Oecologia
https://doi.org/10.1007/s00442-021-05011-9
COMMUNITY ECOLOGY – ORIGINAL RESEARCH
Temperature and productivity distinctly affect the species richness
of ectothermic and endothermic multitrophic guilds along a tropical
elevational gradient
Chaim J. Lasmar1 · Clarissa Rosa2 · Antônio C. M. Queiroz1 · Cássio A. Nunes3 · Mayara M. G. Imata1 ·
Guilherme P. Alves1 · Gabriela B. Nascimento1 · Ludson N. Ázara4 · Letícia Vieira5 · Júlio Louzada6 ·
Rodrigo M. Feitosa7 · Antonio D. Brescovit8 · Marcelo Passamani9 · Carla R. Ribas1
Received: 30 January 2021 / Accepted: 3 August 2021
© The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2021
Abstract
The diversity of endotherms and ectotherms may be differently affected by ambient temperature and net primary productivity (NPP). Additionally, little is known about how these drivers affect the diversity of guilds of different trophic levels. We
assessed the relative role of temperature and NPP in multitrophic guilds of ectothermic (arthropods: ants, ground beetles,
spiders, and harvestmen) and endothermic (large mammals) animals along a tropical elevational gradient. We sampled arthropods at eight elevation belts and large mammals at 14 elevation belts in Atlantic rainforest (ranging from 600 to 2450 m.a.s.l.)
of Itatiaia National Park, Southeast Brazil. Overall arthropod species richness was more associated with temperature than
overall large-mammal species richness, while the latter was more associated with NPP. When separated into trophic guilds,
we found that the species richness associated with NPP increased across arthropod trophic levels from herbivores to predators.
Conversely, although NPP influenced large-mammal herbivore species richness, its effects did not seem to accumulate across
large-mammal trophic levels since the species richness of large-mammal omnivores was more associated with temperature
and none of the variables we studied influenced large-mammal predators. We suggest that thermal physiological differences
between ectotherms and endotherms are responsible for the way in which arthropods and large mammals interact with or
are constrained by the environment. Furthermore, the inconsistency regarding the role of temperature and NPP on species
richness across multitrophic guilds of ectotherms and endotherms could indicate that thermal physiological differences might
also interfere with energy use and flux in the food web.
Keywords Community ecology · Elevational gradient · Species richness · Trophic ecology · Tropical mountain
Introduction
The evaluation of patterns of species diversity across ecological gradients has been a core focus of ecology studies
(MacArthur 1972; Fine 2015). Several hypotheses have
been developed to explain how species diversity varies in
space based on the influence of historical and ecological
factors (Fine 2015). Among them, two main ecological factors, temperature, and net primary productivity (NPP), help
to explain how energy in an ecosystem may drive patterns
Communicated by Nina Farwig.
* Chaim J. Lasmar
chaimlasmar@gmail.com
Extended author information available on the last page of the article
of species richness (Wright 1983; Hawkins et al. 2003a;
Evans et al. 2005; Brown 2014). However, their effects are
complex: for instance, although both temperature and NPP
positively affect the species richness of both endotherms
and ectotherms (Hillebrand 2004; Brown 2014), differences
in the thermal physiology between these organisms lead to
different mechanisms regulating species diversity patterns
(Allen et al. 2002; Buckley et al. 2012). Additionally, as
both temperature and NPP are related to how energy enters
and flows through an ecosystem, their effects vary in relative importance when dealing with diversity patterns among
different trophic guilds (Voigt et al. 2003; Welti et al. 2020).
The high resource assimilation at high temperatures and
the less intense competition caused by the high availability of resources (e.g., high NPP) increase the population
sizes of species, which in turn decreases extinction rates
13
Vol.:(0123456789)
Oecologia
and allow more species to coexist (Evans et al. 2005; Brown
2014). Temperature affects animals through a positive influence on their metabolic rates, up to a point (Gillooly et al.
2001), which also increases speciation rates (Brown 2014).
Endotherms are usually bigger than ectotherms, and smaller
organisms typically lose too much heat (Buckley et al 2012).
This is especially true for ectotherms, such as arthropods,
since their body temperature is dependent on environmental temperature (Lessard et al. 2011; Brown 2014). Even in
environments with high resource availability, ectotherms are
unable to obtain and assimilate energy from the ecosystem at
low temperatures (Sanders et al. 2007; Buckley et al. 2012).
On the other hand, endotherms acquire resources more independently of external temperatures than ectotherms because
they are able to control their body temperature. However,
endothermy comes at a high cost (10 times higher than those
of ectotherms for energetic maintenance), which results in
endotherms obtaining a greater amount of energy from food
resources (Buckley et al. 2012). Therefore, the higher consumption of resources possibly brings endotherms faster
to the limit of ecosystem resources. Thus, we could expect
that physiological constraints caused by temperature would
be more strongly associated with diversity patterns of ectotherms than endotherms (Buckely et al. 2012). In contrast,
endotherms would be more constrained by NPP effects
(Buckely et al. 2012), leading to greater species richness
in more productive environments, as reported for large and
small mammals along elevational gradients (e.g., McCain
et al. 2018; Gebert et al. 2019).
Accordingly, both NPP and temperature may also affect
how energy flows across trophic guilds in an ecosystem
(Evans et al. 2005; Birkhofer and Wolters 2012) and operates among trophic levels (Voigt et al. 2003, 2007; Welti
et al. 2020). Through accelerated kinetics and increased
metabolic rates, temperature positively affects the assimilation of resources for many trophic levels, from herbivores to
top predators (Moorthi et al. 2016; Binkenstein et al. 2018).
In more productive environments, the greater abundance of
animals in low trophic levels (e.g., herbivores) is caused by
the high energy input in the ecosystem; these abundant low
trophic level animals serve as resources for animals at higher
trophic levels (e.g., predators), resulting in an increase in
the abundance of these animals (Evans et al. 2005). In this
bottom-up effect, high animal abundance at all trophic levels
decreases extinction rates and allows for more diversity in
the whole food chain (Turney and Buddle 2016). Although
predators also control the abundance and diversity of lower
trophic levels (Terborgh 2015), they are more sensitive to the
energy input and flux in the food chain (Voigt et al. 2007;
Turney and Buddle 2016; Brose et al. 2017). The energy lost
from one trophic level to another results in the increase of
resource limits, in which predators are the ultimate energy
receptors in the ecosystem. Previous studies have reported
13
that arthropods at higher trophic levels are more sensitive to
changes in NPP than species at lower trophic levels (Kaspari
2001; Haddad et al. 2009). Thus, even if temperature is the
major driver of the diversity of ectotherms, we expect an
increase in the influence of NPP across trophic levels not
only for endotherms but also for ectotherms.
Although few studies have investigated the influence of
both temperature and NPP on multitrophic guilds of animals
(e.g., Xu et al. 2018; Welti et al. 2020), no studies to date
have evaluated whether the effects of temperature and NPP
across trophic guilds are consistent among ectothermic and
endothermic animals. Searching for similar patterns across
the trophic levels of both endothermic and ectothermic animals might be helpful in explaining the influence of environmental drivers on biodiversity and energy flux and ultimately
on ecosystem functioning (Turney and Buddle 2016). In this
context, tropical elevational gradients are a great lab for testing the differences in the influence of temperature and NPP
on biodiversity. Elevational gradients present a large variation in environmental factors on a relatively small spatial
scale, wherein animal diversity is constrained by this environmental variation but not by dispersal limitations (Körner
2007; Sundqvist et al. 2013). Many ecological drivers are
directly linked to elevation besides temperature, such as
land area, atmospheric pressure, and clear-sky turbidity
(Körner 2007). However, temperature and NPP are known
as the main drivers for tropical elevational fauna (Peters et al
2016; Gebert et al. 2019). In addition, because tropical animals have evolved in a more climatically stable environment,
they typically have narrow thermal tolerances (Janzen 1967).
This results in a high species turnover in tropical mountains
due to ecological filters imposed by temperature variation
(Janzen 1967; Rahbek et al. 2019), which is advantageous
for disentangling temperature from any other effects of drivers that are typically not strongly correlated to elevation,
such as NPP. Understanding how temperature and NPP constrain species richness in both ectotherms and endotherms as
well as across different trophic guilds is essential for predicting how biodiversity and ecosystem functioning will respond
to a rapidly changing world.
