Content uploaded by Cristina Martinez-Garza
Author content
All content in this area was uploaded by Cristina Martinez-Garza on Jun 03, 2015
Content may be subject to copyright.
RESEARCH ARTICLE
Recovering More than Tree Cover: Herbivores
and Herbivory in a Restored Tropical Dry
Forest
Iris Juan-Baeza
1☯
, Cristina Martínez-Garza
2☯
, Ek del-Val
3☯
*
1 Facultad de Ciencias Biológicas, Universidad Autónoma del Estado de Morelos, Av. Universidad No.
1001, Col Chamilpa, Cuernavaca, Morelos, C.P. 62209, México, 2 Centro de Investigación en Biodiversidad
y Conservación, Universidad Autónoma del Estado de Morelos, Av. Universidad No. 1001, Col Chamilpa,
Cuernavaca, Morelos, C.P. 62209, México, 3 Instituto de Investigaciones en Ecosistemas y Sustentabilidad,
Universidad Nacional Autónoma de México, Antigua Carretera a Pátzcuaro No. 8701, Col. Ex-Hacienda de
San José de La Huerta, C.P. 58190, Morelia, Michoacán, México
☯ These authors contributed equally to this work.
* ekdelval@cieco.unam.mx
Abstract
Intense and chronic disturbance may arrest natural succession, reduce environmental quality
and lead to ecological interaction losses. Where natural succession does not occur, ecologi-
cal restoration aims to accelerate this process. While plant establishment and diversity is pro-
moted by restoration, few studies have evaluated the effect of restoration activities on
ecological processes and animal diversity. This study assessed herbivory and lepidopteran
diversity associated with two pioneer tree species growing in 4-year-old experimental restora-
tion plots in a tropical dry forest at Sierra de Huautla, in Morelos, Mexico. The study was car-
ried out during the rainy season of 2010 (July-October) in eleven 50 x 50 m plots in three
different habitats: cattle-excluded, cattle-excluded with restoration plantings, and cattle graz-
ing plots. At the beginning of the rainy season, 10 juveniles of Heliocarpus pallidus (Malva-
ceae) and Ipomoea pauciflora (Convolvulaceae) were selected in each plot (N = 110 trees).
Herbivory was measured in 10 leaves per plant at the end of the rainy season. To evaluate
richness and abundance of lepidopteran larvae, all plants were surveyed monthly. Herbivory
was similar among habitats and I. pauciflora showed a higher percentage of herbivory. A total
of 868 lepidopteran larvae from 65 morphospecies were recorded. The family with the highest
number of morphospecies (9 sp.) was Geometridae, while the most abundant family was
Saturnidae, with 427 individuals. Lepidopteran richness and abundance were significantly
higher in H. pallidus than in I. pauciflora. Lepidopteran richness was significantly higher in the
cattle-excluded plots, while abundance was significantly higher in the non-excluded plots.
After four years of cattle exclusion and the establishment of plantings, lepidopteran richness
increased 20 –fold in the excluded plots compared to the disturbed areas, whereas herbivory
levels were equally high in both restored and disturbed sites. Restoration with plantings and
exclusion of cattle and plantings was shown to be a successful strategy for attracting lepidop-
terans and cattle exclusion was the main factor explaining lepidopteran diversity.
PLOS ONE | DOI:10.1371/journal.pone.0128583 June 1, 2015 1/14
OPEN ACCESS
Citation: Juan-Baeza I, Martínez-Garza C, del-Val E
(2015) Recovering More than Tree Cover: Herbivores
and Herbivory in a Restored Tropical Dry Forest.
PLoS ONE 10(6): e0128583. doi:10.1371/journal.
pone.0128583
Academic Editor: Madhur Anand, University of
Guelph, CANADA
Received: January 29, 2015
Accepted: April 28, 2015
Published: June 1, 2015
Copyright: © 2015 Juan-Baeza et al. This is an open
access article distributed under the terms of the
Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any
medium, provided the original author and source are
credited.
Data Availability Statement: All relevant data are
within the paper and its Supporting Information files.
Funding: Universidad Nacional Autónoma de México
(PAPIIT-IN208610) fundes EDV, the Mexican
Program for the Improvement of Professors awarded
to CM-G (PROMEP, 103.5/05/1901), Cuerpo
Académico de Ecología Evolutiva (PROMEP) and the
National Council of Science and Technology of
Mexico (Grant # 80027), awarded to CM-G. The
funders had no role in study design, data collection
and analysis, decision to publish, or preparation of
the manuscript.
Introduction
Intense and chronic disturbance may arrest natural succession, reduce environmental quality
and lead to the loss of ecological interactions [1,2]. Ecological restoration aims to accelerate the
process of natural succession, in cases where it does not occur [ 3, 4]. Manipulation of natural
succession by planting late-successional tree species may increase the richness and density of
plants and animals in early-successional environments. However, there are few studies that
have evaluated the significance of these techniques for faunal diversity and ecological processes
[5–10].
The selective pressure imposed by herbivory determines the survival and fitness of plants
[11,12]. Many restoration efforts are hampered by herbivores that may impede plant develop-
ment; vertebrates in particular have been involved in many restoration failures [13–15]. In
tropical systems lepidopterans are very diverse and function as important herbivores since they
consume significant quantities of leaf tissue [16,17]. As adults, moths and butterflies participate
in the vital process of pollination [18,19]; in fact in tropical dry forests they accoun t for the pol-
lination of at least 10% of the plant species [20]. It is therefore desirable that functional restor ed
ecosystems have a complement of herbivores that will becom e effective pollinators when they
reach adulthood.
