| |

Received: 25 March 2019    Revised: 26 October 2020    Accepted: 9 November 2020

DOI: 10.1111/jbi.14043

RESEARCH ARTICLE

Unveiling the drivers of local dung beetle species richness in

the Neotropics

Marcelo Bruno Pessôa1  | Fernanda Alves-Martins2,3  | Paulo De Marco Júnior1  |

Joaquín Hortal1,2

1
Departamento de Ecologia, Instituto de
Ciências Biológicas, Universidade Federal de Abstract
Goiás, Goiânia, Brazil Aim: Nearly 40 different hypotheses have been put forward to explain the latitudinal
2
Department of Biogeography and Global
diversity gradient, implying that geographical variations of biodiversity may be the
Change, Museo Nacional de Ciencias
Naturales (MNCN-CSIC), Madrid, Spain result of a complex array of factors affecting organisms in different ways. Our main
3
Global Change Ecology & Evolution goal was to identify the most important drivers of local dung beetle species richness
(GLOCEE) Group, Departamento de Ciencias
de la Vida, Universidad de Alcalá, Madrid,
in the Neotropics.
Spain Location: Neotropics.

Correspondence
Taxon: Dung Beetles (Coleoptera: Scarabaeinae).
Marcelo Bruno Pessôa, Departamento de Methods: We used a multi-model approach to identify which potential drivers cor-
Ecologia, Instituto de Ciências Biológicas,
Universidade Federal de Goiás, Avenida
relate better with the variations in local dung beetle species richness. We surveyed
Esperança s/n, Campus Samambaia, ICB 5, published literature on dung beetle communities to extract information on species
CEP 74690-900, Goiânia, Goiás, Brazil.
Email: mbpessoa@yahoo.com.br
richness, abundance, type of bait, type of habitat and sampling effort (as hours/pit-
fall) for different localities, discarding sites with low sampling effort. We used en-
Funding information
Conselho Nacional de Desenvolvimento
vironmental variables to account for six possible explanations of species richness
Científico e Tecnológico, Grant/Award gradients: productivity, water–energy, ambient energy, habitat heterogeneity, re-
Number: 308694/2015-5 and 314523/2014-
6; MCTIC/CNPq/FAPEG, Grant/Award
source heterogeneity and seasonality, as well as spatial data to account for other
Number: 465610/2014-5; Coordenação geographically structured phenomena. We used mixed models—with abundance,
de Aperfeiçoamento de Pessoal de
Nível Superior, Grant/Award Number:
ecoregion and bait type as random factors—to select the best model among the vari-
120147/2016-01, 88881.135489/2016-01 ables accounting for each explanation. Finally, we used structural equation models to
and PROEX 0487; SCENIC–‘Scaling the
effects of niche and interaction dynamics on
assess which explanations are associated with variations in dung beetle diversity and
the ecological and evolutionary outcomes how they interact.
of coexistence’, Grant/Award Number:
PID2019-106840GB-C21
Results: Resource heterogeneity was the best single correlate of dung beetle rich-
ness. However, the best multiple model comprises three different explanations: pro-
Handling Editor: Lyn Cook
ductivity, resource heterogeneity and other spatially structured factors. Structural
equation models show that abundance is directly (positively) associated with rich-
ness, followed by primary productivity and soil variables (a proxy for environmental
heterogeneity), together with mammal richness, (a proxy for resource heterogeneity).
Main conclusions: Several explanations need to be considered to account for
Scarabaeinae local richness patterns. The diversity of dung beetle communities cor-
relates with the interaction of water–energy dynamics and heterogeneity in both re-
sources and habitats. However, while heterogeneity variables are directly associated

Journal of Biogeography. 2020;00:1–11. wileyonlinelibrary.com/journal/jbi © 2020 John Wiley & Sons Ltd     1 |
 |
2       PESSÔA et al.

with richness, energy relates with it through abundance, and water through resource
diversity.

KEYWORDS

habitat heterogeneity, Neotropics, resource availability, Scarabaeinae, species richness,