Here, we aimed to assess the relative role of temperature
and NPP in multitrophic guilds of ectothermic (arthropods:
ants, ground beetles, spiders, and harvestmen) and endothermic (large mammals) animals along a tropical elevational
gradient. Although we expect that both temperature and NPP
will positively influence both endotherms and ectotherms,
we specifically tested whether the relative importance of
NPP concerning temperature is higher for endotherms than
for ectotherms. Additionally, regardless of the most important environmental driver for these patterns, we also tested
whether there was an increase in the association of NPP with
species richness across trophic levels for both ectotherms
(arthropods) and endotherms (large mammals).
Oecologia
Materials and methods
Study area
We conducted our sampling at Itatiaia National Park (INP)
in Southeast Brazil (between 22° 16′–22° 28′ S and 44°
34′–44° 42′ W). INP was established in 1937 and is the
oldest national park in Brazil in the Atlantic rainforest
biome. The park is part of the Itatiaia Massif, which is
within the Mantiqueira Mountain Range. The protected
area starts at 600 m of elevation and peaks at 2878 m.
Several vegetation types occur along INP elevational
gradient varying from forests to grasslands, including lower montane forest (at 0–500 m.a.s.l.), montane forest (500–1500 m.a.s.l.), and upper montane
forest (1500–2000 m.a.s.l.). High-altitude grasslands
(2000–2500 m.a.s.l.) and natural remnants of upper montane forest (Safford 1999) comprise the highest part of the
INP. As vegetation structure, in terms of forest or grassland, affects the observed diversity patterns along elevational gradients (Lasmar et al. 2020), we sampled both
arthropods and mammals only in forest habitats.
Sampling of arthropods and classification
into trophic guilds
We sampled four groups of arthropods to represent ectothermic animals: ants (Insecta: Hymenoptera: Formicidae), ground beetles (Insecta: Coleoptera: Carabidae),
spiders (Arachnida: Araneae), and harvestmen (Arachnida:
Opiliones). All samplings took place during the rainy season between March and April 2015, and we chose eight
elevation belts (at 600, 848, 1134, 1515, 1810, 2000, 2200,
and 2457 m.a.s.l; Fig. 1a) spatially separated by at least
1.4 km. To sample arthropods, we set a 200 m transect in
each elevation belt containing ten sampling points each,
which were 20 m apart from each other. At each sampling
point, we set a 1.5 m × 1.5 m grid of four unbaited pitfall
traps composed of one trap at each grid vertex (Fig. 1b).
The pitfall traps consisted of recipients (11 cm in diameter
and 11 cm in depth) filled with a solution of water, salt,
and liquid soap. They operated for 48 h in the field, as this
period is known to be enough to sample the majority of
arthropod species on the ground (Engelbrecht 2013; Lasmar et al. 2017). We started the samplings instantly after
pitfall installation, as we considered the digging-in effect
(i.e., high captures in the early periods of pitfall operation)
negligible (Lasmar et al. 2017). We also extracted 1 m2
of forest leaf litter leaf using Fisher’s (1998) “miniWinkler” (Fig. 1b). For the Winkler samples, arthropods were
extracted for 72 h in the miniWinkler sacks. Combining
both pitfall traps and Winkler samples allowed us to obtain
the best representation of arthropod diversity (Sabu et al.
2011).
We sorted the ants by genera following Baccaro et al.’s
(2015) auxiliary keys. The ants in each genus were then
sorted into morphospecies and identified at the species
level whenever possible by comparison with identified
species in the ant reference collection of the Laboratório
de Ecologia de Formigas at the Universidade Federal de
Lavras (UFLA), Lavras, Brazil. Ant species identification
was confirmed by taxonomists at the Universidade Federal
do Paraná (UFPR) in Curitiba, Brazil (Dr. Alexandre Casadei-Ferreira, Dr. Thiago R. da Silva, and Weslly Franco).
Voucher specimens were deposited at UFLA and at the
Entomological Collection Padre Jesus Santiago Moure of
UFPR (DZUP) in Curitiba, Brazil. Carabid ground beetles
were identified by auxiliary keys (Straneo and Ball 1989),
and the morphospecies were identified to the lowest level
possible. The specimens were deposited in the Entomological Collection of the Laboratório de Ecologia in the
Departamento de Ciências Florestais at UFLA. For harvestmen and spiders, we considered only adult individuals.
We sorted harvestmen based on the analysis of the specimen exomorphology using relevant taxonomic literature
(Pinto-da-Rocha and Giribet 2007) and by comparison
against specimen in the type material. The specimens were
deposited in the Collection of Arachnida, Myriapoda, and
Onychophora at Museu Nacional at Universidade Federal
do Rio de Janeiro (MNRJ). Unfortunately, all the collected
harvestmen were lost in a fire in 2018 along with the bulk
of the arachnological collection of MNRJ (Kury et al.
2018). The spiders were identified and deposited at the
Arachnological Collection of the Laboratório de Coleções
Zoológicas at the Instituto Butantan. The spider specimens
were determined using the key of Brescovit et al. (2002)
and the auxiliary bibliography of the World Spider Catalog
(2021).
We classified arthropods into three trophic guilds
according to their ecological role in the ecosystem following previous studies (see Basset et al. 2012; Kitching
et al. 2020). Thus, arthropods were assigned to one of the
following trophic guilds: (1) arthropod herbivores, which
feed predominantly on plant resources or through symbiosis with fungi that grow in plant resources; (2) arthropod
omnivores, which feed on a wide range of substances, such
as plant resources, feces, and dead and living organisms;
and (3) arthropod predators, which feed predominantly
on other living animals. In short, herbivore and omnivore
arthropods consisted solely of ants and arthropod predators of ants, ground beetles, harvestmen and spiders. Each
arthropod species and its corresponding trophic guild are
shown in Table S1 of the Online Resource.
13
Oecologia
(a)
(b)
(c)
Fig. 1 a Elevational gradient sampled in Itatiaia National Park in
Southeast Brazil. Black stars represent the arthropod (ants, ground
beetles, harvestmen, and spiders) sampling sites while black circles
represent the large-mammal sampling sites. b Scheme of the arthro-
pod sampling transects and sampling points using pitfall traps and
Winkler extractor. c Scheme of the large mammals sampling with
camera traps
Sampling of large mammals and classification
into trophic guilds
elevation belts (871, 889, 953, 1100, 1175, 1258, 1316,
1422, 1472, 1672, 1813, 1875, 2000, 2246 m.a.s.l.;
Fig. 1a), all in forest habitats that were spatially separated
by at least 730 m. In each elevation belt, we installed one
motion-activated camera trap (Bushnell HD, © Bushnell
We sampled large mammals in the rainy season between
January and April of 2014 and 2015. We chose 14
13
Oecologia
Outdoor Products, California, USA) that remained in
operation for 80 consecutive days with a total sample
effort of 1,120 trap/nights (Fig. 1c). We installed camera
traps in places commonly used by mammals based on
the presence of animal signs and trails (Srbek-Araujo and
Chiarello 2013). We used no bait to avoid artificial animal
attraction and to maintain the premise of equal catchability. We programmed camera traps to take three photos in
30-s intervals when triggered by an animal. We identified large mammals through photographic records, and
we considered as independent observations the records
that presented a time lapse of > 1 h between them (SrbekAraujo and Chiarello 2013). We classified mammals into
three trophic guilds based on their role in the ecosystem
(see Srbek-Araujo and Kierulff 2016): (1) herbivores, (2)
omnivores, and (3) predators. Each large-mammal species
and its corresponding trophic guild are shown in Table S2
in Online Resource.
Measuring temperature and net primary
productivity
To extract net primary productivity (NPP) for each elevation belt for elevation gradients of both arthropods and
large mammals, we assessed the Land Processes Distributed Active Archive Center website from NASA (https://
lpdaac.usgs.gov) to extract MODIS TERRA and AQUA
NPP (MOD17A3HGF and MYD17A3HGF, https://
lpdaac.usgs.gov/products/mod17a3hgfv006) at 500 m of
resolution. As NPP is an annual product and we collected
our samples in 2014 and 2015, we extracted NPP values
for the year that we collected the samples and for the corresponding previous year because we always sampled at
the beginning of the given year; therefore, our response
variable might also be influenced by the NPP of the previous year. We calculated the average values between
MODIS TERRA and AQUA NPP and for two years (i.e.,
the previous year and the year we collected the samples)
to obtain one value for each elevation belt. To extract
temperature data, we obtained the mean annual temperature from the CHELSA database (Karger et al. 2017) with
a 1 km resolution. CHELSA mean annual temperature
corresponds to average values from monthly temperature
for the years 1979–2013. Mean annual temperature was
highly correlated with elevation in both arthropod (Pearson r = − 0.99) and large-mammal gradients (Pearson
r = − 0.98). This is in accord with Barry (2008), who
states that it is well established that temperature decreases
0.6 °C per 100 m elevation gain. Both the NPP and temperature data were extracted with ArcGIS 10.1 software,
and we ensured that all elevation belts were not in the
same grid cell.
