SIT Graduate Institute/SIT Study AbroadSIT Digital Collections

Independent Study Project (ISP) Collection SIT Study Abroad

Spring 2018

Herpetofauna of Sumak Allpa: A BaselineAssessment of an Unstudied Island HerpetofaunalCommunitySara FreimuthSIT Study Abroad

Follow this and additional works at: https://digitalcollections.sit.edu/isp_collection

Part of the Animal Studies Commons, Biodiversity Commons, Biology Commons, LatinAmerican Studies Commons, and the Zoology Commons

This Unpublished Paper is brought to you for free and open access by the SIT Study Abroad at SIT Digital Collections. It has been accepted forinclusion in Independent Study Project (ISP) Collection by an authorized administrator of SIT Digital Collections. For more information, pleasecontact digitalcollections@sit.edu.

Recommended CitationFreimuth, Sara, "Herpetofauna of Sumak Allpa: A Baseline Assessment of an Unstudied Island Herpetofaunal Community" (2018).Independent Study Project (ISP) Collection. 2784.https://digitalcollections.sit.edu/isp_collection/2784

Herpetofauna of Sumak Allpa:

A Baseline Assessment of an Unstudied Island Herpetofaunal Community

Freimuth, Sara

Academic Director: Silva, Xavier, PhD

Project Advisor: Wagner, Martina, MSc

Claremont McKenna College

Organismal Biology and Spanish

South America, Ecuador, Orellana, Sumak Allpa

Submitted in partial fulfillment of the requirements for Ecuador Comparative Ecology and

Conservation, SIT Study Abroad, Spring 2018

Abstract

Sumak Allpa is an island dedicated to the provision and protection of habitat for the

conservation and rehabilitation of primates. As such, the island - a varzea ecosystem located in

the Western Amazon of Ecuador, one of the most biodiverse and also most threatened regions in

the world – consists of protected primary forest that is home not only to a variety of primates, but

also to an even wider variety of other taxa, nearly all of which have gone unstudied on the island.

The present investigation assessed two of those taxa, amphibians and reptiles, in order to

establish a baseline inventory of the island’s herpetofaunal community and provide a preliminary

assessment of its composition, activity, and habitat use as a way to suggest and inform future

investigations for the study and monitoring of the community and its populations. Through 78

hours of visual and acoustic encounter surveys spanned across 16 days, a community of 552

individuals representing 15 species of 5 families of Anura and 11 species of 7 families of

Squamata was observed on the island. Analysis of this community indicated high species

richness and low species diversity and that the majority of species are capable of occupying all of

the island’s habitats, demonstrating preferences for specific habitat characteristics such as leaf

litter microhabitats and high density understory vegetation. An additional analysis of the

population of Adenomera hylaedactyla, a species accounting for 57.4% of all individuals

encountered, also demonstrated these same overall community trends, and this particular

population had an average snout-vent-length (SVL) much smaller than the average SVL reported

for the species overall. Collectively, these findings indicate long-term seasonal monitoring of the

island and future investigation of its individual populations would best inform the conservation

of the island’s herpetofaunal community and fulfill some of the many gaps in the existent

knowledge of Ecuadorian and Neotropical herpetofauna more generally.

Resumen

Sumak Allpa es una isla dedicada a la provisión y protección del hábitat para la

conservación y rehabilitación de primates. Como tal, la isla - un ecosistema de várzea ubicado en

la Amazonía occidental de Ecuador, una de las regiones más biodiversas y más amenazadas del

mundo - consiste en un bosque primario protegido que alberga no solo una variedad de primates,

sino también una variedad aún más amplia de otros taxones, casi todos los cuales no han sido

estudiados en la isla. La presente investigación evaluó dos taxones, anfibios y reptiles, con el fin

de establecer un inventario de línea de base de la comunidad de herpetofauna de la isla y de

proporcionar una evaluación preliminar de su composición, actividad y uso del hábitat como una

forma de sugerir e informar investigaciones futuras para el estudio y monitoreo de la comunidad

y sus poblaciones. A través de 78 horas de encuestas visuales y acústicas durante 16 días, se

observó en la isla una comunidad de 552 individuos que representaban a 15 especies de 5

familias de Anura y 11 especies de 7 familias de Squamata. El análisis de esta comunidad indicó

una alta riqueza de especies y baja diversidad de especies y que la mayoría de las especies son

capaces de ocupar todos los hábitats de la isla, demostrando preferencias por características

específicas del hábitat tales como microhábitats de hojarasca y vegetación de sotobosque de alta

densidad. Un análisis adicional de la población de Adenomera hylaedactyla, una especie que

representa el 57.4% de todas los individuos encontrados, también demostró estas mismas

tendencias generales de la comunidad, y esta población particular tuvo una longitud de rostro-

cloacal promedio (LRC) mucho más pequeña que la LRC promedio declarada para la especie en

general. Colectivamente, estos hallazgos indican un monitoreo estacional a largo plazo de la isla

y la investigación futura de sus poblaciones individuales serían las mejores formas de informar la

conservación de la comunidad herpetofauna de la isla y cubrirían algunas de las brechas en el

conocimiento existente de la herpetofauna ecuatoriana y neotropical en general.

Acknowledgements

A large debt of gratitude is owed to the many people who made this project possible.

First, a special thank you belongs to Diana Serrano, Xavier Silva, and Javier Robayo for their

continued support throughout the ISP process. This project would not have come to fruition

without their constant advice and emotional support. Another special thanks goes to Martina

Wagner for welcoming me to the Amazon, sharing with me a plethora of guides and materials

for the identification of the region’s herpetofauna, introducing me to all of the trails of the island,

and showing me incredible kindness and patience. Further thanks go to Valerio Gualinga for

clearing trails and transects for safer nocturnal surveying and to Hector Vargas for his

hospitality, compassion, and provision of transportation and delicious warm meals. Lastly, many

thanks go to Rachel Meyer and Emily Babb for keeping me sane during my first few days on

Sumak Allpa and to Sophie Cash and Krista Smith-Hanke for their friendship, constant

reassurance, and extraordinary moral support while sharing this experience with me.

Agradecimientos

Se debe una gran deuda de gratitud a las muchas personas que hicieron posible este

proyecto. Primero, un agradecimiento especial les pertenece a Diana Serrano, Xavier Silva y

Javier Robayo por su continuo apoyo durante todo el proceso de ISP. Este proyecto no habría

llegado a buen término sin su constante asesoramiento y apoyo emocional. Otro agradecimiento

especial para Martina Wagner por darme la bienvenida a la Amazonía, compartiendo conmigo

una plétora de guías y materiales para la identificación de la herpetofauna de la región,

presentándome a todos los senderos de la isla y mostrándome una amabilidad y paciencia

increíble. Agradezco a Valerio Gualinga por despejar senderos y transectos para realizar estudios

nocturnos más seguros y a Héctor Vargas por su hospitalidad, compasión y provisión de

transporte y deliciosas comidas calientes. Por último, muchas gracias a Rachel Meyer y Emily

Babb por mantenerme cuerda durante mis primeros días en Sumak Allpa y a Sophie Cash y

Krista Smith-Hanke por su amistad, consuelo constante y extraordinario apoyo moral mientras

compartían esta experiencia conmigo.

Introduction

Global Herpetofauna: Ecological Functions and Population Declines

Although amphibians and reptiles are often considered together in the study of

herpetology, these two vertebrate groups that constitute herpetofauna are actually quite different.

Derived from lineages that have been independent since 300 million years ago, amphibians and

reptiles have quite different life histories with distinct morphological characteristics,

reproductive systems, behavioral traits, and ecological requirements (Zug et al., 2001; Costa et

al., 2016). Consequently, amphibians and reptiles contribute to a diverse range of ecological

functions and herpetofauna constitute abundant and diverse components of many terrestrial and

freshwater ecosystems (Young et al., 2004; Tyler et al., 2007; Pough et al., 2004; Wells, 2007;

Collins & Crump, 2009), particularly in the Neotropics, where they are among the most abundant

and diverse vertebrate taxa (Cortés-Gomez et al., 2016; Stuart et al., 2004).