water–energy dynamics

1 |  I NTRO D U C TI O N both ecto- and endotherms feed and reproduce better, since energy
consumption for maintaining body heat and/or movement is lower,
Geographical gradients of species richness were perhaps the first so more resources can be devoted to reproduction, thus increasing
patterns investigated in ecology and biogeography (Hawkins, 2001). reproductive rates and with it population size and, lastly, species
Why and wherefore some places host more species than others has numbers (Turner, 2004).
long been subject to debate since Alexander von Humboldt (1850) Heterogeneity explanations also include three main groups of
noticed that there are more species in the tropics than in temperate drivers, namely environmental and resource variability, and sea-
regions and tried to explain it with recourse to climate and species sonality. According to niche theory, the higher the environmental
resistance to freezing. Strikingly, almost 40 different hypotheses variability (in terms of habitats or other environmental conditions)
have been put forward to explain the latitudinal diversity gradient within an area, the greater the diversity of species with different
(LDG)—based on climate, habitat, physiological responses and many niches that can coexist, due to reduced effects of competition
other factors (Hawkins, 2008; Hawkins et al., 2003; Pianka, 1966). (Tilman,  1985; Stein & Kreft, 2015; Tews et al., 2004). This argu-
Among these, three major groups of explanations stand out: spe- ment stands for both habitat heterogeneity (Hortal et al., 2009) and
cies–energy hypotheses which assume that richness varies accord- resource heterogeneity (Ricklefs, 1977). In the case of seasonality,
ing to the differences in the amount of energy available or produced however, strong seasonal changes may result in short growing and
(Hutchinson,  1959); heterogeneity hypotheses in which richness reproductive seasons, hence reducing survival and reproductive
gradients are associated with the variance in climate, resources and/ rates, so regions suffering higher seasonality may present poorer
or habitat (Lack, 1969); and historical and evolutionary hypotheses species communities (Gouveia et  al.,  2013; Sanders,  1968). Finally,
in which the LDG would be the outcome of historical and/or evolu- historical events and long-term evolutionary processes also affect
tionary processes (such as climatic stability, diversification rate, evo- species richness gradients, as increased extinction rates or the lack
lutionary origin or temporal changes in extinction rates; Mittelbach of adaptations to face novel environmental conditions may constrain
et al., 2007). species richness in certain regions, while long-term temporal stabil-
Three main groups of drivers have been traditionally ascribed ity and high geological dynamism foster net diversification rates in
to species–energy hypotheses, accounting for productivity, water– other territories (Rangel et al., 2018; Wiens & Donoghue, 2004).
energy dynamics and ambient energy. Productivity may affect Such variety of explanations indicates that geographical diver-
species richness through the production rate of the resources of sity gradients may be the result of a complex array of factors affect-
interest for each particular species group; the higher the produc- ing organisms in different ways. Thus, explaining ecological patterns
tion rate, the greater the number of individuals that an ecosystem requires considering multiple hypotheses and using different ex-
can support, and with it the larger the total number of species it planatory variables (Lawton, 1999). Problems arise when single ex-
can reach in equilibrium (the so-called more-individuals hypoth- planations are interpreted as the sole drivers of diversity gradients
esis; Wright,  1983; Storch et  al.,  2018). Many variables have been and are treated as rivals rather than complementary. Therefore,
used to account for productivity, but for plants it is common to use seeking general explanations for patterns of biodiversity requires
solar radiation or actual evapotranspiration (AET; see Whittaker & adopting an analytical framework that allows incorporating the ef-
Field, 2000), whereas for animals a direct measure of the quantity fects of different potential drivers.
of the resource produced, such as net primary productivity, may be Dung beetles (Coleoptera, Scarabaeinae) are known to be af-
used. In contrast, the explanations based on water–energy dynamics fected by multiple factors across scales. They respond rapidly to
attribute the origin of diversity gradients to the geographical varia- environmental changes, are easy to survey, and play important eco-
tions in the availability of energy and water (generally indicated by system functions (Gardner et al., 2008; Nichols et al., 2008). Indeed,
temperature and AET): in temperate or cold areas species richness is the drivers of dung beetle diversity gradients in temperate regions
mainly constrained by energy inputs, whereas in warmer conditions, are relatively well-known (Hortal et  al.,  2011). However, in the
where energy is not limiting, water availability is the main constrain Neotropics, knowledge on their diversity is limited to local studies,
for hosting species-rich assemblages (Hawkins et al., 2003). Finally, and the determinants of geographical variations in their communi-
ambient energy relies on the direct ecophysiological responses of ties cannot be easily extrapolated from other regions. Neotropical
individuals of different species to temperature as an explanation dung beetles present certain particularities (Scholtz et  al.,  2009)
for species richness gradients (Turner et al., 1987). In warm regions since their evolutionary radiation in this region is linked to forest
PESSÔA et al.       3|
habitats, in contrast with the Afrotropical and Palearctic history of beetle abundance and species richness altogether. This allows us to
grassland evolution (Gunter et  al.,  2016; Monaghan et  al.,  2007). assess the degree of association of each competing explanation with
That said, several climatic and environmental variables are known species richness, and also to assess interactions between concurrent
to correlate with dung beetle diversity, including temperature, hab- explanations and their relationship with the more–individuals hy-
itat changes, mammal diversity and seasonality. This latter factor pothesis (through abundance; see Storch et al., 2018; Wright, 1983).
has a strong influence on dung beetle richness since in dry seasons
dung beetles are less active and represent a sub-sample of rainy
season communities (Hanski & Cambefort,  1991). Unfortunately, 2 | M ATE R I A L S A N D M E TH O DS
the lack of well-resolved phylogenetic information for Neotropical
Scarabaeinae prevents assessing potential evolutionary constrains 2.1 | Data collection and filtering
on the current distribution of this group's diversity. Also, the lim-
ited number of surveys in Austral South America (see data in Pessôa Dung beetles have been surveyed in different places for a long time
et al., 2020) does not allow exploring the eventual association of cur- in the Neotropics, making it possible to compile a database that rep-
rent dung beetle richness with climatic conditions during the glacia- resents the variations in local diversity throughout this region. To
tion. However, there are sufficient data to relate local Scarabaeinae construct the database, we searched for ecological works on dung
richness with variables accounting for several species–energy and beetles in Web of Science and Google Scholar with the following
heterogeneity explanations. keywords: “dung beetle*”, “Scarabaeinae” and the names of all the
Here we aim to identify the main drivers of local dung beetle spe- countries with territory in the neotropical region. We only selected
cies richness in the Neotropics and assess their relative importance. works that were based on field surveys with baited pitfall traps (a
To do this, we use data from a comprehensive compilation of local standard for dung beetle collection), presented a list of all dung bee-
studies to conduct a multi-model evaluation of six groups of drivers tle species recorded (or a reliable account of species richness and
related to species–energy and heterogeneity explanations (Table 1). abundance), and for which information on the location of the surveys
In a first phase, we assess the information provided by the variables could be georeferenced with a spatial precision of at least 10  km.
accounting for each one of these explanations using model selection From each work, we extracted information about the number of spe-
analyses. Then, we use structural equation models to evaluate the cies recorded, total abundance, type of habitat (classified in open or
relationships between the most informative explanations and dung closed), type of bait used (classified into omnivorous, herbivorous,

TA B L E 1   Main explanations for the origin of geographic diversity gradients evaluated in this work, including the reference where they
were originally proposed, the variables initially proposed or studied and the variables used in this work to account for each one of them

Explanations Original reference(s) Originally proposed variables Variables used in this work

Species–energy hypotheses Hutchinson, 1959

Productivity Wright, 1983 AET + TNPP AET + NDVI
Water–energy Currie, 1991 PET AET + Annual mean
Precipitation
+ Precipitation wettest
quarter
+ Precipitation driest quarter
Ambient energy Turner et al., 1987 Temperature + Solar radiation Annual mean temperature
+ Mean temperature
warmest quarter
+ Mean temperature coldest
quarter
Heterogeneity hypotheses Lack, 1969
Habitat heterogeneity Ricklefs, 1977 Gradients of physical factors and Land Cover S + Habitat Type
opening in canopy forests + Soil structure + Elevation
Range
Seasonal/temporal Sanders, 1968 Seasonality Isothermality + Temperature
heterogeneity seasonality
+ Precipitation Seasonality
Resource heterogeneity Tilman, 1985 Resource variability Mammal Species Richness
Other spatially structured factors Griffith, 2003; Diniz-Filho & Spatial eigenvectors Spatial Eigenvectors and
Bini, 2005 the 3-degree polynomial of
latitude and longitude (TSA)

Abbreviations: AET, actual evapotranspiration; PET, potential evapotranspiration; TNPP, total net primary production.
 |
4       PESSÔA et al.

rotten fruit and rotten meat), year, trap hours as a measure of sam- QGIS (QGIS Development Team,  2019). Data on climate were ob-
pling effort (calculated by the number of pitfall traps*time in the tained from WorldClim 2.0 (Fick & Hijmans, 2017) and actual evapo-
field*number of collects) and country. transpiration (AET) from CGIAR (Trabuco & Zomer,  2010), both at
Our literature search retrieved 170 articles with local studies on a resolution of 30 arcseconds (c. 1 km). For Normalized Difference
dung beetle communities in countries with all or part of their terri- Vegetation Index (NDVI), we used the NASA-MODIS database of
tory in the Neotropical realm (Table S1 in Appendix S1). These arti- monthly NDVI values (Didan,  2015) at a resolution of 10  min (c.
cles rendered data on 298 local sampling sites with 20 or more pitfall 20 km). We assigned to each study the annual mean from the year
traps in the survey (Table S2 in Appendix  S1; Pessôa et  al.,  2020). when the survey was conducted: for works performed in more than
Brazil and Mexico had more sites due to their size and long tradition 1  year, we estimated the average NDVI for all years surveyed. To
of research on dung beetle ecology. From this initial list of sites, we account for habitat heterogeneity, we calculated land cover richness
filtered out those that had performed fewer than 960 trap hours (see (i.e. habitat diversity, see Hortal et al., 2009) as the number of land
Appendix S2), surveys that were conducted before 2000 (to match cover categories in a buffer of a 10-km radius around the georefer-
the time frame of the NDVI data, that runs from 2000 to the present; enced point of each study, using the Global Land Cover by National
Didan,  2015), and/or those that lacked precise locations or any of Mapping Organizations (GLCNMO Version 3; Tateishi et al., 2014) at
the required information. After this process, 190 sites remained that a resolution of 15 arcseconds (c. 500 m). In addition, data on habi-
were located throughout most of the Neotropics–except Austral tat type were collected from each article and classified as open or
South America (Figure 1; Table S1 in Appendix S1). closed, or as ‘both’ if the study did not provide a list with separate
types of habitats. The classification was made according to the de-
scription of the surveyed locality (forest, pasture and so on) provided
2.2 | Predictor variables by the original reference. We extracted the maximum and minimum
altitude from the FAO elevation database (Fischer et al., 2012), using
We extracted information on a series of predictor variables from the same buffer as for land cover at a resolution of 30 arcseconds
several widely used climatic and environmental geodatabases using (c. 1  km). For soil structure, we extracted the proportion of sand,