Data analyses
We checked the sample coverage for each of our response
variables by estimating asymptotic species richness from
frequency data (i.e., frequency of occurrence in sampling
points per transect) using a nonparametric species richness
estimator (Chao 1; Colwell and Coddington 1994; Gotelli
and Colwell 2011). We estimated arthropod diversity
through frequency data because an ant worker, as a eusocial
organism, cannot be considered an individual (Gotelli et al.
2011), which led us to standardize this for all arthropods. We
also estimated large-mammal diversity through frequency
data since we only counted photographs that were registered on camera traps. We checked for correlation between
observed species richness and Chao 1‐estimated diversity.
If we found high correlations (r ≥ 0.7), we only performed
the analyses with observed values of species richness. If
those parameters were not strongly correlated (r < 0.7), we
performed analyses with both observed species richness and
Chao 1-estimated diversity.
We verified whether observed species richness followed a
linear or unimodal function in response to elevation for the
total species richness of arthropods (n = 8) and large mammals (n = 14) as well as for each of the three trophic guilds of
arthropods and mammals separately. To do so, we performed
a theoretical approach following the second-order Akaike
information criterion corrected for small samples (AICc—
Burnham and Anderson 2002). To perform the model selection, we built, for each observed species richness (response
variables), three models that presented elevation as a linear
function, elevation2, to represent a unimodal function and
the null model as explanatory variables. We used the function ‘aictab’ from the AICcmodavg package in R (Mazerolle
2020) to test which function best explained our response
variables. The explanatory variables with the lowest AICc
indicated the function (linear or unimodal) that best fit our
response variable pattern along the elevational gradient.
To test the influence of temperature and NPP on the species richness of ectotherms (all arthropods) and endotherms
(all large mammals) as well as on the different trophic guilds,
we first verified the correlation between our explanatory variables, NPP and temperature. For both gradients, NPP and
temperature were not strongly correlated (arthropod gradient, Spearman r = 0.63; large-mammal gradient, Spearman
r = 0.68). Thus, to test whether temperature was more important than NPP in driving the species richness of ectotherms
and whether NPP was more important than temperature in
driving the species richness of endotherms, we compared
the responses of overall arthropod species richness (n = 8)
and overall species richness of large mammals (n = 14) by
constructing generalized linear models (GLMs) with temperature and NPP as explanatory variables. We simplified
both models by performing a backward approach, consisting
13
Oecologia
of removing nonsignificant explanatory variables from the
models until only the significant variables were present in
the model. To verify that the collinearity between temperature and NPP was not a problem for fitting and interpreting
our model, we calculated the variance inflation factor (VIF)
for each model using the vif function in the ‘car’ package in
R (Fox and Weisberg 2019). If the VIF values were lower
than 5, the collinearity did not require attention, but it did
require attention for VIF values greater than 5. We used the
“Quasipoisson” family for all our models, and we verified
which of those significant variables better explained our
response variables by assessing the 95% confidence intervals of their proportion of variation explained (R2) in the
regression models through a resample from 1000 bootstrap
samples using the ‘boot.ci()’ function from the ‘boot’ package (Canty and Ripley 2012).
To test whether the influence of NPP on species richness
increases across trophic levels in both ectotherms (arthropods) and endotherms (large mammals), we constructed
GLMs for each trophic guild as a response variable and
performed a backward approach with elevation and NPP
as explanatory variables. We extracted the variance inflator
factor (VIF) for each model. We also compared the 1000
bootstrapped R2 values from significant explanatory variables from GLMs of trophic guilds of arthropods and large
mammals. When performing the GLMs, we assessed the
residual to obtain the adequacy of the error distribution
(Crawley 2002).
We also performed variance partitioning analyses for all
models (i.e., overall arthropods and large mammals, as well
as for all trophic guilds), to verify the unique and shared proportion of variance explained by temperature and NPP. We
performed this using the ‘var.part’ function in the ‘vegan’
package in R (Oksanen et al. 2020).
We confirmed the adequacy of the error distribution family for each model as well as VIF values lower than 5 (see
Table S5 in the Online Resource). We performed all analyses
in R version 4.0.3 (R Development Core Team 2020).
Results
In the arthropod elevational gradient, we identified a total
of 286 species from 969 ant occurrences, 322 ground beetles, 179 harvestmen (145 adults and 34 juveniles), and 1229
spiders (686 adults and 543 juveniles). Per trophic guild, we
identified seven species of herbivores (seven ant species), 96
species of omnivores (96 ant species), and 183 predators (72
ant species, 16 harvestmen species, 23 ground beetle species, and 72 spider species). In the large-mammal gradient,
we recorded 21 species from 383 occurrences. One of these
species was Sus scrofa, a nonnative animal of the American continent that was introduced in the highest part of the
13
Mantiqueira Mountain Range; therefore, we removed S.
scrofa from the analyses since all diversity patterns changed
when including this introduced species (See Table S3 and
S4 in Online Resource). We recorded five species of largemammal herbivores, six large-mammal omnivores and eight
large-mammal predators. The list of all recorded species as
well as their trophic guilds can be found in Table S1 and S2
in Online Resource. We found strong correlations between
observed species richness and Chao 1-estimated diversity
for all arthropods combined (r = 0.99), arthropod herbivores
(r = 0.99), arthropod omnivores (r = 0.98), and arthropod
predators (r = 0.99). We also found strong correlations
between observed species richness and Chao 1-estimated
diversity for total large mammals (r = 0.96), large-mammal
herbivores (r = 1), large-mammal omnivores (r = 0.96) and
large-mammal predators (r = 0.99). Therefore, we used only
observed species richness for both arthropods and largemammal parameters for all analyses.
Elevational gradient patterns
Overall, arthropods, arthropod omnivores, and arthropod
predators better followed a unimodal pattern along elevation,
while arthropod herbivores linearly declined with elevation
(Fig. 2a–d and Table S3 in Online Resource). We found that
the unimodal function best explained the elevational gradient for species richness of overall large mammals, largemammal herbivores and omnivores (Fig. 2e–h and Table S3
in Online Resource). However, neither the linear nor unimodal function explained the elevational gradient for largemammal predators.
Effects of temperature and net primary productivity
on all arthropods and all large mammals
We found that overall arthropod species richness was
positively related to temperature and to NPP (Table 1 and
Fig. 3a, b). On the other hand, overall species richness of
large mammals was positively related to NPP but not to
temperature (Table 1, Fig. 3c, d). As expected, our bootstrap R2 analyses shows that temperature had the greatest
explanatory power for overall arthropod species richness,
while NPP was the major ecological driver influencing
the species richness of all large mammals (Fig. 4a). The
variance partitioning showed that the majority of the
overall explained variance was shared between temperature and NPP (Fig. S1a and b in the Online Resource).
However, we also observed that the unique explained
variance of temperature was higher for overall arthropods species richness (42% of the overall explained variance) in comparison to large mammals (31% of the overall explained variance). On the other hand, the shared
contribution, which also accounts for NPP effects, was
Oecologia
140
4
(a)
60
(b)
Species richness
120
80
(c)
3
100
60
40
80
2
40
30
60
20
40
1
20
10
20
0
0
600
1000
1400
1800
2200
0
600
Elevation (m.a.s.l.)
14
1000
1400
1800
2200
0
600
4
1400
1800
2200
600
1000
Elevation (m.a.s.l.)
Elevation (m.a.s.l.)
(e)
1000
4
(f)
1400
1800
2200
Elevation (m.a.s.l.)
6
(g)
12
Species richness
(d)
50
(h)
5
3
10
3
4
8
2
2
1
1
0
0
3
6
2
4
1
2
0
900
1200
1500
1800
2100
900
Elevation (m.a.s.l.)
1200
1500
1800
2100
0
900
Elevation (m.a.s.l.)
1200
1500
1800
2100
900
Elevation (m.a.s.l.)
