Effects of livestock on
arthropod biodiversity in Iberian Holm oak savannas revealed by
metabarcoding
Tara Caneloa,b*, Daniel
Marquinac,d*, Sergio Chozase,
Johannes Bergstenf, Álvaro
Gaytáng,, Carlos Pérez-Izquierdoa,
Raúl Bonala, h
a Dpto. Ingeniería del Medio Agronómico y Forestal.
Grupo de Investigación Forestal (INDEHESA), Centro Universitario de
Plasencia, Universidad de Extremadura, Avda. Virgen del Puerto 2, 10600
Plasencia, Cáceres, Spain
b Centro de Ecologia Aplicada “Prof. Baeta Neves”
(CEABN), Instituto Superior de Agronomia, Universidade de Lisboa, Tapada
da Ajuda, 1349-017 Lisboa, Portugal
c Department of Bioinformatics and Genetics, Swedish
Museum of Natural History, Stockholm, Sweden
d Department of Zoology, Stockholm University,
Stockholm, Sweden
e cE3c - Centre for Ecology, Evolution and
Environmental Changes & CHANGE - Global Change and Sustainability
Institute. FCUL: Sciences Faculty of the University of Lisbon Campo
Grande, Portugal
f Department of Zoology, Swedish Museum of Natural
History, Stockholm, Sweden
g Instituto de Recursos Naturales y Agrobiología de
Sevilla, Consejo Superior de Investigaciones Científicas (IRNAS-CSIC),
Sevilla, Spain
h Department of Biodiversity, Ecology and Evolution.
Complutense University of Madrid, Spain
* These authors contributed equally to the study.
Corresponding author: Tara Canelo, taracanelo@unex.es
Running head: Effects of livestock on arthropod biodiversity
ABSTRACT
Increasing food production while avoiding negative impacts on
biodiversity constitutes one of the main challenges of our time.
Traditional silvopastoral systems like Iberian oak savannas
(“dehesas”) set an example, where free-range livestock has been reared
for centuries while preserving a high natural value. Nevertheless,
factors decreasing productivity need to be addressed, one being acorn
losses provoked by pest insects. An increased and focalized grazing by
livestock on infested acorns would kill the larvae inside and decrease
pest numbers, but increased livestock densities could have undesired
side effects on ground arthropod communities as a whole. We designed an
experimental setup including areas under trees with livestock exclosures
of different ages along with controls, using DNA metabarcoding
(mitochondrial markers COI and 16S) to rapidly assess arthropod
communities’ composition. Livestock removal quickly increased grass
cover and arthropod taxonomic richness and diversity, which was already
higher in short-term (1-year exclosures) than beneath the canopies of
control trees. Interestingly, arthropod diversity was not highest at
long-term exclosures (≥10 years), although their community composition
was the most distinct. The diversity peak at short term exclosures would
support the intermediate disturbance hypothesis, which relates it with
the higher microhabitat heterogeneity at moderately disturbed areas.
Thus, we propose a rotatory livestock management in dehesas: plots with
increased grazing should co-exist with temporal short-term exclosures.
Ideally, a few long-term excluded areas should be also kept for the
singularity of their arthropod communities. This strategy would make
possible the combination of biological pest control and arthropod
conservation in Iberian dehesas.
Key-words : wood-pastures, grazing, livestock exclusion,
metabarcoding, arthropods, intermediate disturbance hypothesis
INTRODUCTION
One of the great human challenges for the 21st century is to produce
enough food for an increasing population, but to do so in an
environmentally-friendly way (FAO 2022; IPBES 2019). In this context,
traditional agroecosystems play a very important role combining the
preservation of natural values with economic activities. Iberian
silvopastoral oak savannas (so called “dehesas” in Spain and
“montados” in Portugal) are a good example (Sucena-Paiva et al.,
2022). Century-long shrub and tree clearing of Mediterranean forests has
produced landscapes with oaks interspersed within a grassland matrix, in
which livestock rearing is the main use (Bugalho et al., 2011; Moreno &
Pulido, 2009; Sá-Sousa, 2014). Dehesas are also included in the Habitats
Directive of the European Union, as they host a wide range of rare,
threatened, or endemic animal and plant species (EC 2013).
As in any other agricultural system, pest insects reduce the
productivity in oak dehesas (Canelo et al., 2021a), but the preservation
of their high natural value preclude the use of any chemical treatment.
Current approaches follow the logic of ecological intensification, based
on managing service-providing organisms that make a quantifiable direct
or indirect contribution to agricultural and forest productivity
(Bommarco et al., 2013; Montesinos, 2019). Parasitoids (frequently
species of the order Hymenoptera), or vertebrates (e.g. birds), are
among them (Diaz-Siefer et al., 2021; García et al., 2021; Redlich et
al., 2018; Zhu et al., 2019). However, for oak dehesas, we previously
proposed a novel way of biological control based on livestock management
(Canelo et al., 2021b).
In dehesas, acorns are a key food resource for livestock, but their
availability is limited by insect pre-dispersal seed predators, mainlyCurculio weevils (Coleoptera: Curculionidae), which can infest
half of the year’s acorn production (Bogdziewicz et al., 2019; Bonal et
al., 2007). The weevil larvae, however, are vulnerable to predation by
large vertebrates for a short time, namely the period of 20 days that
they spend inside the acorns on the ground after the premature
abscission of the seed (Bonal & Muñoz, 2007). Thus, we proposed an
increased livestock grazing on infested acorns during that period that
proved successful to reduce pest numbers (Canelo et al., 2021b). But it
is unclear whether this increased grazing could have undesired negative
side effects on arthropod communities coexisting with the pest and the
livestock.
