity and Geosciences
Program, Queensland Museum, South
Brisbane, Qld, Australia
6The New Guinea Binatang Research Centre,
Madang, Papua New Guinea
7Environmental Futures Research
Institute, School of Environment, Griffith
University, Nathan, Qld, Australia
Nicholas J. Clark, School of Veterinary Science,
University of Queensland, Gatton, Qld,
Funding information
Griffith University New Researcher Grant;
National Geographic Society Committee for
Research and Exploration Grant, Grant/Award
Number: 9383-13; Birds Queensland; Czech
Science Foundation, Grant/Award Number:
14-36098G; staff of The New Guinea
Binatang Research Centre; Darwin Initiative
for the Survival of Species, Grant/Award
Number: 22-00
Editor: Jeremy Austin
Aim: Identifying barriers that govern parasite community assembly and parasite inva-
sion risk is critical to understand how shifting host ranges impact disease emergence.
We studied regional variation in the phylogenetic compositions of bird species and
their blood parasites (Plasmodium and Haemoproteus spp.) to identify barriers that
shape parasite community assembly.
Location: Australasia and Oceania.
Methods: We used a data set of parasite infections from >10,000 host individuals
sampled across 29 bioregions. Hierarchical models and matrix regressions were used
to assess the relative influences of interspecies (host community connectivity and
local phylogenetic distinctiveness), climate and geographic barriers on parasite local
distinctiveness and composition.
Results: Parasites were more locally distinct (co- occurred with distantly related para-
sites) when infecting locally distinct hosts, but less distinct (co- occurred with closely
related parasites) in areas with increased host diversity and community connectivity (a
proxy for parasite dispersal potential). Turnover and the phylogenetic symmetry of
parasite communities were jointly driven by host turnover, climate similarity and geo-
graphic distance.
Main conclusions: Interspecies barriers linked to host phylogeny and dispersal shape
parasite assembly, perhaps by limiting parasite establishment or local diversification.
Infecting hosts that co- occur with few related species decreases a parasite’s likelihood
of encountering related competitors, perhaps increasing invasion potential but de-
creasing diversification opportunity. While climate partially constrains parasite distri-
butions, future host range expansions that spread distinct parasites and diminish
barriers to host shifting will likely be key drivers of parasite invasions.
community assembly, host shifting, host specificity, interspecies barriers, parasite invasion,
Regional variation in community composition is a central property in
nature (Kraft, Cornwell, Webb, & Ackerly, 2007; Wallace, 1876). With
increasing environmental destabilization and biotic homogenization,
predicting how ecosystems will function following disturbance relies
on identifying processes that govern community assembly (Ricklefs,
1987; Barnagaud et al., 2014; see Table 1 for bold term definitions).
Understanding parasite community assembly is crucial, as changes to
parasite composition or the frequency of host–parasite interactions
   CLARK et AL.
can alter risks of parasite invasions and emerging disease (Adlard,
Miller, & Smit, 2015; Agosta, Janz, & Brooks, 2010; Brooks & Hoberg,
2007; Hoberg & Brooks, 2008; Lafferty, 2009).
A strong incentive exists to identify barriers to species estab-
lishment and determine how these barriers modulate invasion risk
(Hoberg, 2010; Kelly, Paterson, Townsend, Poulin, & Tompkins, 2009;
Springborn et al., 2015). For parasites, geographic barriers (such as
distance or mountain ranges) are known to constrain species’ distri-
butions (Brooks & Ferrao, 2005; Krasnov, Shenbrot, Khokhlova, &
Degen, 2016; Lafferty, 2009; Warburton, Kohler, & Vonhof, 2016).
In addition, environmental barriers (such as temperature and precip-
itation) drive development or transmission rates for many parasites,
especially vector- borne parasites such as those causing malaria and
lyme disease (Epstein, 2001; Githeko, Lindsay, Confalonieri, & Patz,
2000; Patz, Campbell- Lendrum, Holloway, & Foley, 2005). However,
parasite distributions are also linked to host life histories and distribu-
tions (Poulin, Krasnov, & Mouillot, 2011; Olsson- Pons, Clark, Ishtiaq,
& Clegg, 2015; Fecchio et al., 2017). Such interspecies barriers are
increasingly recognized to govern local assembly (HilleRisLambers,
Adler, Harpole, Levine, & Mayfield, 2012; Mayfield & Stouffer, 2017;
Wisz et al., 2013). Predicting how parasite composition may change in
the future relies on defining a consistent framework to identify pat-
terns that improve knowledge of assembly and elucidate underlying
mechanisms acting as barriers. Such patterns may be driven by a hier-
archical process, where parasites must first break through geographic
and/or environmental barriers to initially colonize a new range (Agosta
et al., 2010; Brooks & Hoberg, 2007). Following colonization, assem-
bly may be limited by interspecies barriers that govern parasite spread
and diversification (Figure 1). This process, termed “ecological fitting”
(Janzen, 1985), suggests many parasites are capable of infecting a
broader range of hosts than is currently realized, with changes to host
and/or parasite distributions producing new associations that may be
limited by host phylogenetic relationships (Araujo et al., 2015; Brooks
& Ferrao, 2005; Radtke, McLennan, & Brooks, 2002).
For parasites that rely on host dispersal to colonize new areas, re-
gions comprising a diversity of host species whose ranges overlap with
other potential hosts (i.e., high distributional connectivity to other re-
gions; “host community connectivity”) should support broader parasite
diversity due to increased niche space (Hector, Dobson, Minns, Bazeley-
White, & Hartley Lawton, 2001, Viana, D. S., Santamaría, L., & Figuerola,
J. (2016)) and a higher likelihood for parasites to break geographic and/
or environmental barriers (Figure 1). However, biotic barriers could still
limit parasite invasions in phylogenetically diverse systems, particularly
if invasion success is positively related to the invader’s local phyloge-
netic distinctiveness (i.e., more locally distinct invaders are less likely
to be limited by related competitors; HilleRisLambers et al., 2012). Yet,
while host community connectivity can overcome geographic dispersal
barriers, few studies recognize this aspect as a potential driver of para-
site assembly (but see Buckee, Danon, & Gupta, 2007).
Parasites are often restricted to hosts with phylogenetically con-
served ecological or physiological traits (Janzen, 1968; Rohde, 1980;
Schulze- Lefert & Panstruga, 2011; Streicker et al., 2010), a phenome-
non that has powerful consequences for species interactions and eco-
system functioning (Ehrlich & Raven, 1964; Hoberg & Brooks, 2008).
As parasites with high host specificity may be unable to shift hosts,
the local availability of suitable hosts can present an invasion barrier
following initial dispersal, especially if parasites are adapted to hosts
that do not commonly co- occur with closely related species (Brooks,
1979; Ewen et al., 2012; Clark & Clegg, 2015; Ellis et al., 2015; Mata,
da Silva, Lopes, & Drovetski, 2015; Figure 1).
While ecological fitting (governed at least partly by parasite
host-specificity and host evolutionary history) and host dispersal po-
tential are clearly important mechanisms impacting parasite establish-
ment and diversification, identifying their roles in natural host–parasite
systems is challenging. We develop a framework to identify relative
influences of barriers to regional parasite community assembly and
apply this framework to naturally occurring parasite infections from
Australasian bird communities. Haemosporidians (genera Plasmodium
TABLE1 Glossary of definitions for proposed community
assembly barriers and metrics used in analyses
Community assembly: The establishment and maintenance of local
communities through arrival of potential colonists from external
species pools.
Environmental barriers: Environmental differences between regions
that may govern species’ distributions, including variation in
macroclimate, habitat and altitude.
Geographic barriers: Physical barriers to between- region parasite
dispersal, including geographic distance, mountain ranges and water
Host community connectivity: The distributional overlap of host
communities among regions, taking into account host species
richness and host geographic range sizes. Here, Sampled.ConH
describes host community connectivity while considering only
sampled avian host species, and Total.ConH describes connectivity
for all occurring avian species within a local assemblage.
Host specificity: the range and diversity of host species observed to
be infected by a parasite. Here, d’ describes parasite host-specificity
using host–parasite interaction networks, while STD* describes
phylogenetic host specificity using host phylogenetic distances.
Interspecies barriers: for parasites, interspecies barriers relate to
variation in host species attributes that prevent parasite spread and
diversification. These may include host phylogenetic relatedness and
ecological similarity (e.g., microhabitat use, nesting behaviour and
feeding behaviour).
Local phylogenetic distinctiveness: the average pairwise phylogenetic
distance between a focal taxon and co- occurring taxa within a local
assemblage. Here, DisP describes parasite species distinctiveness,
Sampled.DisH describes host species distinctiveness with respect to
co- occurring sampled host species, and Total.DisH describes host
species distinctiveness with respect to all co- occurring sampled avian
Phylogenetic community skewness: a measure of the asymmetry of
species’ pairwise phylogenetic distances, where a left skew indicates
relatively more distantly than closely related species in a community,
while a right skew indicates the opposite.
Phylogenetic turnover (β): shifts in phylogenetic diversity between
communities. Here, βP describes parasite phylogenetic turnover,
Sampled.βH describes turnover of sampled host assemblages, and
Total.βH describes turnover of total avian assemblages.
and Haemoproteus) are vector- borne blood parasites that display a
vector dispersal (Ejiri et al., 2011), avian hosts are the primary vehi-
cles by which these parasites disperse (Pérez- Tris & Bensch, 2005).
Avian haemosporidians have been introduced to numerous bioregions,
sometimes with devastating effects on native birds, raising questions
about how interspecies and geographic barriers regulate parasite as-
sembly and invasion potential (Hellgren et al., 2014; van Riper, van
Riper, Goff, & Laird, 1986, Clark, Clegg & Lima 2014).
We assess barriers that may govern parasite local coexistence at
the species level by estimating effects of host community connectivity
and interspecies barriers (host phylogeny and parasite host-specificity)
on parasite local phylogenetic distinctiveness. We then address bar-
riers at the community level by (1) exploring effects of host phyloge-
netic turnover, environmental variation and geographic distance on
parasite turnover, and (2) testing if host connectivity or environmental
variation influence parasite phylogenetic community skewness. We
expect that increased host community connectivity reduces barriers
to parasite establishment, leading to phylogenetically homogenized
parasite communities. If host phylogeny acts as a relatively strong in-
terspecies barrier to parasite assembly, we expect that distinct hosts
carry distinct parasites and that between- region host turnover pre-
dicts parasite turnover. We also expect host- specialist parasites to
be more locally distinct than generalists, as specialists may have less
opportunity to diversify through host range expansions. Alternatively,
if higher diversities of host specialists are able to co- occur through
extensive niche packing (Ricklefs, 2010), then we expect specialists to
be less distinct than generalists.