Fig. 2 Elevational species richness patterns in Itatiaia National Park,
Southeast Brazil. Graphs show the linear or unimodal relationship
of elevation and the observed species richness for a all arthropods, b
arthropod herbivores, c arthropod omnivores, d arthropod predators,
e all large mammals, f large-mammal herbivores, g large-mammal
1200
1500
1800
2100
Elevation (m.a.s.l.)
omnivores, and h large-mammal predators. Points show the observed
species richness according to elevation, and the lines are model
predictions of the lowest AICc function (i.e., linear or unimodal).
In graph h, the null model best represents the relationship between
large-mammal predators and elevation
Table 1 Generalized linear models (GLMs) for two elevational gradients of arthropods and large mammals in Itatiaia National Park in Southeast
Brazil
Dependent
Independent
df
Pseudo R2
Standardized rc
2.5% CI
97.5% CI
F
p
All arthropods
Net primary productivity
Temperature
Net primary productivity
Temperature
Net primary productivity
Temperature
Net primary productivity
Temperature
Net primary productivity
Temperature
Net primary productivity
Temperature
Net primary productivity
Temperature
Net primary productivity
Temperature
5
0.88
6
0.77
5
0.84
6
0.75
12
0.30
11
0.53
12
0.36
–
–
0.326
0.325
Excluded
1.393
0.409
0.532
0.411
Excluded
0.449
Excluded
0.295
0.631
Excluded
0.367
Excluded
Excluded
0.086
0.143
–
0.733
− 0.025
0.238
0.208
–
0.061
–
− 0.578
0.095
–
0.114
–
–
0.581
0.506
–
2.230
0.923
0.829
0.637
–
1.015
–
1.645
1.245
–
0.634
–
–
25.83
12.17
3.21
20.65
14.12
12.49
17.46
5.20
5.51
1.96
9.57
5.45
3.01
8.31
2.84
< 0.01
0.004
0.017
0.132
0.003
0.013
0.016
0.005
0.071
0.036
0.188
0.010
0.039
0.111
0.013
0.117
0.930
Arthropod herbivores
Arthropod omnivores
Arthropod predators
All large mammals
Large-mammal herbivores
Large-mammal omnivores
Large-mammal predators
The GLMs were separately built for the observed species richness of all arthropods, large mammals and their trophic guilds as the dependent
variables, and we used temperature and NPP as independent variables. We simplified all the GLMs by performing a backward approach. We
reported the GLM degrees of freedom (d.f.) and Pseudo R2 for the final models and the standardized regression coefficients (rc) and their confidence intervals (CI) only for independent variables that were present in the final models. We also reported statistical test values (F) and p values
for the excluded independent variables from the final (excluded) model and for the remaining independent variables in the final model. Bold values indicate significant relationships (p ≤ 0.05)
13
Oecologia
All large mammals
All arthropods
140
(a)
120
100
100
80
80
60
60
40
40
20
0
Species richness
120
12
14
16
18
14
(b)
12
Species richness
140
14
(c)
12
10
10
8
8
6
6
4
4
20
2
2
0
0
20
12000
Mean annual temperature (°C)
14000
16000
0
12
18000
14
Arthropod herbivores
4
(e)
3
3
2
2
1
1
0
14
16
18
20
Mean annual temperature (°C)
12000
14000
3
3
2
2
1
1
16000
Net primary producitivity (g_C/m²/year)
60
(i)
14
40
30
30
20
20
10
10
0
18000
(h)
18
12000
14000
16000
18000
Net primary producitivity (g_C/m²/year)
Large-mammal omnivores
4
(j)
4
(k)
14
16
18
20
Mean annual temperature (°C)
3
3
2
2
1
1
0
0
12
12000
14000
16000
(l)
0
12
18000
14
(m)
60
60
40
40
20
20
0
14
16
18
20
Mean annual temperature (°C)
12000
14000
12000
14000
16000
18000
Net primary producitivity (g_C/m²/year)
6
(0)
5
5
4
4
3
3
2
2
1
1
0
0
12
18
Large-mammal predators
6
(n)
Species richness
80
16
Mean annual temperature (°C)
Net primary producitivity (g_C/m²/year)
Arthropod predators
Species richness
16000
50
40
80
16
Mean annual temperature (°C)
Species richness
Species richness
50
14000
0
12
18000
Arthropod omnivores
60
12000
Net primary producitivity (g_C/m²/year)
4
(g)
0
0
12
18
Large-mammal herbivores
4
(f)
Species richness
Species richness
4
16
Mean annual temperature (°C)
Net primary producitivity (g_C/m²/year)
(d)
16000
18000
Net primary producitivity (g_C/m²/year)
(p)
0
12
14
16
18
Mean annual temperature (°C)
12000
14000
16000
18000
Net primary producitivity (g_C/m²/year)
Fig. 3 Relationship between temperature and net primary productivity with arthropods and large mammals along a tropical elevational
gradient at Itatiaia National Park, Southeast Brazil. Graphs show the
comparison of the relationship of temperature and net primary productivity with the observed species richness of all arthropods (a, b),
all large mammals (c, d), arthropod herbivores (e, f), large-mammal
herbivores (g, h), arthropod omnivores (i, j), large-mammal omni-
vores (k, l), arthropod predators (m, n), and large-mammal predators
(o, p). Note that in c, d, k, and l, we only presented graphs without
the nonnative species S. scrofa. Points show the observed species
richness according to a given value of temperature or net primary
productivity, and the lines are model predictions of significant relationships (e.g., p ≤ 0.05)
higher for species richness of large mammals (69% of the
overall explained variance) than for arthropods (57% of
the overall explained variance). This was in accordance
with our bootstraps R 2 analyses and indicated that the
effect of temperature is strongly associated with arthropods while the NPP effect was more strongly associated
with large mammals.
13
Oecologia
Fig. 4 Comparison of the proportion of variance explained
(R2) by the significant effects
of temperature and net primary
productivity (NPP) from the
GLM regressions. Comparisons of temperature and NPP
effects on observed species
richness were made for a all
arthropods and for all largemammal species richness and
for b arthropods and large
mammals separated in trophic
guilds (herbivores, omnivores,
and predators). White boxplots
represent temperature effects,
while blue boxplots represent
NPP effects. Vertical dashed
lines are bootstrapped confidence intervals constructed by
1000 bootstrap samples with
replacement, and black dots
represent data outliers. Notch
areas on boxplots mark the
95% confidence intervals of the
median value (shown as black
horizontal lines)
Temperature effect
1.0
Net primary productivity effect
(a)
0.8
0.6
R²
0.4
0.2
0.0
All arthropods
1.0
All arthropods
All large mammals
(b)
0.8
0.6
R²
0.4
0.2
0.0
Arthropod Arthropod Arthropod Arthropod L. mammal L. mammal L. mammal
herbivores omnivores omnivores predators herbivores herbivores omnivores
Effects of temperature and net primary productivity
on multitrophic guilds of arthropods and large
mammals
When verifying the influence of temperature and NPP
across trophic guilds, we found that the species richness
of arthropod herbivores was positively related to temperature, but we detected no relationship with NPP (Table 1,
Fig. 3e, f). For the species richness of arthropod omnivores, we found a positive relationship with both temperature and NPP (Table 1, Fig. 3i, j). For the species richness
of arthropod predators, we found a positive relationship
with NPP, but we detected no relationship with temperature (Table 1, Fig. 3m, n).
For large-mammal herbivores, we found a positive relationship with both temperature and NPP, while for largemammal omnivores, we found a positive relationship with
temperature, but we detected no relationship with NPP
(Table 1, Fig. 3g, h, k, l). We found that large-mammal
predators were not related to temperature or NPP (Table 1,
Fig. 3o, p).
Comparing the bootstrap R2 values from the GLMs, we
found an increase in NPP associated with an increase in the
trophic guild level of arthropods, as we expected (Fig. 4b).
However, distinctly from what we expected, NPP association with species richness did not increase across largemammal trophic guilds, which indicated a more complex
pattern (Fig. 4b). The variance partitioning analysis also
showed that increasing arthropods trophic levels increased
NPP association (Fig S1c, e, g in Online Resource). From
arthropod herbivores to predators, the shared explained variance increased from 0 to 65% while the unique explained
variance of NPP increased from 8 to 24% in relation to
the overall explained variance. However, we observed no
evidence that increasing trophic levels also increased NPP
association for large-mammal trophic guilds (Fig. S1d, f,
h in Online Resource). Therefore, bootstrap analysis was
also in accordance with variance partition analysis, showing
13
Oecologia
that increasing trophic levels of ectotherms (arthropods) also
increased NPP association while endotherms (large mammals) presented complex patterns about NPP association
across trophic levels.