F I G U R E 1   Location of all dung beetle

studies in the Neotropics found through
our literature search. Studies used for our
analyses are in red, and those that were
discarded in black
PESSÔA et al. |
      5

silt, clay, coarse fragments and bulk density at three depths to ac- and letsR (Vilela & Villalobos, 2020) for the spatial data. All variables
count for species that nest at different depths: 0.15, 0.6 and 2  m, were centred trough the Scale function from the ‘Vegan’ package.
at a resolution of 250 m from the World Soil Information database
(Hengl et al., 2017). Then we performed a PCA and used the broken
stick criterion to choose the most representative axes as measures 2.3 | Statistical analyses
of soil structure (see Table S1 in Appendix S2). To account for mam-
mal species richness in each study we used Biodiversitymapping We used generalized linear mixed models (GLMM) to assess the as-
data (Pimm et al., 2014) at 10 km resolution, excluding all volant and sociation of all explanations with species richness in a multi-model
marine mammals. Finally, since several ecological and historical pro- approach (Burnham & Anderson,  2002; Harrison et  al.,  2018). Since
cesses may shape the patterns analysed beyond environmental vari- abundance, type of bait and ecoregion can strongly influence the
ations by creating spatial structure in the data, we used the latitude observed richness of the studied local communities, we included the
and longitude of each study to construct a Trend Surface Analysis first variable as a covariate and the latter two as random factors in
(Hortal et al., 2008; Legendre & Legendre, 1998) and spatial eigen- the GLMM to control their effects on the multi-model selection. We
vectors (Diniz-Filho & Bini, 2005). did not use sampling effort as a covariate because, in a preliminary
We then selected the predictor accounting for the six explana- analysis, traps/hour had a low correlation with richness and abun-
tions evaluated (Table 1) from the environmental variables described dance (r2 = 0.06 for richness, r2 = 0.05 for abundance), once the lo-
above, as follows. For primary productivity, we selected AET plus calities with low sampling effort were discarded (see above). Models
NDVI. For water–energy, we used AET plus annual precipitation, were constructed by subsequentially adding hypotheses. That is, we
precipitation of wettest quarter and precipitation of the driest quar- evaluated each hypothesis individually and then all subsequent com-
ter. For ambient energy, we selected annual mean temperature plus binations of two, three, four, five and six hypotheses. In each step,
mean temperature of the warmest quarter and mean temperature we selected the most informative model as the one with the lowest
of the coldest month. For habitat heterogeneity, we used land cover Akaike Information Criterion (AIC; Burnham & Anderson, 2002). In the
richness plus altitudinal range, PCA axis of soil structure and type last step, we compared the lowest AIC of each model in the previous
of habitat (Table 1). For temporal heterogeneity and seasonality, we step with the model including all the hypotheses. We included all hy-
used Isothermality plus Temperature Seasonality and Precipitation potheses in a conceptual model that accounts for the relationships be-
Seasonality. For resource heterogeneity, we used Mammal richness. tween dung beetle richness and abundance, and the variables used to
And to account for other spatially structured factors, we used Trend describe all six hypotheses (Figure 2). These analyses were conducted
Surface Analysis and eigenvector-based spatial filtering to include the in R environment using the ‘nlme’ (Pinheiro et al., 2020) package.
spatial data in the models (Diniz-Filho & Bini, 2005). See Table 1 for We used structural equation models (SEM; Grace,  2006;
further details. We processed all these variables in R environment, Shipley, 2016) to evaluate the relationships among the different con-
using the packages ‘vegan’ (Oksanen et al., 2019) for PCA analyses, current hypotheses and dung beetle diversity. We used generalized

Habitat Condition Landscape

Heterogeneity
Habitat type Soil structure Productivity Climate
Open Closed Land Cover
Soil1 Soil2 Soil3 AET NDVI Climate1 Climate2
Habitat Habitat Richnes

Resource
Diversity
Scarabaeinae
Mammal
Abundance
Richness

Scarabaeinae
Richness

F I G U R E 2   Prior conceptual model of the structure of the relationships between dung species beetle richness and the variables
accounting for the main hypotheses on the origin of geographical diversity gradients.
 |
6       PESSÔA et al.

F I G U R E 3   Dung beetle local

species abundance and richness in the
Neotropical Region. Increasing species
abundance is depicted in progressively
larger circles and on a continuous scale
from white (lowest richness) to red
(highest richness)

linear models as the component models, without including random new relationships) until it fit the observed data (reached by a d-
factors to avoid losing many degrees of freedom. In addition, we set sep > 0.05, Shipley, 2009) and excluded non-significant paths, until
trap hours as offset in the abundance model to control for the effect all remaining variables were informative (see Calatayud et al., 2016
of sampling effort in abundance estimates. Unlike the multi-model for a similar approach). Then we interpreted the final clean structural
approach, for SEM analyses we did not use the isolated climatic vari- model. SEM analyses were conducted in R environment using the
ables. Instead, we used a Principal Components Analysis to summa- package ‘piecewiseSEM’ (Lefcheck et al., 2019).
rize total climatic variation into a limited set of uncorrelated factors
using the broken stick criterion to determine the number of axes re-
tained (Appendix S2), in this case the first and the second PCA axes. 3 | R E S U LT S
To construct the conceptual model, we took into consideration the
resolution of variables and the known relationships between them The 190 local communities used showed higher species richness
(see Table 1 and Hawkins et al., 2003; Schemske et al., 2009; Storch in tropical latitudes, with a decrease towards Mesoamerica and
et al., 2018). Based on this, we assumed that mammal richness, the Subtropical South America (Figure  3). Local richness varies from 1
variable with the coarsest spatial resolution, was only related with to 105—the highest being at a single site in Ecuador, in the Eastern
climate and land cover richness (this variable was calculated with a Cordillera Real Montane Forest ecoregion. Seventy-eight per cent of
buffer and summed up the distinct land covers). Importantly, since these studies presented data from forest habitats (including works
our data on resource heterogeneity did not have any measure of that surveyed both forest and pastures), and omnivore faeces was
quantity, we did not assume a relationship between mammal rich- the most common type of bait (65%).
ness and dung beetle abundance. We performed the SEM frame- Resource heterogeneity was the most informative single correlate
work in a piecewise approach (Lefcheck, 2016; Shipley, 2009). We of species richness (lowest AIC, Table 2). However, after evaluating
applied a test of directional separation (d-separation) to evaluate if all possible combinations of variables (117 models), a model includ-
the relationships hypothesized in this initial conceptual model (see ing Productivity, Resource Heterogeneity and Spatial Structure was
Figure  2) fit the observed data (Shipley,  2000, 2009). If any rela- selected as the most parsimonious, as it provided lower AIC values
tionships were missing from the model, we added new paths (i.e. with less complexity (Table 2; TableS1 in Appendix S3).
PESSÔA et al. |
      7