Despite their plenitude in such a wide range of terrestrial and freshwater ecosystems,

there has been a generally low level of understanding regarding the ecological functions and

services amphibians and reptiles provide (Bickford et al, 2010; Hocking & Babbitt, 2014).

Recent studies, however, cite how in spite of their biological and ecological differences,

amphibians and reptiles often play very similar ecological roles, contributing to nutrient cycling,

bioturbation, seed dispersal, pollination, and energy flow through trophic chains as both predator

and prey species (Cortés-Gomez et al., 2016). Other studies also emphasize their provision of

ecosystem services to humans, specifically, evidencing their regulation of pest outbreaks and

alteration of disease transmission through predation and competition with mosquitoes and

disease-carrying fly species, in addition to their use as food, in cultural practices across the

globe, and in both Western and traditional medicines (Hocking & Babbitt, 2014).

This recent surge in efforts to understand the ecological roles herpetofauna play within

ecosystems, nevertheless, has been prompted by a sense of urgency and out of necessity.

Amphibian and reptile populations in recent decades have experienced rapid declines and

reduction is still ongoing (Gibbons et al., 2000; Lips et al., 2006; Reading et al., 2010; Heatwole

2013). In addition to prompting efforts to better understand the ecological functions of

amphibians and reptiles, these declines also have impelled the investigation of specific threats to

herpetofauna and the establishment of criteria for the evaluation of the degree of threat to species

of amphibians and reptiles (Gibbons et al., 2000; Todd et al., 2010; Stuart et al., 2004; Böhm et

al., 2012).

Among the most significant threats to amphibian and reptilian populations at the global

scale are habitat loss and degradation, introduced invasive species, pollution, disease,

unsustainable use, and climate change (Lips et al., 2006; Gibbons et al., 2000; Todd et al., 2010).

In the Neotropics, these threats are particularly imminent. Not only does this region contain

49.2% of the world’s amphibian species and significant reptile diversity and richness, but also it

has suffered among highest rates of population decline (63.1% for amphibians and about 20% for

reptiles) (Stuart et al., 2004; Böhm et al., 2012).

Herpetofauna of Ecuador

As one of the 17 megadiverse countries in the world that collectively contain 70% of the

animal and plant species on the planet, Ecuador is well represented by a variety of taxa,

particularly amphibians and reptiles (Centro Jambatu, 2011). Home to 615 species of

amphibians, 236 of which are endemic, Ecuador ranks fourth in the world in amphibian species

richness behind Brazil, Colombia, and Peru, respectively (AmphibiaWeb). Given its small

relative size, however, Ecuador ranks first in the world for amphibian species richness per unit

area, and with 464 recorded species of reptiles, it ranks the same in terms of reptilian species

richness per unit area, too (AmphibiaWeb; Reyes-Puig, Almendáriz, & Torres-Carvajal, 2017).

Thus, with such species richness condensed in such a small region and high levels of endemism,

Ecuador is an important area to consider for the investigation herpetofauna in general, but is

especially significant considering the recent trends of Neotropical amphibian and reptile

population declines.

Despite this significance, few studies have been conducted to assess amphibian and

reptilian populations in Ecuador, and very little is known about many of the species that

constitute its richness and diversity. The first quantitative study on amphibian decline in Ecuador

was not published until 2003, and the first comprehensive quantitative study of reptile

conservation in Ecuador was not published until 2017 (Bustamante et al., 2005; Reyes-Puig,

Almendáriz, & Torres-Carvajal, 2017).

The lack of understanding of the status of Ecuador’s amphibian and reptilian diversity is

even further evidenced by how much of it has only been recently discovered. Nearly 10% of all

reptilian species in Ecuador have been reported or described in this century alone (Reyes-Puig,

Almendáriz, & Torres-Carvajal, 2017), and a large number of amphibian species have been

described in the past 5 years alone (e.g. Batallas & Brito, 2014; Reyes-Puig & Yánez-Muñoz,

2012; Brito, Ojala-Barbour, Batallas & Almendáriz, 2016). Assessing these species, especially

their baseline population levels and distributions, is crucial to monitoring these new populations

for their conservation.

Newly described species, however, are not the only understudied populations of

herpetofauna in Ecuador. While amphibian species recently have been assessed more extensively

(Rodrigues et al., 2006), only about 25% of the species of reptiles from continental Ecuador are

listed on the IUCN Red List of Threatened Species and 17% of those listed are Data Deficient

(Reyes-Puig, Almendáriz, & Torres-Carvajal, 2017). Ultimately, one of the biggest deficits in the

conservation of herpetofauna in Ecuador, and the Neotropics accordingly, is simply the lack

baseline knowledge and continued monitoring of its many different species’ populations.

The Ecuadorian Amazon

The Amazon is one of the most biodiverse biomes on the planet, and where Ecuador falls

in western Amazon, in particular, is especially rich in biodiversity and contains some of the

highest levels of amphibian and reptilian species richness in Ecuador (AmphibiaWeb; Reyes-

Puig, Almendáriz, & Torres-Carvajal, 2017). The Ecuadorian Amazon is not only species rich,

however – the region also happens to be rich in untapped oil reserves, many of which are located

in Yasuní National Park, a major protected area within the western Amazon of Ecuador (Bass et

al., 2010). Consequently, the ecosystems of Yasuní and the surrounding areas confront numerous

threats characteristic of the entire Ecuadorian Amazonian region. Most prominent among these

threats is petroleum exploration and exploitation; however, accompanying these projects are

transportation projects that open up the region to numerous other threats, including mining,

illegal logging, oil palm plantations, and rapid resource exploitation and human development

(Bass et al., 2010). Thus, the region’s populations of herpetofauna are at risk for or already are

experiencing consequences of these anthropogenic activities, such as habitat loss and degradation

and pollution, in addition more general threats such as disease and climate change.

Sumak Allpa and the Present Investigation

Sumak Allpa is an island situated in the Ecuadorian Amazon downstream the River Napo

from Coca in the vicinity of Yasuní National Park and the namesake of the non-profit primate

rehabilitation center that was founded in 2006 by Hector Vargas and is housed on the island

(Figure 1). The project concentrates on providing education for the global and local community

on the importance of the protection and conservation of the Amazon and also is centered on the

protection of native primates from anthropogenic destruction through the use of the island as a

sanctuary for primates rescued principally from animal-trafficking in the surrounding area

(Zewdie, 2017). As a result of the island’s use for primate rehabilitation, Sumak Allpa contains

protected primary forest subject to little human activity beyond infrequent, small-scale tourism

that is home not only to primates but also to a wide variety of other taxa that contribute to the

health and maintenance of the ecosystem. Few of these other taxa have been investigated

formally, if at all, and the island’s herpetofauna had never been assessed before the present

investigation.

Figure 1. Satellite imagery of demonstrating where Sumak Allpa is located in Ecuador and how it is

situated downstream from Coca, Puerto Francisco de Orellana, Orellana, Ecuador along the Napo

River.

The conservation scheme of Sumak Allpa provided by its status as a primate

rehabilitation center presents a unique situation for the assessment and continued monitoring of

herpetofauna. While the herpetofaunal community is nearly entirely protected from direct

anthropogenic threats, such as habitat loss and degradation, due the protection of the ecosystem

for primate rehabilitation, the isolation provided by the ecosystem occurring on an island, and the

absence of frequent human use, the herpetofauna on the island still may be threatened by

generalized threats to the surrounding region. For example, while reduced compared to other

sites along the river, Sumak Allpa is still subject to the effects of pollution in both the water and

air as a result of petroleum exploitation in the mainland regions around the island. The island

also is not exempt from climate change and risk of exposure to invasive species and disease.

Thus, Sumak Allpa provides the unique opportunity to study the impacts of such threats using a

small and isolated community of herptofauna.

Being situated in the heart of the northern Ecuadorian Amazon and within the vicinity of

the biodiversity hotspot, Yasuní National Park, the island also contains a small portion of the

Sumak Allpa

amphibian and reptile diversity of this important Neotropical region. Though the populations of

this diversity are small and made up of common and relatively non-threatened species, Sumak

Allpa still carries significance for the conservation of the region’s herpetofauna - some of the

Ecuadorian amphibian and reptile species that have poorly understood ecologies and unassessed

populations by the IUCN are known to exist on the island. Therefore, as small, contained area,

Sumak Allpa serves as an ideal site to study the ecological niches and functions of these species

and fill knowledge gaps to better strategize the conservation of these species throughout their

entire ranges.