Discussion
As expected, temperature was strongly associated with
arthropod diversity, while NPP was associated with largemammal diversity. However, distinctly from the expected
results, we found no consistency of stronger NPP association with species richness of higher trophic levels between
arthropods and large mammals. The importance of NPP
association increased across trophic levels of arthropods,
while for large-mammal trophic guilds, we did not observe
clear patterns. Our results indicate that differences in animal
thermal physiology might affect the way diversity is constrained by environmental temperature and NPP, as reviewed
in Buckley et al. (2012). In addition, our findings also indicate that thermal physiology probably affects the energy flux
in the food web, which we think is a step forward in understanding the differences in the responses of ectotherms and
endotherms to the environment.
Effects of temperature and net primary productivity
on all arthropods and all large mammals
Our results are in accordance with studies that have shown a
decline in arthropod species richness with increasing elevation due to temperature decrease (e.g., Peters et al. 2016)
and those that have reported a positive association of NPP
with large mammals in an African elevational gradient (e.g.,
Gebert et al. 2019). NPP was also positively associated with
arthropods, indicating that resource availability is important
for increasing arthropod species richness, possibly by reducing species competition and local extinction (Evans et al.
2005). However, as arthropod diversity was more strongly
associated with temperature, we suggest that temperature
limits arthropod species richness through physiological
constraints and that it seems to impose greater constraints
on species richness of arthropods than NPP. As arthropods
are dependent on external temperatures, low temperatures
decrease their foraging activity (Cerdá et al. 1998; Lasmar
et al. 2021), which influences their population sizes at high
elevations, resulting in a decrease in arthropod species richness (Sanders et al. 2007; Lasmar et al. 2020). On the other
hand, endotherms are able to harvest food resources even in
cold temperatures at the expense of a high amount of food
resources to maintain their vital functions (Buckley et al.
2012). Therefore, resource availability is a key driver that
limits the number of coexisting species of large mammals
in the tropics (Gebert et al. 2019). Considering our limited
13
sample size (arthropods n = 8; large mammals n = 14), we
should be cautious about generalizing our results, and more
studies in other tropical regions could be helpful in consolidating our findings and inferences. This is especially
true when considering that we evaluated only one group of
endotherms (i.e., only large mammals), and, therefore, the
causes of the stronger positive association of NPP on large
mammals could be due to both their endothermy and to other
specific traits of large mammals. However, the stronger association between NPP concerning temperature on other endotherms was also supported by Buckley et al. (2012) as well
as in other case studies (e.g., birds, Hawkins et al. 2003b;
Acharya et al. 2011; bats and rodents, Owen 1990; McCain
et al. 2018). In this sense, differences in thermal physiology
between ectotherms and endotherms may be responsible for
the way arthropods and large mammals interact with or are
constrained by the environmental temperature and NPP.
Effects of temperature and net primary productivity
on multitrophic guilds of arthropods and large
mammals
For arthropod trophic guilds, temperature is strongly associated with the diversity of lower trophic levels, while NPP
is strongly associated with the diversity of higher trophic
levels. In addition to the fact that accelerated metabolic rates
at high temperatures increase arthropod foraging activity
(Cerdá et al. 1998; Lasmar et al. 2021) and, consequently,
energy use from the ecosystem (Twomey et al. 2012), primary consumers are extremely sensitive to changes in plant
phenology, which are temperature dependent (Thackeray
et al. 2016). However, the greater association of arthropod
herbivores with temperature in this study could also occur
because most of the species are represented by fungus-garden ants, which are extremely influenced by temperature
(Bollazzi and Roces 2002). Thus, because the species richness of arthropod herbivores is underrepresented in our data,
the generalization of our results to other important arthropod
herbivores (e.g., Hemiptera, Lepidoptera, Orthoptera) needs
to be tested. Nevertheless, we consider our representation of
arthropod omnivores and predators to be highly satisfactory
and enough to confirm our prediction that the association
of NPP with species richness increases across arthropod
trophic levels. This is in accordance with previous studies
that found that the abundance and species richness of arthropod predators are more constrained by resource availability
than primary consumers (e.g., Kaspari 2001; Xu et al. 2018).
Abundance across trophic levels accumulates along the NPP
gradient (Oksanen et al. 1981; Kaspari 2001), which makes
predators more sensitive to the energy input in the ecosystem
than organisms at lower trophic levels (Turney and Buddle
2016; Brose et al. 2017). The higher the trophic level, the
more energy is needed to maintain viable population sizes
Oecologia
and then high species richness (Turney and Buddle 2016).
Our results are partially following Mayr et al. (2020), who
found that Hymenoptera at the low trophic level are mainly
influenced by temperature while higher trophic levels are
influenced by both temperature and resource availability.
Our limited sampling size may have caused the lack of statistical support for the influence of temperature on the species
richness of predators in this study, even though it is clear
that the association of species richness with NPP increases
from lower to higher trophic levels. Therefore, even though
temperature is one of the major constraints on arthropod
diversity (Gillooly et al. 2001; Lessard et al. 2011), we suggest that the sensitivity to energy input in the ecosystem
increases across arthropod trophic levels, which positively
affects species richness. However, this seems not to be the
rule among ectotherms and endotherms since the trophic
guilds of large mammals followed distinct patterns.
The species richness of large-mammal trophic guilds
presented a complex pattern concerning temperature and
NPP. NPP was only associated with the species richness
of large-mammal herbivores. As endotherms may use the
energy from the ecosystem more independently of climate
than ectotherms (Buckley et al. 2012), the availability of
resources could constrain primary consumers’ diversity of
large mammals (Beck 2006). However, this effect of the
energy input in the ecosystem on species richness seems
not to be strong when trophic levels increase. This is because
although NPP influenced large-mammal herbivores, the
effect of temperature seemed to be stronger and the diversity of large-mammal omnivores was only associated with
temperature, while no variables were associated with largemammal predators. NPP may present stronger effects on
communities with more phylogenetic diversity than on communities with narrowly defined clades, because the latter
typically use a smaller fraction of NPP in comparison to a
phylogenetically broader community (Peters et al. 2020).
Considering that we had a narrower number of clades distributed among large-mammal trophic guilds, the correlation
of NPP with their true food resources might be less strong.
This may explain why we found an effect of NPP on overall
large mammals (i.e., a more broad phylogenetic community),
a weak effect of NPP for large-mammal herbivores and no
effect of NPP for other trophic groups. On the other hand,
temperature is known to also increase trophic interactions
and resource assimilation (Fine 2015; Peters et al. 2020).
For large-mammal omnivores, we should consider that these
animals feed on plant resources, such as fruits and seeds, and
on other animals, including many arthropods (e.g., PinedaMunoz and Alroy 2014; Galetti et al. 2015), all extremely
influenced by temperature (Peters et al. 2016). Therefore,
we suggest resource availability as an important driver of
omnivore species richness. However, this may be not simply dependent on the energy input provided by producers
but also on other factors that influence the food resources
obtained from both producers and other prey at lower trophic
levels. In conclusion, it is possible that trophic interactions,
which are positively influenced by temperature (Fine 2015)
could outweigh the effects of environmental NPP on species richness at low and intermediate trophic levels (e.g.,
large-mammal herbivores and omnivores). The dependence of trophic interactions could also explain the lack of
association of large-mammal predators with environmental
temperature and NPP.
Due to the lack of a relationship between large-mammal
predators and our variables, we have two not mutually exclusive hypotheses. First, large-mammal predators occur naturally at low densities in the environment and forage for large
distances (Smith et al. 2017); in addition, human-related
actions reduce or dislocate individuals from their original
areas (Rogala et al. 2011). Despite our study area being a
protected area, it historically suffered from deforestation,
poaching and criminal fires. Therefore, human-related variables that influence carnivores may not have been considered
in this study, and we may need biogeographical scales to
better understand patterns of large-mammal predators. Second, the prey of predator animals could also obscure their
association with temperature and NPP. This is because largemammal predators are strongly influenced by their prey
abundance (Smith et al. 2017), in which at least their species
richness in this study was associated with both temperature
and NPP. In addition, Sus scrofa is known to influence both
the spatial and temporal patterns of native mammals (Hegel
et al. 2019; Galetti et al. 2015), such as the prey of predators,
and may also be a new source of prey (Hegel et al. 2018).