TA B L E 2   Results of the multi-model analyses evaluating each attributed to a single, simple explanation. The explicit treatment of
explanation in isolation and the most explicative combination of multiple sets of potential drivers through multi-model and structural
them
equation modelling analytical frameworks allowed us to uncover
df AIC BIC logLik the complexity of the interactions between climate, soil, habitat
PP 10 227.9764 260.4466 −103.988 and mammal richness in their potential influence on the diversity of
Neotropical dung beetle communities. Strikingly, the strong direct
WE 12 225.3951 264.3594 −100.698
associations of abundance and, to a lesser extent, productivity indi-
AE 11 232.5041 268.2213 −105.252
cate that local richness might primarily be related by the amount of
HH 16 232.9178 284.8702 −100.459
energy and resources. However, soil, habitat and mammal richness
SE 11 229.7065 265.4238 −103.853
also potentially present multiple direct and indirect influences on
RH 9 219.4671 248.6903 −100.734
dung beetle species richness. In fact, resource heterogeneity stands
SP 10 231.2128 263.683 −105.606
out as the most informative single explanation, rather than one of
PP.RH.SP 13 210.4454 252.6568 −92.2227 the species–energy explanations that would be expected given the
The Bold Values are the lowest AIC and referenced in the text. strong association between richness and abundance (which is una-
Abbreviations: AE, ambient energy; HH, habitat heterogeneity; PP, voidably determined by productivity; Wright, 1983).
primary productivity; RH, resource heterogeneity; SE, seasonality; SP, That habitat heterogeneity is an important influence on dung
spatial structure; WE, water–energy.
beetle local richness is not surprising. Both climate and microclimatic
conditions exert a strong influence on dung beetle activity, affecting
The final Structural Equation Model incorporating all the evalu- their metabolic rates and altering resource volatility and availabil-
ated macroecological explanations accounted for 89% of local dung ity time (Davis et  al.,  2013; Medina & Lopes,  2014). Furthermore,
beetle richness variations in the Neotropics. An explicit analysis of most Neotropical species of Scarabaeinae have evolved to exploit
the relationships between the different sets of drivers allowed us to tropical and subtropical forest environments (Gunter et  al.,  2016;
identify the existence of a complex array of interactions between Monaghan et al., 2007; Scholtz et al., 2009). As a result, dung beetle
predictors, from which the strong positive association of abundance communities are strongly dependent on habitat structure (Gardner
and the negative association of open habitats with dung beetle spe- et  al.,  2008; Hanski & Cambefort,  1991; Martello et  al.,  2016;
cies richness stand out (Figure 4; Table S2 in Appendix S3). Besides Nichols et  al.,  2007) and the microclimatic differences induced by
these major relationships, Soil4 (a PCA axis related to volume of the variations in forest structure (da Silva & Hernández, 2016) and
coarse fragments) and Mammal Richness also showed direct posi- elevation (Nunes et  al.,  2016). Furthermore, dung beetles also re-
tive associations with dung beetle richness. The association of ambi- spond to soil conditions, since their soil burial behaviour (Halffter
ent energy (a PCA axis related to temperature WorldClim variables) & Edmonds,  1982) imposes a selection pressure in relation to soil
and NDVI with richness was positive, though indirect through their structure (i.e. the amount of clay, sand and silt and soil compac-
direct influence on abundance. Furthermore, while resolving the tion; Davis,  1996). It is important to note, however, that we found
model, a new relationship was added to fit the data based on the an unexpected negative relationship between altitudinal range and
d-separation test, which identified a negative association between scarabaeine abundance. We have not been able to identify any possi-
altitudinal range and dung beetle abundance (Figure 4). ble reason why dung beetle populations should be smaller in steeper
areas, except perhaps that cattle herds are generally more abundant
in the plains so areas with smaller altitudinal ranges may host more
4 | D I S CU S S I O N diverse communities due to the higher availability of resources. In
any case, this intriguing relationship deserves further exploration.
Our results evidence that the geographical variations in the species Scarabaeine dung beetles use mammal dung as their primary
richness of dung beetle communities in the Neotropics cannot be resource for feeding and nesting (Halffter & Matthews,  1966), so

Resource Habitat Altitudinal Ambient Energy Productivity

Open habitats (mammal) heterogeneity
F I G U R E 4   Structural equation model richness (Soil4) range (Temp) (NDVI)
for dung beetle species richness on the -0.3884***
Neotropics. The arrows indicate the 0.2609***

significant paths with their respective

standardized coefficients (numbers). -0.217***
0.7317***
0.0568***
Blue and red arrows denote positive 0.02***
0.0967***
Abundance
and negative relationships, respectively. R2=0.28
Soil4 is a PCA axis mainly representing
the volume of coarse fragments in the 0.29***

soil (see Table S1 in Appendix S2)

Richness
R2=0.89
 |
8       PESSÔA et al.