2.1 | Host–parasite occurrence data and avian
community connectivity
We surveyed published literature and queried the MalAvi database
(; accessed September
2016; Bensch, Hellgren, & Pérez- Tris, 2009) to compile data from
>10,000 sampled host individuals (from 297 avian species) across
83 sites, ranging across latitudes −50.77 to 14.27 and longitudes
tified using PCR targeting the cytochrome- b (cyt- b) gene (Hellgren,
Waldenström, & Bensch, 2004; Waldenström, Bensch, Hasselquist,
& Östman, 2004). Evidence indicates lineages differing by as little
as one base pair may be reproductively isolated (Bensch, Pérez- Tris,
Waldenström, & Hellgren, 2004). We thus regard each unique se-
quence as a parasite “species”. Low numbers of recovered parasites
at some sites meant we could not assess within- site composition. We
thus grouped sites into 29 regions. Australian mainland sites were
grouped by climate zone using the Bureau of Meteorology’s Köppen
classification, which defines zones using temperature, precipitation
and vegetation data (
ages/climate-classifications/; accessed November 2016). Papua New
Guinea mainland sites were grouped based on elevation (highlands,
mean altitude = 2,500 m; and lowlands, mean altitude = 60 m). Island
sites were either grouped by island (if at least three parasite species
were recovered) or into regions representing nearby islands in an ar-
chipelago (Figure 2; Data set S1).
FIGURE1 Schematic illustrating potential barriers to regional spread and diversification for parasites that rely on host movement for
dispersal. Plates represent different bioregions, while zones (forest, mountain) within plates represent different habitat types. At the bottom
left is a sectional zoom of the forested habitat in the left- hand plate, illustrating within- region parasite diversification where closely related
host species enable the breakdown of interspecies barriers. Shown in black is the focal host of a given parasite species, with ecologically or
phylogenetically similar host species depicted as similar shapes in varying shades of grey. A distantly related host species is depicted as a
different body shape. Concentric oval shapes represent parasites, with different shapes and colours representing different parasite species
   CLARK et AL.
We downloaded range maps for all avian species occurring in
the study area (N = 3,024 species) from BirdLife International and
NatureServe (; accessed October
2016). For each region, we obtained lists of occurring avian species (de-
fined as the “total” assemblage) by recording all species whose ranges
overlapped 111 km buffers (1° at the equator) around sites. Bird range
sizes were calculated as the total area of range polygons. Range sizes var-
ied from 1 km2 (island endemics) to 28,000 km2 (wide- ranging seabirds).
Avian community connectivity was calculated as an inverse
Simpson diversity index (Simpson 1949) using species’ range sizes as
weights (instead of using species abundances). Here, increased species
richness, larger species range sizes and more even range size distri-
butions all lead to increased collective mobility of a local host assem-
blage. Two connectivity indices were created, one using sampled hosts
(Sampled.ConH) and second using total assemblages (all occurring avian
species; Total.ConH). We included Total.ConH because many haemospo-
ridians infect a diversity of avian species (Ewen et al., 2012; Olsson-
Pons et al., 2015), suggesting unsampled but present host species
impact parasite assembly. This will be especially relevant for generalist
parasites, whereas sampled hosts should be representative for spe-
cialized parasites that are unlikely to occur in unsampled host species.
2.2 | Parasite and host phylogenetic reconstructions
Parasite cyt- b sequences (205 Haemoproteus and 80 Plasmodium par-
asites) were used to reconstruct phylogenetic relationships in beast
v1.8.1 (Drummond & Rambaut, 2007; See Fig. S1). We identified the
best evolutionary model (HKR+G) using maximum likelihood in mega
v7.0 (Tamura, Dudley, Nei, & Kumar, 2007). We specified a Yule spe-
ciation prior and ran two chains of 17,500,000 iterations, sampling
every 100,000 and removing 2,500,000 samples as burn- in. Chains
were examined visually for stationarity and convergence.
Avian phylogenies were gathered from (http://; accessed September 2016), which contains a Bayesian
posterior distribution of phylogenies for 9,993 avian species (Jetz,
Thomas, Joy, Hartmann, & Mooers, 2012). We gathered 100 trees
from the “Ericsson All Species Trees” data set for the 297 sampled
host species, and another 100 trees for the 3,024 avian species oc-
curring in the sample area. For all trees, branch lengths represented
substitutions per site and were scaled (dividing branch lengths by the
maximum) prior to analyses.
2.3 | Species- level analyses
2.3.1 | Host and parasite phylogenetic
For sampled host species, local phylogenetic distinctiveness (Sampled.
DisH) was calculated as mean pairwise phylogenetic distance between
a focal species and all other sampled host species in a region. This dis-
tance was divided by the mean of all pairwise distances in the region,
resulting in region- specific distinctiveness (higher values indicating
FIGURE2 Distribution of parasites across the study area. Lines connect phylogenetic parasite lineages to the region where they were most
frequently observed. Circle sizes are inversely proportional to mean phylogenetic turnover (βP) between the region and remaining regions,
accounting for geographic distance. Hence, larger circles show communities with lower mean turnover to surrounding regions, which can be
thought of as having more “connected” parasite communities. Lines and circles are coloured according to region, with closely situated regions
grouped to improve clarity
more distinct species). We calculated total host distinctiveness (Total.
DisH) using mean phylogenetic distance between a sampled host
and all occurring avian species (sampled and unsampled) in a region.
Parasite distinctiveness (DisP) was calculated separately for each para-
site genus.
2.3.2 | Parasite host-specificity
Two indices described parasite host-specificity. First, we built bipar-
tite networks (using numbers of infected individuals for each host spe-
cies) and calculated the d’ specialization index using Kullback–Leibler
distances (Blüthgen, Menzel, & Blüthgen, 2006). Ranging from zero
(no specialization; i.e., using all available hosts) to one (perfect special-
ist), d’ quantifies how strongly a parasite is “specialized” compared to
other parasites in terms of host range and interaction frequencies. We
calculated phylospecificity for each parasite (STD*; Poulin & Mouillot,
2005), which accounts for the number of infected host species and
their phylogenetic distances. Because STD* ranges from one (special-
ist) to greater than one, we used inverse STD* so both metrics could
be interpreted in the same scale and direction. Parasite STD* and d’
were uncorrelated (Pearson correlation; t = −1.41,p = .16), suggesting
they capture different aspects of parasite host-specificity (d’ capturing
the level of host sharing by parasites and STD* capturing phylogenetic
relationships of infected hosts).
2.3.3 | Influences of host community connectivity,
host phylogeny and host specificity on parasite
We tested whether interspecies barriers influenced parasite distinc-
tiveness (DisP) with a hierarchical linear model, using 548 unique
parasite*host*region combinations as data points (Data set S2).
Because DisP indices were non- negative and positively skewed, we
log- transformed values and specified a Gaussian error distribu-
tion. Continuous predictors were the two host distinctiveness met-
rics (Sampled.DisH, Total.DisH), the two host connectivity metrics
(Sampled.ConH, Total.ConH), host geographic range and both parasite
host-specificity metrics (d’, STD*). Because parasite genera showed
different phylogenetic patterns (see Results) and Total.DisH explained
a significant proportion of variance in DisP in preliminary analyses, we
tested a Total.DisH*parasite genus interaction. To decompose varia-
tion among covariates and account for underlying phylogeographic
structure, host phylogeny and sample region were included as random
grouping terms, allowing inferences for group- specific slopes whilst
estimating between- group variation (Gelman & Hill, 2007).
The model was fitted in a Bayesian framework using R package mc-
mcglmm (Hadfield, 2010). We used a flat prior for residual variance and
parameter expansion (redundant multiplicative reparameterization
of the linear model) for grouping terms, which reduces dependence
among parameters and improves mixing (Gelman, 2006). To account
for phylogenetic uncertainty, we ran separate models across 50 host
trees (Guillerme & Healy, 2014). Models were run using two chains
of 100,000 iterations with burn- in of 10,000 and thinning interval
of 300. Chains were inspected for mixing/convergence both visu-
ally and with the Gelman–Rubin diagnostic (Gelman & Rubin, 1992).
Autocorrelations were calculated to ensure independence of coeffi-
cient estimates (all autocorrelations <0.1).
2.4 | Community analyses
2.4.1 | Interspecies and geographic barriers to
parasite phylogenetic turnover
To describe shifts in diversity among regions, parasite phylogenetic
turnover (βP) was calculated (using binary occurrence data; Tsirogiannis
& Sandel, 2015) between regions where three or more parasites oc-
curred. Host turnover was calculated using either sampled hosts
(Sampled.βH) or total avian assemblages (Total.βH). Distances between
paired regions were calculated as beeline distance (km) between cen-
tral points (mean latitude and longitude of regions). Regional climate
dissimilarity was captured by three Gower’s distance matrices (Gower,
1971) to describe temperature and precipitation variation (both of
which are thought to influence haemosporidian distributions; Sehgal
et al., 2010; Sehgal, 2015). We used minimum temperature of the
coldest month and mean temperature of the coldest quarter in a min.
temp matrix, while a max.temp matrix included maximum temperature
of the warmest month and mean temperature of the warmest quarter.
Mean yearly precipitation and precipitations of the wettest and driest
quarters were included in a precip matrix. For climate matrices, vari-
ables were sourced from (; accessed November
2016) and were continuous, unweighted and scaled by range (dividing
by the maximum).
We tested if βP was correlated with Sampled.βH, Total.βH, geographic
distance or climate dissimilarity matrices using multiple regressions on
distance matrices (MRM; Goslee & Urban, 2007). Phylogenetic uncer-
tainty was captured by repeating regressions over 1,000 iterations,
where β values were re- calculated in each iteration using randomly
sampled (with replacement) trees. To account for sampling variation
that could bias turnover estimates (rare species may be more likely to
be observed with larger sample sizes), we randomly removed subsets
of species from well- sampled regions (>8 observed parasite species)
prior to regression. We arbitrarily allowed the proportion of removed
species to vary across a uniform distribution from zero to 30% in each
iteration. Regression coefficients and R2 values were gathered from
the 1,000 iterations.
2.4.2 | Barriers to parasite phylogenetic
community skewness
Host and parasite phylogenetic community skewness were calculated
using pairwise phylogenetic distance distributions. A measure of sym-
metry, this index will be less than zero (right- skewed) if communities
are made up of relatively more closely than distantly related species
(Schweiger, Klotz, Durka, & Kühn, 2008), suggesting future coloniz-
ing parasites have a greater likelihood of being locally distinct. Thus,
regions with right- skewed communities may be more vulnerable to
   CLARK et AL.
invasions by distantly related species if parasites are able to overcome
environmental barriers and colonize. Skewness was calculated for re-
gions where three or more parasites occurred.