Both could affect the presence of large mammals along the
elevational gradient.
Conclusion
We report here that the species richness of ectothermic
arthropods is more associated with temperature, while that
of endothermic large mammals is associated with productivity (NPP) along a tropical elevational gradient. Most of
the comparisons reporting that the relative importance of
temperature and NPP varies between ectotherms and endotherms were assessed between vertebrates (reviewed in
Buckley et al. 2012). Here, we provide evidence that those
differences in the role of NPP and temperature in driving
species richness also apply when comparing ectothermic
invertebrates with endothermic vertebrates. Additionally,
we found no consistency regarding the role of temperature
and NPP on the species richness structure across multitrophic guilds of ectotherms and endotherms. The association of species richness with NPP increased across trophic
levels for arthropods; however, although NPP is important
13
Oecologia
for large-mammal herbivores, its influence does not seem
to accumulate across trophic levels. Large-mammal trophic
guilds seem to be more indirectly dependent on temperature
via other trophic interactions than simply on environmental
NPP. Based on that, it is possible that besides global temperature change might strongly affect arthropods, it might
also affect large-mammal trophic guilds in the tropics. Here,
we also suggest that thermal physiological differences might
also interfere with energy use and flux in food webs, which
differently reflect species richness across trophic levels
between ectotherms and endotherms. Several studies propose several mechanisms explaining how species abundance,
biomass and richness are distributed across trophic levels in
the food web (e.g., Oksanen et al. 1981; Turney and Buddle
2016; Wang and Brose 2018). Therefore, it would be useful
for future studies to consider thermal physiological differences in animals when assessing the distinct facets of trophic
guilds, such as the pyramid shape of abundance, biomass,
and species richness.
Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s00442-021-05011-9.
Acknowledgements This work was funded by Fundação de Amparo
à Pesquisa de Minas Gerais (FAPEMIG, Grant PPM 00243/14), and
Tropical Conservation Act/Fundo Brasileiro da Biodiversidade (TFCA/
FUNBIO). We thank Marcell K. Peters and an anonymous referee for
their helpful comments. We are thankful to the staff of National Itatiaia Park, especially to Leo Nascimento and Marcelo Motta, who permitted the sampling in the park area. We are also indebted to Maria
Regina de Souza, Tobias R. Silva, Luiza Santiago, Edson Guilherme
de Souza, Daniel Q. Domingos, Ernesto O. Canedo–Júnior, Graziele
Santiago, Luana Zurlo, Fernando H. Puertas, Thamíris C. K. de Abreu,
and José Cristiano for their help with logistics and fieldwork. Thanks
to Mariana Rabelo, Ícaro Carvalho, and Felipe Lopes for helping with
laboratory work. We are grateful to the staff at the Laboratório de
Sistemática de Formigas da Universidade Federal do Paraná for confirming ant identification, especially to Alexandre Ferreira. CJL and
CAN received a postdoctoral fellowship from PNPD/CAPES. ACMQ
received a postdoctoral fellowship from CEMIG—Companhia Energética de Minas Gerais S.A. (P&D 611—Descomissionamento da PCH
Pandeiros: Uma experiência inédita na América do Sul). RMF and
ADB were supported by the Conselho Nacional de Desenvolvimento
Científico e Tecnológico (CNPq grant 302462/2016–3 and CNPq grant
303903/2019-8). This study was part of Chaim J. Lasmar’s MSc. thesis
at the Universidade Federal de Lavras that was supported by Coordenação de Aperfeiçoamento Pessoal (CAPES, Finance code: 001).
Author contribution statement CJL CR, and CRR conceived and
designed the experiments. CJL, CR, MMGI, GPA, and GBN conducted
the field and lab work; CJL, CR, RMF, LNZ, LV, and ADB identified
the collected species. CJL, ACMQ, and CAN analysed the data. CJL
wrote the manuscript; all other authors provided editorial advice.
Funding This study was funded by Fundação de Amparo à Pesquisa de
Minas Gerais (FAPEMIG, Grant PPM 00243/14), Tropical Forest Conservation Act/Fundo Brasileiro Para a Diversidade (TFCA)/FUNBIO).
Data availability All data produced from this study are provided in the
Electronic Supplementary Material of this manuscript.
13
Declarations
Conflict of interest The authors declare that they have no conflict of
interest.
Ethical approval All applicable institutional and/or national guidelines
for the care and use of animals were followed. All animal experiments
were approved by Instituto Chico Mendes de Conservação da Biodiversidade (ICMBio), reference number 46564-1.
References
Acharya BK, Sanders NJ, Vijayan L, Chettri B (2011) Elevational gradients in bird diversity in the Eastern Himalaya: an evaluation
of distribution patterns and their underlying mechanisms. PLoS
ONE 6(12):e29097. https://doi.org/10.1371/journal.pone.0029097
Allen AP, Brown JH, Gillooly JF (2002) Global biodiversity, biochemical kinetics, and the energetic-equivalence rule. Science
297:1545–1548. https://doi.org/10.1126/science.1072380
Baccaro FB, Feitosa RM, Fernandez F, Fernandes IO, Izzo TJ, Souza
JLP et al (2015) Guia para gêneros de formigas do Brasil. Editora
INPA Manaus. https://doi.org/10.5281/zenodo.32912
Barry RG (2008) Mountain weather and climate, 3rd edn. Cambridge
University Press, Cambridge
Basset Y, Cizek L, Cuénoud P, Didham RK, Guilhaumon F, Missa
O, Tishechkin AK et al (2012) Arthropod diversity in a tropical
forest. Science 338:1481–1484. https://doi.org/10.1126/science.
1226727
Beck H (2006) A review of peccary–palm interactions and their ecological ramifications across the Neotropics. J Mammal 87:519–530
Binkenstein J, Klein AM, Assmann T, Buscot F, Erfmeier A, Ma K,
Pietsch KA, Schmidt K, Scholten T, Wubet T, Bruelheide H,
Schuldt A, Staab M (2018) Multi-trophic guilds respond differently to changing elevation in a subtropical forest. Ecography
41:1013–1023. https://doi.org/10.1111/ecog.03086
Birkhofer K, Wolters V (2012) The global relationship between climate, net primary production and the diet of spiders. Glob Ecol
Biogeogr 21:100–108. https://doi.org/10.1111/j.1466-8238.2011.
00654.x
Bollazzi M, Roces F (2002) Thermal preference for fungus culturing
and brood location by workers of the thatching grass-cutting ant
Acromyrmex heyeri. Insectes Soc 49:153–157. https://doi.org/10.
1007/s00040-002-8295-x
Brescovit AD, Bonaldo AB, Bertani R, Rheims CA (2002) Araneae.
In: Adis J (ed) Amazonian Arachnida and Myriapoda. Pensoft
Publishers, Sofia-Moscow, pp 303–343
Brose U, Blanchard JL, Eklöf A, Galiana N, Hartvig M, Hirt MR,
Kalinkat G et al (2017) Predicting the consequences of species
loss using size-structured biodiversity approaches. Biol Rev
92:684–697. https://doi.org/10.1111/brv.12250
Brown JH (2014) Why are there so many species in the tropics? J Biogeogr 41:8–22. https://doi.org/10.1111/jbi.12228
Buckley LB, Hurlbert AH, Jetz W (2012) Broad-scale ecological implications of ectothermy and endothermy in changing environments.
Glob Ecol Biogeogr 21:873–885. https://doi.org/10.1111/j.14668238.2011.00737.x
Burnham KP, Anderson DR (2002) Model selection and inference:
a practical information—theoretic approach, 2nd edn. Springer,
New York
Canty A, Ripley B (2012) Boot, Bootstrap R (S–Plus) Functions. R
Package version1.3–4. https://CRAN.R-project.org/package=boot.
Accessed 1 Feb 2020
Oecologia
Cerdá X, Retana J, Cros S (1998) Critical thermal limits in Mediterranean ant species: trade-off between mortality risk and foraging performance. Funct Ecol 12:45–55. https://doi.org/10.1046/j.