their communities are often related with mammal diversity and since abundance was included as a cofactor in all these models.
abundance (Raine & Slade, 2019). In general, dung beetles present Given the SEM results, we can assume that the general influence of
low levels of trophic specialization (Raine et  al.,  2018), and faeces abundance on dung beetle diversity extends over all the hypotheses
from humans or other omnivores are used by most species in the evaluated in this work, thus providing support for its role as the main
Neotropics. However, a number of species present a certain degree regulator of the effects of productivity on species richness (Storch
of specialization, so Neotropical mammals have some dung beetle et al., 2018).
species specialized to their excrements (Halffter & Matthews, 1966; The community data compiled in our database comes from local
Howden & Nealis, 1975). Perhaps due to this, dung beetle richness dung beetle inventories sampled using a large variability of survey
and abundance decline when mammal richness declines (Nichols methodologies and sampling efforts. In many cases, surveys were
et al., 2009) or when native mammal species are substituted by ex- poorly described and lacked a clear description of the locality, hab-
otic mammals (Filgueiras et al., 2009; see also Raine & Slade, 2019). itat and/or method of capture. In some cases, such limitations pre-
This interdependence with mammals is confirmed by the direct rela- vented us from distinguishing the abundances of each species in
tionships found in our structural equation models and the presence each type of habitat or the individuals captured with different kinds
of resource heterogeneity in the set of explanations selected in the of bait. This hampers a more in-depth assessment of several import-
multi-model analysis. ant factors that operate at the local scale, but it does not affect our
Besides habitat structure and resource availability, energy af- ability to assess variables operating at larges scales. Indeed, the a
fects richness by regulating productivity, altering individual meta- priori selection of sites surveyed with a sufficiently large sampling
bolic rates and increasing niche packing. The relative importance effort (in our case, 960  trap/hours) ensures that all inventories
of these factors varies geographically, so the inclusion of spatial considered for our analyses were reasonably complete. The use of
factors in our final model could be partly supporting the influence sampling effort as an offset of abundance on the SEM analyses also
of water–energy dynamics on dung beetle diversity (i.e. species accounts for any possible spurious effect of unevenness in sampling
richness is determined by a balance of water availability and ambi- effort in the data analysed. This makes unlikely the existence of any
ent energy; Hawkins et al., 2003). However, according to this ex- consistent biases in our data and results, beyond the exclusion of
planation, tropical areas would be expected to be more affected by Austral South America. Thus, although future works based on more
water availability because population growth is typically not lim- detailed data on the location and characteristics of the surveyed
ited by temperature in these areas (Hawkins et al., 2003; Tshikae sites may enhance our knowledge on the local effects on dung bee-
et  al.,  2013), whereas in temperate regions temperature is the tle diversity, we do not expect significant changes in the structure
main constraint to dung beetle diversity (Hortal et al., 2011; Lobo of relationships among large-scale factors and the local diversity of
et  al.,  2002). Indeed, the effects of rain on the composition and dung beetles identified in this work.
richness of Neotropical dung beetle communities are well-known
(Liberal et  al.,  2011; Novais et  al.,  2016). However, in our SEM,
only temperature variables (Climate PCA axis1) present an indirect 5 | CO N C LU S I O N S
correlation on richness. Therefore, although our results do not call
for a change in the hierarchy of the importance of the relationships Ecological communities are complex systems where multiple mecha-
with water and energy variables for this group, they do challenge nisms operate at the same time. Such complexity may apparently
the simple interpretation that limitations to water metabolism of create incongruent patterns in different localities and regions.
dung beetles impose a major constraint to their diversity. Rather, Therefore, a joint analysis of the different hypotheses raised to
such influence may be related to either the constraints imposed by account for diversity gradients can improve our understanding of
water metabolism to mammal diversity, the effects of water avail- their origin. In this work, at least three of the six explanations evalu-
ability on the attractiveness, texture and nutrient accessibility of ated need to be considered to account for the local species richness
mammal dung, or both. patterns of Neotropical Scarabaeinae. The diversity of dung beetle
Nonetheless, the strong association of scarabaeine abundance communities in the region studied can be explained by a combina-
with richness in our structural equation model mediates most of tion of primary productivity, ambient energy balance, habitat het-
the relationships between diversity and productivity, ambient en- erogeneity and resource heterogeneity hypotheses. The influence
ergy and altitudinal range. Regardless of the importance of other of energy, temperature and altitudinal range apparently occurs in-
factors, this indicates that the main mechanism driving dung beetle directly through the relationships of these factors with abundance
local richness in the Neotropics may be the species–energy relation- (in accordance with Storch et al., 2018), which is the most important
ship, as outlined by the more-individuals hypothesis (Wright, 1983). direct correlate of richness in our analyses. In contrast, habitat and
According to this hypothesis, high energy permits sustaining larger resource heterogeneity present direct associations with dung bee-
populations, diminishing extinction rates and increasing niche pack- tle diversity. Taking this comprehensive picture into consideration,
ing, allowing the coexistence of a larger number of species (Evans in the tropical and subtropical areas of Meso- and South America,
et al., 2005). Note here that this strong effect of abundance was not the species richness of dung beetle communities can be primar-
explicitly evaluated in our multi-hypothesis evaluation framework ily explained by the more-individuals hypothesis (Wright,  1983),
PESSÔA et al. |
      9