We tested if parasite skewness was predicted by host connectiv-
ity (Sampled.ConH, Total.ConH) using linear regression with Gaussian
error distribution. Mean annual precipitation and mean temperatures
of the warmest and coldest quarters were included as continuous co-
variates to account for possible climate influences, while sampled and
total host skewness were included to account for influences of host
phylogenetic symmetry. Parasite genus was included as a categorical
covariate. The model was fitted using mcmcglmm with a flat prior for re-
sidual variance. We ran two chains of 100,000 iterations with burn- in
of 10,000 and thinning interval of 300, following procedures above to
examine convergence and estimate autocorrelations.
For all phylogenetic metrics (skewness, distinctiveness and STD*),
we accounted for phylogenetic uncertainty by calculating median
indices across 1,000 randomly sampled host and parasite trees.
Significance of model effects was determined by examining if 95%
quantiles (for MRM models) or 95% credible intervals (CI; for Bayesian
models) of regression coefficients did not overlap zero. Continuous
predictors were scaled (centred and divided by one standard devia-
tion), and variances explained were calculated following Nakagawa
and Schielzeth (2013). Data were analysed in R v3.2.1 (R Core Team,
2016; R: A language and environment for statistical computing). Data
and R code are presented in Supplementary Data and the Dryad Digital
Repository: (
3.1 | Host phylogeny, local distinctiveness and
connectivity drive parasite distinctiveness
Parasite distinctiveness (DisP) was strongly related to host phylogeny
(variance explained = 46.8%–78.3%), with hosts from certain clades
more likely to carry distinct parasites (Figure 3). These included car-
riers of distinct Haemoproteus spp. such as doves (Columbidae),
kingfishers (Alcedinidae) and corvoids such as crows (Corvidae) and
whistlers (Pachycephalidae; Figure 3), all of which occupy a range
of regions yet rarely co- occur with sympatric sister species (Dutson,
2012; Jønsson et al., 2014). After accounting for the strong influ-
ence of host phylogeny, DisP was also positively predicted by local
host total distinctiveness (Total.DisH; coefficient 95% CI = 0.04–0.12;
variance explained = 2.48%–6.38%; Figure 3), suggesting host relat-
edness to the local avian assemblage acts as an interspecies barrier to
parasite assembly. This relationship varied between parasite genera,
as increases in Total.DisH lead to a 1.95 times higher increase in DisP
for Haemoproteus than for Plasmodium parasites.
DisP decreased with increasing total host connectivity (Total.ConH;
coefficient = 0.01–0.09; variance explained = 0.04%–7.7%; See Fig.
S2), indicating greater host diversity and collective mobility increase
a parasite’s chance of encountering related parasites. Total.ConH was
highest in Malaysia (509 avian species; Total.ConH = 83.60) and south-
east Australia (468 avian species; Total.ConH = 80.42), moderate in
Papua New Guinea where many endemic avian species occur (mean
species = 520.5; mean Total.ConH = 42.62) and lowest in Vanuatu and
New Caledonia (mean species = 115 and 110; mean Total.ConH = 32.3
and 31.6, respectively). DisP was not influenced by Sampled.ConH,
Sampled.DisH or individual host range (coefficient CIs overlapped zero).
We observed considerable variation in host specificity for both
parasite genera, though neither specificity metric influenced DisP (co-
efficients overlapped with zero). For both genera, STD* (phylospeci-
ficity) ranged from 0.41 to 1 (mean = 0.79 and 0.87 for Plasmodium
and Haemoproteus, respectively), while d’ (network specificity) ranged
from 0 to 1 (means = 0.65 and 0.67). In total, fixed effects (d’, STD*,
host range size, Total.ConH, Sampled.ConH, Total.DisH, Sampled.DisH,) ex-
plained 5.7%–13.2% of variance in DisP while the full model (including
host phylogeny and region grouping terms) explained 69.8%–88.9%.
3.2 | Host phylogeny and climate shape parasite
community structure
We found evidence that both environmental and interspecies bar-
riers influence parasite turnover. For Plasmodium, βP was positively
correlated with Sample.βH (MRM coefficient = 1.01–1.86), indicating
host phylogeny influences shifts in parasite diversity. Plasmodium
βP also correlated positively with geographic distance (0.56–1.21),
but negatively with max.temp (−0.09 to −0.18). For Haemoproteus,
βP correlated positively with both host turnover metrics (Sampled.βH
coefficient = 0.30–0.61; Total.βH = 0.58–1.13), and with geographic
distance and max.temp (0.04–1.37; 0.16–0.45, respectively), but
FIGURE3 Distribution of local phylogenetic distinctiveness for
hosts (Total.DisH) and their parasites (DisP) across the host phylogeny.
Distinctiveness represents mean phylogenetic distance between the
focal species and all co- occurring species within a region. Values are
scaled so values >0 indicate taxa that are more distinct, while those
<0 indicate less distinct taxa
negatively with min.temp(−0.11to−0.28).Varianceexplainedbypre-
dictors ranged from 47% to 57% for Haemoproteus βP and from 4% to
11% for Plasmodium βP.
Mainland communities such as Papua New Guinea and eastern
Australia showed low mean parasite turnover among paired regions
(low average pairwise βP after accounting for geographic distance;
Figure 2; Data set S3), suggesting these assemblages were less phy-
logenetically unique within the study area. Parasite assemblages on
Melanesian islands (New Caledonia and Vanuatu) showed moderate
mean turnover, while relatively isolated and less well- sampled com-
munities such as Christmas Island and north- west Australia showed
high turnover (Figure 2). Plasmodium communities in New Zealand
and Micronesia, where many occurring parasites are known to be in-
troduced (Beadell et al., 2006; Ewen et al., 2012), showed high mean
turnover (Figure 2).
Parasite community skewness indices were predominantly nega-
tive (right- skewed; Figure 4), with assemblages generally made up of
more closely than distantly related parasites. Parasite skewness was
not influenced by host community connectivity or host skewness
but was driven by mean temperature of the coldest quarter (coeffi-
cient = 0.02–2.98; variance explained = 0.2%–10.6%), with colder
regions harbouring more negatively skewed communities (Figure 4).
Parasiteskewness also differedbetween genera(coefficient=−0.91
to −0.02; variance explained=7.5%–27.10%), with Plasmodium
more negatively skewed than Haemoproteus communities (Figure 4).
Interestingly, Haemoproteus communities in Papua New Guinea were
positively skewed, while those in eastern Australian were negatively
skewed (Figure 4), suggesting neighbouring parasite assemblages with
low phylogenetic turnover (Figure 2) can vary substantially in commu-
nity structure.
We illustrate a framework for identifying relative influences of inter-
species, environmental and geographic barriers to parasite community
assembly. Using this framework, we show that host phylogeny is a key
driver of local parasite assembly, while climate and the regional con-
nectivity of host assemblages play lesser but nonetheless important
roles. Moreover, host phylogeny and geographic distance were more
important than environmental barriers in shaping parasite turnover,
indicating alterations to host movement and community composition
may strongly affect parasite dispersal and invasion potential across
biogeographic scales.
4.1 | Barriers to parasite community
assembly and their roles in parasite spread
Host phylogeny was an important driver of parasite distinctiveness
and species turnover, supporting suggestions that host identity drives
shifts in haemosporidian diversity and implicating host evolution-
ary history as a determinant of regional parasite assembly (Fecchio
et al., 2017; Scordato & Kardish, 2014). Phylogenetic signals are a
proxy for physical (i.e., physiological, morphological, biochemical) and
ecological traits, where closely related species resemble each other
more than random pairs, indicating conserved attributes likely play a
role in modulating interspecies barriers to regional parasite assembly
(Huang, Bininda- Emonds, Stephens, Gittleman, & Altizer, 2014). Yet,
an important consideration here is that we do not know which shared
host traits influence blood parasite assembly patterns. Determining
underlying interspecies barriers to parasite composition will require
additional interdisciplinary work, combining data on host traits with
FIGURE4 Parasite phylogenetic
community skewness across regions.
Skewness >0 indicates co- occurring
parasites are relatively distantly related
(left- skewed pairwise distance distribution),
while <0 indicates parasites are relatively
closely related (right- skewed distance
distribution). Regions are ordered based on
mean temperature of the coldest quarter,
with numbers in parentheses indicating
the number of parasites recovered in each
region. NZ, New Zealand; AUS, Australia;
NC, New Caledonia; VAN, Vanuatu; PNG,
Papua New Guinea
NZ South Island (3
NZ North Island (7)
AUS south-east (14)
AUS central east (38)
NC Grande Terre (21)
AUS north-east (60)
NC Mare (5)
NC Lifou (8)
VAN southern islands (17)
NC Ouvea (7)
VAN west
ern islands (11)
AUS north-west (8)
VAN cent
ral islands (12)
VAN Espiritu Santo (8
PNG Highl
ands (18)
VAN northern islands (4)
PNG Lowlan
ds (42)
Philippines (58)
Malaysia (16)
Warmer min tempColder min temp
More closely relatedLess closely related
   CLARK et AL.
methods that can decompose phylogenetic and ecological similarity
to improve inference (Cadotte, Albert, & Walker, 2013; Clark & Clegg,
Future host range shifts may considerably impact parasite spread
and disease emergence, both by breaking down existing barriers to
host shifting and by increasing parasite dispersal (Atkinson & LaPointe,
2009; Young, Parker, Gilbert, Guerra, & Nunn, 2017). Here, a positive
relationship between host and parasite distinctiveness indicates that
diminishing phylogeographic barriers (where host range shifts may
alter local host distinctiveness) could present more opportunities for
parasites to shift between related hosts. Yet, a strong host phyloge-
netic signal, where distinct parasites are more strongly associated with
certain host clades, suggests alterations to host species’ distributions
may have different effects on parasite spread depending on host
evolutionary history. For instance, we identified multiple host clades
as prominent carriers of distinct parasites, including non- passerines
(kingfishers and doves) as well as certain passerine groups (crows and
whistlers), indicating that future range shifts for these host groups
could lead to novel parasite introductions. Our work therefore cor-
roborates a large body of literature to show that interactions between
ecological fitting and shifting geographic distributions will have pow-
erful influences on parasite assembly and emergence potential (Agosta
et al., 2010; Araujo et al., 2015; Brooks & Hoberg, 2007; Hoberg,
2010; Hoberg & Brooks, 2008). However, a significant influence of
host community connectivity suggests that parasite distinctiveness is
not only driven by host phylogeny, but also by forces that limit host
diversity and distributional overlap (i.e., competitive exclusion or dis-
persal barriers; Ricklefs, 2010; Ewen et al., 2012). This finding gener-
ates exciting new avenues for studying parasite assembly, particularly
as few studies relate the connectivity of host communities to parasite
dispersal opportunity (but see Buckee et al., 2007).