Habitat perturbation has shown a significant effect on plant-herbivore interactions [21]; dis-
turbed habitats present greater levels of herbivory when generalist herbivores are promoted
[22,23] or, conversely, present reduced herbivory where the main herbivores are specialists and
have disappeared following habitat perturbation [24]. Few studies have evaluated the presence
of lepidopterans in restoration projects (but see [5,25]) and therefore there is an urgent need to
determine whether lepidopterans are able to rea ch to restored sites and, if so, whether they are
able to provide the same ecological services as is the case in natural habitats.
The effect of cattle in dry forest has been evaluated for plants and animals. Cattle browse un-
derstory herbs and grasses and also the lower foliage of trees negatively affecting tree diversity
[26,27] and bird populations [28,29]. Cattle ranching activities are also known to affect the
composition of trees [30], cacti [31], herbs and grasses [32]. Decreases in biodiversity also have
a negative effects on ecosystem processes but changes in composition are expected to have
higher negative impacts [33]. For example, changes in tree composition in dry forest as a result
of cattle ranching can alter seed dispersal patterns [34,35] and also favor the establishment of
weeds dispersed by cattle [32,36]. Furthermore, cattle may decrease the size of large cacti, due
to trampling, affecting the frugivorous animals that depend on th em [31]. Cattle affect dry for-
est functioning by changing the composition of plants and indirectly through their effects on
plant-animal interactions, such as dispersal.
We evaluated Lepidopteran richness, abundance and composition during the rainy season
of 2010 (July-November) in a 30 year-old secondary tropical dry forest at Sierra de Huautla in
Central Mexico in order to estimate the potential recovery of ecosystem function after 4 years
of exclusion of chronic disturbance by cattle. We addressed the following specific questions: 1)
How does the lepidopteran larval diversity associated with two pioneer species differ between
habitats? and 2) how does accumulated leaf herbivory in two pioneer species differ between
three habitats: two restoration settings (planting s and cattle exclus ions) and disturbed areas?
We proposed the following hypotheses: 1) increased lepidopteran richness and density will
occur in the individuals growing in the restoration plantings and the lepidopteran communities
will differ in terms of species and habitats and 2) there will be higher accumulated herbivory in
the individuals growing in the restoration plantings,
Lepidopterans in a Restored Mexican Dry Forest
PLOS ONE | DOI:10.1371/journal.pone.0128583 June 1, 2015 2/14
Competing Interests: The authors have declared
that no competing interests exist.
Materials and Methods
Study site
This study was carried out close to the town of El Limón de Cuauchichinola (1220 m a.s.l.), lo-
cated within the Sierra de Huautla Biosphere Reserve (SHBR), in the south of the state of More-
los, central Mexico. The SHBR (18° 20’ 10”, 18° 34’ 20” N and 98° 51‘ 20”, 98° 08‘ 15” W)
comprises 59,030 ha, in which the main vegetation type is tropical dry forest. Mean annual
temperature is 24.5°C and average total rainfall (average for 1971–2000) is 817.5 mm (CONA-
GUA, Gerencia Regional Balsas, http://smn.cna.gob.mx/climatologia/normales/estacion/mor/
NORMAL17057.TXT), with ~90% of this rainfall occurring between late May and October.
Most of the trees shed their leaves during the dry season (November to May). The soils are
shallow (< 30 cm in depth) and sandy-loam in texture.
In the SHBR, 939 native species of vascular plants from 478 genera and 130 families have
been recorded [37]. Of these, 157 species are trees or shrubs [38]. This forest has one stratum
of trees 8 to 12 m in height with convex or flat canopies. Most of the leaves of these trees are
compound with small leaflets [39]. In terms of number of species, the most important families
are Fabaceae, Poaceae, Asteraceae and Burseraceae. The most common canopy trees are Con-
zattia multiflora (B.L. Rob.) Standl., Lysiloma acapulcense (Kunth) Benth., Lysiloma divarica-
tum (Jacq.) J.F. Macbr. (Fabaceae), Bursera spp. (Burseraceae) and Ceiba spp. (Bombacaceae)
[37].
Land history and description
The landscape of the SHBR comprises a mosaic of primary and secondary dry tropical forest
surrounded by agricultural land and small towns. Forty one percent of the area of the reserve
has been classified as intact or under good conservation status, whith 22.4% classified as well-
conserved, based on aerial images and vegetation surveys conduceted at selected points. The re-
maining 36.2% presents different degrees of degradation and most of this fraction is used for
economic activities [37,40]. In El Limón de Cuauchichinola (referred to hereafter as El Limón),
large parts of the forest were cleared > 30 years ago, used for maize cultivation for ca. 6 years
and subsequently abandoned. Since abandonment, the resulting secondary forest has been
used for wood extraction and extensive cattle ranching (G. Pacheco, pers. comm). In El Limon,
56% of the area is currently covered with intact forest or is under good conservation status,
19% is perturbed dry forest, 12% is secondary vegetation and the remain ing 13% is dedicated
to agriculture and housing [41]. In this forest, approximately 50% of the organic matter is con-
centrated within the upper 10 cm of the soil profile (E. Solís, unpub. data) and the tree density
of individuals 5 cm of diameter at breast height (1.3 m) is 264 individuals/ha from 14 tree
species. Of these, Acacia cochliacantha (Hum b&Bonpl) (Fabaceae), Ipomoea pauciflora
(Mart&Gal) (Convolvulaceae), Acacia farnesiana (Willd) (Fabaceae) and Mimosa benthamii
(Macbride) (Fabaceae) are dominant [42]. During the rainy season, ca. 600 head of livestock
forage in this secondary forest (~ 7 head/ha); most of these animals are brought to the site from
neighboring towns. Cattle are maintained in farms during the dry season but goats, pigs and
horses are left to forage in the forest daily throughout the year. Grazing reduces the biomass in
the understory by more than 70% [32]. As part of an experimental restoration project to pro-
mote the coexistence of cattle grazing activities and forest biodiversity, eight 50 x 50 m plots
were excluded from disturbance in the form of wood extraction and grazing by large domestic
livestock since January 2006. Exclusion was accomplished with a fence of 4 lines of barbed
wire, with an additional 75 cm of chicken wire netting attached to the lower part of the fence to
exclude smaller domestic animals. In 2010 an additional electric fence was installed outside the
Lepidopterans in a Restored Mexican Dry Forest
PLOS ONE | DOI:10.1371/journal.pone.0128583 June 1, 2015 3/14
original fence. Distances among plots ranged from 0.08 to 1.59 km (0.72 ± 0.46;
mean ± standard deviation). The closest old-growth forests were located at a range from 0.11
to 0.26 km (0.21 km ± 0.05) from the plots.