Livestock plays a major role in ecosystem functioning (Eldridge et al.,
2016, 2017; Eskelinen et al., 2022), but grazing consequences on
biodiversity remain not fully understood (Filazzola et al., 2020; Yuan
et al., 2016). Some effects of large herbivores are direct, like
unintentional predation on insects while feeding (Bonal & Muñoz, 2007;
Gómez & González-Megías, 2002, 2007; Retamosa et al., 2004). By
contrast, others are indirect and mediated by biotic and abiotic
variables (Maestre et al., 2022). For example, grazing can alter the
richness, abundance, and species composition of different animal
communities through changes in plant cover and/or diversity (Adler et
al., 2001; Brambila et al., 2020; Kirk et al., 2019; Magnano et al.,
2019; Sims et al., 2019; et al., Song et al., 2020). Livestock also
increases soil hardness and changes its physicochemical properties
(Armas-Herrera et al., 2019; Zhang et al., 2020), as well as water
composition and quality (Kilgarriff et al., 2020; O’Callaghan et al.,
2018; O’Sullivan et al., 2019). Grazing effects are not always easy to
predict because they result from complex cascade effects (Abdala-Roberts
et al., 2019; Eldridge et al., 2016, 2017; Evans et al., 2015; Sankaran
& Augustine, 2004; Vandegehuchte et al., 2017), which transcend
ecosystem boundaries (Knight et al., 2005). Moreover, the variability of
direct and indirect effects, along with their interactions with
environmental factors (Vojta et al., 2020), may make the consequences of
grazing seem counterintuitive, as different organisms may respond
differently to the same disturbance events (Didham et al., 2009; Gossner
et al., 2016; Jackson et al., 2015). For oak dehesas, we proposed a
temporary localized increased grazing to reduce acorn pests, rotated
among different farm areas every 2 -3 years (Canelo et al. 2021b). In
principle, grazing would be expected to reduce arthropod numbers and
diversity compared to areas with no livestock. Nonetheless, it could
depend on grazing intensity, as generally predicted by the intermediate
disturbance hypothesis (IDH).
The intermediate disturbance hypothesis (IDH) states that species
diversity peaks at intermediate levels of environmental stress (Connell,
1978; Gao & Carmel, 2020; Roxburgh et al., 2004; Svensson et al., 2007;
Yan et al., 2015). On the one hand, high disturbances result in constant
selection pressure allowing only a few species to survive; on the other
hand, the lack of disturbance benefits just a few strong competitors.
However, in between, the competition between existing species decreases
and new species can rapidly colonize. Consequently, species richness
rises under intermediate disturbance regimes. This hypothesis has been
proposed to explain the high arthropod diversity recorded in areas with
moderate levels of grazing (Joubert et al., 2016; Kaltsas et al., 2013;
Kati et al., 2012; Lazaro et al., 2016a, 2016b; Qin et al., 2017; Winck
et al., 2019).
Livestock exclusion is seen as a good strategy to restore degraded
ecosystems (Kröpfl et al. 2011; Prober et al. 2011; Su et al. 2015). At
the same time, it is a very efficient way to evaluate the effects of
grazing on other organisms (Adams, 1975; Trigo et al., 2020; Wassie et
al., 2009). Such effects usually appear over time, and not always in a
linear fashion (Filazzola et al., 2020), hence, the use of exclosures of
different age is recommendable to study them. We did so in the present
study, building exclosures and, after one year, comparing them with
adjacent grazed areas and places in which livestock had been absent for
a longer period of time. We performed extensive and detailed analyses on
the effects on arthropod communities considering all trophic and
taxonomic levels. To achieve this ambitious goal, we used DNA
metabarcoding approach.
In arthropod biodiversity studies, species identification is always the
main challenge (Moretti et al., 2004). Gaps in taxonomic identification
are usually related to the nature of taxonomic knowledge: due to the
outstanding arthropod diversity, researchers need to focus on specific
taxonomic groups (Didham et al., 2009; Jackson et al., 2015), and
taxonomic expertise is not always available for every study.
Metabarcoding overcomes this taxonomic impediment, allowing species
identification at a community level by delimiting molecular operational
taxonomic units (MOTUs), which correspond to putative species (Gaytán et
al., 2020). These MOTUs can be determined to the Order, Family, Genus or
Species level depending on the availability of identified reference
barcodes. In this sense, using more than one marker for species
identification enhances detection likelihood (Kaunisto et al., 2017;
Wangensteen et al., 2018). For terrestrial arthropods, the two
mitochondrial genes cytochrome oxidase I (COI) and the 16S rRNA gene
(16S) have shown to complement each other very well (Marquina et al.,
2019a; Roger et al., 2022).
The general objective of this study is to assess, for the first time,
the effects of grazing on arthropod communities of Iberian oak dehesas
using DNA metabarcoding, along with the environmental factors that may
drive them. Our specific goals were: i) to compare the success of two
genetic markers (mitochondrial genes COI and 16S) used for arthropod
species detection by DNA metabarcoding; ii) evaluate changes in
vegetation structure after livestock exclusion and its effects on
arthropod diversity; and iii) assess arthropod species richness and
community composition across trophic levels throughout a gradient of
grazing exclusion.
MATERIAL AND METHODS