Given the unique suitability of Sumak Allpa as a locality for the study of herpetofauna

and lack of prior assessment, the present investigation serves two primary purposes: 1) to

establish a baseline inventory of the island’s herpetofauna for future studies and 2) to provide a

preliminary assessment of the composition, activity, and habitat use of the herpetofaunal

community that can be used to inform future studies and monitoring of the community and

populations of its species.

Materials & Methods

Study Site

Sumak Allpa is an island situated in the Napo River 19 kilometers east of Coca, Puerto

Francisco de Orellana, Orellana, Ecuador at about 315 m above sea level (0°26'24" S 76°49'02"

W). The climate of the island and surrounding area is characterized by high temperatures

averaging between 24°C and 27°C, a high annual rainfall of 3,200 mm, high humidity averaging

between 80 and 94%, and a mild dry season (Bass et al., 2010). The island contains a varzea

ecosystem and experiences various levels of flooding: areas in the northeast are most regularly

flooded and more central areas flood only intermittently, while the southwest areas of the island

remain dry year-round. The island experienced one inundation event during the time of this

study. The forest is recovered from cacao and coffee plantations and now is between 70% and

80% primary forest and 20% and 30% secondary forest (Zewdie, 2017). A small portion of the

southwestern shore of the island has been developed for four cabins and a small dock, and aside

from the seven maintained trails for research and tourism, the rest of the island is permitted to

grow freely and continues to recover (Figure 2).

Figure 2. Satellite image of Sumak Allpa with the seven maintained trails and three

convening transects highlighted and labeled.

Selection of Surveying Techniques

Searches were conducted using a combination of visual encounter surveying and acoustic

encounter surveying, two standard techniques for the inventory and monitoring of herpetofauna

(Heyer, 1994; Eekhout, 2011; Lambert, 1984). Visual encounter surveying (VES) was selected

over other active search techniques, such as quadrat and patch sampling, and passive search

techniques, such as pitfall trapping and PVC pipe refugia sampling, as the principle surveying

method for this study with the consideration of both the present assessment and future studies.

VES has the lowest relative cost and time investment for the inventory and monitoring of

the broadest spectrum of herpetofauna and can still be employed successfully when personnel is

limited (Heyer, 1994). When utilized in short-term studies, such as this, the VES also has

demonstrated significantly higher encounter rates of diurnal amphibians and nocturnal reptiles

than other active search techniques and was comparable to such techniques for the detection of

nocturnal amphibians and diurnal reptiles in the short-term (Doan 2003). VES is even more

effective in combination with other inventory and/or monitoring techniques, to the point where it

may even be able to obtain a complete species inventory when paired with an appropriate

secondary technique (Eekhout, 2011).

Thus, to account for canopy-dwelling and more cryptic calling species, VES was paired

with acoustic encounter surveying (AES), another low-cost, low-personnel, and low-time-

investment monitoring technique (Heyer, 1994). When used together, these methods have been

demonstrated to detect the highest species numbers across families, guilds, and microhabitats

among active and passive search techniques that can be used for short-term surveys (Rödel &

Ernst, 2004). All factors considered, this combination was decidedly the most effective

methodology not only for conducting this short-term baseline assessment but also for

establishing a feasible monitoring scheme for the continued long-term assessment of

herpetofauna on Sumak Allpa.

VES and AES

As much of the island as was accessible and possible was surveyed between April 22 and

May 7, 2018. Using VES and AES, areas of all seven maintained trails and three convening

temporary transects were searched in three, four, and five-hour blocks between the hours of 8:00-

17:15 for what were termed diurnal searches and 19:30- 1:45 for nocturnal searches for a total of

39 hours each of diurnal and nocturnal surveys. Surveys were conducted at an average speed of

2.5 meters per minute. Temperature and relative humidity were recorded at the start and finish of

each survey and averaged to describe conditions at the time of each survey.

Herpetofauna located within 1 meter of the edge of each side of trails and transects were

searched thoroughly from the ground up to 2 meters high. Trails wider than 2 meters were

searched one side at a time. A Black Diamond Spot headlamp was utilized to provide light for all

nocturnal surveys.

Each time an individual was located visually, a photo was taken while it remained at its

perch if possible, and immediately following the individual was captured by hand if possible.

Once captured, a series of 5 photos - top view, ventral view, right side view, front facial view,

and bird’s eye view - was taken of each individual with additional photographs of notable

morphological features taken as necessary. Once photographed, snout-vent-length (SVL) and any

other notable identification features were measured using a digital caliper. Individuals unable to

be handled or successfully captured were photographed to the best extent possible and described

thoroughly in the field immediately after being sighted.

After recording these data and other relevant observations, each individual handled was

released at or near the exact site of capture. All individuals were handled with disposable latex

gloves which were changed after the release of each individual and all equipment to come into

contact with any given organism was sanitized after each use in order to avoid vectoring fungi

and other pathogens between organisms, particularly amphibians.

For these visual encounters, the time of sighting was recorded for each individual, and the

location of each encounter was recorded using the offline smartphone GPS application,

Maps.ME. The perch height of each individual also was measured with measuring tape and

microhabitat, surrounding understory vegetation density, and habitat type of each individual

were noted. Microhabitats were described in the following categories: exposed soil, mud, grass,

leaf litter, leaf, branch, log, and man-made structure. Leaf litter also was further described as wet

or dry and loose or compact. Understory vegetation density was defined as low, medium, or

high.

Habitat types were classified as river edge, high varzea, low varzea, or cabin area. River

edge represented areas within about 20 meters of the edge of the Napo River. High varzea

designated areas unaffected by flooding, while low varzea signified areas subject to any level of

flooding during an inundation event. Cabin area described the man-made clearing of high varzea

and river edge habitat on the southwest shore of the island containing the dock and cabins that is

subject to the most human alteration and highest levels of human activity on the island.

Acoustic encounter surveys were conducted concurrently with VES. When individuals

were heard calling within about 1-2 meters of the trail or transect being searched, they were

recorded for 30-60 seconds using an Olympus VN-721PC Digital Voice Recorder, holding the

recorder as close to the source of the sound as possible. For each of these encounters, time of

call, density of understory vegetation, and habitat type also were documented.

Identification

Given the small size of Sumak Allpa and the lack of knowledge regarding the exact size

and status of populations of herpetofauna on the island, no specimens were collected and

preserved for identification in this study. Species encountered visually were identified instead

entirely based upon the morphological species concept to the most taxonomically specific level

possible using Guía Dinámica de los Anfibios del Ecuador and Guía Dinámica de los Reptiles

del Ecuador from BIOWEB Ecuador, Guía de Campo de Anfibios del Ecuador, Guía de Campo

de Reptiles del Ecuador, Reptiles and Amphibians of the Amazon: An Ecotourist's Guide, the

Frogs of the Yachana Reserve field guide, and additional online resources such as AmphibiaWeb

and BIOWEB Ecuador as necessary. Recorded calls were identified using Frogs of the

Ecuadorian Amazon and online databases with species-specific recordings of calls, such as

AmphibiaWeb and BIOWEB Ecuador.

Community Analysis

The overall herpetofaunal community surveyed was analyzed in a variety of ways. First,

all species encountered were inventoried taxonomically by order, family, and species and by

their IUCN Red List of Threatened Species status. The community was then analyzed first in

terms of the composition of encounters: the proportions of acoustic versus visual and nocturnal

versus diurnal encounters were calculated for each species, and the percentage of overall

encounters of each species was calculated to demonstrate the relative size of each population

within the community.

Sumak Allpa’s herpetofaunal community was then analyzed for its species composition

overall and in terms of various sub-communities. Sub-communities were classified within two

main categories: habitat type, which contained the river edge, high varzea, low varzea, and cabin

area sub-communities, and period of activity, which was categorized as diurnal or nocturnal. The

sample coverage, species richness, and species diversity were estimated for the overall

community and various sub-communities through asymptotic analysis using the software iNext

Online and quantified using Hill numbers. Hill numbers (or the effective number of species) are

parameterized by a diversity order q, which determines the measures’ sensitivity to species

relative abundances (Chao et al., 2015; see A1 in Appendix for formulas). The species richness

(q = 0), Shannon diversity (the exponential of Shannon entropy, q = 1) and Simpson diversity

(the inverse of Simpson concentration, q = 2), all of which are expressed in units of “species

equivalents,” were calculated for the overall community, the 4 defined habitat types, and the

diurnal and nocturnal communities within the overall island community.