1365-2435.1998.00160.x
Colwell RK, Coddington JA (1994) Estimating terrestrial biodiversity through extrapolation. Philos Trans R Soc Lond B Biol Sci
345:101–118. https://doi.org/10.1098/rstb.1994.0091
Crawley MJ (2002) Statistical computing, an introduction to data analysis using S-Plus, 1st edn. Wiley, London
Engelbrecht I (2013) Pitfall trapping for surveying trapdoor spiders: the
importance of timing, conditions and effort. J Arachnol 41:133–
142. https://doi.org/10.1636/P12-57.1
Evans KL, Warren PH, Gaston KJ (2005) Species-energy relationships
at the macroecological scale: a review of the mechanisms. Biol
Rev 80:1–25. https://doi.org/10.1017/S1464793104006517
Fine PV (2015) Ecological and evolutionary drivers of geographic variation in species diversity. Annu Rev Ecol Evol Syst 46:369–392.
https://doi.org/10.1146/annurev-ecolsys-112414-054102
Fox J, Weisberg S (2019) An R companion to applied regression, 3rd
edn. Sage, Thousand Oaks. http://socialsciences.mcmaster.ca/jfox/
Books/Companion/. Accessed 01 Dec 2020
Galetti M, Camargo H, Siqueira T, Keuroghlian A, Donatti CI, Jorge
MLS, Ribeiro MC (2015) Diet overlap and foraging activity
between feral pigs and native peccaries in the Pantanal. PLoS
ONE 10:1–10. https://doi.org/10.1371/journal.pone.0141459
Gebert F, Njovu HK, Treydte AC, Steffan-Dewenter I, Peters MK
(2019) Primary productivity and habitat protection predict elevational species richness and community biomass of large mammals
on Mt. Kilimanjaro. J Anim Ecol 88:1860–1872. https://doi.org/
10.1111/1365-2656.13074
Gillooly JF, Brown JH, West GB, Savage VM, Charnov EL (2001)
Effects of size and temperature on metabolic rate. Science
293:2248–2251. https://doi.org/10.1126/science.1061967
Gotelli NJ, Colwell RK (2011) Estimating species richness. In: Magurran AE, McGill BJ (eds) Frontiers in measuring biodiversity.
Oxford University Press, New York, pp 39–54
Gotelli NJ, Ellison AM, Dunn RR, Sanders NJ (2011) Counting ants
(Hymenoptera: Formicidae): biodiversity sampling and statistical
analysis for myrmecologists. Myrmecol News 15:13–19
Haddad NM, Crutsinger GM, Gross K, Haarstad J, Knops JM, Tilman D (2009) Plant species loss decreases arthropod diversity
and shifts trophic structure. Ecol Lett 12:1029–1039. https://doi.
org/10.1111/j.1461-0248.2009.01356.x
Hawkins BA, Field R, Cornell HV, Currie DJ, Guégan JF, Kaufman
DM et al (2003a) Energy, water, and broad-scale geographic patterns of species richness. Ecology 84:3105–3117. https://doi.org/
10.1890/03-8006
Hawkins BA, Porter EE, Felizola Diniz-Filho JA (2003b) Productivity
and history as predictors of the latitudinal diversity gradient of
terrestrial birds. Ecology 84:1608–1623. https://doi.org/10.1890/
0012-9658(2003)084[1608:PAHAPO]2.0.CO;2
Hegel CGZ, Marini MÂ (2018) Large felids as predators of wild boars
in the Atlantic forest: reconciling Verdade et al. and Rosa et al.
Anim Conserv 21:363–364. https://doi.org/10.1111/acv.12402
Hegel CGZ, Santos LR, Marinho JR, Marini MÂ (2019) Is the wild pig
the real “big bad wolf”? Negative effects of wild pig on Atlantic
forest mammals. Biol Invas 21:3561–3574. https:// doi. org/ 10.
1007/s10530-019-02068-9
Hillebrand H (2004) On the generality of the latitudinal diversity gradient. Am Nat 163:192–211. https://doi.org/10.1086/381004
Janzen DH (1967) Why mountain passes are higher in the tropics. Am
Nat 101:233–249. https://doi.org/10.1086/282487
Karger DN, Conrad O, Böhner J, Kawohl T, Kreft H, Soria-Auza
RW, Zimmermann NE, Linder P, Kessler M (2017) Climatologies at high resolution for the Earth land surface areas. Sci Data
4:170122. https://doi.org/10.1038/sdata.2017.122
Kaspari M (2001) Taxonomic level, trophic biology, and the regulation
of local abundance. Glob Ecol Biogeogr 10:229–244. https://doi.
org/10.1046/j.1466-822X.2001.00214.x
Kitching RL, Dahlsjö CAL, Eggleton P (2020) Invertebrates and
the complexity of tropical ecosystems. Biotropica 52:207–214.
https://doi.org/10.1111/btp.12768
Körner C (2007) The use of ‘altitude’ in ecological research. Trends
Ecol Evol 22:569–574. https://doi.org/10.1016/j.tree.2007.09.006
Kury AB, Giupponi APL, Mendes AC (2018) Immolation of Museu
Nacional, Rio de Janeiro—unforgettable fire and irreplaceable
loss. J Arachnol 46:556–558
Lasmar CJ, Queiroz ACM, Rabello AM, Feitosa RM, Canedo-Júnior
EO, Schmidt FA, Cuissi RG, Ribas CR (2017) Testing the effect of
pitfall-trap installation on ant sampling. Insectes Soc 64:445–451.
https://doi.org/10.1007/s00040-017-0558-7
Lasmar CJ, Ribas CR, Louzada J, Queiroz ACM, Feitosa RM, Imata
MM, Alves GP, Nascimento GB, Neves FS, Domingos DQ (2020)
Disentangling elevational and vegetational effects on ant diversity
patterns. Acta Oecol 102:103489. https://doi.org/10.1016/j.actao.
2019.103489
Lasmar CJ, Bishop TR, Parr CL, Queiroz ACM, Schmidt FA, Ribas
CR (2021) Geographical variation in ant foraging activity and
resource use is driven by climate and net primary productivity. J
Biogeogr 48:1448–1459. https://doi.org/10.1111/JBI.14089
Lessard JP, Sackett TE, Reynolds WN, Fowler DA, Sanders NJ (2011)
Determinants of the detrital arthropod community structure: the
effects of temperature and resources along an environmental gradient. Oikos 120:333–343. https://doi.org/10.1111/j.1600-0706.
2010.18772.x
MacArthur RH (1972) Geographical ecology: patterns in the distribution of species. Princeton University Press, Princeton
Mayr AV, Peters MK, Eardley CD, Renner ME, Röder J, SteffanDewenter I (2020) Climate and food resources shape species
richness and trophic interactions of cavity-nesting Hymenoptera.
J Biogeogr 47:854–865. https://doi.org/10.1111/jbi.13753
Mazerolle MJ (2020) AICcmodavg: model selection and multimodel
inference based on (Q)AIC(c). R package version 1.30. http://
CRAN.R-project.org/package=AICcmodavg. Accessed 01 Dec
2020
McCain CM, King SRB, Szewczyk T, Beck J (2018) Small mammal
species richness is directly linked to regional productivity, but
decoupled from food resources, abundance, or habitat complexity. J Biogeogr 45:2533–2545. https://doi.org/10.1111/jbi.13432
Moorthi SD, Schmitt JA, Ryabov A, Tsakalakis I, Blasius B, Prelle
L, Tiedemann M, Hodapp D (2016) Unifying ecological stoichiometry and metabolic theory to predict production and trophic
transfer in a marine planktonic food web. Philos Trans R Soc B
371:20150270. https://doi.org/10.1098/rstb.2015.0270
Oksanen L, Fretwell SD, Arruda J, Niemela P (1981) Exploitation ecosystems in gradients of primary productivity. Am Nat 118:240–
261. https://doi.org/10.1086/283817
Oksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, McGlinn
D, Minchin P, O’Hara R, Simpson G, Solymos P, Stevens M,
Szoecs E, Wagner H (2020) vegan: Community Ecology Package.
R package version 2.5-7. https://CRAN.R-project.org/package=
vegan. Accessed 01 Dec 2020
Owen JG (1990) Patterns of mammalian species richness in relation to
temperature, productivity, and variance in elevation. J Mammal
71(1):1–13
Peters MK, Hemp A, Appelhans T, Behler C, Classen A, Detsch F
et al (2016) Predictors of elevational biodiversity gradients change
from single taxa to the multi-taxa community level. Nat Commun
7:13736. https://doi.org/10.1038/ncomms13736
Peters MK, Classen A, Müller J, Steffan-Dewenter I (2020) Increasing the phylogenetic coverage for understanding broad-scale
13
Oecologia
diversity gradients. Oecologia 192:629–639. https://doi.org/10.