Natural succession was manipulated in half of the sites with enrichment plantings of 18 na-
tive tree species found in late-successional env ironments and two early-successional species
(for experimental details see [23][22][21][20]). The entire experiment consisted of four plots
fenced against cattle where natural succession took place, four plots fenced against cattle with
additional plantings and three control plots where cat tle freely grazed with some remaining
adult trees.
In June of 2010, two early successional species that occur naturally in all of the sites (Ipo-
moea pauciflora [Convolvulaceae] and Heliocarpus pallidus [Tiliaceae]) were selected for this
study. Ten naturally recruited juveniles of each species (average height 0.5–2 m) per plot were
selected in the eight plots. Due to cattle ranching activities, no natural recruitment occurred
outside the plots and consequently only adult individuals of the species selected were found.
Herbivore sampling
Lepidopteran larvae were sampled monthly during the wet season (July—November 2010),
when both leaves and larvae are present. We surveyed all the leaves in the chosen trees from.
H. pallidus and I. pauciflora up to a height of 2 m in order to find lepidopteran larvae. In taller
trees, we sub-sampled 3 branches per individual [43]. In each survey, unknown lepidopteran
larvae were reared in the laboratory in order to taxonomically identify the adults. When there
were several larvae of the same species eating in the same plant only three larvae were taken to
the lab. When the larvae were already identified, they were left in the field.
Herbivory rates
At the end of the growing season, before the trees started to shed their leaves, 10 leaves per in-
dividual tree were collected and transported to the laboratory in order to evaluate herbivory
rates at the different sites. Each leaf was measured with a portable leaf area scanner (CI-202
Leaf Area Meter–CID, Inc) to calculate the area of leaf removed and accumulated herbivory
percentage values were calculated per plant . All sampling procedures and experimental manip-
ulations were approved as part of obtaining the field permit from the General Direction of
Mexican Wildlife (permit number SGPA/DGVS/ 02693/10 to Cristina Martínez-Garza).
Statistical analysis
Statistical analyses for herbivory were performed with a factorial analysis of variance
(ANOVA). The dependent variable was the average leaf area consumed per plant and the inde-
pendent variables were species with two levels (I. pauciflora and H. pallidus), habitat with three
levels: non-excluded (N = 3 sites), excluded (N = 4 sites) and excluded with plantings (N = 4
sites) and we had ten individuals per species per habitat. The average leaf area per plant con-
sumed was transformed with the arcsine square root of the proportion of herbivory + 0.1 in
order to comply with the assumptions for ANOVA.
In order to analyze Lepidopteran diversity, two repeated measures ANOVAs were per-
formed. The dependent variables were richness and abundance of lepidopteran larvae and the
independent variables were species identity with two levels and habitat with three levels (exclu-
sions with planting, exclusions with no planting and non-excluded plots). The repeated mea-
sure was sampling month, with four levels (July, August, September and October)
To analyze changes in lepidopteran species composition associated with restoration treat-
ment, we also performed a clustering analysis using a distance matrix obtained with Bray-
Lepidopterans in a Restored Mexican Dry Forest
PLOS ONE | DOI:10.1371/journal.pone.0128583 June 1, 2015 4/14
Curtis index of total lepidopteran abundance per species per treatment and plotted the result-
ing dendrogram showing Bray-Curtis distances and, in order to obtain the tree significance, we
performed a Mantel test with 100 permutations using the “vegan” library in R 2.14.0 [44].
Results
We recorded 868 caterpillars from 64 morphospecies during the four months of study. To date,
15 of these have been identified to family level, four to genus level and nine to species level (S1
Table). We recorded species belonging to 11 families; the highest number of species (nine spe-
cies) was from the Geometridae, whereas the highest caterpillar abundance (429) was from the
Saturnidae. Most of these individuals (422) were from one species: Arsenura armida. Only two
species were found to be associated with both plants: Orgyia sp. (Lymantridae) and Hyper-
compe suffusa (Arctidae).
Lepidopteran richness
Richness of lepidopterans associated with H. pallidus (3.05 ± 0.45 morphospecies) did not dif-
fer significantly from those associated with I. pauciflora (2.09 ± 0.32 morphospecies; F
(1, 16)
=
3.81, P > 0.07) and this was true for all of the habitats (species
habitat: F
(2,16)
= 1.21,
P > 0.32). However, plants growing in the excluded plots and those growing in the plantings
presented 20 times higher richness (3.63 ± 0.45 and 3.31 ± 0.45 morphospecies respectively)
than those found in plants growing in the non-excluded plots (0.17 ± 0.08 morphospecies).
The post hoc Tukey test revealed that excluded plots and plantings showed similar lepidopteran
richness and this was found to be significantly higher than that of the non-excluded plots
(F
(2,16)
= 27.65, P<< 0.001, Fig 1).
Lepidopteran richness decreased throughout the rainy season (F
(3, 48)
= 23.91, p<<0.0001).
The post Hoc Tukey test revealed that lepidopteran richness was similar in July (4.00 ± 0.71
Fig 1. Lepidopteran richness associated with tree species in different experimental treatments.