through both productivity (Storch et  al.,  2018) and niche packing Insect Conservation, 17(3), 565–576. https://doi.org/10.1007/s1084​
1-012-9542-8
(Hutchinson, 1959).
Didan, K. (2015). MOD13C2 MODIS/Terra Vegetation Indices Monthly
L3 Global 0.05Deg CMG V006. NASA EOSDIS Land Processes DAAC.
AC K N OW L E D G E M E N T S https://doi.org/10.5067/MODIS/​MOD13​C2.006
This work was supported by the projects ‘Predicting diversity vari- Diniz-Filho, J. A. F., & Bini, L. M. (2005). Modelling geographical pat-
terns in species richness using eigenvector-based spatial filters.
ations across scales through process-based models linking commu-
Global Ecology and Biogeography, 14(2), 177–185. https://doi.
nity ecology and biogeography’ (CNPq PVE 314523/2014-6), and org/10.1111/j.1466-822X.2005.00147.x
SCENIC–‘Scaling the effects of niche and interaction dynamics on Evans, K. L., Warren, P. H., & Gaston, K. J. (2005). Species–energy rela-
the ecological and evolutionary outcomes of coexistence’ (PID2019- tionships at the macroecological scale: A review of the mechanisms.
106840GB-C21, funded by AEI/FEDER, UE). MBP was supported Biological Reviews, 80(1), 1–25. https://doi.org/10.1017/S1464​79310​
4006517
by CAPES grants PROEX-0487 and 88881.135489/2016-01, and
Fick, S. E., & Hijmans, R. J. (2017). WorldClim 2: New 1-km spatial
FAM by CAPES Postdoctoral scholarship 120147/2016-01. PDMJ resolution climate surfaces for global land areas. International
research is funded by CNPq (grant 308694/2015-5). This paper is Journal of Climatology, 37(12), 4302–4315. https://doi.org/10.1002/
a contribution of the INCT in Ecology, Evolution and Biodiversity joc.5086
Filgueiras, B. K. C., Liberal, C. N., Aguiar, C. D. M., Hernández, M. I. M., &
Conservation funded by MCTIC/CNPq/FAPEG (grant 465610/2014-
Iannuzzi, L. (2009). Attractivity of omnivore, carnivore and herbivore
5). No collection permits were needed for this work. mammalian dung to Scarabaeinae (Coleoptera, Scarabaeidae) in a
tropical Atlantic rainforest remnant. Revista Brasileira de Entomologia,
DATA AVA I L A B I L I T Y S TAT E M E N T 53(3), 422–427. https://doi.org/10.1590/S0085​-56262​0 0900​
0300017
All data used in this study comes from published sources. We pro-
Fischer, G., Nachtergaele, F. O., Prieler, S., Teixeira, E., Tóth, G., Van
vide Supporting Information that include the final data used for the Velthuizen, H., Verelst, L., & Wiberg, D. (2012). IIASA/FAO Global
analyses (Appendix S1), after cleaning and processing. A fully docu- Agro-ecological Zones (GAEZ v3. 0). IIASA and FAO.
mented version of this database is available under an open access Gardner, T. A., Barlow, J., Araujo, I. S., Ávila-Pires, T. C., Bonaldo, A. B.,
Costa, J. E., Esposito, M. C., Ferreira, L. V., Hawes, J., Hernandez, M.
license in DIGITAL.CSIC public repository (Pessôa et al., 2020; avail-
I. M., Hoogmoed, M. S., Leite, R. N., Lo-Man-Hung, N. F., Malcolm, J.
able at http://hdl.handle.net/10261/​220248). R., Martins, M. B., Mestre, L. A. M., Miranda-Santos, R., Overal, W. L.,
Parry, L., … Peres, C. A. (2008). The cost-effectiveness of biodiversity
ORCID surveys in tropical forests. Ecology Letters, 11(2), 139–150. https://
doi.org/10.1111/j.1461-0248.2007.01133.x
Marcelo Bruno Pessôa  https://orcid.org/0000-0002-2601-8511
Gouveia, S. F., Hortal, J., Cassemiro, F. A. S., Rangel, T. F., & Diniz-Filho, J.
Fernanda Alves-Martins  https://orcid. A. F. (2013). Nonstationary effects of productivity, seasonality, and
org/0000-0003-4269-586X historical climate changes on global amphibian diversity. Ecography,
Paulo De Marco Júnior  https://orcid.org/0000-0002-3628-6405 36, 104–113. https://doi.org/10.1111/j.1600-0587.2012.07553.x
Joaquín Hortal  https://orcid.org/0000-0002-8370-8877 Grace, J. B. (2006). Structural equation modeling and natural systems.
Cambridge University Press.
Griffith, D. A. (2003). Spatial autocorrelation and spatial filtering: Gaining
REFERENCES understanding through theory and scientific visualization. Springer
Burnham, K. P., & Anderson, D. R. (2002). Model selection and multi- Science & Business Media.
model inference: A practical information-theoretic approach (2nd ed.). Gunter, N. L., Weir, T. A., Slipinksi, A., Bocak, L., & Cameron, S. L. (2016).
Springer-Verlag. If Dung Beetles (Scarabaeidae: Scarabaeinae) arose in association
Calatayud, J., Hortal, J., Medina, N. G., Turin, H., Bernard, R., Casale, with dinosaurs, did they also suffer a mass co-extinction at the K-Pg
A., Ortuño, V. M., Penev, L., & Rodríguez, M. Á. (2016). Glaciations, boundary? PLoS ONE, 11(5), e0153570. https://doi.org/10.1371/
deciduous forests, water availability and current geographical journ​al.pone.0153570
patterns in the diversity of European Carabus species. Journal Halffter, G., & Edmonds, W. D. (1982). The nesting behavior of dung bee-
of Biogeography, 43(12), 2343–2353. https://doi.org/10.1111/ tles (Scarabaeinae). An ecological and evolutive approach. The Nesting
jbi.12811 Behavior of Dung Beetles (Scarabaeinae). An Ecological and Evolutive
Currie, D. J. (1991). Energy and large-scale patterns of animal-and Approach. https://www.cabdi​rect.org/cabdi​rect/abstr​act/19830​
plant-species richness. The American Naturalist, 137(1), 27–49. 503784
https://doi.org/10.1086/285144 Halffter, G., & Matthews, E. G. (1966). The Natural History of Dung
da Silva, P. G., & Hernández, M. I. M. (2016). Spatial variation of dung Beetles of the Subfamily Scarabaeinae (Coleoptera:Scarabaeidae).
beetle assemblages associated with forest structure in remnants of Folia Entomológica Mexicana, 12, 312.
southern Brazilian Atlantic Forest. Revista Brasileira de Entomologia, Hanski, I., & Cambefort, Y. (1991). Dung beetle ecology. Princeton
60(1), 73–81. https://doi.org/10.1016/j.rbe.2015.11.001 University Press.
Davis, A. L. V. (1996). Seasonal dung beetle activity and dung dispersal Harrison, X. A., Donaldson, L., Correa-Cano, M. E., Evans, J., Fisher, D.
in selected South African habitats: Implications for pasture improve- N., Goodwin, C. E. D., Robinson, B. S., Hodgson, D. J., & Inger, R.
ment in Australia. Agriculture, Ecosystems & Environment, 58(2), 157– (2018). A brief introduction to mixed effects modelling and multi-
169. https://doi.org/10.1016/0167-8809(96)01030​- 4 model inference in ecology. PeerJ, 6, e4794. https://doi.org/10.7717/
Davis, A. L. V., van Aarde, R. J., Scholtz, C. H., Guldemond, R. A. R., peerj.4794
Fourie, J., & Deschodt, C. M. (2013). Is microclimate-driven turnover Hawkins, B. A. (2001). Ecology’s oldest pattern? Trends in Ecology &
of dung beetle assemblage structure in regenerating coastal vege- Evolution, 16(8), 470. https://doi.org/10.1016/S0169​-5347(01)02197​
tation a precursor to re-establishment of a forest fauna? Journal of -8
 |
10       PESSÔA et al.