1007/s00442-020-04615-x
Pineda-Munoz S, Alroy J (2014) Dietary characterization of terrestrial
mammals. Proc R Soc B Biol Sci 281:20141173. https://doi.org/
10.1098/rspb.2014.1173
Pinto-da-Rocha R, Giribet G (2007) Harvestmen: the biology of the
Opiliones. Harvard University Press, Cambridge
R Development Core Team (2020) R: a language and environment
for statistical computing. R Foundation for Statistical Computing,
Vienna. http://www.rproject.org. Accessed 01 May 2021
Rahbek C, Borregaard MK, Colwell RK, Dalsgaard B, Holt BG, Morueta-Holme N, Nogues-Bravo D, Whittaker RJ, Fjeldså J (2019)
Humboldt’s enigma: what causes global patterns of mountain
biodiversity? Science 365:1108–1113. https://doi.org/10.1126/
science.aax0149
Rogala JK, Hebblewhite M, Whittington J, White CA, Coleshill J,
Musiani M (2011) Human activity differentially redistributes
large mammals in the Canadian Rockies National Parks. Ecol
Soc 16(3):16
Sabu TK, Shiju RT, Vinod KV, Nithya S (2011) A comparison of the
pitfall trap, Winkler extractor and Berlese funnel for sampling
ground-dwelling arthropods in tropical montane cloud forests. J
Insect Sci 11:2011–2028. https://doi.org/10.1673/031.011.0128
Safford HD (1999) Brazilian Páramos I. An introduction to the physical
environment and vegetation of the campos de altitude. J Biogeogr
26:693–712. https://doi.org/10.1046/j.1365-2699.1999.00313.x
Sanders NJ, Lessard JP, Fitzpatrick MC, Dunn RR (2007) Temperature, but not productivity or geometry, predicts elevational diversity gradients in ants across spatial grains. Glob Ecol Biogeogr
16:640–649. https://doi.org/10.1111/j.1466-8238.2007.00316.x
Smith JE, Lehmann KD, Montgomery TM, Strauss ED, Holekamp KE
(2017) Insights from long-term field studies of mammalian carnivores. J Mammal 98:631–641. https://doi.org/10.1093/jmamm
al/gyw194
Srbek-Araujo AC, Chiarello AG (2013) Influence of camera-trap sampling design on mammal species capture rates and community
structures in southeastern Brazil. Biota Neotrop 13:51–62. https://
doi.org/10.1590/S1676-06032013000200005
Srbek-Araujo AC, Kierulff MCM (2016) Mamíferos de médio e grande
porte das florestas de Tabuleiro do norte do Espírito Santo: grupos
funcionais e principais ameaças. In: Rolim SG, Menezes LFT,
Srbek-Araujo AC (eds) Floresta Atlântica de Tabuleiro: diversidade e endemismos na Reserva Natural da Vale. Rupestre, Belo
Horizonte, pp 469–479
Straneo SL, Ball GB (1989) Synopsis of the genera and subgenera of
the Tribe Peleciini, and revision of the neotropical and oriental
species (Coleoptera: Carabidae). Insecta Mundi 3(2):73–178
Sundqvist MK, Sanders NJ, Wardle DA (2013) Community and ecosystem responses to elevational gradients: processes, mechanisms,
and insights for global change. Annu Rev Ecol Evol Syst 44:261–
280. https://doi.org/10.1146/annurev-ecolsys-110512-135750
Terborgh JW (2015) Toward a trophic theory of species diversity. Proc
Natl Acad Sci USA 112:11415–11422. https://doi.org/10.1073/
pnas.1501070112
Thackeray SJ, Henrys PA, Hemming D, Bell JR, Botham MS, Burthe
S et al (2016) Phenological sensitivity to climate across taxa and
trophic levels. Nature 535:241–245. https://doi.org/10.1038/natur
e18608
Turney S, Buddle CM (2016) Pyramids of species richness: the determinants and distribution of species diversity across trophic levels.
Oikos 125:1224–1232. https://doi.org/10.1111/oik.03404
Twomey M, Brodte E, Jacob U, Brose U, Crowe TP, Emmerson MC
(2012) Idiosyncratic species effects confound size-based predictions of responses to climate change. Philos Trans R Soc B
367:2971–2978. https://doi.org/10.1098/rstb.2012.0244
Voigt W, Perner J, Davis AJ, Eggers T, Schumacher J, Bährmann R,
Fabian B, Heinrich W, Köhler G, Lichter D, Marstaller R, Sander
FW (2003) Trophic levels are differentially sensitive to climate.
Ecology 84:2444–2453. https://doi.org/10.1890/02-0266
Voigt W, Perner J, Hefin Jones T (2007) Using functional groups to
investigate community response to environmental changes: two
grassland case studies. Glob Change Biol 13:1710–1721. https://
doi.org/10.1111/j.1365-2486.2007.01398.x
Wang S, Brose U (2018) Biodiversity and ecosystem functioning in
food webs: the vertical diversity hypothesis. Ecol Lett 21:9–20.
https://doi.org/10.1111/ele.12865
Welti EAR, Kuczynski L, Marske KA, Sanders NJ, de Beurs KM,
Kaspari M (2020) Salty, mild, and low plant biomass grasslands
increase top-heaviness of invertebrate trophic pyramids. Glob
Ecol Biogeogr 29:1474–1485. https://doi.org/10.1111/geb.13119
World Spider Catalog (2021) World Spider Catalog. Version 22.0.
Natural History Museum Bern. http://wsc.nmbe.ch. Accessed 20
May 2021. https://doi.org/10.24436/2
Wright DH (1983) Species-energy theory: an extension of species-area
theory. Oikos 558(41):496–506. https://doi.org/10.2307/3544109
Xu G, Zhang S, Zhang Y, Ma K (2018) Environmental correlates
underlying elevational richness, abundance, and biomass patterns
of multi-feeding guilds in litter invertebrates across the treeline.
Sci Total Environ 633:529–538. https://doi.org/10.1016/j.scito
tenv.2018.03.212
Authors and Affiliations
Chaim J. Lasmar1 · Clarissa Rosa2 · Antônio C. M. Queiroz1 · Cássio A. Nunes3 · Mayara M. G. Imata1 ·
Guilherme P. Alves1 · Gabriela B. Nascimento1 · Ludson N. Ázara4 · Letícia Vieira5 · Júlio Louzada6 ·
Rodrigo M. Feitosa7 · Antonio D. Brescovit8 · Marcelo Passamani9 · Carla R. Ribas1
1
2
Programa de Pós-Graduação em Ecologia Aplicada,
Departamento de Ecologia e Conservação, Instituto de
Ciências Naturais, Laboratório de Ecologia de Formigas,
Universidade Federal de Lavras, PO Box 3037, Lavras,
MG 37200-900, Brazil
Coordenação de Biodiversidade, Instituto Nacional de
Pesquisas da Amazônia, Manaus, Amazonas 69067-375,
Brazil
13
3
Programa de Pós-Graduação em Ecologia Aplicada,
Departamento de Ecologia e Conservação, Universidade
Federal de Lavras, Lavras, Brazil
4
Laboratório de Aracnologia, Departamento de Invertebrados,
Museu Nacional, Universidade Federal do Rio de Janeiro,
Quinta da Boa Vista, São Cristóvão, 20, Rio de Janeiro,
RJ 940-040, Brazil
Oecologia
5
Laboratório de Ecologia Florestal, Departamento de
Ciências Florestais, Universidade Federal de Lavras, Lavras,
MG 37200-000, Brazil
6
Laboratório de Ecologia de Invertebrados, Departamento
de Ecologia e Conservação, Instituto de Ciências Naturais,
Universidade Federal de Lavras, Lavras, MG 37200-000,
Brazil
7
Departamento de Zoologia, Universidade Federal do Paraná,
CP 19020, Curitiba, PR 81531-980, Brazil
8
Laboratório de Coleções Zoológicas, Instituto Butantan,
São Paulo, SP 05503-900, Brazil
9
Laboratório de Ecologia e Conservação de Mamíferos,
Departamento de Ecologia e Conservação, Instituto de
Ciências Naturais, Universidade Federal de Lavras, Lavras,
MG 37200-000, Brazil
13