Lepidopteran richness (log10 +1) associated with trees of I. pauciflora and H. pallidus in excluded sites, with
and without plantations, and disturbed sites with cattle grazing in the tropical dry forest of Huautla, Morelos,
Mexico. Vertical lines represent a 95% confidence interval. Different letters represent significant differences
evaluated with a post hoc Tukey test.
doi:10.1371/journal.pone.0128583.g001
Lepidopterans in a Restored Mexican Dry Forest
PLOS ONE | DOI:10.1371/journal.pone.0128583 June 1, 2015 5/14
morphospecies), August (2.77 ± 0.37) and September (2.86 ± 0.49) but significantly lower in
October (0.63 ± 0.24). Lepidopteran richness showed a similar monthly pattern for the two
tree species evaluated (F
(3,48)
= 1.79, p> 0.16). Interaction between habitat and month was sig-
nificant (F
(6, 48)
= 6.15, p<< 0.001), revealing that the non-excluded plots had a significantly
lower lepidopteran richness in July and September whereas, in August and October, richness
was similar among the three habitats (Fig 2). The interaction between species and habitat was
not significant (F
(6,48)
= 0.57, p> 0.76).
Lepidopteran abundance
Lepidopterans were almost four times more abundant in H. pallidus (15.61 ± 4.98 individuals)
compared to I. pauciflora (4.32 ± 0.76 individuals; F
(1, 16)
= 32.17, p<< 0.001; Fig 3). Lepidop-
teran abundance was similar in the three habitats (F
(2, 16)
= 3.08, p = 0.07). Statistical analysis
showed that lepidopteran abundance differed over the course of the four months (F
(3,48)
=
47.96, p<<0.001). The post hoc Tukey test revealed that the plants had a higher and similar
abundance during July and August (12.41 ± 2.78 and 22.59 ± 8.95 individuals respectively),
while abundance was intermediate in September (3.95 ± 0.76 individuals) and lowest in Octo-
ber (0.91 ± 0.37 individuals). Interaction between species and habitat was significant (F
(2, 16)
=
10.70, p<0.001) and the post hoc Tukey test revealed that H. pallidus presented a similar lepi-
dopteran abundance in all three habitats, whereas I. pauciflora presented a lower lepidopteran
abundance in the non-excluded plots (Fig 4).
Similarity index
Similarity analysis of the lepidopteran commun ity showed a tree with three main branches; the
first and the second grouped Lepidopteran communities associated with H. pallidus and I. pau-
ciflora growing in perturbed plots, the third branch grouped Lepidopterans associated with
Fig 2. Temporal richness of Lepidopteran larvae. Temporal pattern of lepidopteran larvae richness
associated with trees of I. pauciflora and H. pallidus in excluded sites, with and without plantations, and
perturbed sites with access for livestock during the rainy season from July to October 2010 in the Sierra
seasonal forest Huautla, Morelos, Mexico. Vertical lines represent a 95% confidence interval. Asterisks
represent significant differences between treatments in the same month evaluated with a post hoc
Tukey test.
doi:10.1371/journal.pone.0128583.g002
Lepidopterans in a Restored Mexican Dry Forest
PLOS ONE | DOI:10.1371/journal.pone.0128583 June 1, 2015 6/14
plants in cattle-excluded plots and it is further subdivided in the communities associated with
I. pauciflora or H. pallidus. A Mantel permutation test (100 permutations) showed the dendro-
gram to be significant (r = 0.98, p = 0.02, Fig 5).
Fig 3. Lepidopteran abundance associated with tree species. Abundance (Log10 (number of individuals
per tree + 1) of lepidopteran larvae associated with the trees H. pallidus and I. pauciflora in a seasonal forest
in Sierra de Huautla, Morelos, Mexico. Vertical lines represent a 95% confidence interval.
doi:10.1371/journal.pone.0128583.g003
Fig 4. Lepidopteran abundance associated with tree species in different experimental treatments.
Abundance (Log10 (num ber of individuals per tree + 1) of lepidopteran larvae associated with t rees of
I. pauciflora and H. pa llidus in excluded sites, with and without plantations, and perturbed sit es with access
for livestock during the rainy season from July to October 2010 in the seasonal forest of Sierra de Huautla,
Morelos, Mexico. Vert ical lines represent 95 % of the data . The asterisk represents significant differences
evaluated with post hoc Tukey test, mainly due to t he super abundance of A. armida.
doi:10.1371/journal.pone.0128583.g004
Lepidopterans in a Restored Mexican Dry Forest
PLOS ONE | DOI:10.1371/journal.pone.0128583 June 1, 2015 7/14
Herbivory
The percentage of accumulated herbivory was significantly higher in I . pauciflora
(30.81 ± 1.38%) than in H. pallidus (26.14 ± 0.96%; F
(1,16)
= 7.55, P < 0.01; Fig 6). Herbivory
did not differ significantly among habitats (F
(2,16)
= 1.14, P > 0.34) and the interaction between
species and habitat was also non-significant (F
(1,16)
= 2.22, P > 0.14).
Discussion
Ecosystem restoration is urgently required in many places of the world where land degradation
is pervasive. The tropical dry forests (TDF) of Mexico ha ve suffered greatly from deforestation
and land use change over the last 50 years and at present only 37% of the original area remains
in good condition [40,45]. Since TDFs in Mexico contain 20% of the national biodiversity [46],
the recovery of such ecosystems is a national priority; however, restoration in TDFs in Mexico
is still incipient and few examples are cited in the literature [5, 35,47,48].