The similarity of these two types of sub-communities also was analyzed to determine the

overlap of species and individuals in habitat use and hours of activity. The similarity between

habitat type communities was estimated with SpadeR Online’s Multiple-Community Measures

tool using following measures: Sørensen similarity index, Sørensen pairwise similarity estimates,

and the Bray-Curtis similarity index. The Sørensen similarity index is a classic local species

richness-based similarity measure (Chao et al., 2015) and was utilized to assess the percent of

species overlap in all habitat types (see A2 in Appendix for formula). Sørensen pairwise

similarity estimates are similarity estimates of the same measure but between specific pairs

within the set of local communities compared and were used to evaluate species overlap in

specific pairs of habitat types. The Bray-Curtis similarity index compares absolute abundances

among communities and thus was used to assess the overlap of individuals in habitat types (see

A3 in Appendix formula). The Sørensen similarity index and the Bray-Curtis similarity index

also were used to assess the similarity and overlap of species and individuals in diurnal and

nocturnal communities. For this comparison, these indices were estimated using SpadeR

Online’s Two-Community Measure tool, as the Sørensen similarity index is calculated

differently for comparisons of only two communities (see A4 in Appendix for formula).

The community’s overall microhabitat preference was then analyzed looking at the

overall community usage and species composition of each microhabitat and level of understory

vegetation density.

Population Analysis of Adenomera hylaedactyla

Adenomera hylaedactyla constituted a significant portion of the island’s overall

herpetofaunal community. Thus, in addition to the community analysis of Sumak Allpa, the

population of A. hylaedactyla, specifically, was analyzed in more depth first through the

consideration of population-specific microhabitat preference. The distributions of both acoustic

and visual encounters of the species among levels of understory vegetation and habitat types also

were assessed to determine the population’s preferred habitat conditions for both calling and

general activity. Lastly, the size distribution of individuals within the population was evaluated

using snout-vent-length measurements.

Results

VES and AES Inventory

During the study period, 373 individuals were identified by visual encounter and 179

individuals were identified by acoustic encounter for a total of 552 individuals, representing 15

species from 5 families of the order Anura and 11 species from 7 families of the order Squamata

(Table 1). Of the total of 26 distinct species, 22 were identified taxonomically to species level

and 4 were identified to the genus level as distinct morphospecies. Of the 22 species identified to

species level, 7 were classified as Least Concern on the IUCN Red List of Threatened Species, 5

were as Least Concern but needs updating, indicating their assessment is older than 10 years

(CITE IUCN). None of reptilian species encountered on Sumak Allpa had been evaluated by the

IUCN (Table 1).

Table 1. Families and species encountered during the study period, including their status on the

International Union for Conservation of Nature (IUCN) Red List of Threatened Species (LC = Least

Concern, LC* = Least Concern but needs updating, NE = Not Evaluated, -- = not identified to

species level), and combined number of visual and acoustic encounters. ANURA

FAMILY SPECIES IUCN Status Number of Encounters

BUFONIDAE Rhinella aff. margaritifera LC 1

Rhinella marina LC 2

HYLIDAE

Boana boans NE 8

Dendropsophus parviceps LC* 1

Osteocephalus taurinus LC 1

LEPTODACTYLIDAE

Adenomera andreae LC* 86

Adenomera hylaedactyla LC 317

Adenomera sp. -- 11

Leptodactylus knudseni LC 5

Leptodactylus mystaceus LC 59

Leptodactylus pentadactylus LC 4

MICROHYLIDAE Chiasmocleis bassleri LC* 17

STRABOMANTIDAE

Oreobates quixensis LC* 2

Pristimantis altamazonicus LC* 7

Pristimantis sp. -- 2

SQUAMATA

FAMILY SPECIES IUCN Status #

COLUBRIDAE

Drepanoides anomalus NE 1

Imantodes cenchoa NE 2

Leptodeira annulata NE 1

ELAPIDAE Micrurus lemniscatus helleri NE 1

VIPERIDAE Bothrops atrox NE 1

GEKKONIDAE Thecadactylus solimoensis NE 2

IGUANIDAE

Anolis bombiceps NE 1

Anolis sp. 1 -- 1

Anolis sp. 2 -- 1

SPHAERODACTYLIDAE Pseudogonatodes guianensis NE 15

TEIIDAE Tupinambis cuzcoensis NE 3

TOTAL

FAMILY SPECIES #

12 26 552

0% 50% 100%

Diurnal Nocturnal

Types and Frequency of Encounters

Of the 26 species observed on Sumak Allpa, 19 were encountered only visually, 3 were

encountered only acoustically, and 4 were both seen and heard calling (Figure 3A). Additionally,

18 species were encountered exclusively nocturnally, 5 exclusively diurnally, and 3 were

encountered both nocturnally and diurnally (Figure 3B).

Figure 3. Percentage of (A) acoustic versus visual and (B) diurnal versus nocturnal encounters for

each species observed on Sumak Allpa.

Adenomera hylaedactyla, a principally nocturnal species, was encountered most

frequently in both visual and acoustic surveys, accounting for 57.4% of all encounters (Figure 4).

Adenomera andreae, a principally diurnal species, was the most encountered diurnal species and

second most encountered species overall, accounting for 15.6% of all encounters (Figure 4).

Leptodactylidae was the most encountered family in both VES and AES.

0% 50% 100%

T. cuzcoensis

T. solimoensis

R. marina

R. aff. margaritifera

P. guianensis

Pristimantis sp.

P. altamazonicus

O. taurinus

O. quixensis

M.l. helleri

L. annulata

L. pentadactylus

L. mystaceus

L. knudseni

I.cenchoa

D. anomalus

D. parviceps

C. bassleri

B. atrox

B.boans

Anolis sp. 2

Anolis sp. 1

A.bombiceps

Adenomera sp.

A. hylaedactyla

A. andreae

Spec

ies

Visual Acoustic

Percentage of Type of Encounter

(A) (B)

Figure 4. Percentage of overall encounters of each species surveyed on Sumak Allpa.

Community Composition

The sample coverage of the island overall was a high estimate of 0.982, and each of the

sub-communities were also estimated to be well-surveyed (Figure 5). Low varzea was the best-

sampled habitat type with a sample coverage of 0.975, and the nocturnal community was

estimated to be slightly better sampled than the diurnal community (Figure 5).

Figure 5. Estimated sample coverage for the entire island and the 6 sub-communities of the island,

where 1.00 indicates 100% sample coverage.

The overall community of herpetofauna was characterized by high species richness and

low species diversity (Figure 6). Species richness of the overall community was estimated at

34.98 species equivalents, while the Shannon diversity index estimated 1.60 species equivalents,

or about 2 common species in the overall community, and the Simpson diversity index estimated

0.633 species equivalents, or effectively 1 dominant species in the overall community (Figure 6).

0.9

0.91

0.92

0.93

0.94

0.95

0.96

0.97

0.98

0.99

1

Overall River Edge High Varzea Low Varzea Cabin Area Diurnal Nocturnal

Sam

ple

Co

vera

ge E

stim

ate

Sub-Community

A. andreae, 15.6% A. hylaedactyla, 57.4%Adenomera sp., 2.0% A. bombiceps, 0.2%Anolis sp. 1, 0.2% Anolis sp. 2, 0.2%B. boans, 1.4% B. atrox, 0.2%C. bassleri, 3.1% D. parviceps, 0.2%D. anomalus, 0.2% I. cenchoa, 0.4%L. knudseni, 0.9% L. mystaceus, 10.7%L. pentadactylus, 0.7% L. annulata, 0.2%M.l. helleri, 0.2% O. quixensis, 0.4%O. taurinus, 0.2% P. altamazonicus, 1.3%Pristimantis sp., 0.4% P. guianensis, 2.7%R. aff. margaritifera, 0.2% R. marina, 0.4%T. solimoensis, 0.4% T. cuzcoensis, 0.5%

Likewise, each habitat type demonstrated the same trend of much higher species richness

than species diversity. High varzea was estimated to have the highest habitat species richness at

48.89 species equivalents, while the cabin area was the least species rich habitat estimated at

only 5.93 species equivalents (Figure 6). In other words, high varzea habitats are used by the

largest number of species, while the cabin area is occupied by only few species. Meanwhile, all

habitat types had effectively 2 or fewer common species and 1 or fewer dominant ones, as given

by Shannon and Simpson diversity, respectively (Figure 6).