Our findings that environmental effects influence parasite turn-
over and community skewness agree with previous studies to suggest
that even if dispersal barriers break down, climate and perhaps other
environmental conditions may constrain parasite distributions (Clark,
Clegg, & Klaassen, 2016; Clark, Wells, Dimitrov, & Clegg, 2016; Kutz,
Hoberg, Molnár, Dobson, & Verocai, 2014; Sehgal, 2015). Indeed, re-
gional temperature similarity impacted shifts in diversity for both para-
site genera, albeit with different directional relationships. One possible
explanation could be that haemosporidians are subject to influences
of external temperature changes on ectothermic vectors (Paaijmans,
Imbahale, Thomas, & Takken, 2010), and Plasmodium and Haemoproteus
parasites are transmitted by different arthropods (mosquitoes from
family Culicidae and midges from family Ceratopogonidae, respec-
tively; Santiago- Alarcon, Palinauskas, & Schaefer, 2012). However,
little is known about the particular vector species transmitting avian
haemosporidians in the South Pacific (but see Ishtiaq et al., 2008),
and so, drawing conclusions from these different patterns remains
challenging. Intriguingly, regions with colder temperatures harboured
more closely related communities for both parasite genera, perhaps
indicating minimum temperatures act as a strong filter for haemospo-
ridian diversity, a finding that warrants future study. Regardless of the
biological mechanism, accounting for interspecies interactions and
environmental conditions can improve predictions of species distri-
butions following climate shifts (Choler, Michalet, & Callaway, 2001;
Mayfield & Stouffer, 2017; Wells, Feldhaar, & O’Hara, 2014).
Determining which species are likely to be introduced and be-
come invasive are prominent ecological questions (Springborn et al.,
2015; Wiens, 2011). Our results suggest that parasites introduced to
regions with low host community connectivity, high host turnover and
low minimum temperatures may be more likely to invade the com-
munity. These patterns highlight that New Zealand, which showed
high rates of host and parasite turnover and contained distantly re-
lated (phylogenetically left- skewed) Plasmodium communities, may
be particularly vulnerable to invasions. Distinct invaders can have key
competitive advantages and a greater chance of becoming invasive
(HilleRisLambers et al., 2012), as has been the case in the Galapágos
where the invasive fly, Philornis downsi, parasitizes a diversity of en-
demic bird species (Fessl, Sinclair, & Kleindorfer, 2006). Indeed, inva-
sive avian malaria parasites have already been recorded infecting a
diversity of native New Zealand birds, with evidence suggesting that
introduced birds play key roles in driving parasite spread (Ewen et al.,
2012; Schoener, Banda, Howe, Castro, & Alley, 2013). Parasites intro-
duced to highly connected host regions, on the other hand, may be
more likely to experience competition with closely related parasites,
perhaps curbing invasion potential. Under this consideration, areas
such as eastern Australia and mainland Papua New Guinea may be
less vulnerable to parasite invasions (though not immune; see Clark,
Olsson- Pons, Ishtiaq, & Clegg, 2015), as these regions contain a rela-
tively balanced phylogenetic diversity of parasites and experience high
host community connectivity.
4.2 | Accounting for unsampled host species in
parasite assembly studies
Our study raises a critical point for assessing parasite composition, as
measures of host relationships were more important in driving para-
site assembly when considering the total host assemblage rather than
only sampled hosts. A host’s distinctiveness with respect to the entire
avian community positively predicted parasite distinctiveness, while
considering only sampled hosts had no influence on parasite distinc-
tiveness. Phylogenetic turnover of the total avian assemblage was also
a stronger predictor of Haemoproteus turnover than was sampled host
turnover. These findings imply that variation in unsampled but locally
present host species is important for driving parasite establishment.
Inferences beyond those obtained from sampled hosts are clearly
needed, a process which is rarely considered in host–parasite interac-
tions (but see Wells et al., 2012), despite being a well- known problem
in the sample survey literature (Little, 2004).
4.3 | Caveats and conclusions
There are several ways in which our study framework can be im-
proved. First, we did not consider individual sites in our study as our
data were limited by small sample sizes for many sites. Inclusion of
site- specific species and climate data could be used as an additional
source of information to examine possible impacts of sampling bias on
regional community inferences. Second, consideration of sampling dis-
tribution across regions may have an impact on community turnover
estimates, as regions such as Christmas Island and Micronesia had a
relatively high turnover that could have been influenced by low overall
sample sizes and large geographic distances to many other study re-
gions. Future studies that sample smaller and more regular geographic
intervals could help to address this drawback. Finally, our phylogenetic
metrics relied only on binary species occurrences (present or absent)
and may be improved with better consideration of species’ relative
abundances, as host abundance plays a role in host reservoir poten-
tial and cross- species parasite transmission (Kilpatrick, Kramer, Jones,
Marra, & Daszak, 2006). Unfortunately, such data for host abundance
were not available and would require additional field survey efforts.
In summary, our study agrees with previous work to suggest that
in addition to identifying environmental barriers, considering host
phylogenetic relationships and dispersal abilities is key to understand-
ing regional parasite assembly (Agosta et al., 2010; Brooks & Ferrao,
2005; Sehgal, 2015; Wells, O’Hara, Morand, Lessard, & Ribas, 2015).
Moreover, we show that accounting for the overall connectivity of the
host community, rather than solely focussing on individual host spe-
cies’ dispersal potentials, may be crucial to predicting future parasite
invasions. With the pervasive need to understand how interspecies
interactions shape species distributions (Wisz et al., 2013), our study
represents an important step towards predicting how parasite assem-
blages will be shaped following future global change.
We thank S. Olsson- Pons, D. Treby, K. Lowe, J. LeBreton, S. Oghino, F.
Cugny, W. Waheoneme, T. Read, Association pour la Sauvegarde de la
Biodiversité d’Ouvéa and Société Calédonienne d’Ornithologie for field
and/or logistical assistance in Australia and New Caledonia. Australian
and Melanesian fieldwork was conducted under Griffith University
ethics clearance ENV/01/12/AEC (SC). Papua New Guinea fieldwork
was conducted under UQ animal ethics permit GPEM/172/13/APA
(WG). SC acknowledges financial support from a Griffith University
New Researcher Grant, a National Geographic Society Committee for
Research and Exploration Grant (#9383- 13) and Birds Queensland.
KS acknowledges financial support from Czech Science Foundation
14- 36098G and staff of The New Guinea Binatang Research Centre.
BK was funded by Darwin Initiative for the Survival of Species No.
22- 00. Work of KS and BK was conducted under University of South
Bohemia animal ethic clearance 515- 20424/2012- 30 and PNG NRI
Research Permit No. 99902077829. We thank two anonymous refer-
ees for comments that improved the manuscript.
Newly reported parasite sequences will be uploaded to GenBank
and the MalAvi avian malaria database upon acceptance. R code
and raw data sets are uploaded to figshare
Nicholas J. Clark
Adlard, R. D., Miller, T. L., & Smit, N. J. (2015). The butterfly effect: Parasite
diversity, environment, and emerging disease in aquatic wildlife. Trends
in Parasitology, 31, 160–166.
Agosta, S. J., Janz, N., & Brooks, D. R. (2010). How specialists can be gen-
eralists: Resolving the “parasite paradox” and implications for emerging
infectious disease. Zoologia, 27, 151–162.
Araujo, S. B., Braga, M. P., Brooks, D. R., Agosta, S. J., Hoberg, E. P., von
Hartenthal, F. W., & Boeger, W. A. (2015). Understanding host-
switching by ecological fitting. PLoS ONE, 10, e0139225.
Atkinson, C. T., & LaPointe, D. A. (2009). Introduced avian diseases, climate
change, and the future of Hawaiian Honeycreepers. Journal of Avian
Medicine and Surgery, 23, 53–63.
Barnagaud, J. Y., Daniel Kissling, W., Sandel, B., Eiserhardt, W. L., Sekercioglu,
C. H., Enquist, B. J., … Svenning, J. C. (2014). Ecological traits influence
the phylogenetic structure of bird species co- occurrences worldwide.
Ecology Letters, 17, 811–820.
Beadell, J. S., Ishtiaq, F., Covas, R., Melo, M., Warren, B. H., Atkinson, C.
T., … Fleischer, R. C. (2006). Global phylogeographic limits of Hawaii’s
avian malaria. Proceedings of the Royal Society B: Biological Sciences, 273,
Bensch, S., Hellgren, O., & Pérez-Tris, J. (2009). MalAvi: A public database
of malaria parasites and related haemosporidians in avian hosts based
on mitochondrial cytochrome b lineages. Molecular Ecology Resources,
9, 1353–1358.
Bensch, S., Pérez-Tris, J., Waldenström, J., & Hellgren, O. (2004). Linkage
between nuclear and mitochondrial DNA sequences in avian ma-
laria parasites: Multiple cases of cryptic speciation? Evolution, 58,
Blüthgen, N., Menzel, F., & Blüthgen, N. (2006). Measuring specialization in
species interaction networks. BMC Ecology, 6, 9.
Brooks, D. R. (1979). Testing the context and extent of host- parasite coevo-
lution. Systematic Biology, 28, 299–307.
Brooks, D. R., & Ferrao, A. L. (2005). The historical biogeography of co-
evolution: Emerging infectious diseases are evolutionary accidents
waiting to happen. Journal of Biogeography, 32, 1291–1299.
Brooks, D. R., & Hoberg, E. P. (2007). How will global climate change affect
parasite–host assemblages? Trends in Parasitology, 23, 571–574.
Buckee, C., Danon, L., & Gupta, S. (2007). Host community structure and
the maintenance of pathogen diversity. Proceedings of the Royal Society
B: Biological Sciences, 274, 1715–1721.
Cadotte, M., Albert, C. H., & Walker, S. C. (2013). The ecology of differ-
ences: Assessing community assembly with trait and evolutionary dis-
tances. Ecology Letters, 16, 1234–1244.
Choler, P., Michalet, R., & Callaway, R. M. (2001). Facilitation and competi-
tion on gradients in alpine plant communities. Ecology, 82, 3295–3308.