In this study, we evaluated a restoration project experience that began in 2006, where cattle
were excluded from several plots and a tree enrichment treatment was incorporated. We found
that cattle exclusion is crucial to the return of lepidopteran diversity to the area; the fenced
plots had 20 times more lepidopteran species than the control cattle grazed sites. Other studies
Fig 5. Bray- Curtis similarities between Lepidopterans associated with different plant species and
experimental treatments. Dendrogram showing Bray-Curtis simila rities in lepi dopteran community
between host plants and sites. Ip = I. pauciflora,Hp=H. pallidus; perturbed = control sites with cattle
grazing; w/p = excluded plots with plantation; wo/p = excl uded plots with out plantation
doi:10.1371/journal.pone.0128583.g005
Lepidopterans in a Restored Mexican Dry Forest
PLOS ONE | DOI:10.1371/journal.pone.0128583 June 1, 2015 8/14
evaluating forest restoration have also found that lepidopterans are able to recover following
restoration; Hernández et al. [5] found that after eigh t years of restoration in a TDF on the
Mexican Pacific coast, the lepidopteran community was similar to that of mature forests. In
other ecosystems, Waltz and Covington [10] restored the understory of a ponderosa-pine for-
est and found a 3.5% increase in butterfly species richness after 3 years of treatment. Several
studies have evaluated whether selectively logged tropical forests could host butterfly commu-
nities similar to those of pristine forests and have found that, although species richness per
family was equivalent, species composition replacement took place [49,50]. In this study, we
also found species replacement between sites (high beta diversity). Other studies assessing the
impact of human intervention in forests using butterflies as indicators have found that mem-
bers of the family Nymphalidae (Satyrinae and Morphinae) in Borneo were absent in logged
forests [51], highlighting the importance of habitat heterogeneity for community conservation.
Summerville et al [8,9] compared lepidopteran community assemblages in restored prairies of
different ages and found very high inter-year variation but that the rarefaction curves showed
that older restored site diversity converges with that of undisturbed sites. However, in a previ-
ous study, Holl [7] argued that the utility of adult butterflies as indic ators of restoration may be
limited because of their high mobility. Lepidopteran differences in restored vs. perturbed sites
when plant food is available may be related to predator abundance [52,53] and/or changes in
microenvironmental conditions, particularly temperature increases and humidity decreases
with perturbation [23,54,55]. The higher richness in restored sites of lepidopteran larvae when
their mobility is minimal represents an increase in the habitat quality and availability of re-
sources for this group of insects.
In this study, fencing against cattle was more important than enrichment plantings for the
lepidopteran community; there were no significant differences between lepidopteran richness
and abundance under natural successional vs. manipulated succession with plantings. Surpris-
ingly, the most abundant species was not found within the fenced area where higher lepidop-
teran diversity was found, but in the control area where cattle were grazing. The species A.
Fig 6. Herbivory in I. pauciflora and H. pallidus. Herbivory (arcsin square root transformed (proportion of
leaf area removed +0.1)) recorded in plants of I. pauciflora and H. pallidus during the rainy season of 2010 in
the tropical dry forest of Huautla, Morelos, Mexico. Vertical lines represent a 95% confidence interval.
doi:10.1371/journal.pone.0128583.g006
Lepidopterans in a Restored Mexican Dry Forest
PLOS ONE | DOI:10.1371/journal.pone.0128583 June 1, 2015 9/14
armida was favored by disturbance in H. pallidus. Coincidently this species is reported to be
used as food by local people [56]; larvae are harvested during the wet season and consumed.
Arsenura armida is highly appreciated because its larvae are large (10cm) with high protein
content (51.3%) and occur in large numbers [56]. In the state of Veracruz, Mexico, H. pallidus
trees are left to grow in perturbed areas to ensure a harvest of A. armida [57]. In this case, A.
armida could be considered an indicator of disturbance, and its higher abundances may im-
pede other lepidopterans from establishing in colonizing trees.
Lepidopteran community comp osition associated with the two tree species studied con-
firmed that lepidopterans select their food according to host plant identity more than habitat
type. The similarity analysis showed that lepidopteran communities associated with I. pauci-
flora are grouped and separated from those associated with H. pallidus. Another study with lep-
idopteran larvae comparing natural succession vs. restored sites found similar results in the
tropical dry forest of Chamela [5]; moths and butterflies select food for their young according
to plant identity regardless of habitat quality. Chemical plant composition appears to dictate
plant-herbivore associations independently of site conditions [17], therefore if lepidopteran
population sources exist in the surrounding (conserved) areas, herbivores can arrive to the re-
stored site following establishment of their host plants. Since we only studied lepidopterans as-
sociated with two tree species, more investigations considering larvae associated with all tree
species are required to determine the response of the whole lepidopteran community.
Temporal patterns of lepidopteran richness and abundance followed the rainy season; a
greater number of species and individuals were found at the beginning of the season and then
presented a constant decline throughout the season, with the exception of plants in the per-
turbed sites where associated lepidopteran richness was constant since only one species was
found. Lepidopteran diversity explored in other tropical dry forests of Mexico has been found
to present a peak in abundance and richness in the middle of the rainy season [58], thus con-
trasting with the findings of this study where the peak occurred at the beginning of the rainy
season; this may be due to the difference in precipitation since in Huautla there is 100 mm less
rain than in Chamela.
Herbivory
Apart from Lepidopteran diversity, this study showed that herbivory rates were different be-
tween plant species. Ipomoea pauciflora had higher levels of herbivory regardless of its report-
edly strong chemical defenses [59]. A specialist herbivore may therefore be responsible for the
observed removal of plant tissue [60]. Other studies have found that herbivory rates depend on
habitat resource availability [61] because plants in richer habitats invest their resources in
growth instead of defense. A study with H. pallidus in the TDF of Chamela showed higher lev-
els of herbivory in riparian compared to deciduous habitats [62], which contrasts with our
findings; however, the highest percentage of leaf removal in Chamela (26%) was similar to our
results (25%). In this study, both species showed similar herbivory rates throughout the sites,
implying that a) environmental resources are similar between studied habitats and plants have
similar nutritional characteristics, or b) the lepidopteran communities differ among sites. The
mechanisms behind this pattern remain to be explored. It is interesting that we found a peak in
lepidopteran abundance in August in the H. pallidus plants growing in perturbed sites, but this
peak did not translate into an increased herbivory rate in th ose plants. It is possible that the
plants compensated for herbivore attack by producing more leaves throughout the season [63]
and, since we only measured leaf consumption at the end of the year, we may have missed
these more subtle differences in the patterns of herbivory.