Figure 6. Diversity profile estimates of the overall and sub-communities of herpetofauna on Sumak

Allpa in terms of species richness (q = 0), Shannon diversity (q = 1), and Simpson diversity (q = 2)

(Species Equivalents ± SE).

Diurnal and nocturnal communities also had high species richness relative to species

diversity (Figure 6). The nocturnal community was estimated to be more than twice as rich as the

diurnal community in species with 30.10 species equivalents, yet both communities had roughly

only one common species and about 1 or fewer dominant species.

The Sørensen index, a richness-based similarity measure, for the comparison of all

habitat types was 0.794, indicating that about 79.4% of species on Sumak Allpa can be

encountered in all four of the classified habitat types (Figure 7). The pairwise similarity

estimates indicate the highest levels of overlap in species use of river edge and low varzea and of

high varzea and low varzea, with 31.3% of species estimated to utilize both river edge and low

varzea habitats and 27.9% of species estimated to make use of both high and low varzea habitats

(Figure 7). Meanwhile, the lowest level of species habitat overlap was between high varzea and

the cabin area with an estimated 10.6% of species being encountered in both (Figure 7).

The Bray-Curtis index, a measure for comparing species absolute abundances, estimated

only 48.8% of individuals in the community make use of all four habitat type (Figure 7).

0 10 20 30 40 50 60 70

q = 0

q = 1

q = 2

Species Equivalents

Hill

Ord

er

Nocturnal Diurnal

Cabin Area Low Varzea

High Varzea River Edge

Overall

Figure 7. Estimated similarity of sub-communities based on habitat type (Edge = river edge, HVz =

high varzea, LVz = low varzea, CA = cabin area) given by the classic richness-based Sørensen

similarity index, richness-based Sørensen pairwise similarity estimates, and species absolute

abundance-based Bray-Curtis index (Similarity Estimate ± SE).

There was very little overlap between the diurnal and nocturnal communities on Sumak

Allpa. The richness-based Sørensen similarity index of the communities was 0.177, estimating

that only 17.7% of species on the island could be encountered both diurnally and nocturnally

(Figure 8). The abundance-based Bray-Curtis similarity index was even lower, suggesting that

only about 7.6% of individuals on the island are active during both what were defined as daytime

and nighttime hours for the purposes of this study (Figure 8).

Figure 8. Estimated similarity of diurnal and nocturnal sub-communities given

by the Sørensen similarity index and Bray-Curtis index (Similarity Estimate ±

SE).

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

Sørensen index Bray-Curtis index

Sim

ilari

ty E

stim

ate

Similarity Measure

0

0.2

0.4

0.6

0.8

1

Sørensen Edge, HVz Edge, LVz Edge, CA HVz, LVz HVz, CA LVz, CA Bray-Curtis

Sim

ilari

ty E

stim

ate

Similarity Measure

Sørensen index Sørensen pairwise similarity Bray-Curtis index

Microhabitat and Understory Vegetation Density Preference

In visual encounter surveys, 339 individuals representing 12 different species were

encountered in leaf litter, making it the most utilized microhabitat by the island’s community

(Figure 9). More specifically, leaf-litter dwelling individuals exhibited preferences for loose over

compact leaf litter and dry over wet leaf litter (Figure 9). The next most preferred microhabitat

was leaves, upon which 16 individuals of 8 different species were encountered (Figure 9).

Figure 9. Community use of microhabitats by number of visual encounters in terms of species

composition (Soil = exposed soil, MMS = man-made structure, DL = dry loose leaf litter, WL = wet

loose leaf litter, DC = dry compact leaf litter, WC = wet compact leaf litter).

High density understory vegetation was most preferred by the community for both calling

and general dwelling, followed by medium density and low density, respectively (Figure 10).

Figure 10. Community preference for level of understory vegetation density in terms of number of

visual and number of acoustic encounters for each level.

0 20 40 60 80 100 120 140 160 180

DL

WL

DC

WC

Soil

Mud

Grass

Leaf

Branch

Log

MMS

Number of Encounters

Mic

roh

abit

at

A. andreae A. hylaedactyla Adenomera sp.A. bombiceps Anolis sp. 1 Anolis sp. 2B. atrox C. bassleri D. parvicepsD. anomalus I.cenchoa L. mystaceusL. annulata M.l. helleri O. quixensisO. taurinus P. altamazonicus Pristimantis sp.P. guianensis R. aff. margaritifera R. marinaT. solimoensis T. cuzcoensis

0

20

40

60

80

100

120

140

160

180

200

Low Medium High

Nu

mb

er o

f En

cou

nte

rs

Understory Vegetation Density

Visual Acoustic

Additionally, the widest array of species was harbored by high density understory

vegetation with 19 different species encountered, versus 16 in medium density and 6 in low

density (Figure 11). Seven of the species found in high density understory vegetation were found

exclusively in that level of density of understory vegetation.

Figure 11. Community preference for level of understory vegetation density by number of overall

encounters in terms of species composition.

Population Analyses of Adenomera hylaedactyla

Accounting for 57.4% of all visual and acoustic encounters, Adenomera hylaedactyla was

by far the most frequently observed species on the island with the highest relative abundance

(Figure 4). The species is primarily nocturnal; however, of the 317 overall encounters, it was

encountered visually 4 times in the last 30 minutes of what were considered diurnal searches for

the purposes of this study.

The population of A. hylaedactyla followed many of the same trends for the community

overall. Individuals among the population preferred leaf litter habitats - dry over wet leaf litter

and loose over compact leaf litter, in addition to showing overall preference for high density

understory vegetation over medium and low density (Figure 9; Figure 11). In terms visual versus

acoustic encounters, high density understory vegetation was preferred for calling sites, but there

were slightly more visual encounters in medium density than high density understory vegetation

(Figure 12).

The population also showed differences in the number of visual versus acoustic

encounters among habitat types (Figure 13). High varzea was the most preferred habitat for both

calling and general activity, but there were more acoustic encounters than visual encounters in

low varzea areas and more visual than acoustic encounters at the river edge habitat (Figure 13).

Not a single vocalization of the species was recorded in the cabin area, although calls could be

heard from forest areas within the vicinity of it (Figure 13).

0 50 100 150 200 250 300

Low

Medium

High

Number of Encounters

Un

der

sto

ry V

eget

atio

n D

ensi

ty

A. andreae A. hylaedactyla Adenomera sp. A.bombiceps

Anolis sp. 1 Anolis sp. 2 B. atrox B.boans

C. bassleri D. anomalus D. parviceps I.cenchoa

L. annulata L. knudseni L. mystaceus L. pentadactylus

M.l. helleri O. quixensis O. taurinus P. altamazonicus

P. guianensis Pristimantis sp. R. aff. margaritifera R. marina

T. cuzcoensis T. solimoensis

Figure 12. Distribution of Adenomera hylaedactyla visual and acoustic encounters among understory

vegetation densities.

Figure 13. Distribution of Adenomera hylaedactyla visual and acoustic encounters among habitat

types.

The body size of individuals in the population ranged from 5.01 mm to 27.60 mm and on

average was 19.75 mm (Figure 14). The average body size of Adenomera hylaedactyla is 23.5

mm for males and 25.6 mm for females (Angulo et al. 2003; Aichinger 1992), which were the

73rd and 87th percentile, respectively, in the population encountered on Sumak Allpa (Figure 14).

0

20

40

60

80

100

120

Low Medium High

Nu

mb

er o

f En

cou

nte

rs

Understory Vegetation Density

Visual Audio

0

10

20

30

40

50

60

70

80

90

100

River Edge Low Varzea High Varzea Cabin Area

Nu

mb

er o

f En

cou

nte

rs

Habitat Type

Visual Audio

Figure 14. Distribution of body size in terms of snout-vent-length of Adenomera

hylaedactlya individuals encountered on Sumak Allpa.