Hawkins, B. A. (2008). Recent progress toward understanding the global H. A., McCain, C. M., McCune, A. R., McDade, L. A., McPeek, M. A.,
diversity gradient. IBS Newsletter, 6(1). Near, T. J., Price, T. D., Ricklefs, R. E., Roy, K., Sax, D. F., … Turelli, M.
Hawkins, B. A., Field, R., Cornell, H. V., Currie, D. J., Guégan, J.-F., (2007). Evolution and the latitudinal diversity gradient: Speciation,
Kaufman, D. M., Kerr, J. T., Mittelbach, G. G., Oberdorff, T., O'Brien, extinction and biogeography. Ecology Letters, 10(4), 315–331. https://
E. M., Porter, E. E., & Turner, J. R. G. (2003). Energy, water, and broad- doi.org/10.1111/j.1461-0248.2007.01020.x
scale geographic patterns of species richness. Ecology, 84(12), 3105– Monaghan, M. T., Inward, D. J. G., Hunt, T., & Vogler, A. P. (2007). A mo-
3117. https://doi.org/10.1890/03-8006 lecular phylogenetic analysis of the Scarabaeinae (dung beetles).
Hengl, T., de Jesus, J. M., Heuvelink, G. B. M., Gonzalez, M. R., Kilibarda, Molecular Phylogenetics and Evolution, 45(2), 674–692. https://doi.
M., Blagotić, A., Shangguan, W., Wright, M. N., Geng, X., Bauer- org/10.1016/j.ympev.2007.06.009
Marschallinger, B., Guevara, M. A., Vargas, R., MacMillan, R. A., Nichols, E., Gardner, T. A., Peres, C. A., & Spector, S. (2009). Co-declining
Batjes, N. H., Leenaars, J. G. B., Ribeiro, E., Wheeler, I., Mantel, S., & mammals and dung beetles: An impending ecological cascade. Oikos,
Kempen, B. (2017). SoilGrids250m: Global gridded soil information 118(4), 481–487. https://doi.org/10.1111/j.1600-0706.2009.17268.x
based on machine learning. PLoS ONE, 12(2), e0169748. https://doi. Nichols, E., Larsen, T., Spector, S., Davis, A. L., Escobar, F., Favila, M.,
org/10.1371/journ​al.pone.0169748 & Vulinec, K. (2007). Global dung beetle response to tropical for-
Hortal, J., Diniz-Filho, J. A. F., Bini, L. M., Rodríguez, M. Á., Baselga, est modification and fragmentation: A quantitative literature review
A., Nogués-Bravo, D., Rangel, T. F., Hawkins, B. A., & Lobo, J. M. and meta-analysis. Biological Conservation, 137(1), 1–19. https://doi.
(2011). Ice age climate, evolutionary constraints and diversity pat- org/10.1016/j.biocon.2007.01.023
terns of European dung beetles: Ice age determines European Nichols, E., Spector, S., Louzada, J., Larsen, T., Amezquita, S., & Favila,
scarab diversity. Ecology Letters, 14(8), 741–748. https://doi. M. E. (2008). Ecological functions and ecosystem services provided
org/10.1111/j.1461-0248.2011.01634.x by Scarabaeinae dung beetles. Biological Conservation, 141(6), 1461–
Hortal, J., Rodríguez, J., Nieto-Díaz, M., & Lobo, J. M. (2008). Regional 1474. https://doi.org/10.1016/j.biocon.2008.04.011
and environmental effects on the species richness of mammal as- Novais, S. M. A., Evangelista, L. A., Reis-Júnior, R., & Neves, F. S. (2016).
semblages. Journal of Biogeography, 35(7), 1202–1214. https://doi. How does dung beetle (Coleoptera: Scarabaeidae) diversity vary
org/10.1111/j.1365-2699.2007.01850.x along a rainy season in a tropical dry forest? Journal of Insect Science,
Hortal, J., Triantis, K. A., Meiri, S., Thébault, E., & Sfenthourakis, S. (2009). 16(1), 81. https://doi.org/10.1093/jises​a/iew069
Island species richness increases with habitat diversity. The American Nunes, C. A., Braga, R. F., Figueira, J. E. C., Neves, F. D. S., & Fernandes,
Naturalist, 173, E205–E217. https://doi.org/10.1086/645085 G. W. (2016). Dung beetles along a tropical altitudinal gradi-
Howden, H. F., & Nealis, V. G. (1975). Effects of clearing in a tropi- ent: Environmental filtering on taxonomic and functional diver-
cal rain forest on the composition of the Coprophagous Scarab sity. PLoS ONE, 11(6), e0157442. https://doi.org/10.1371/journ​
Beetle Fauna (Coleoptera). Biotropica, 7(2), 77–83. https://doi. al.pone.0157442
org/10.2307/2989750 Oksanen, J., Blanchet, F. G., Friendly, M., Kindt, R., Legendre, P., McGlinn,
Hutchinson, G. E. (1959). Homage to Santa Rosalia or why are there so D., Minchin, P. R., O'Hara, R. B., Simpson, G. L., Solymos, P., Stevens,
many kinds of animals? The American Naturalist, 93(870,), 145–159. M. H. H., Szoecs, E., & Wagner, H. (2019). ‘vegan’ – Community ecology
https://doi.org/10.1086/282070 R package. Retrieved from https://cran.r-proje​c t.org/web/packa​ges/
Lack, D. (1969). The numbers of bird species on islands. Bird Study, 16(4), vegan/
193–209. https://doi.org/10.1080/00063​65690​9476244 Pessôa, M. B., Alves-Martins, F., De Marco Júnior, P., & Hortal, J. (2020).
Lawton, J. H. (1999). Are there general laws in ecology? Oikos, 84(2), A database of Neotropical dung beetle local richness from standard-
177–192. https://doi.org/10.2307/3546712 ized inventories. DIGITAL.CSIC. https://doi.org/10.20350/​digit​alCSI​
Lefcheck, J. S. (2016). PiecewiseSEM: Piecewise structural equa- C/12635
tion modelling in R for ecology, evolution, and systematics. Pianka, E. R. (1966). Latitudinal gradients in species diversity: A review
Methods in Ecology and Evolution, 7(5), 573–579. https://doi. of concepts. The American Naturalist, 100(910), 33–46. https://doi.
org/10.1111/2041-210X.12512 org/10.1086/282398
Lefcheck, J., Byrnes, J., & Grace, J. (2019). 'piecewiseSEM' R package – Pimm, S. L., Jenkins, C. N., Abell, R., Brooks, T. M., Gittleman, J. L., Joppa,
Piecewise structural equation modeling v. 2.1.0. Available at https:// L. N., Raven, P. H., Roberts, C. M., & Sexton, J. O. (2014). The bio-
cran.r-proje​c t.org/web/packa​ges/piece​wiseS​EM/ diversity of species and their rates of extinction, distribution, and
Legendre, P., & Legendre, L. (1998). Numerical ecology (2nd ed.). Elsevier. protection. Science, 344(6187), 1246752. https://doi.org/10.1126/
Liberal, C. N., Farias, D., Isidro, Â. M., Meiado, M. V., Filgueiras, B. K. C., & scien​ce.1246752
Iannuzzi, L. (2011). How habitat change and rainfall affect dung bee- Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D., EISPACK authors,
tle diversity in Caatinga, a Brazilian semi-arid ecosystem. Journal of Heisterkamp, S., & Van Willigen, B.R.-C.T. (2020) 'nlme' R package- lin-
Insect Science, 11(1) , 1–11. https://doi.org/10.1673/031.011.11401 ear and nonlinear mixed effects models. Retrieved from https://cran.r-
Lobo, J. M., Lumaret, J.-P., & Jay-Robert, P. (2002). Modelling the spe- proje​c t.org/web/packa​ges/nlme/
cies richness distribution of French dung beetles (Coleoptera, QGIS Development Team. (2019). QGIS Geographic Information
Scarabaeidae) and delimiting the predictive capacity of different System (Version 3.4) [Computer software]. Open Source Geospatial
groups of explanatory variables. Global. Ecology, 13. Foundation. Retrieved from https://www.qgis.org/
Martello, F., Andriolli, F., de Souza, T. B., Dodonov, P., & Ribeiro, M. Raine, E. H., Mikich, S. B., Lewis, O. T., Riordan, P., Vaz-de-Mello, F. Z.,
C. (2016). Edge and land use effects on dung beetles (Coleoptera: & Slade, E. M. (2018). Extinctions of interactions: Quantifying a
Scarabaeidae: Scarabaeinae) in Brazilian cerrado vegetation. Journal dung beetle-mammal network. Ecosphere, 9, e02491. https://doi.
of Insect Conservation, 20(6), 957–970. https://doi.org/10.1007/ org/10.1002/ecs2.2491
s1084​1-016-9928-0 Raine, E. H., & Slade, E. M. (2019). Dung beetle–mammal associations:
Medina, A. M., & Lopes, P. P. (2014). Seasonality in the dung beetle com- Methods, research trends and future directions. Proceedings of the
munity in a Brazilian tropical dry forest: Do small changes make a Royal Society B: Biological Sciences, 286, 28620182002. https://doi.
difference? Journal of Insect Science, 14(1). https://doi.org/10.1093/ org/10.1098/rspb.2018.2002
jis/14.1.123 Rangel, T. F., Edwards, N. R., Holden, P. B., Diniz-Filho, J. A. F., Gosling,
Mittelbach, G. G., Schemske, D. W., Cornell, H. V., Allen, A. P., Brown, J. W. D., Coelho, M. T. P., Cassemiro, F. A. S., Rahbek, C., & Colwell,
M., Bush, M. B., Harrison, S. P., Hurlbert, A. H., Knowlton, N., Lessios, R. K. (2018). Modeling the ecology and evolution of biodiversity:
PESSÔA et al. |
      11

Biogeographical cradles, museums, and graves. Science, 361, Turner, J. R. G. (2004). Explaining the global biodiversity gradient:
eaar5452. https://doi.org/10.1126/scien​ce.aar5452 Energy, area, history and natural selection. Basic and Applied Ecology,
Ricklefs, R. E. (1977). Environmental heterogeneity and plant species 5(5), 435–448. https://doi.org/10.1016/j.baae.2004.08.004
diversity: A hypothesis. The American Naturalist, 111(978), 376–381. Turner, J. R. G., Gatehouse, C. M., & Corey, C. A. (1987). Does solar en-
https://doi.org/10.1086/283169 ergy control organic diversity? Butterflies, moths and the British cli-
Sanders, H. L. (1968). Marine benthic diversity: A comparative mate. Oikos, 48(2), 195. https://doi.org/10.2307/3565855
study. The American Naturalist, 102(925), 243–282. https://doi. Vilela, B., & Villalobos, F. (2020). letsR' R package – Data handling and anal-
org/10.1086/282541 ysis in macroecology v. 3.2. Retrieved from https://cran.r-proje​c t.org/
Schemske, D. W., Mittelbach, G. G., Cornell, H. V., Sobel, J. M., & web/packa​ges/letsR/
Roy, K. (2009). Is there a latitudinal gradient in the importance von Humboldt, A. (1850). Views of Nature: Or Contemplations on the
of biotic interactions? Annual Review of Ecology, Evolution, and Sublime Phenomena of Creation; with Scientific Illustrations, Translated
Systematics, 40(1), 245–269. https://doi.org/10.1146/annur​ev.ecols​ from the German by E.C. Otté and Henry G. Bohn; with a frontispiece
ys.39.110707.173430 from a sketch by the author, a fac-simile of his handwriting, and a com-
Scholtz, C. H., Davis, A. L. V., Kryger, U., & EBSCOhost. (2009). Evolutionary prehensive index. H.G. Bohn.
biology and conservation of dung beetles. Pensoft Publishers Coronet Whittaker, R. J., & Field, R. (2000). Tree species richness modelling: An
Books. Distributor http://search.ebsco​host.com/login.aspx?direc​ approach of global applicability? Oikos, 89(2), 399–402. https://doi.
t=true&scope​=site&db=nlebk​&db=nlabk​& AN=320509 org/10.1034/j.1600-0706.2000.890222.x
Shipley, B. (2000). A new inferential test for path models based on di- Wiens, J. J., & Donoghue, M. J. (2004). Historical biogeography, ecol-
rected acyclic graphs. Structural Equation Modeling: A Multidisciplinary ogy and species richness. Trends in Ecology & Evolution, 19, 639–644.
Journal, 7(2), 206–218. https://doi.org/10.1207/S1532​8 007S​ https://doi.org/10.1016/j.tree.2004.09.011
EM0702_4 Wright, D. H. (1983). Species-energy theory: An extension of spe-
Shipley, B. (2009). Confirmatory path analysis in a generalized multilevel cies-area theory. Oikos, 41(3), 496. https://doi.org/10.2307/3544109
context. Ecology, 90(2), 363–368. https://doi.org/10.1890/08-1034.1
Shipley, B. (2016). Cause and correlation in biology a user’s guide to path
analysis, structural equations and causal inference with R 2nd edition B I O S K E TC H
(2nd ed., Vol. 314). Oxford University Press. Marcelo Bruno Pessôa is an ecologist interested in the ecology
Stein, A., & Kreft, H. (2015). Terminology and quantification of envi- and biogeography of neotropical dung beetles, and the patterns
ronmental heterogeneity in species-richness research. Biological
and processes behind community assembly. Fernanda Alves
Reviews, 90, 815–836. https://doi.org/10.1111/brv.12135
Storch, D., Bohdalková, E., & Okie, J. (2018). The more-individuals hy-
Martins, Paulo De Marco Júnior and Joaquín Hortal work on
pothesis revisited: The role of community abundance in species rich- biogeography and community ecology, with emphasis on several
ness regulation and the productivity-diversity relationship. Ecology groups of insects.
Letters, 21(6), 920–937. https://doi.org/10.1111/ele.12941
Tateishi, R., Hoan, N. T., Kobayashi, T., Alsaaideh, B., Tana, G., & Phong,
Author contributions: MBP, PDMJ and JH conceived the ideas;
D. X. (2014). Production of global land cover data – GLCNMO2008.
Journal of Geography and Geology, 6(3). https://doi.org/10.5539/jgg. MBP collected the data; MBP and FAM analysed the data; MBP
v6n3p99 and JH wrote the paper, with FAM and PDMJ.
Tews, J., Brose, U., Grimm, V., Tielbörger, K., Wichmann, M. C.,
Schwager, M., & Jeltsch, F. (2004). Animal species diversity driven
S U P P O R T I N G I N FO R M AT I O N
by habitat heterogeneity/diversity: The importance of key-
stone structures: Animal species diversity driven by habitat het- Additional supporting information may be found online in the
erogeneity. Journal of Biogeography, 31(1), 79–92. https://doi. Supporting Information section.
org/10.1046/j.0305-0270.2003.00994.x
Tilman, D. (1985). The resource-ratio hypothesis of plant succes-
sion. The American Naturalist, 125(6), 827–852. https://doi.
How to cite this article: Pessôa MB, Alves-Martins F, De
org/10.1086/284382
Trabuco, A., & Zomer, R. J. (2010). Global high-resolution soil-water balance
Marco Júnior P, Hortal J. Unveiling the drivers of local dung
| CGIAR-CSI [Map]. Retrieved from http://www.cgiar​-csi.org/data/ beetle species richness in the Neotropics. J Biogeogr.
globa​l-high-resol​ution​-soil-water​-balance 2020;00:1–11. https://doi.org/10.1111/jbi.14043
Tshikae, B. P., Davis, A. L. V., & Scholtz, C. H. (2013). Species richness –
Energy relationships and dung beetle diversity across an aridity and
trophic resource gradient. Acta Oecologica, 49, 71–82. https://doi.
org/10.1016/j.actao.2013.02.011