Lepidopterans in a Restored Mexican Dry Forest
PLOS ONE | DOI:10.1371/journal.pone.0128583 June 1, 2015 10 / 14
Regarding restoration, this study shows that lepidopteran larvae attracted to sites with en-
riched plantings perform similarly to those attract ed to sites with natural recruitment and
planting is therefore an appropriate strategy for the more rapid incorporation of biodiversity
into perturbed areas. While lepidopterans in restored sites showed no difference with succes-
sional plots, investigations with other groups have found that more species are arriving to the
restoration plots [32]. Vertebrates are visiting restored sites more frequently and dispersing
more seeds from different species [64], therefore increases in plant diversity will eventually
translate in higher lepidopteran diversity. The entire lepidopteran community associated with
all plant species under the different management treatments remains to be explored in order to
corroborate our hypothesis. This information would provide tools with which to accelerate
ecosystem restoration including other trophic guilds.
Studies of the impact of cattle ranching activities in tropical dry forest suggest the exclusion
of cattle in order to favor ecosystem recovery [2,48,65,66], and therefore cattle grazing should
be prevented. In our study, following four years of cattle exclusion and the establishment of
plantings, lepidopteran richness was found to have increased 20-fold in the excluded plots
compared to perturbed areas, whereas herbivory levels were equally high in restored and per-
turbed sites. Restoration was a successful strategy for attracting lepidopterans and cattle exclu-
sion was the main factor that explained lepidopteran diversity
Supporting Information
S1 Table. Abundance of each identified lepidopteran morphospecies associated with H. pal-
lidus and I. pauciflora in Sierra de Huautla, Morelos, Mexico.
(DOCX)
Acknowledgments
We are grateful to Gerardo Pacheco and René Gadea for expert field assistance. The assistance
of Keith MacMillan as English reviewer significantly improved the manuscript.
Author Contributions
Conceived and designed the experiments: EDV CMG. Performed the experiments: IJB CMG.
Analyzed the data: IJB CMG EDV. Contributed reagents/materials/analysis tools: CMG EDV.
Wrote the paper: CMG EDV.
References
1. Olff H, Ritchie ME (1998) Effects of herbivores on grassland plant diversity. Trends in Ecology & Evolu-
tion 13: 261–265.
2. Burgos A, Maass JM (2004) Vegetation change associated with land-use in tropical dry forest areas of
Western Mexico. Agriculture Ecosystems and Envrionment 104: 475–481.
3. Howe HF, Martínez-Garza C (2014) Restoration as experiment. Botanical Sciences 92.
4. SER (2007) Principios de SER Internacional sobre la Restauración Ecológica. USA: Society for Eco-
logical Restoration International. 15 p.
5. Hernández Y, Boege K, Lindig-Cisneros R, del-Val E (2014) Are ecological processes reestablished
after restoration? A comparison of herbivory rates and herbivore communities in a restored vs. Succes-
sional site of a tropical dry forest. Southwestern Naturalist.
6. Chapman CA, Chapman LJ (1999) Forest restoration in abandoned agricultural land: a case study from
East Africa. Conservation Biology 13: 1301–1311.
7. Holl KD (1995) Nectar Resources and Their Influence on Butterfly Communities on Reclaimed Coal
Surface Mines. Restoration Ecology 3: 76–85.
Lepidopterans in a Restored Mexican Dry Forest
PLOS ONE | DOI:10.1371/journal.pone.0128583 June 1, 2015 11 / 14
8. Summerville KS, Steichen M, Lewis MN (2005) Restoring Lepidopteran Communities to Oak Savan-
nas: Contrasting Influences of Habitat Quantity and Quality. Restoration Ecology 13: 120–128.
9. Summerville KS, Bonte AC, Fox LC (2007) Short-Term Temporal Effects on Community Structure of
Lepidoptera in Restored and Remnant Tallgrass Prairies. Restoration Ecology 15: 179–188.
10. Waltz AEM, Covington WW (2004) Ecological restoration treatments increase butterfly richness and
abundance: mechanisms of response. Restoration Ecology 12: 85–96.
11. Marquis R (1984) Leaf Herbivores Decrease Fitness of a Tropical Plant. Science 226: 537–539. PMID:
17821511
12. Crawley MJ (1983) Herbivory: the dynamics of animal-plant interactions. Oxford: Blackwell Scientific
Publications.
13. Blanco-García A, Lindig-Cisneros R (2005) Incorporating Restoration in Sustainable Forestry Manage-
ment: Using Pine-Bark Mulch to Improve Native Species Establishment on TephraDeposits. Restora-
tion Ecology 13: 703–709.
14. King SL, Keeland BD (1999) Evaluation of reforestation in the lower Mississippi river alluvial valley.
Restoration Ecology 7: 348–359.
15. Sweeney BW, Czapka SJ, Petrow CA (2007) How planting method, weed abatement, and herbivory af-
fect afforestation success. Southern Journal of Applied Forestry 31: 85–97.
16. Novotny V, Drozd P, Miller SE, Kulfan M, Janda M, Basset Y, et al. (2006) Why are there so many spe-
cies of herbivorous insects in tropical rainforests? Science 313: 1115–1118. PMID: 16840659
17. Dyer LA, Singer MS, Lill JT, Stireman JOI, Gentry GL, Marquis RJ, et al. (2007) Host specificity of Lepi-
doptera in tropical and temperate forests. Nature 448: 696–700. PMID: 17687325
18. Bawa KS, Bullock SH, Perry DR, Coville RE, Grayum MH (1985) Reproductive biology of tropical low-
land rain forest trees. II. Pollination s ystems. American Journal of Botany 72: 346–356.