Discussion

Island Biogeography Theory proposes that the number of species of any island reflects a

balance between the rate at which new species colonize it and the rate at which populations of

established species decline and become extinct posits that the larger and/or closer to the

mainland an island is, the more species it will be able to support (MacArthur & Wilson, 2016).

Thus, a small island close to mainland, such as Sumak Allpa, should in theory be able to support

a medium to high level of species richness but very few individuals within each. Considering the

trends of high species richness and low species diversity in the diversity profile estimates of the

community surveyed in the current study, the herpetofaunal assemblage on Sumak Allpa appears

to be a case in which Island Biogeography Theory holds true.

This conclusion, however, should not be made definitively without the consideration of a

variety of factors. First, although the methodology of combining VES and AES for this study

was effective in detecting a large number of species in a wide range of niches across multiple

taxa, including a number of species that would not have been registered if only a visually-

oriented technique had been applied, there was still bias in the species the methodology

encountered. Because only the understory and visible areas of the forest floor could effectively

searched, aquatic species, fossorial species, and arboreal species dwelling above 2 meters,

especially non-vocalizing species, were heavily biased against in this study. Additionally,

because surveys were conducted on the maintained trails and transects of the island, the study

design also was biased against species particularly sensitive to human use impacts. Furthermore,

as in many active search techniques, species encountered are limited by number of observers and

observer ability, so cryptic species may be detected significantly less, if detected at all, and

species that dwell in open areas or have loud and highly distinguishable calls may be detected

significantly more. Thus, in order to generate a complete species inventory of the island, future

studies should be conducted in areas both on and off trails using multiple observers with

experience in herpetofaunal surveying and also consider the inclusion of additional microhabitat-

specific search techniques to account for species biased against in the methodology of this study.

Though the sample coverage estimates reflected effective and near-complete sampling of

Sumak Allpa’s herpetofaunal community, a variety of observed field conditions and other factors

suggest these estimates may not reflect the true coverage of sampling. The study, for example,

was conducted over a period of only 16 consecutive days all in the same season, and amphibian

and reptile activity can vary temporally, resulting differences in the abundance and number of

species detected in different seasons and even at different times within one season (Menin,

Waldez, & Lima, 2008).

There also was relatively variation in rainfall, temperatures, and percent humidity

throughout the period of study, and all of these factors are known to affect herpetofaunal activity

(Paladino, 1985; Brown & Shine, 2002). Furthermore, by giving rise to cues such as light

intensity, geomagnetism, and gravity, lunar cycles have been associated with the reproductive

phenologies (including mating vocalizations) of a number of anuran and urodele species to

different effects (Grant, Chadwick, & Halliday, 2009). Thus, the full moon occurring in the

middle of the study also may have played a role in the activity and detectability of species on

Sumak Allpa.

Inundation also may have influenced the abundance and number of species detected.

Differences in the number and types of anuran vocalizations heard before and after flooding were

observed informally during surveying; however, inundation occurred too early in the study

period for these differences to be effectively quantified and analyzed. Any number of these

variables could have influenced the high number of Adenomera hylaedactyla and Adenomera

andreae individuals detected and/or the low number of encounters of other species.

Therefore, under the consideration of all of these factors, more long-term future studies

should be conducted throughout different times of year and under a wider array of conditions,

not only to account for the potential impacts of temporality, climate, and lunar cycles on the

composition of the island’s herpetofaunal community but also to investigate how each of these

variables may influence specific populations within it.

These factors and potential limitations aside, a number of important conclusions still can

be made about the composition, activity, and habitat use of the island’s herpetofaunal

community. As previously mentioned, the island supports a high level of species richness

relative to species diversity, and although this trend was consistent across habitat types, species

richness was not the same for each habitat’s community. High varzea, for instance, had much

higher species richness than the river edge, low varzea, and cabin area communities, indicating it

may provide higher resource quality, quantity, and/or accessibility to a wider variety of species

and support a wider variety of niches than other habitats on the island. Additionally, the cabin

area demonstrated the lowest species richness and diversity, suggesting trends opposite to high

varzea in terms of resource and niche availability and also that human activity may severely

inhibit the habitat from supporting large populations and numbers of species.

The similarity indices of communities within these habitats also serve to characterize

much of the habitat use of the island’s herpetofaunal community. The Sørensen index score of

0.794 indicates a high level of overlap in the species utilizing all of the island’s habitats. This

significant overlap not only is indicative of the small size of the island and contiguity of its

habitats but also suggests that a high percentage of the island’s species are habitat generalists.

The pairwise similarity estimates also help characterize the community’s use of specific

habitats. The higher percentages of overlap of river edge with low varzea habitat and high varzea

with low varzea habitat are to be expected, given the river edge and inundated low varzea both

provide access to water, which is necessary for the reproduction of many anuran species, and low

varzea can be very similar to high varzea habitat when it is not inundated.

Furthermore, the richness-based Sørensen similarity index was higher than the

abundance-based Bray-Curtis similarity index, signifying that there is less overlap in habitat use

by specific individuals within populations than there is by overall species. This relationship

possibly suggests that although a species may be able to utilize more than one habitat, not all

individuals in its population necessarily do.

Ultimately, the composition of the herpetofaunal community did vary among habitat

types, but understanding the exact causes of this variation will require further investigation.

Future studies, therefore, should investigate more of the specific shared versus distinguishing

characteristics of each of these habitats in terms of vegetation cover and type, prey and other

resource diversity and availability, microclimate and microhabitat variation, and even

geomorphology, as all of these factors have been evidenced to affect herpetofaunal assemblages

in other Amazonian communities (Doan & Arriaga, 2002; Deichmann, 2009).

The relationships between diurnal and nocturnal communities also provide important

characterizations of the herpetofaunal community composition of Sumak Allpa. The nocturnal

community had more than double the species richness than the diurnal community; however, the

Sørensen similarity index indicated a small percentage of overlap in species between these

communities, and the Bray-Curtis similarity index demonstrated an even smaller percentage of

overlap in individuals. This apparent species richness and lack of overlap signifies the

importance of continuing to monitor herpetofauna during all times of day in order to best

conserve the entire community of herpetofauna on the island. Additionally, the possible

misrepresentation of crepuscular species as nocturnal or diurnal species suggests that future

analysis of communities in terms of periods of activity should be done with more specificity,

blocking surveys during more and different hours of the day to determine more specific hours of

activity for assemblages and specific populations within the overall community.

The trends in microhabitat and understory vegetation density use by the community also

can serve to guide future studies. In terms of microhabitat, leaf litter appeared to be the preferred

by the largest portion of the community, and dry and loose leaf litter was preferred over wet and

compact leaf litter. Given the significance of the microhabitat to so many individuals on the

island and the limitations of using only qualitative descriptions to classify leaf litter, further

analysis of more quantitative leaf litter characteristics with respect to herpetofaunal assemblages

would contribute greatly to the understanding of the herpetofaunal community of Sumak Allpa,

Studying small invertebrate communities inhabiting the leaf litter may also be beneficial to

understanding different herpetofaunal populations and how and why the leaf litter supports them

differently, especially since other Amazonian studies have demonstrated significant differences

in resource utilization and guild structure among leaf-litter dwelling species of frogs and lizards

(Vitt & Caldwell, 1994). Likewise, a further understanding of the relationship between

understory vegetation density and herpetofaunal assemblages would benefit from a more

quantitative analysis of said vegetation, its composition, and potentially associated characteristics

that differentiate it from medium and low density understory vegetation other than visibility by

predators, such as microclimate and prey diversity and availability.

While the population of Adenomera hylaedactyla in the community was analyzed briefly

in this study, there was no conclusive evidence for why it was so much more abundant than all

other species encountered on Sumak Allpa. Though there was evidence that they are able to

occupy all habitat types with all levels of understory vegetation density, this does not explain

their proliferation, as other, less-abundant species also demonstrated these trends. Based off the

observed population trends and observations of other investigations of the species, a variety of

hypotheses may explain the high number of individuals in this population relative to other

species on the island.