Clark, N. J., & Clegg, S. M. (2015). The influence of vagrant hosts and
weather patterns on the colonization and persistence of blood para-
sites in an island bird. Journal of Biogeography, 42, 641–651.
Clark, N. J., & Clegg, S. M. (2017). Integrating phylogenetic and ecological
distances reveals new insights into parasite host-specificity. Molecular
Ecology, 26, 3074–3086.
Clark, N. J., Clegg, S. M., & Klaassen, M. (2016). Migration strategy and
pathogen risk: Non- breeding distribution drives malaria prevalence in
migratory waders. Oikos, 125, 1358–1368.
Clark, N. J., Clegg, S. M., & Lima, M. R. (2014). A review of global di-
versity in avian haemosporidians (Plasmodium and Haemoproteus:
Haemosporida): New insights from molecular data. International Journal
for Parasitology, 44, 329–338.
   CLARK et AL.
Clark, N. J., Olsson-Pons, S., Ishtiaq, F., & Clegg, S. M. (2015). Specialist
enemies, generalist weapons and the potential spread of exotic patho-
gens: Malaria parasites in a highly invasive bird. International Journal for
Parasitology, 45, 891–899.
Clark, N. J., Wells, K., Dimitrov, D., & Clegg, S. M. (2016). Co- infections and
environmental conditions drive the distributions of blood parasites in
wild birds. Journal of Animal Ecology, 85, 1461–1470.
Drummond, A. J., & Rambaut, A. (2007). BEAST: Bayesian evolutionary
analysis by sampling trees. BMC Evolutionary Biology, 7(1), 214. https://
Dutson, G. (2012). Birds of Melanesia: Bismarcks, Solomons, Vanuatu, and
New Caledonia. New Jersey: A&C Black.
Ehrlich, P. R., & Raven, P. H. (1964). Butterflies and plants: A study in coevo-
lution. Evolution, 18, 586–608.
Ejiri, H., Sato, Y., Kim, K.-S., Hara, T., Tsuda, Y., Imura, T., … Yukawa, M.
(2011). Entomological study on transmission of avian malaria parasites
in a zoological garden in Japan: Bloodmeal identification and detection
of avian malaria parasite DNA from blood- fed mosquitoes. Journal of
Medical Entomology, 48, 600–607.
Ellis, V. A., Collins, M. D., Medeiros, M. C., Sari, E. H., Coffey, E. D., Dickerson,
R. C., … Merrill, L. (2015). Local host specialization, host- switching, and
dispersal shape the regional distributions of avian haemosporidian
parasites. Proceedings of the National Academy of Sciences of the United
States of America, 112, 11294–11299.
Epstein, P. R. (2001). Climate change and emerging infectious diseases.
Microbes and Infection, 3, 747–754.
Ewen, J. G., Bensch, S., Blackburn, T. M., Bonneaud, C., Brown, R., Cassey,
P., … Pérez-Tris, J. (2012). Establishment of exotic parasites: The origins
and characteristics of an avian malaria community in an isolated island
avifauna. Ecology Letters, 15, 1112–1119.
Fecchio, A., Pinheiro, R., Felix, G., Faria, I., Pinho, J., Lacorte, G., … Tkach,
V. (2017). Host community similarity and geography shape the diver-
sity and distribution of haemosporidian parasites in Amazonian birds.
Fessl, B., Sinclair, B., & Kleindorfer, S. (2006). The life- cycle of Philornis
downsi (Diptera: Muscidae) parasitizing Darwin’s finches and its im-
pacts on nestling survival. Parasitology, 133, 739–747.
Gelman, A. (2006). Prior distributions for variance parameters in hierar-
chical models (comment on article by Browne and Draper). Bayesian
Analysis, 1, 515–534.
Gelman, A., & Hill, J. (2007). Data analysis using regression and multilevel/
hierarchical models. Cambridge: Cambridge University Press.
Gelman, A., & Rubin, D. B. (1992). Inference from iterative simulation using
multiple sequences. Statistical Science, 7(4), 457–472.
Githeko, A. K., Lindsay, S. W., Confalonieri, U. E., & Patz, J. A. (2000).
Climate change and vector- borne diseases: A regional analysis. Bulletin
of the World Health Organization, 78, 1136–1147.
Goslee, S. C., & Urban, D. L. (2007). The ecodist package for dissimilarity-
based analysis of ecological data. Journal of Statistical Software, 22, 1–19.
Gower, J. C. (1971). A general coefficient of similarity and some of its prop-
erties. Biometrics, 27(4), 857–871.
Guillerme, T., & Healy, K. (2014). mulTree: A package for running MCMCglmm
analysis on multiple trees. Zenodo. Retrieved from https://doi.
Hadfield, J. D. (2010). MCMC methods for multi- response generalized lin-
ear mixed models: The MCMCglmm R package. Journal of Statistical
Software, 33, 1–22.
Hector, A., Dobson, K., Minns, A., Bazeley-White, E., & Hartley Lawton, J.
(2001). Community diversity and invasion resistance: An experimen-
tal test in a grassland ecosystem and a review of comparable studies.
Ecological Research, 16, 819–831.
Hellgren, O., Atkinson, C. T., Bensch, S., Albayrak, T., Dimitrov, D., Ewen, J.
G., … Palinauskas, V. (2014). Global phylogeography of the avian ma-
laria pathogen Plasmodium relictum based on MSP1 allelic diversity.
Ecography, 38(8), 842–50.
Hellgren, O., Waldenström, J., & Bensch, S. (2004). A new PCR assay for
simultaneous studies of Leucocytozoon, Plasmodium, and Haemoproteus
from avian blood. Journal of Parasitology, 90, 797–802.
HilleRisLambers, J., Adler, P., Harpole, W., Levine, J., & Mayfield, M. (2012).
Rethinking community assembly through the lens of coexistence the-
ory. Annual Review of Ecology, Evolution, and Systematics, 43, 227–248.
Hoberg, E. (2010). Invasive processes, mosaics and the structure of hel-
minth parasite faunas. Revue Scientifique et Technique, 29, 255.
Hoberg, E. P., & Brooks, D. R. (2008). A macroevolutionary mosaic: Episodic
host- switching, geographical colonization and diversification in com-
plex host–parasite systems. Journal of Biogeography, 35, 1533–1550.
Huang, S., Bininda-Emonds, O. R., Stephens, P. R., Gittleman, J. L., & Altizer,
S. (2014). Phylogenetically related and ecologically similar carnivores har-
bour similar parasite assemblages. Journal of Animal Ecology, 83, 671–680.
Ishtiaq, F., Guillaumot, L., Clegg, S. M., Phillimore, A. B., Black, R. A., Owens,
I. P. F., … Sheldon, B. C. (2008). Avian haematozoan parasites and
their associations with mosquitoes across Southwest Pacific Islands.
Molecular Ecology, 17, 4545–4555.
Janzen, D. H. (1968). Host plants as islands in evolutionary and contempo-
rary time. The American Naturalist, 102, 592–595.
Janzen, D. H. (1985). On ecological fitting. Oikos, 45, 308–310.
Jetz, W., Thomas, G., Joy, J., Hartmann, K., & Mooers, A. (2012). The global
diversity of birds in space and time. Nature, 491, 444–448.
Jønsson, K. A., Irestedt, M., Christidis, L., Clegg, S. M., Holt, B. G., & Fjeldsa,
J. (2014). Evidence of taxon cycles in an Indo- Pacific passerine bird
radiation (Aves: Pachycephala). Proceedings of the Royal Society B:
Biological Sciences, 281, 20131727.
Kelly, D. W., Paterson, R. A., Townsend, C. R., Poulin, R., & Tompkins, D.
M. (2009). Parasite spillback: A neglected concept in invasion ecology?
Ecology, 90, 2047–2056.
Kilpatrick, A. M., Kramer, L. D., Jones, M. J., Marra, P. P., & Daszak, P. (2006).
West Nile virus epidemics in North America are driven by shifts in mos-
quito feeding behavior. PLoS Biology, 4, e82.
Kraft, N. J., Cornwell, W. K., Webb, C. O., & Ackerly, D. D. (2007). Trait
evolution, community assembly, and the phylogenetic structure of eco-
logical communities. The American Naturalist, 170, 271–283.
Krasnov, B. R., Shenbrot, G. I., Khokhlova, I. S., & Degen, A. A. (2016). Trait- based
and phylogenetic associations between parasites and their hosts: A case
study with small mammals and fleas in the Palearctic. Oikos, 125, 29–38.
Križanauskienė, A., Hellgren, O., Kosarev, V., Sokolov, L., Bensch, S., &
Valkiūnas,G. (2006).Variation inhost specificity betweenspeciesof
avian hemosporidian parasites: Evidence from parasite morphology and
cytochrome B gene sequences. Journal of Parasitology, 92, 1319–1324.
Kutz, S. J., Hoberg, E. P., Molnár, P. K., Dobson, A., & Verocai, G. G. (2014). A
walk on the tundra: Host–parasite interactions in an extreme environment.
International Journal for Parasitology: Parasites and Wildlife, 3, 198–208.
Lafferty, K. D. (2009). The ecology of climate change and infectious dis-
eases. Ecology, 90, 888–900.
Little, R. J. (2004). To model or not to model? Competing modes of infer-
ence for finite population sampling. Journal of the American Statistical
Association, 99, 546–556.
Mata, V. A., da Silva, L. P., Lopes, R. J., & Drovetski, S. V. (2015). The Strait
of Gibraltar poses an effective barrier to host- specialized but not
to host- generalized lineages of avian Haemosporidia. International
Journal for Parasitology, 45(11), 711–9.
Mayfield, M., & Stouffer, D. (2017). Higher- order interactions capture
unexplained complexity in diverse communities. Nature Ecology and
Evolution, 1, 0062.
Nakagawa, S., & Schielzeth, H. (2013). A general and simple method for
obtaining R2 from generalized linear mixed- effects models. Methods in
Ecology and Evolution, 4, 133–142.
Olsson-Pons, S., Clark, N. J., Ishtiaq, F., & Clegg, S. M. (2015). Differences
in host species relationships and biogeographic influences produce
contrasting patterns of prevalence, community composition and
genetic structure in two genera of avian malaria parasites in southern
Melanesia. Journal of Animal Ecology, 84, 985–998.
Paaijmans, K. P., Imbahale, S. S., Thomas, M. B., & Takken, W. (2010). Relevant
microclimate for determining the development rate of malaria mosqui-
toes and possible implications of climate change. Malaria Journal, 9, 196.
Patz, J. A., Campbell-Lendrum, D., Holloway, T., & Foley, J. A. (2005). Impact
of regional climate change on human health. Nature, 438, 310–317.
Pérez-Tris, J., & Bensch, S. (2005). Dispersal increases local transmission of
avian malarial parasites. Ecology Letters, 8, 838–845.
Poulin, R., Krasnov, B. R., & Mouillot, D. (2011). Host specificity in phyloge-
netic and geographic space. Trends in Parasitology, 27, 355–361.
Poulin, R., & Mouillot, D. (2005). Combining phylogenetic and ecological informa-
tion into a new index of host specificity. Journal of Parasitology, 91, 511–514.
Radtke, A., McLennan, D., & Brooks, D. (2002). Evolution of host specific-
ity in Telorchis spp. (Digenea: Plagiorchiformes: Telorchiidae). Journal of
Parasitology, 88, 874–879.
R Core Team (2016). R: A language and environment for statistical comput-
ing. Vienna, Austria: R Foundation for Statistical Computing. Retrieved
Ricklefs, R. E. (1987). Community diversity: Relative roles of local and re-
gional processes. Science, 235, 167–171.
Ricklefs, R. E. (2010). Evolutionary diversification, coevolution between
populations and their antagonists, and the filling of niche space.
Proceedings of the National Academy of Sciences of the United States of
America, 107, 1265–1272.
Rohde, K. (1980). Host specificity indices of parasites and their application.
Experientia, 36, 1369–1371.
Santiago-Alarcon, D., Palinauskas, V., & Schaefer, H. M. (2012). Diptera vec-
tors of avian Haemosporidian parasites: Untangling parasite life cycles
and their taxonomy. Biological Reviews, 87, 928–964.
Schoener, E. R., Banda, M., Howe, L., Castro, I. C., & Alley, M. R. (2013). Avian
malaria in New Zealand. New Zealand Veterinary Journal, 62, 189–198.
Schulze-Lefert, P., & Panstruga, R. (2011). A molecular evolutionary con-
cept connecting nonhost resistance, pathogen host range, and patho-
gen speciation. Trends in Plant Science, 16, 117–125.
Schweiger, O., Klotz, S., Durka, W., & Kühn, I. (2008). A comparative test of
phylogenetic diversity indices. Oecologia, 157, 485–495.
Scordato, E. S., & Kardish, M. R. (2014). Prevalence and beta diversity in
avian malaria communities: Host species is a better predictor than ge-
ography. Journal of Animal Ecology, 83, 1387–1397.
Sehgal, R. N. (2015). Manifold habitat effects on the prevalence and di-
versity of avian blood parasites. International Journal for Parasitology:
Parasites and Wildlife, 4, 421–430.
Sehgal, R. N. M., Buermann, W., Harrigan, R. J., Bonneaud, C., Loiseau, C.,
Chasar, A., … Smith, T. B. (2010). Spatially explicit predictions of blood
parasites in a widely distributed African rainforest bird. Proceedings of
the Royal Society B- Biological Sciences, 278(1708), 1025–33. https://doi.
Simpson, E. H. (1949). Measurement of diversity. Nature, 163, 688.
Springborn, M. R., Keller, R. P., Elwood, S., Romagosa, C. M., Zambrana-
Torrelio, C., & Daszak, P. (2015). Integrating invasion and disease in the
risk assessment of live bird trade. Diversity and Distributions, 21, 101–110.
Streicker, D. G., Turmelle, A. S., Vonhof, M. J., Kuzmin, I. V., McCracken, G. F., &
Rupprecht, C. E. (2010). Host phylogeny constrains cross- species emer-
gence and establishment of rabies virus in bats. Science, 329, 676–679.
Tamura, K., Dudley, J., Nei, M., & Kumar, S. (2007). MEGA4: Molecular
Evolutionary Genetics Analysis (MEGA) software version 4.0. Molecular
Biology and Evolution, 24, 1596–1599.
Tsirogiannis, C., & Sandel, B. (2015). PhyloMeasures: A package for com-
puting phylogenetic biodiversity measures and their statistical mo-
ments. Ecography, 39, 709–714.
van Riper III, C., van Riper, S. G., Goff, M. L., & Laird, M. (1986). The epizo-
otiology and ecological significance of malaria in Hawaiian land birds.
Ecological Monographs, 56, 327–344.
Viana, D. S., Santamaría, L., & Figuerola, J. (2016). Migratory birds as global
dispersal vectors. Trends in Ecology & Evolution, 31, 763–775.
Waldenström, J., Bensch, S., Hasselquist, D., & Östman, Ö. (2004). A new
nested polymerase chain reaction method very efficient in detecting
Plasmodium and Haemoproteus infections from avian blood. Journal of
Parasitology, 90, 191–194.
Wallace, A. R. (1876). The geographical distribution of animals with a study of
the relations of living and extinct faunas as elucidating the past changes of
the earth’s surface: By Alfred Russel Wallace… with maps and illustrations.
In two volumes. New York: Harper and Brothers, Publishers.
Warburton, E. M., Kohler, S. L., & Vonhof, M. J. (2016). Patterns of para-
site community dissimilarity: The significant role of land use and lack of
distance- decay in a bat–helminth system. Oikos, 125, 374–385.
Wells, K., Feldhaar, H., & O’Hara, R. B. (2014). Population fluctuations
affect inference in ecological networks of multi- species interactions.
Oikos, 123, 589–598.
Wells, K., O’Hara, R. B., Morand, S., Lessard, J.-P., & Ribas, A. (2015). The
importance of parasite geography and spillover effects for global pat-
terns of host–parasite associations in two invasive species. Diversity
and Distributions, 21, 477–486.
Wells, K., O’Hara, R. B., Pfeiffer, M., Lakim, M. B., Petney, T. N., & Durden,
L. A. (2012). Inferring host specificity and network formation through
agent- based models: Tick–mammal interactions in Borneo. Oecologia,
172, 307–316.
Wiens, J. J. (2011). The niche, biogeography and species interactions.
Philosophical Transactions of the Royal Society B: Biological Sciences, 366,
Wisz, M. S., Pottier, J., Kissling, W. D., Pellissier, L., Lenoir, J., Damgaard, C.
F., … Guisan, A. (2013). The role of biotic interactions in shaping distri-
butions and realised assemblages of species: Implications for species
distribution modelling. Biological Reviews, 88, 15–30.
Young, H. S., Parker, I. M., Gilbert, G. S., Guerra, A. S., & Nunn, C. L. (2017).
Introduced species, disease ecology, and biodiversity–disease relation-
ships. Trends in Ecology & Evolution, 32, 41–54.
Nicholas Clark is a disease ecologist interested in evolutionary
ecology and the biogeography of wildlife pathogens. His research
interests concern topics in computational phylogenetics, biodiver-
sity and host–pathogen interactions, as well as large- scale patterns
in community assembly.
Author contributions: N.J.C. and K.W. conceived the ideas; all au-
thors contributed to data collection; N.J.C. and K.W. analysed the
data and led the writing.
Additional Supporting Information may be found online in the sup-
porting information tab for this article.
How to cite this article: Clark NJ, Clegg SM, Sam K, Goulding
W, Koane B, Wells K. Climate, host phylogeny and the
connectivity of host communities govern regional parasite
assembly. Divers Distrib. 2017;00:1–11. https://doi.
... Besides the important role that host-related traits may have in structuring parasite metacommunities, abiotic factors can also have important effects on parasite community composition (Krasnov et al., 2010). Some studies have shown that both environmental and geographical barriers can govern the assembly and distributional patterns of parasite species across space (Clark et al., 2017;Fecchio et al., 2017). For example, mean annual temperature, precipitation, and fluctuations in diurnal temperature range can directly affect the reproduction rate of some parasites within hosts (i.e., particularly vector-borne parasites) (Paaijmans et al., 2009) and thus limit their colonization and establishment in certain areas (Clark et al., 2017). ...
... Some studies have shown that both environmental and geographical barriers can govern the assembly and distributional patterns of parasite species across space (Clark et al., 2017;Fecchio et al., 2017). For example, mean annual temperature, precipitation, and fluctuations in diurnal temperature range can directly affect the reproduction rate of some parasites within hosts (i.e., particularly vector-borne parasites) (Paaijmans et al., 2009) and thus limit their colonization and establishment in certain areas (Clark et al., 2017). In addition, the geographical distance among sites has been observed to produce a decay in compositional similarity for some parasite communities (i.e., distance-decay pattern) (Poulin 2003;Ishtiaq et al., 2010). ...
... Conversely, annual precipitation and precipitation seasonality did not contribute to explain the variation in lineage community composition through the metacommunity (0.4%, p > 0.05). Possibly, the low predictive power of both sets of variables, compared to a higher explanatory power in other avian haemosporidian studies (Clark et al., 2017), relates to their differential effect on species richness and distribution at distinct spatial scales (Rahbek, 2005). Indeed, macroclimatic variables (e.g., broad climatic conditions such as mean annual temperature and precipitation) have been suggested to have an effect on large spatial scales (i.e., continental/regional; Clark et al., 2017), whereas microclimatic variables associated to habitat heterogeneity and structure (e.g., vegetation structure) have been found to have more important effects on smaller spatial scales (i.e., local and landscape scales) (Goetz et al., 2014). ...
Metacommunity ecology studies how species compositional patterns and their distributions vary across local and regional scales and provides insights on processes driving the distribution of communities. Avian haemo-sporidians comprise a diverse and widely distributed parasite taxon; some studies have analyzed their alpha and beta diversity patterns. Yet, metacommunity structures of avian haemosporidians and thus relevant biotic and abiotic variables explaining such structures at the landscape scale (i.e., 10-200 km) have not been assessed. We studied the metacommunity structure of avian haemosporidian mtDNA cyt b lineages and the infected avian host assemblage across four different elevations in Central Veracruz, Mexico. We performed variation-partitioning analyses to evaluate the contribution of host-related traits and climatic variables to the metacommunity. We found a richness of 78 lineages within 38 infected species of birds. At the component community level, we observed that bird species infected with a lower number of parasite lineages (e.g., <3) represented a nested subset of those with a higher number of parasite lineages (e.g., >8) (i.e., nested structure). However, this nested pattern was due to the restricted spatiotemporal co-occurrence of hosts and parasites, given the high degree of turnover across elevations. Host-related traits (functional, transmission-associated, and phylogenetic relationships) only explained a small fraction of the variation (4.4%) in parasite lineage composition across avian hosts. At the habitat level, there was a group turnover by parasite genera across elevation (i.e., quasi-Clementsian structure), which was partly explained by climatic variables (mean annual temperature and annual diurnal range; 27.6%) that may constrain parasite reproduction and vector distribution across the environmental gradient. At the scale of our study, environmental conditions represented a more important driver of avian haemosporidian metacommunity structure than host-related traits, suggesting an important role of environmental filtering structuring parasite assemblages at the landscape level.
... Critical transmission windows typically occur when climate variation or land use modifications facilitate the development and dispersion of haemosporidian parasites within avian communities (Brooks and Hoberg, 2007;Møller, 2010;Clark et al., 2016Clark et al., , 2018. Recent studies demonstrate that long-term inter-annual and seasonal trends in temperature and precipitation (called bioclimatic variables) influence haemosporidian prevalence, host specificity and distribution within a geographic region (Sehgal et al., 2011;Clark et al., 2016;Ferraguti et al., 2020;Fecchio et al., 2019a). ...
... Though common in wild bird communities, haemosporidian infections may represent a potential conservation issue due to unforeseen effects of climate change on parasitehost interactions (Khasnis and Nettleman, 2005;Stresman, 2010). Several factors could potentially influence how haemosporidian infection trends are expressed within avian communities (Benning et al., 2002), including potential changes in transmission risk (Huijben et al., 2016;Garamszegi, 2011), changes in prevalence and coinfection rates (Zamora-Vilchis et al., 2012;Clark et al., 2016), species turnover Clark et al., 2018) and changes in distribution and development thresholds (Marcogliese, 2008;Paaijmans and Thomas, 2013). For instance, avian malaria (i.e., Plasmodium spp.) prevalence correlates positively with temperature and humidity (Lindsay and Martens, 1998;Benning et al., 2002;Valkiūnas, 2005;Stresman, 2010;Garamszegi, 2011;Atkinson et al., 2014). ...
Long-term, inter-annual and seasonal variation in temperature and precipitation influence the distribution and prevalence of intraerythrocytic haemosporidian parasites. We characterized the climatic niche behind the prevalence of the three main haemosporidian genera (Haemoproteus, Plasmodium and Leucocytozoon) in central-eastern Mexico, to understand their main climate drivers. Then, we projected the influence of climate change over prevalence distribution in the region. Using the MaxEnt modelling algorithm, we assessed the relative contribution of bioclimatic predictor variables to identify those most influential to haemosporidian prevalence in different avian communities within the region. Two contrasting climate change scenarios for 2070 were used to create distribution models to explain spatial turnover in prevalence caused by climate change. We assigned our study sites into polygonal operational climatic units (OCUs) and used the general haemosporidian prevalence for each OCU to indirectly measure environmental suitability for these parasites. A high statistical association between global prevalence and the bioclimatic variables ‘mean diurnal temperature range’ and ‘annual temperature range’ was found. Climate change projections for 2070 showed a significant modification of the current distribution of suitable climate areas for haemosporidians in the study region.
... Despite some exceptions [18], most large-scale studies on parasite β-diversity are either focused on ectoparasites, mammals or cold regions ( particularly the Palearctic), most likely due to data availability. However, the observed patterns and identified drivers of parasite diversity are likely to change in response to all those factors. ...
... Parasites rely on hosts to disperse, and there is evidence that spatial connectivity among host assemblages is an essential driver of parasite similarity [18]. As a result, spatial distance can be even more influential in host-parasite systems where both parasites and hosts are dispersal limited. ...
A robust understanding of what drives parasite β-diversity is an essential step towards explaining what limits pathogens' geographical spread. We used a novel global dataset (latitude −39.8 to 61.05 and longitude −117.84 to 151.49) on helminths of anurans to investigate how the relative roles of climate, host composition and spatial distance to parasite β-diversity vary with spatial scale (global, Nearctic and Neotropical), parasite group (nematodes and trematodes) and host taxonomic subset (family). We found that spatial distance is the most important driver of parasite β-diversity at the global scale. Additionally, we showed that the relative effects of climate concerning distance increase at the regional scale when compared with the global scale and that trematodes are generally more responsive to climate than nematodes. Unlike previous studies done at the regional scale, we did not find an effect of host composition on parasite β-diversity. Our study presents a new contribution to parasite macroecological theory, evidencing spatial and taxonomic contingencies of parasite β-diversity patterns, which are related to the zoogeographical realm and host taxonomic subset, respectively. This article is part of the theme issue ‘Infectious disease macroecology: parasite diversity and dynamics across the globe’.
... In community level studies, involving several avian host species, temperature also seems to be a good predictor of Plasmodium prevalence, such as in northeastern Brazil (Rodrigues et al., 2021), and in the Spanish Iberian Peninsula (Illera et al., 2017). However, Parahaemoproteus prevalence has shown contrasting results (associated with colder environments) in comparison with Plasmodium (Clark, 2018;Clark et al., 2018Clark et al., , 2020, which may be related to the different life histories of the primary vectors of Plasmodium and Parahaemoproteus parasites. ...
Full-text available
Vector-borne parasites are important ecological drivers influencing life-history evolution in birds by increasing host mortality or susceptibility to new diseases. Therefore, understanding why vulnerability to infection varies within a host clade is a crucial task for conservation biology and for understanding macroecological life-history patterns. Here, we studied the relationship of avian life-history traits and climate on the prevalence of Plasmodium and Parahaemoproteus parasites. We sampled 3569 individual birds belonging to 53 species of the family Thraupidae. Individuals were captured from 2007 to 2018 at 92 locations. We created 2 phylogenetic generalized least-squares models with Plasmodium and Parahaemoproteus prevalence as our response variables, and with the following predictor variables: climate PC1, climate PC2, body size, mixed-species flock participation, incubation period, migration, nest height, foraging height, forest cover, and diet. We found that Parahaemoproteus and Plasmodium prevalence was higher in species inhabiting open habitats. Tanager species with longer incubation periods had higher Parahaemoproteus prevalence as well, and we hypothesize that these longer incubation periods overlap with maximum vector abundances, resulting in a higher probability of infection among adult hosts during their incubation period and among chicks. Lastly, we found that Plasmodium prevalence was higher in species without migratory behaviour, with mixed-species flock participation, and with an omnivorous or animal-derived diet. We discuss the consequences of higher infection prevalence in relation to life-history traits in tanagers.
... Furthermore, hemosporidians are able to infect a large array of host species, but they usually infect phylogenetically closely related hosts and not necessarily hosts with similar ecological niches (21). Temperature and rainfall seasonality predict a higher parasite host specialization and assemblage uniqueness (32); for Plasmodium, there is a negative association between maximum temperature and phylobeta diversity, suggesting that as temperature increases, communities become more homogeneous (33). Thus, for avian malaria we predict that phylogenetic and environmental variables must be the most important factors predicting bird species susceptibility. ...
Disease transmission prediction across wildlife is crucial for risk assessment of emerging infectious diseases. Susceptibility of host species to pathogens is influenced by the geographic, environmental, and phylogenetic context of the specific system under study. We used machine learning to analyze how such variables influence pathogen incidence for multihost pathogen assemblages, including one of direct transmission (coronaviruses and bats) and two vector-borne systems (West Nile Virus [WNV] and birds, and malaria and birds). Here we show that this methodology is able to provide reliable global spatial susceptibility predictions for the studied host–pathogen systems, even when using a small amount of incidence information (i.e., < 20 % of information in a database). We found that avian malaria was mostly affected by environmental factors and by an interaction between phylogeny and geography, and WNV susceptibility was mostly influenced by phylogeny and by the interaction between geographic and environmental distances, whereas coronavirus susceptibility was mostly affected by geography. This approach will help to direct surveillance and field efforts providing cost-effective decisions on where to invest limited resources.
... Second, overlap networks seem particularly useful to understand probabilities of host shifts in host-parasite interactions. In particular, host species with highly connected range overlap networks might serve as hubs with a high probability of host shifts [39,40]. Finally, range overlap networks might facilitate the study of the relationship between species coexistence and phenotypic evolution by providing a means to directly assess which species are more likely to experience interspecific biotic interactions [41,42]. ...
Full-text available
Direct interactions among species are only possible if there is some overlap in their geographical distributions. However, despite intense focus of macroecological research on species geographical ranges, relatively little theoretical and empirical work has been done on the evolution of range overlap. In this study we explore a simple model of range overlap based on a log-normal distribution of species range sizes along a one-dimensional domain, with or without absorbing boundary conditions. In particular, we focus on the mean and variance of range overlap distributions, as well as the topology of the resulting overlap networks with respect to their degree distribution, evenness, and betweenness scores. According to the model, there is an approximately linear relationship between many aspects of the distribution of range overlaps and their underlying species distributions, such as their mean and variance. However, the expected mean number of non-zero range overlaps for a given species varied from linear to convex depending on the variance of the underlying geographical range distribution. The expected topology of range overlap networks varied substantially depending on the mean and variance in the corresponding geographical distributions, particularly in the case of the degree and closeness distributions. Finally, we test the expectations of our model against five datasets of altitudinal distributions of Neotropical birds. We found strong departures from the expectations based on our model, which could potentially result from phylogenetic niche conservatism related to altitudinal gradients in environmental conditions, or from the asymmetric colonization of mountains by species from lowlands. Potential applications of range overlap networks to a variety of ecological and evolutionary phenomena are discussed.
... Although some island-parasite systems supported island biogeography theory Ishtiaq et al. 2008;Olsson-Pons et al. 2015;Clark et al. 2016;Ellis et al. 2017;Padilla et al. 2017), parasites on other island systems did not show a correlation with geography or environmental conditions. These parasite communities were, instead, significantly correlated with host community composition (Fallon et al. 2003a, b;Beadell et al. 2004;Santiago-Alarcon et al. 2008;Svensson-Coelho and Ricklefs 2011;Olsson-Pons et al. 2015;Clark et al. 2016;Clark et al. 2018;Humphries et al. 2019). To accurately understand the role of island biogeography on parasite communities, we must conduct additional island-specific studies that incorporate geography, environmental conditions and host communities. ...
Full-text available
The taxonomically diverse and relatively understudied avifauna of Papua New Guinea’s (PNG) island archipelagos provide a unique ecological framework for studying haemosporidian parasite differentiation and geographic structure. We implemented molecular and phylogenetic analyses of partial mitochondrial DNA sequences to assess the host distribution of 3 genera of vector-transmitted avian blood parasites ( Plasmodium , Leucocytozoon and Haemoproteus ) across a range of islands off the southeastern tip of PNG. We identified 40 new lineages of haemosporidians, including five lineages belonging to Leucocytozoon , a genus not previously described in this region. Leucocytozoon infections were only observed on the larger, human-inhabited islands. Lineages belonging to Haemoproteus were diverse and had broad geographic distribution. Compared to the mainland, Haemoproteus parasites on the smaller, more distant islands had greater host specificity and lower infection prevalence. The black sunbird ( Leptocoma aspasia ), a commonly caught species, was shown to be a rare host for Haemoproteus spp. infections. Moreover, although birds of the genus Pitohui harbor a neurotoxin (homobatrachotoxin), they demonstrated an infection prevalence comparable to other bird species. The islands of PNG display heterogeneous patterns of haemosporidian diversity, distribution and host-specificity and serve as a valuable model system for studying host-parasite-vector interactions.
... Consequently, to explore dissimilar sets of hosts and environments successfully, widespread parasites might present generic, locally suboptimal adaptations (Futuyma & Moreno, 1988). For instance, Clark et al. (2018) showed that host phylogeny and climate shape haemosporidian parasite assemblages and limit parasite distribution, respectively. Furthermore, specialized haemosporidian lineages colonizing new sites with diverse host communities may be less prone to find suitable hosts and persist in this new community (Pérez-Tris & Lima, 2020). ...
Aim Despite the wide distribution of many parasites around the globe, the range of individual species varies significantly, even among phylogenetically related taxa. Given that parasites need suitable hosts to complete their development, parasite geographical and environmental ranges should be limited to communities where their hosts are found. Parasites might also suffer from a trade-off between being locally abundant or widely dispersed. We hypothesize that the geographical and environmental ranges of parasites are negatively associated with their host specificity and their local abundance. Location World-wide. Time period 2009–2021. Major taxa studied Avian haemosporidian parasites. Methods We tested these hypotheses using a global database that comprises data on avian haemosporidian parasites from across the world. For each parasite lineage, we computed five metrics, namely phylogenetic host range, environmental range, geographical range and the mean local and total number of observations in the database. Phylogenetic generalized least squares models were run to evaluate the influence of phylogenetic host range and the total and local abundances on geographical and environmental ranges. In addition, we analysed separately the two regions with the largest amount of available data: Europe and South America. Results We evaluated 401 lineages from 757 localities and observed that generalism (i.e., phylogenetic host range) was associated positively with both the geographical and environmental ranges of the parasites at global and European scales. For South America, generalism was associated only with geographical range. Finally, mean local abundance (mean local number of parasite occurrences) was negatively related to geographical and environmental ranges. This pattern was detected world-wide and in South America, but not in Europe. Main conclusions We demonstrate that parasite specificity is linked to both their geographical and environmental ranges. The fact that locally abundant parasites present restricted ranges indicates a trade-off between these two traits. This trade-off, however, becomes evident only when sufficient heterogeneous host communities are considered.
Biodiversity is a manifestation of an ecological system specific to a geographical region and shapes the social systems established and pursued by the human community rooted in that region. Biodiversity is thus deeply embedded in the material, social, and even spiritual perceptions of people sharing a particular ecosystem. West Bengal, though is predominantly a region of mono-crop culture growing rice, is endowed with several subregions rich in biodiversity. These are the wetlands of riparian, estuarine, and coastal areas, foothill forest area of the north, mangrove forest area of the deltaic south, and the deciduous forest area of the south-west plateau-fringe districts. The last one among them is unique in a sense that the indigenous species of plants and animals have been the basis of social systems followed for long by a number of tribal communities. The ecological systems interactive therein are protective to the unique social system of the “first people,” and the social systems enriched with “traditional ecological knowledge” are equally protective to the ecology and biodiversity. An integration of administrative, judicial, economic, and social systems formed of traditional ecological knowledge can conserve forest and biodiversity. This paper is an endeavor to explore the chances of conservation of forest and diversity in a combined management system. Analyses have been made with field-based primary and secondary data available for the last five decades; methodology includes qualitative, quantitative, and remote sensing techniques.KeywordsBiodiversityEcological systemIndigenous speciesSocial systemTraditional ecological knowledgeTribal communities
Full-text available
Host-parasite metacommunities are influenced by a myriad of factors, although little is known about which processes affect this relationship at different scales. Here, we tested how local habitat characteristics and host traits explained the parasite metacommunity of a migratory fish in a large Brazilian river floodplain. The parasite metacommunity structure showed a Clementsian pattern, which indicates a more deterministic assembly pattern, in accordance with partial Redundancy Analysis results. Results indicated that species filtering is the predominant mechanism driving community assembly. Patterns were clearer in the dry season of the floodplain. Environmental determinism seems to explain ectoparasite metacommunities in the dry season, in contrast with endoparasites that were more correlated to host traits. Overall, our results indicated that ectoparasitism is an interaction marked by opportunity, whereas endoparasitism is likely related to host features. Thus, we argue that metacommunity structuring of parasites depends on the infection strategy. Our results show that floodplain dynamics are central not only for free-living animal organizations but also for symbiotic interactions. Here, we highlight the importance of understanding the factors influencing the distribution of parasites to predict their transmission, as well as the importance of floodplain dynamics and its hydrological regime on the maintenance of ecological interactions.
Full-text available
Identifying the mechanisms driving the distribution and diversity of parasitic organisms and characterizing the structure of parasite assemblages are critical to understanding host-parasite evolution, community dynamics, and disease transmission risk. Haemosporidian parasites of the genera Plasmodium and Haemoproteus are a diverse and cosmopolitan group of bird pathogens. Despite their global distribution, the ecological and historical factors shaping the diversity and distribution of these protozoan parasites across avian communities and geographic regions remains unclear. Here we used a region of the mitochondrial cytochrome b gene to characterize the diversity, biogeographical patterns and phylogenetic relationships of Plasmodium and Haemoproteus infecting Amazonian birds. Specifically, we asked whether, and how, host community similarity and geography (latitude and area of endemism) structure parasite assemblages across 15 avian communities in the Amazon Basin. We identified 265 lineages of haemosporidians recovered from 2661 sampled birds from 330 species. Infection prevalence varied widely among host species, avian communities, areas of endemism, and latitude. Composition analysis demonstrated that both malarial parasites and host communities differed across areas of endemism and as a function of latitude. Thus, areas with similar avian community composition were similar in their parasite communities. Our analyses, within a regional biogeographic context, imply that host switching is the main event promoting diversification in malarial parasites. Although dispersal of haemosporidian parasites was constrained across six areas of endemism, these pathogens are not dispersal-limited among communities within the same area of endemism. Our findings indicate that the distribution of malarial parasites in Amazonian birds is largely dependent on local ecological conditions and host evolutionary relationships. This article is protected by copyright. All rights reserved.
The range of hosts a pathogen infects (host specificity) is a key element of disease risk that may be influenced by both shared phylogenetic history and shared ecological attributes of prospective hosts. Phylospecificity indices quantify host specificity in terms of host relatedness, but can fail to capture ecological attributes that increase susceptibility. For instance, similarity in habitat niche may expose phylogenetically unrelated host species to similar pathogen assemblages. Using a recently proposed method that integrates multiple distances, we develop a new index to assess the relative contributions of host phylogenetic and functional distances to pathogen host specificity (functional-phylogenetic host specificity). We apply this index to a dataset of avian malaria parasite (Plasmodium and Haemoproteus spp.) infections from Melanesian birds to show that multi-host parasites generally use hosts that are closely related, not hosts with similar habitat niches. We also show that host community phylogenetic ß diversity (Pßd) predicts parasite Pßd, and that individual host species carry phylogenetically clustered Haemoproteus parasite assemblages. Our findings were robust to phylogenetic uncertainty, and suggest that phylogenetic ancestry of both hosts and parasites play important roles in driving avian malaria host specificity and community assembly. However, restricting host specificity analyses to either recent or historical timescales identified notable exceptions, including a 'habitat specialist' parasite that infects a diversity of unrelated host species with similar habitat niches. This work highlights that integrating ecological and phylogenetic distances provides a powerful approach to better understand drivers of pathogen host specificity and community assembly. This article is protected by copyright. All rights reserved.
Natural communities are well known to be maintained by many complex processes. Despite this, the practical aspects of studying them often require some simplification, such as the widespread assumption that direct, additive competition captures the important details about how interactions between species impact community diversity. More complex non-additive ‘higher-order’ interactions are assumed to be negligible or absent. Notably, these assumptions are poorly supported and have major consequences for the accuracy with which patterns of natural diversity are modelled and explained. We present a mathematically simple framework for incorporating biologically meaningful complexity into models of diversity by including non-additive higher-order interactions. We further provide empirical evidence that such higher-order interactions strongly influence species’ performance in natural plant communities, with variation in seed production (as a proxy for per capita fitness) explained dramatically better when at least some higher-order interactions are considered. Our study lays the groundwork for a long overdue shift in how species interactions are used to study the diversity of natural communities.
Taxa co‐occurring in communities often represent a nonrandom sample, in phenotypic or phylogenetic terms, of the regional species pool. While heuristic arguments have identified processes that create community phylogenetic patterns, further progress hinges on a more comprehensive understanding of the interactions between underlying ecological and evolutionary processes. We created a simulation framework to model trait evolution, assemble communities (via competition, habitat filtering, or neutral assembly), and test the phylogenetic pattern of the resulting communities. We found that phylogenetic community structure is greatest when traits are highly conserved and when multiple traits influence species membership in communities. Habitat filtering produces stronger phylogenetic structure when taxa with derived (as opposed to ancestral) traits are favored in the community. Nearest‐relative tests have greater power to detect patterns due to competition, while total community relatedness tests perform better with habitat filtering. The size of the local community relative to the regional pool strongly influences statistical power; in general, power increases with larger pool sizes for communities created by filtering but decreases for communities created by competition. Our results deepen our understanding of processes that contribute to phylogenetic community structure and provide guidance for the design and interpretation of empirical research.
Species introductions are a dominant component of biodiversity change but are not explicitly included in most discussions of biodiversity-disease relationships. This is a major oversight given the multitude of effects that introduced species have on both parasitism and native hosts. Drawing on both animal and plant systems, we review the competing mechanistic pathways by which biological introductions influence parasite diversity and prevalence. While some mechanisms - such as local changes in phylogenetic composition and global homogenization - have strong explanatory potential, the net effects of introduced species, especially at local scales, remain poorly understood. Integrative, community-scale studies that explicitly incorporate introduced species are needed to make effective predictions about the effects of realistic biodiversity change and conservation action on disease. Introduced species can have strong effects on parasite prevalence and richness.Introductions primarily affect disease via changes in host composition, not richness.Local effects of introduced species can amplify or dilute parasite prevalence.Global homogenization and spread of human commensals systematically increase disease.