19. MacGreggor CJ, Pocock MJO, Fox R, Evans DM (2014) Pollination by nocturnal Lepidoptera, and the
effects of light pollution: a review. Ecological Entomology.
20. Haber WA, FRankie GW (1989) A Tropical Hawkmoth Community: Costa Rican Dry Forest Sphingidae.
Biotropica 21: 155–172.
21. Kremen C, Colwell RK, Erwin TL, Murphy DD, Noss R, Sanjayan MA (1993) Terrestrial arthropod as-
semblages: their use in conservation planning. Conservation Biology 7: 796–808.
22. Gutiérrez-Granados G, Dirzo R (2009) Remoción de semillas, herbivoría y reclutamiento de plántulas
de Brosimum alicastrum (Moraceae) en sitios con manejo forestal contrastante de la selva maya, Quin-
tana Roo, México. Boletín de la Sociedad Botánica de México 85: 51–58.
23. Ruiz-Guerra B, Guevara R, Mariano AN, Dirzo R (2009) Insect herbivory declines with forest fragmen-
tation and covaries with plant regeneration mode: evidence from a Mexican tropical rain forest. Oikos
119: 317–325.
24. Arnold EA, Asquith NM (2002) Herbivory in a fragmented tropical forest: patterns from islands at Lago
Gatún, Panama. Biodiversity and Conservation 11: 1663–1680.
25. Forup ML, Memmott J (2005) The restoration of plant pollinator interactions in hay meadows. Restora-
tion Ecology 13: 265–274.
26. Roth LC (1999) Anthropogenic change in subtropical dry forest during a century of settlement in Jaiqui
Picado, Santiago Province, Dominican Republic. Journal of Biogeography 26: 739–759.
27. Sánchez-Velázquez LR, Hernández-Vargas G, Carranza-M MA, Pineda-López MR, Cuevas-G R, Ara-
gón-C F (2002) Estructura arbórea del bosque tropical caducifolio usado para la ganadería extensiva
en el norte de la Sierra de Manantlán, México. Antagonismos de usos. Polibotanica 12: 25–
46.
28. VanderWerf EA (2004) Demography of Hawai'i 'Elepaio: Variation with habitat disturbance and popula-
tion density. Ecology 85: 770–783.
29. Piana RP, Marsden SJ (2014) Impacts of cattle grazing on forest structure and raptor distribution within
a neotropical protected area. Biodiversity and Conservation 23: 559–572.
30. Stern M, Quesada M, Stoner KE (2002) Changes in composition and structure of a tropical dry forest
following intermittent cattle grazing. Revista De Biologia Tropical 50: 1021–1034. PMID: 12947586
31. Arias-Medellín LA, Flores-Palacios A, Martínez-Garza C (2014) Cacti community structure in a tropical
Mexican dry forest under chronic disturbance. Botanical Sciences 92: 405–415.
32. de la O-Toriz J, Maldonado B, Martínez-Garza C (2012) Efecto de la perturbación en la comunidad de
herbáceas nativas y ruderales de una Selva estacional Mexicana. Botanical Sciences 90: 469–480.
33. Díaz S, Tilman D, Fargione J (2005) Biodiversity Regulation of Ecosystem Services. In: Hassan R,
Scholes R, Ash N, editors. Ecosystems and Human Well-being: Current State and Trends Washington,
USA: Island Press.
Lepidopterans in a Restored Mexican Dry Forest
PLOS ONE | DOI:10.1371/journal.pone.0128583 June 1, 2015 12 / 14
34. Ceccon E, Hernádez P (2009) Seed rain dynamics following disturbance exclusion in a secondary trop-
ical dry forest in Morelos, Mexico. Revista De Biologia Tropical 57: 257–269. PMID: 19637705
35. Martínez-Garza C, Osorio-Beristain M, Valenzuela-Galván D, Nicolás-Medina A (2011) Intra and inter-
annual variation in seed rain in a secondary dry tropical forest excluded from chronic disturbance. For-
est Ecology and Management 262: 2207–2218.
36. Miceli-Mendez CL, Ferguson BG, Ramírez-Marcial N (2008) Seed dispersal by cattle: Natural, history
and applications to Neotropical Forest Restoration and Agroforestry. In: Myster RW, M-M C.L., Fergu-
son BG, Ramirez-Marcial N, editors. Post-Agricultural Succession in the Neotropics. New York:
Springer. pp. 165–191.
37. Dorado O, Maldonado B, Arias DM, Sorani V, Ramírez R, Leyva E, et al. (2005) Programa de Conser-
vación y Manejo Reserva de la Biosfera Sierra Huautla México. Mexico: CONANP. 210 p.
38. Conanp-Semarnat (2005) Programa de Conservación y Manejo de la Reserva de la Biosfera Sierra de
Huautla, México. 202 p.
39. Rzedowski J (1978) Vegetación de México. Mexico: Limusa. 432 p. PMID: 4419188
40. Trejo I, Dirzo R (2000) Deforestation of seasonally dry tropical forest: a national and local analysis in
Mexico. Biological Conservation 94: 133–142.
41. De Leon-Ibarra MA (2005) Fenología de especies de plantas con frutos carnosos y disponibilidad espa-
cial y temporal de este recurso en la Reserva de la Biosfera Sierra de Huautla: Implicaciones para los
vertebrados [Bachelor Dissertation]. Cuernavaca, Morelos, Mexico: Universidad Autónoma del Estado
de Morelos. 76 p.
42. Martínez-Garza C, Osorio-Beristain M, Valenzuela-Galván D, Nicolás-Medina A (2011) Intra and inter-
annual variation in seed rain in a secondary dry tropical forest excluded from chronic disturbance. For-
est Ecology and Management 262: 2207–2218.
43. Villa-Galaviz E, Boege K, del-Val E (2012) Resilience in plant-herbivore networks during secondary
succession. PLOS-one 7: e53009. doi: 10.1371/journal.pone.0053009 PMID: 23300846
44. RDevelopmentCoreTeam (2011) R: a language and environment for statistical computing R foundation
for statistical computing. Viena, Austria.
45. Trejo I, Dirzo R (2002) Floristic diversity of Mexican seasonally dry tropical forests. Biodiversity and
Conservation 11: 2048–2063.
46. Challenger A, Soberón J (2008) Los ecosistemas terrestres. In: Conabio, editor. Capital natural de Mé-
xico, vol 1: Conocimiento actual de la biodiversidad: Conabio. pp. 87–108.
47. Vieira DLM, Scariot A (2006) Principles of Natural Regeneration of Tropical Dry Forests for Restoration.
Restoration Ecology 14: 11–20.
48. Ceccon E, Hernández P (2009) Seed rain dynamics following disturbance exclusion in a secondary
tropical dry forest in Morelos, Mexico. Revista De Biologia Tropical 57: 257–269. PMID: 19637705
49. Lewis OT (2001) Effect of experimental selective logging on tropical butterflies. Conservation Biology
15: 389–400.
50. Hamer KC, Hill JK, Benedick S, Mustaffa N, Sherratt TN, Maryati M, et al. (2003) Ecology of butterflies
in natural and selectively logged forests of northern Borneo: the importance of habitat heterogeneity.
Journal of Applied Ecology 40: 150–162.
51. Willott SJ, Lim DC, Compton SG, Sutton SL (2000) Effects of selective logging on the butterflies of a
Bornean rainforest. Conservation Biology 14: 1055–1065.
52. Kareiva P (1987) Habitat fragmentation and the stability of predator–prey interactions. Nature 326:
388–390.
53. Elzinga JA, van Nouhuys S, van Leeuwen DJ, Biere A (2007) Distribution and colonisation ability of
three parasitoids and their herbivorous host in a fragmented landscape. Basic and applied ecology 8:
75
–88.
54. Simonetti JA, Grez AA, Celis-Diez JL, Bustamante RO (2007) Herbivory and seedling performance in a
fragmented temperate forest of Chile. Acta Oecologica 32: 312–318.
55. Vasquez PA, Grez AA, Bustamante RO, Simonetti JA (2007) Herbivory, foliar survival and shoot growth
in fragmented populations of Aristotelia chilensis. Acta Oecologica 31: 48–53.
56. Ramos-Elorduy J, Moreno JMP, Vázquez AI, Landero I, Oliva-Rivera H, Camacho VHM (2011) Edible
Lepidoptera in Mexico: Geographic distribution, ethnicity, economic and nutritional importance for rural
people. Journal of Ethnobiology and Ethnomedicine 7: 3–22. doi: 10.1186/1746-4269-7-3 PMID:
21457504
57. Landeros-Torres I, Oliva-Rivera H, Galindo-Tovar ME, Balcazar-Lara A, Murguía-González J, Ramos-
Elorduy J (2012) Uso de la larva de Arsenura armida armida (Cramer, 1779) (Lepidoptera: Saturniidae),
“cuecla” en Ixcohuapa, Veracruz, México. Cuadernos de Biodiversidad 38: 4–8.
Lepidopterans in a Restored Mexican Dry Forest
PLOS ONE | DOI:10.1371/journal.pone.0128583 June 1, 2015 13 / 14
58. López-Carretero A (2010) Composición y diversidad de lepidópteros en la cronosecuencia sucesional
del bosque tropical caducifolio, consecuencias sobre la herbivoría de Casearia nitida. Mexico DF:
UNAM. 86 p.
59. Steward JL, Keeler KH (1988) Are there trade-offs among antiherbivore defenses in Ipomoea (Convol-
vulaceae)? Oikos 53: 79–86.
60. Sánchez-Montoya G (2002) Variación intraespecífica en la herbivoría y su impacto en algunos compo-
nentes del éxito reproductivo masculino y femenino de Ipomoea pauciflora en una selva baja caduci.
Mexico DF: Universidad Nacional Autónoma de México. 23 p.
61. Coley PD, Bryant JP, Chapin FS III (1985) Resource availability and plant anti-herbivore defense. Sci-
ence 230: 895–899. PMID: 17739203
62. Cuevas-Reyes P, Oyama K, González-Rodríguez A, Fernandes W, Mendoza-Cuenca L (2011) Con-
trasting herbivory patterns and leaf fluctuating asymmetry in Heliocarpus pallidus between different
habitat types within a Mexican tropical dry forest. Journal of Tropical Ecology 27: 383–391.
63. Marquis RJ (1996) Plant architecture, sectoriality and plant tolerance to herbivores. Vegetatio 127: 85–
97.
64. Gamboa Villa L (2012) Efecto de plantaciones experimentales en la lluvia de semillas en la Sierra de
Huautla, Morelos. Cuernavaca, Morelos: Universidad Autónoma del Estado de Morelos. 84 p.
65. Griscom HP, Griscom BW, Ashton MS (2009) Forest Regeneration from Pasture in the Dry Tropics of
Panama: Effects of Cattle, Exotic Grass, and Forested Riparia. Restoration Ecology 17: 117–126.
66. Griscom HP, Ashton MS (2011) Restoration of dry tropical forests in Central America: A review of pat-
tern and process. Forest Ecology and Management 261: 1564–1579.
Lepidopterans in a Restored Mexican Dry Forest
PLOS ONE | DOI:10.1371/journal.pone.0128583 June 1, 2015 14 / 14