The first of these hypotheses is that A. hyladactyla was significantly more influenced than

all other species by one or more of the aforementioned biotic and abiotic factors influencing frog

activity, such as temporality, climate, lunar cycle, and inundation. There also exists the

possibility that they may be able to consume a wider variety of prey or that they may have more

of their type of prey available to them than other species on the island. They also may be

significantly less sensitive than other species to human activity and thus were more likely than

other species to be encountered along the island’s maintained trails. Increased visual detectability

compared to other species, though less likely than the aforementioned hypotheses, is also another

possible explanation for the high frequency of their encounters, and any number of these

hypotheses could be tested simply with any of the aforementioned future study suggestions for

the community by simply changing their scope to the population.

Another hypothesis is that the population of A. hylaedactlya is actually an assemblage of

multiple morphs of the species, subspecies, or even distinct species. A study conducted in the

Peruvian Amazon Basin by Angulo, Cocroft, & Reichile (2003) summarized how the genus

Adenomera has been a difficult group for systematic studies and investigated A. andreae and A.

hylaedactyla, the two most encountered species on Sumak Allpa. The researchers examined

advertisement calls in relation to the frogs’ morphological characteristics and determined

significant enough differences in the morphologies and acoustic parameters of the calls of

associated individuals to indicate potentially four sympatric species derived from A. hylaedactyla

existing at the study site. Given subtle differences in morphological features of Adenomera

hylaedactyla were observed among individuals in Sumak Allpa’s population, a similar study to

that of Angulo, Cocroft, & Reichile (2003) may be merited on the island.

Finally, it also could be proposed that Sumak Allpa has the highest numbers of

individuals of small species and lowest numbers of individuals of large species because it is

experiencing a phenomenon known as excess density compensation. The theory posits that island

assemblages could be partitioned differently (few species or smaller individuals) from mainland

sites without differing in aggregate biomass (Rodda & Dean-Bradley, 2002). Such “excess

density compensation” may be understood as either a lack of interference competition, a lack of

overexploitation, or a release from predation on these islands (Case, Gilpin, & Diamond, 1979;

Wright, 1980).

Lack of interference competition means that this overcompensation of biomass density is

caused by exclusion of more efficient competitors from some portion of the resource spectrum

by inefficient species, allowing species-poor island populations or assemblages to harvest

resources more efficiently and support higher total population densities than species-rich

mainland communities (Wright, 1980).

Resource overexploitation occurs when consumers are so efficient on the mainland that

their resource abundances are reduced to the point that consumer abundances also must decline

until the consumer efficiency declines to the point where resource abundances stop being

reduced enough to also stop reduction of consumer abundances (Wright, 1980). If this

overexploitation occurs on the mainland but not the island, species-poor islands may support

higher resource abundances and higher total population densities of consumers than the species-

rich mainland (Wright, 1980).

Lastly, release from predation indicates that islands containing fewer predators than

corresponding mainland sites would allow for higher prey densities on islands than the mainland

(Wright, 1980).

This final hypothesis is by far the most involved and also probably the least likely,

especially given much of the debate surrounding the concept of excess density compensation in

general, yet there may be merit in investigating it as a potential alternative if the other

hypotheses do not account for the high numbers of A. hylaedactyla and other small-bodied

species on Sumak Allpa.

Ultimately, given its high density and abundance on Sumak Allpa, A. hylaedactyla should

be a priority species for population studies to determine why the island supports so many

individuals compared to other species. Population studies, nevertheless, should not be limited to

this species alone. There is merit in the assessment of all populations of the island, especially

considering the number of outdated IUCN Red List listings of species and unevaluated species

encountered on the island.

Though comprehensive studies of global and national assemblages of herpetofauna are

necessary to generate awareness of the world’s significant amphibian and reptile declines and

call to action the large-scale conservation efforts needed to combat them, these studies ultimately

are informed by the aggregation of small-scale studies monitoring local communities and

populations. Despite the immense species richness and diversity of herpetofauna across Ecuador

and the Neotropics, the study of specific herpetofaunal communities and populations overall is

pretty lacking. Thus, gaining a better understanding of the local interactions, functions,

behaviors, and ecologies of specific species, populations, and communities of herpetofauna in

threatened regions such as the western Amazon and the Neotropics is one of the biggest strides

that needs to be made for the conservation of the immense richness and diversity of these

regions. Thus, especially considering the lack of evaluation and continued monitoring of the

majority of species encountered at Sumak Allpa by the IUCN – the global authority on the status

of the natural world – any investigation filling one of the many knowledge gaps of herpetofauna

on Sumak Allpa is merited, warranted, and valuable work for the conservation of herpetofauna

on the island and in the region overall.

Bibliography

Aichinger, M. (1992) Fecundity and breeding sites of an anuran community in a seasonal tropical

environment. Studies on Neotropical Fauna and Environment 27:9-18.

AmphibiaWeb Ecuador. Diversidad Y Biogeografia. N.p., n.d. Web.

http://zoologia.puce.edu.ec/Vertebrados/Anfibios/AnfibiosEcuador/DiversidadEndemism

o.aspx accessed 11 May 2018

Angulo, A., Cocroft, R. B., & Reichle, S. (2003). Species identity in the genus Adenomera

(Anura: Leptodactylidae) in Southeastern Peru. Herpetologica, 59(4), 490–504.

https://doi.org/10.1655/20-104

Bartlett, R. D., & Bartlett, P. (2003). Reptiles and Amphibians of the Amazon: An Ecotourist’s

Guide. Gainesville: University Press of Florida.

Bass, M. S., Finer, M., Jenkins, C. N., Kreft, H., Cisneros-Heredia, D. F., McCracken, S. F., …

Kunz, T. H. (2010). Global Conservation Significance of Ecuador’s Yasuní National

Park. PLoS ONE, 5(1), e8767. https://doi.org/10.1371/journal.pone.0008767

Beirne, C., & Whitworth, A. (2011). Frogs of the Yachana Reserve. Exeter: Global Vision

International.

Bickford, D., Howard, S. D., Ng, D. J. J., & Sheridan, J. A. (2010). Impacts of climate change on

the amphibians and reptiles of Southeast Asia. Biodiversity and Conservation, 19(4),

1043–1062. https://doi.org/10.1007/s10531-010-9782-4

Böhm M, Collen B, Baillie EMJ, Chanson J, Cox N. et al. (2012) The conservation status of the

world’s reptiles. Biological Conservation 157: 372-385.

BIOWEB. (2018) BIOWEB Ecuador. https://bioweb.bio/. Accessed 11 May 2018

Brito M., J., Ojala-Barbour, R., Batallas R., D., & Almendáriz C., A. (2016). A New Species of

Pristimantis (Amphibia: Strabomantidae) from the Cloud Forest of Sangay National

Park, Ecuador. Journal of Herpetology, 50(2), 337–344. https://doi.org/10.1670/13-103

Brown, G. P., & Shine, R. (2002). Influence of weather conditions on activity of tropical snakes:

Tropical Snake Activity. Austral Ecology, 27(6), 596–605.

https://doi.org/10.1046/j.1442-9993.2002.01218.x

Bustamante, M. R., Ron, S. R., & Coloma, L. A. (2005). Cambios en la Diversidad en Siete

Comunidades de Anuros en los Andes de Ecuador1. Biotropica, 37(2), 180–189.

https://doi.org/10.1111/j.1744-7429.2005.00025.x

Case, T., Gilpin, M., & Diamond, J. (1979). Overexploitation, Interference Competition, and

Excess Density Compensation in Insular Faunas. The American Naturalist, 113(6), 843-

854. Retrieved from http://www.jstor.org.ccl.idm.oclc.org/stable/2460307

Centro Jambatu Fundación Otonga. (2011). Strategic plan for the conservation of the Ecuadorian

amphibians in extinction risk.

Chao, A., Ma, K. H., and Hsieh, T. C. (2016) iNEXT (iNterpolation and EXTrapolation) Online:

Software for Interpolation and Extrapolation of Species Diversity. Program and User’s

Guide published at http://chao.stat.nthu.edu.tw/wordpress/software_download/.

Chao, A., Ma, K. H., Hsieh, T. C. and Chiu, C. H. (2015) Online Program SpadeR (Species-

richness Prediction And Diversity Estimation in R). Program and User’s Guide published

at http://chao.stat.nthu.edu.tw/wordpress/software_download/.

Collins, JP, Crump, ML (2009) Extinction in our times: global amphibian decline. London,

Oxford University Press.

Cortes, A. M., Ruiz-Agudelo, C. A., Valencia-Aguilar, A., & Ladle, R. J. (2014). Ecological

functions of neotropical amphibians and reptiles: a review. Universitas Scientiarum,

20(2), 229. https://doi.org/10.11144/Javeriana.SC20-2.efna

Costa, A., Posillico, M., Basile, M., Romano, A., 2016. Conservation of herpetofauna as a

function of forestry. Ital. J. Agr. 11 (s1), 38–41

Deichmann, J. L. (2009). The effects of geomorphology and primary productivity on neotropical

leaf litter herpetofauna: implications for Amazonian rainforest conservation. LSU

Doctoral Dissertation, Baton Rouge.

Doan, T. M. (2003). Which Methods Are Most Effective for Surveying Rain Forest

Herpetofauna? Journal of Herpetology, 37(1), 72–81. https://doi.org/10.1670/0022-

1511(2003)037[0072:WMAMEF]2.0.CO;2

Doan, T. M., & Arriaga, W. A. (2002). Microgeographic Variation in Species Composition of

the Herpetofaunal Communities of Tambopata Region, Peru1. Biotropica, 34(1), 101–

117. https://doi.org/10.1111/j.1744-7429.2002.tb00246.x

Eekhout, X. (2011). Sampling Amphibians and Reptiles. In Manual on Field Recording

Techniques and Protocols for ATBI+M (Vol. 8). Madrid.

Grant, R. A., Chadwick, E. A., & Halliday, T. (2009). The lunar cycle: a cue for amphibian

reproductive phenology? Animal Behaviour, 78(2), 349–357.

https://doi.org/10.1016/j.anbehav.2009.05.007

Gibbons JW, Scott DE, Ryan TJ, Buhlmann KA, Tuberville TD, Metts BS, Greene JL, Mills T,

Leiden Y, Poppy S, Winne CT (2000) The global decline of reptiles, deja-vu amphibians.

Bioscience 50: 653-667.

Heatwole, Harrold. (2013). Worldwide decline and extinction of amphibians. Chapter 18, pg.

259-278 in K. Rohde, The Balance of Nature and Human Impact, Cambridge University

Press, UK.

Heyer, W. R. (Ed.). (1994). Measuring and monitoring biological diversity. Standard methods

for amphibians. Washington: Smithsonian Institution Press.

Hocking, D. J., & Babbitt, K. J. (2014). Effects of Red-Backed Salamanders on Ecosystem

Functions. PLoS ONE, 9(1), e86854. https://doi.org/10.1371/journal.pone.0086854

IUCN. (2018) The IUCN red list of threatened species.

http://www.iucnredlist.org/apps/redlist/search. Accessed 11 May 2018.

J. Vitt, L., & Caldwell, J. P. (1994). Resource utilization and guild structure of small vertebrates

in the Amazon forest leaf litter. Journal of Zoology, 234(3), 463–476.

https://doi.org/10.1111/j.1469-7998.1994.tb04860.x

Lambert, M. R. K., 1984. Amphibians and Reptiles. In: J. L. Cloudsley-Thompson. (ed.). Sahara

Desert. Key Environments. Pergamon Press. Oxford: 205-227.

Lips KR, Brem F, Brenes R, Reeve JD, Alford RA, Voyle J, Carey C (2006) Emerging infectious

disease and the loss of biodiversity in a Neotropical amphibian community. PNAS

103:3165-3170.

MacArthur, R. H., & Wilson, E. O. (2016). The Theory of Island Biogeography. Princeton:

Princeton University Press.

Menin, M., Waldez, F., & Lima, A. P. (2008). Temporal variation in the abundance and number

of species of frogs in 10,000 ha of a forest in central amazonia, Brazil. South American

Journal of Herpetology, 3(1), 68–81. https://doi.org/10.2994/1808-

9798(2008)3[68:TVITAA]2.0.CO;2

Paladino, F. V. (1985). Temperature Effects on Locomotion and Activity Bioenergetics of

Amphibians, Reptiles, and Birds. American Zoologist, 25(4), 965–972.

https://doi.org/10.1093/icb/25.4.965

Pough FH, Andrews RM, Cadle JE, Crump ML, Savitzky AH, Wells KD (2004) Herpetology.

Upper Saddle River, Prentice-Hall.

Read, M. (2011). Frogs of the Ecuadorian Amazon (Vol. Version 1.2).

Reading CJ, Luiselli LM, Akani GC, Bonnet X, Amori G, Ballouard JM, Filippi E, Naulleau G,

Pearson D, Rugiero L (2010) Are snake populations in widespread decline? Biology

Letters 6:1-4.

Reyes-Puig C, Almendáriz C A, Torres-Carvajal O. 2017. Diversity, threat, and conservation of

reptiles from continental Ecuador. Amphibian & Reptile Conservation 11(2) [General

Section]: 51–58 (e147).

Reyes-Puig, J. P., & Yánez-Muñoz, M. H. (2012). Una nueva especie de Pristimantis (Anura:

Craugastoridae) del corredor ecológico Llangantes-Sangay, Andes de Ecuador. Papéis

Avulsos de Zoologia (São Paulo), 52(6), 81-91.

Rodda, G. H., & Dean-Bradley, K. (2002). Excess density compensation of island herpetofaunal

assemblages. Journal of Biogeography, 29(5-6), 623–632. https://doi.org/10.1046/j.1365-

2699.2002.00711.x

Rödel MO, Ernst R (2004) Measuring and monitoring amphibian diversity in tropical forests. An

evaluation of methods with recommendations for standardization. Ecotropica 10(1): 1–

14.

Rodrigues, A., Pilgrim, J., Lamoreux, J., Hoffmann, M., & Brooks, T. (2006). The value of the

IUCN Red List for conservation. Trends in Ecology & Evolution, 21(2), 71–76.

https://doi.org/10.1016/j.tree.2005.10.010

Stuart SN, Chanson JS, Cox NA, Young BE, Rodrigues AS, Fischman DL, Waller RW (2004)

Status and trends of amphibian declines and extinctions worldwide. Science 306:1783-

1786.

Todd BD, Luhring TM, Rothermel BB, Gibbons JW, 2010. Effects of forest removal on

amphibian migrations: implications for habitat and landscape connectivity. J. Appl. Ecol.

46:554-561.

Tyler MJ, Wassersug R, Smith B (2007) How frogs and humans interact: influences beyond

habitat destruction, epidemics and global warming. Applied Herpetology 4:1-18.

Valencia, J. H., Toral, E., Morales, M., Betancourt, R., & Barahona, A. (2008a). Guía de Campo

de Anfibios del Ecuador. Fundación Herpetológica Gustavo Orcés.

Valencia, J. H., Toral, E., Morales, M., Betancourt, R., & Barahona, A. (2008b). Guía de Campo

de Reptiles del Ecuador. Quito: Fundación Herpetológica Gustavo Orcés.

Wells KD (2007) The ecology and behavior of amphibians. USA, The University of Chicago

Press.

Wright, S. J. (1980). Density compensation in island avifaunas. Oecologia, 45(3), 385–389.

https://doi.org/10.1007/BF00540211

Young, B.E., Stuart, S.N., Chanson, J.S., Cox, N.A., Boucher, T.M. (2004) Joyas que están

desapareciendo: el estado de los anfibios en el nuevo mundo. Arlington-Virginia,

NatureServe.

Zewdie, S. (2017). Development of a Wild Infant Woolly Monkey: Social Interactions, Time

Allocation and Behavior of a Wild Lagothrix lagotricha poeppigii Infant and Her

Mother. Quito.

Zug, GR., Vitt, LJ, Caldwell J,. 2001. Herpetology: An Introductory Biology of Amphibians and

Reptiles. Academic Press.

Appendix

A1. Hill’s number of order q, q ≥ 0:

and

A2. Sørensen similarity index for multiple community measures:

A3. Classic Bray-Curtis similarity index:

A4. Sørensen similarity index for the special case of two communities: