*Corresponding author: molinam@unesur.edu.ve

Misael Molina

1,2*

Marina Mazón

Rev. Fac. Agron. (LUZ). 2022, 39(2): e223951

ISSN 2477-9407

DOI: https://doi.org/10.47280/RevFacAgron(LUZ).v39.n4.06

Crop production

Associate editor: Dr. Rosa Razz

University of Zulia, Faculty of Agronomy

Bolivarian Republic of Venezuela

Keywords:

Biomass

Connectivity

Merida

Notosciurus granatensis

Shade

Theobroma cacao

Venezuela

Agrolandscape structure and damage caused by red squirrels to cocoa pods

Estructura del agropaisaje y daño causado por ardillas rojas a mazorcas de cacao

Instrutura da agropaisagem y dano causado por esquilos vermelhos às vagens de cacau

Universidad Nacional Experimental Sur del Lago.

Venezuela.

Programa de Doctorado en Ciencias Agrarias. Facultad de

Agronomía, Universidad del Zulia. Maracaibo, Venezuela.

Centro de Investigaciones Tropicales del Ambiente y

Biodiversidad. Universidad Nacional de Loja. Loja, Ecuador.

Received: 01-09-2022

Accepted: 18-10-2022

Published: 15-11-2022

Abstract

Cocoa is a strategic resource because it hosts high biodiversity, and it is

a reliable source of foreign exchange. In Venezuela, fungi and red squirrels

mainly affect its productivity. The objective was to evaluate the inuence

of the cocoa crop structure on the level of damage caused by red squirrels

on the pods. The vegetation structure was dened in 15 farms assessing the

cocoa tree biomass, the magnitude of shade tree cover, number of associated

crops, presence of alternative fruit trees, and the type of ecological

succession present in the surrounding. The % of damage was calculated, and

nonparametric tests were used to process data. The damage was higher in

crops with less shade cover, a more abundant number of associated crops,

and alternative fruit trees. Squirrels used the cocoa pods opportunistically but

preferred other fruits. It is concluded that pods can be a complementary food

for squirrels to the extent that the cocoa plantation is more complex and has a

connection with the forest. Shade cover did not directly inuenced squirrels,

but it is crucial for bird predators which are more sensitive to disturbance.

This scientic publication in digital format is a continuation of the Printed Review: Legal Deposit pp 196802ZU42, ISSN 0378-7818.

Rev. Fac. Agron. (LUZ). 2022, 39(4): e223951. October-December. ISSN 2477-9407.2-6 |

Resumen

El cacao es un recurso estratégico porque aloja una biodiversidad

alta y es una fuente conable de divisas. En Venezuela su productividad

se ve afectada principalmente por hongos y ardillas rojas. El objetivo

fue evaluar la inuencia de la estructura del cultivo de cacao sobre el

nivel de daño causado por ardillas a las mazorcas. La estructura de la

vegetación se denió en 15 ncas midiendo la biomasa del árbol de

cacao, la magnitud de la cobertura de los árboles de sombra, el número

de cultivos asociados, la presencia de árboles frutales alternativos y el

tipo de sucesión ecológica presente en el entorno. Se calculó el % de

daño y se usaron pruebas no paramétricas para procesar los datos. El

daño fue mayor en cultivos con menor cobertura de sombra, mayor

número de cultivos asociados y de frutales alternativos. Las ardillas

usaron las bayas de cacao de manera oportunista, pero prerieron otros

frutos. Las mazorcas pueden constituir alimentos complementarios

para las ardillas en la medida en que la plantación de cacao es más

compleja y tiene una conexión con el bosque. La cobertura de sombra

no inuyó directamente en las ardillas, pero puede ser crucial para las

aves depredadoras que son más sensibles a las perturbaciones.

Palabras clave: biomasa, conectividad, Notosciurus granatensis,

Mérida, Theobroma cacao, sombra, Venezuela

Resumo

O cacau é um recurso estratégico porque abriga alta biodiversidade

e é uma fonte conável de divisas. En Venezuela su productividad se

ve afectada principalmente por hongos y squilos. O objetivo é avaliar

a inuência da estrutura do cultivo de cacau no nível do dano causado

por esquilos às mazorcas. A estrutura da vegetação é denida em

15 fazendas midiendo a biomasa do árbol de cacau, a magnitude da

cobertura das árboles de sombra, o número de cultivos associados, a

presença de árboles frutas alternativas e o tipo de sucessão ecológica

presente no entorno. Foi calculado el % de daño e foram utilizadas

testes não paramétricas para processar os dados. El daño fue mayor en

cultivos com menor cobertura de sombra, maior número de cultivos

associados e de frutas alternativas. Os squilos ussaram as bagas de

cacau de manera oportunista, pero prerieron outros frutos. Conclui-

se que as vainas podem ser o alimento complementar para los squilos

na medida em que a plantação de cacau é mais completa e tem uma

conexão com o bosque. A cobertura de sombra não inuenciou

diretamente nos squilos, mas é crucial para as aves depredadoras que

são mais sensíveis às perturbações.

Palavras-chave: biomassa, conectividade, Mérida, Notosciurus

granatensis, sombra, Theobroma cacao, Venezuela

Introduction

The vegetation structure in the agrolandscape refers to the

composition of species and the spatial setup of plants, both in the

horizontal and vertical planes (Gidoin et al., 2015). Plant species have

different growth forms (Verbeeck et al., 2019), and a greater diversity

of them implies increased complexity (Gough, 2020). The vegetation

structure and the biodiversity are linked, and they control critical

ecosystem’s attributes (Walter et al., 2021).

The agrolandscape attributes inuence the abundance, density,

and magnitude of the damage caused by vertebrates (Holloway et al.,

2012; Daghela et al., 2013; Garcés-Restrepo and Saavedra-Rodríguez,

2013). On the other hand, since the agrolandscape can be shaped by

agricultural management, this management might affect the damage

caused by pests and diseases, thus understanding how these animals

respond to crop management becomes essential (Canavelli et al.,

2014; Milosavljevic et al., 2016).

Although few works relate the vegetation structure with the level

of damage caused by small arboreal rodents, this parameter seems to

determine the abundance of many species (Craig and Gese, 2013). In

the specic case of squirrels, a higher number of shade trees contribute

to forming a greater structural complexity that favors colonization

and permanence (Wilson et al., 2008). In the same way, the tree

branches facilitate their mobility (Flaherty et al., 2012; Holloway et

al., 2012), as well as the construction of their nests (Holloway and

Malcolm, 2007).

Cocoa crop hosts a higher biodiversity in agroforestry systems

(Cassano et al., 2011; Sonwa et al., 2018) compared to traditional

crops. However, how the structure of the cocoa farm determines the

level of rodent damage to the pods is almost unknown.

Smith and Nott (1988) found that squirrels nd shelter in cocoa

crops with extended shade cover. Monge and Hilje (2006) noted

that the squirrel Sciurus variegatoides varies its diet depending on

the tree’s architecture, which determines its protection and access to

food. On the other hand, Alves de Sousa Silva et al. (2008) found that

the small mammal diversity in cocoa agroforestry systems in Brazil,

was favored by the structural complexity of the crops conditioned by

a greater diversity of niches. In their study, structural complexity was

limited to shade cover.

In the neotropics, the red squirrel (Notosciurus granatensis) causes

considerable damage to cocoa pods, and its geographic distribution

coincides with that of this crop (Warren and Emmandie, 1993).

The objective was to evaluate the inuence of the cocoa plantation

structure on the level of damage caused to the pods by red squirrels

using the following response variables: the cocoa biomass, the

magnitude of shade cover, the number of associated crops, the presence

of alternative fruit trees, and the type of ecological succession present

in the surrounding vegetation. The study started from the hypothesis

that the magnitude of the damage caused by squirrels increases with

the structural complexity.

Materials and methods

Agroecological area

The study was carried out in cocoa farms at the Cordillera de

Mérida characterized by mountainous landscapes with hills, cones,

and valleys with slopes ranging 3-30°, shaped by quaternary erosion

on tertiary formations. The soils are weakly acidic, with good

drainage, quite weathered and moderately fertile in the steepest areas,

but very fertile in the plains (Vivas, 1992). According to Huber et al.

(2010), the area was initially composed of two kinds of vegetation:

semi-deciduous forests, with trees up to 35 m tall, and two tree strata,

up to 800 m.a.s.l.; and evergreen forests, located between 800 and

1800-2000 m.a.s.l.

Population and sample

The population was 1,068 cocoa-producing farms established in

the study area (Portillo et al., 1995). The study was carried out in 15

farms located in Mesa Bolívar, Mesa Julia, Río Frio and Zea, located

between 300 and 900 m.a.s.l (table 1). Farms were selected by their

accessibility and by having the consent of the farm owner.

This scientic publication in digital format is a continuation of the Printed Review: Legal Deposit pp 196802ZU42, ISSN 0378-7818.

Molina and Mazón. Rev. Fac. Agron. (LUZ). 2022, 39(4): e2239513-6 |

Table 1. Relative location of the cacao farms in Merida state,

Venezuela.

Farm name Coordinates N/W Parish Municipality

Santa Lucia 8°28’50’’/71°37’80’’

Mesa

Bolivar

Antonio Pinto Salinas

San Rafael 8°28’41’’/71°37’84’’

Mesa

Bolivar

Antonio Pinto Salinas

Mi Refugio 8°27’88’’/71°38’12’’

Mesa

Bolivar

Antonio Pinto Salinas

La Vega 8°30’92’’/71°38’74’’

Mesa

Bolivar

Antonio Pinto Salinas

La Escondida 8°27’94’’/71°38’76’’

Mesa

Bolivar

Antonio Pinto Salinas

Los Manzanos 8°30’92’’/71°38’75’’

Mesa

Bolivar

Antonio Pinto Salinas

La Florida 8°28’33’’/71°39’28’’

Mesa

Bolivar

Antonio Pinto Salinas

La Trinidad 8°27’54’’/71°39’47’’

Mesa

Bolivar

Antonio Pinto Salinas

Canaima 8°58’25’’/71°15’59’’

Mesa

Julia

Caracciolo Parra y Olmedo

La Esperanza 8°54’27’’/71°18’42’’ Río Frio Caracciolo Parra y Olmedo

El Rosal 8°53’26’’/71°18’35’’ Río Frio Caracciolo Parra y Olmedo

La Montaña 8°52’40’’/71°18’03’’ Río Frio Caracciolo Parra y Olmedo

Los Taguanes 8°56’70’’ /71°14’75’’ Río Frio Caracciolo Parra y Olmedo

La Hedionda 8°24’44’’/71°47’37’’ Zea Zea

Labradores 8°23’25’’/71°47’88’’ Zea Zea

Crop characteristics

Most crops have 2 ha or fewer; criollo cocoas predominate,

although it is common to nd hybrids and, less frequently, foreign

cocoas, with plants separated 3 x 3 m, and weeds controlled mainly

using mechanical methods. Most farms have a reduced shade cover

(gure 1A), low richness of shade trees in two strata: the upper one

made up of native trees (gure 1B), mainly cedar (Cedrela odorata),

pardillo (Cordia alliodora), and bucare (Erythrina poeppigiana),

although caraño (Protium sp.), jobo (Spondias mombin) and higueron

(Ficus spp.) trees are sometimes found; and a lower stratum made up

mainly of E. poeppigiana, and guamo (Inga spp.), with some sparse

surure (Myrcia fallax) and yagrumo (Cecropia spp.), and even exotic

species such as bamboo (Bambusa vulgaris).

In some cases, a third vertical layer in the shade vegetation

was found (gure 1C), made up of fruit trees: avocado (Persea

americana), soursop (Annona muricata), orange (Citrus cinensis),

mandarin (Citrus reticulata), lemon (Citrus auratiifolia), banana

(Musa AAA) and plantain banana (Musa AAB).

Figure 1. Cocoa crops with permanent shade made up of timber

trees in the upper stratum (A) and intermediate (B),

and of fruit trees in the lower stratum (C).

Source: Misael Molina.

Observing the shape and size of the holes left by the squirrels after

biting the cocoa pods (gure 2A) and by woodpeckers (gure 2B)

after pecking the fruits, the damage was attributed to these rodents,

discarding the birds.

Sampling and evaluation

Twelve plants were randomly selected in each farm, and during the

rainy season of 2018 and the dry season of 2021, the number of pods

bitten by the squirrels and the total number of pods occurring in the

plant were counted twice (once in each season). Then, the percentage

of damage was calculated. Likewise, during the dry season of 2021

the following independent variables were measured: biomass of cocoa

trees, shade vegetation cover, number of associated crops and number

of alternative fruit trees. Also, the type of ecological succession in

the surrounding vegetation was observed and was classied in three

categories: other cocoa plantations, stubble, secondary forests, and

grasslands; when different types were occurring, the dominant one

(i.e., more than 50% of crop perimeter) was considered.

The cocoa biomass was calculated using the equation derived by Huy

et al. (2016) for tropical trees: AGB = N*0.06*DBH

2.30

0.50

*WD

0.80

where AGB is aboveground biomass, N is the number of plants.ha

-1

DBH is plant diameter at breast height, H is plant height, and WD is

cocoa wood density (0.42 g.(cm

)

-1

Figure 2. Cocoa pods damaged by squirrels (A) and woodpeckers

(B), note size and shape of hole.

Source: Misael Molina.

The shade vegetation cover was measured selecting randomly

one hectare in each farm which was then divided into 100 grids of

10 x 10 m. Then 15 of these grids were randomly selected to take

a low-angle photograph from the center of each grid. The same

person always took the photographs using a ZTE Blade V20 cell

phone, leveling the phone horizontally and consistently placing it at

face height. Subsequently, using the ImageJ program (Ferreira and

Dasband, 2012), each photograph was converted to a binary format

to have black and white images, and the percentage of black pixels

represents the surface covered by shade vegetation. The average of

these 15 values was taken as the percentage of shade vegetation cover

in every farm.

Furthermore, the number of associated crops on each farm was

counted, and the presence of alternative fruit trees (trees other than

cocoa that produce fruits consumed by squirrels) was veried.

Statistical data processing

The independent quantitative variables were standardized to

reduce differences due to scales: mean and standard deviation were

calculated, and then the original value was subtracted from the mean

This scientic publication in digital format is a continuation of the Printed Review: Legal Deposit pp 196802ZU42, ISSN 0378-7818.

Rev. Fac. Agron. (LUZ). 2022, 39(4): e223951. October-December. ISSN 2477-9407.4-6 |

Bss: biomass (kg.ha

-1

). Succession categories: 1 other cocoa

crops; 2 stubbles; 3 secondary forests; 4 grasslands).

Cocoa biomass

The biomass of cocoa pods ranged 12,810-424,550 kg.ha

-1

, but

there were not statistically signicant differences (p>0.05) in the

level of damage among the four categories.

Shade coverage

Similarly, there were not statistically signicant differences

(p>0.05) in the level of damage among the categories of shade

trees (gure 3). However, shade cover and damage level showed a

negative correlation (r = -0.046). Shade trees contribute to forming

a greater structural complexity that favors the presence and mobility

of squirrels through the branches (Flaherty et al., 2012; Holloway

et al., 2012), and the construction of nests (Holloway and Malcolm,

2007). Therefore, a higher level of damage could be expected in

crops with great shade cover. But our results indicate the opposite,

also contrasting with the work of Smith and Nott (1998), who

argued that squirrels nd protection in cocoa crops with extensive

shade cover.

Figura 3. Relation between the level of damage (%) by red squirrels

and the shade cover in cacao farms of Merida state,

Venezuela. The categories of shade cover are: I: 18-34%, II:

35-51%, III: 52-68%, IV: >69%. Same letters indicate non-

signicant statistical differences (p>0,05).

However, the results might be explained within the theoretical

framework of the prey-predator relationship. On one hand, birds of

prey are known to be primarily predators of squirrels (wildlifeonline.

me.uk/article/squirrel-predators), and on the other hand, some

squirrels can get used to interacting with humans (Engelhardt and

Weladji, 2011; Kittendorf and Danzer, 2021). Nevertheless, while

squirrels exhibit behavioral elasticity to take advantage of resources

in human-altered landscapes, raptors are sensitive to the scarcity of

shaded areas (Sagorski and Swihart, 2021; Tinajero et al., 2017).

Therefore, squirrels learn to recognize the risk of both predation by

birds and direct harm from humans and can weigh the predation risks

against the advantage of staying close to a safe and good-quality

food source (Badrie et al., 2015) and transmit that information to

the progeny (Engelhardt and Weladji, 2011; Uchida et al., 2017;

Uchida, 2019). As a result, a greater abundance of squirrels can be

found in these places (Engel et al., 2020), thus it is reasonable to

nd a higher level of damage in areas with less shade cover but

and this result was divided by the standard deviation. Then, variables

were organized into four categories each as follows: four groups for

the biomass of the cocoa plants (I: 12,810-103,000; II: 104,000-

207,000; III: 208,000-311,000; IV: 312,000-425,000 kg.ha

-1

), four

groups for the cover of shade vegetation (I: 18-34; II: 35-51; III:

52-68; IV: >69%), four groups for the number of associated crops

(I: 0-1; II: 2-3; III: 4-5; IV: 6-7), and four groups for the ecological

succession present in the surrounding vegetation (I: other cocoa

crops; II: stubbles; III: secondary forests; IV: grasslands). Only

presence of alternative fruits was organized into two groups (I:

present; II: absent).

A Kruskal-Wallis Analysis of Variance for independent samples

was run for the variables with four categories, and a Kolmogorov test

for the independent samples of the presence/absence of alternative

fruits. The program SPSS20 was used to process data (IBM, 2011).

Results and discussion

Level of damage by red squirrels

The average level of damage was 6.33% for the farms located

in Mesa Bolivar, 13% for Mesa Julia, 3.20% for Rio Frio, and 5.30

for Zea (table 2). These results contrast with those of López et al.

(2014), who found losses between 26 and 34% for Nicaragua, and

those of Mollineau et al. (2008), who reported losses higher than

30% for Trinidad and Tobago, despite including farms with similar

management conditions than those evaluated in this research.

Although our values are lower than those reported by these authors,

it is signicant for Merida producers since they face a crisis reected

through scarcity and high costs of agro-inputs and labor.

Table 2. Calculated values for the structural variables of cocoa

farms and the damage caused by red squirrels to the

cocoa pods in Merida state, Venezuela.

Location

Farm

name

damage

Bss*

cocoa

Nº

associated

crops

shade

cover

Succession

category

Mesa

Bolivar

Santa Lucia 3.73 78930 2 49.10 2

Candelaria

4.77 140950 0 66.03 2

Mi Refugio 24.94 12810 2 42.33 3

La Vega 4.56 314110 7 35.31 3

Escondida

4.48 232560 4 66.19 2

Los

Manzanos

5.61 74440 1 46.06 2

La Florida 2.51 145650 2 41.76 4

La Trinidad 0.00 247300 5 18.60 4

Mesa

Julia

Canaima 13 107800 1 34.80 3

Rio Frio

Esperanza

2.27 215400 1 54.24 3

El Rosal 1.60 248630 3 83.93 2

Los

Taguanes

1.86 424550 4 67.58 1

La Montaña 7.06 92550 4 63.23 3

Zea

Hedionda

7.56 85860 3 58.34 3

Labradores 3.04 132670 4 44.65 3

This scientic publication in digital format is a continuation of the Printed Review: Legal Deposit pp 196802ZU42, ISSN 0378-7818.

Molina and Mazón. Rev. Fac. Agron. (LUZ). 2022, 39(4): e2239515-6 |

having more abundant food. Also, they might nd safe escape routes

against predators since the high connectivity between cocoa plants

and surrounding vegetation guarantees it. However, an extensive

shade cover does not seem to guarantee a higher connectivity to

facilitate red squirrel mobility. In this study that mobility occurred

between the cocoa plants, but not through the branches of the middle

and upper strata.

Number of associated crops

The magnitude of the damage did not differ (p>0.05) between the

categories of associated crops. However, it was higher in farms with

two to three associated crops (category II, gure 4). It was expected

that the damage would increase as the amount of intercropping was

reduced; a correlation test corroborated this trend (r = -0.20) but

was not signicant (p>0.05). This result seems to indicate that not

the number, but the type of crops may shape the rodent’s occurrence,

since some of them are fruit trees highly valuable for them (Reher et

al., 2016), such as avocado (P. americana). That is why damage was

more distributed among the different crops.

Figure 4. Relationship between the level of damage (%) by red squirrels

and the number of associated crops in cocoa farms of Merida

state, Venezuela. Same letters indicate non-signicant statistical

differences (p>0.05). The categories of associated crops are I: 0 to

1, II: 2 to 3, III: 4 to 5, IV: 6 to 7.

Presence of alternative fruit trees

The magnitude of the damage was higher in crops with alternative

fruits, but differences were non-signicant (p>0.05) to that in farms

without these fruits. Squirrels are known to respond to the presence

of fruit trees (Monge and Hilje, 2006). For N. granatensis, there

are no detailed studies published yet, although Reher et al. (2016)

pointed out that the abundance of Sciurus vulgaris is related to the

existence of food sources. However, according to Monge and Hilje

(2006) the damage is diluted when crops are diverse because squirrels

are opportunistic and take advantage of easily accessible resources.

Therefore, cocoa pods might appear only as alternative food for

squirrels when their preferred fruits are not available.

Ecological succession present in the surrounding vegetation

The level of damage did not differ (p>0.05) among the categories

of the surrounding vegetation. The damage was minimal or inexistent

when pastures surrounded the cocoa plantations, but it was greater

when surrounded by forest. Particularly, damage was signicantly

higher when farms were surrounded by some kind of forest (gure

5). This result is consistent with the highest damage founded in crops

with less shade cover. On the other hand, the damage observed in

crops surrounded by secondary forests had a similar value to that of

other cocoa crops. Forest-dwelling squirrels feed opportunistically

on cocoa pods and alternative fruits, as occurs with Sciurus aestuans

(Alvarenga and Talamoni, 2006) and S. variegatoides (Monge and

Hilje, 2006). Silva-Ferreira et al. (2020) founded that forest remnants

are essential to support mammalian populations in cocoa crops.

Mobility within and between the crop and the forest is critical (Monge

and Hilje, 2006; Mäkeläinen et al., 2016), which in the crops studied

here were guaranteed by the overlap between branches of different

plants.

Figure 5. Relationship between the magnitude of the damage

(%) by red squirrels and the type of surrounding

vegetation of cocoa farms in Merida state, Venezuela. I:

other cocoa plantations, II: stubble, III: secondary forests.

Different letters indicate signicant statistical differences

(p<0.05).

Because these rodents play the role of prey for several vertebrate

species, mainly birds of prey, they avoid open habitats like the

grasslands, where the risk of predation is high (Potash et al., 2019).

Conclusions

The damage to cocoa pods was higher in crops with less

shade cover, with more associated crops, and with the presence of

alternative fruit trees. Shade cover did seem to indirectly inuence

squirrels by favoring the presence of bird predators, which are more

sensitive to disturbance. It seemed that red squirrels use the cocoa

pods opportunistically, only when their preferred fruits are not

available, and therefore ripe cocoa fruits are a complementary food

for squirrels.

Literature cited

Alvarenga, C. A. and Talamoni, S. (2006). Foraging behaviour of the Brazilian

squirrel Sciurus aestuans (Rodentia, Sciuridae). Acta Theriologica, 51(1),

69-74. https://doi.org/10.1007/BF03192657

Alves de Sousa Silva, A., Del Valle-Álvarez, M., Mariano-Neto, E. & Righetto-Cassano,

C. (2018). Is shadier better? The effect of agroforestry management on small

mammal diversity. Biotropica, 00, 1-10. https://doi.org/10.1111/btp.12750

This scientic publication in digital format is a continuation of the Printed Review: Legal Deposit pp 196802ZU42, ISSN 0378-7818.

Rev. Fac. Agron. (LUZ). 2022, 39(4): e223951. October-December. ISSN 2477-9407.6-6 |

Badrie, N., Bekele, F., Sikora, E. & Sikora, M. (2015). Cocoa agronomy, quality,

nutritional, and health aspects. Critical Reviews in Food Science and

Nutrition, 55, 620-659. https://doi.org/10.1080/10408398.2012.669428.

Canavelli S. B., Branch L. C., Cavallero O., González C. & Zaccagnini M. E.

(2014). Multi-level analysis of bird abundance and damage to crop elds.

Agriculture, Ecosystems and Environment, 197, 128-136. https://doi.

org/10.1016/J.AGEE.2014.07.024.

Cassano, C., Kierulff, M. C. M. & Chiarello, A. G. (2011). The cacao agroforests

of the Brazilian Atlantic Forest as habitat for the endangered maned

sloth Bradypus torquatus. Mammalian Biology, 76, 243-250. https://doi.

org/10.1016/j.mambio.2010.06.008.

Craig, T. and Gese, E. (2013). Inuence of vegetation structure on the small

community in a shortgrass prairie ecosystem. Acta Theriologica, 58, 55-

61. https://doi.org/10.1007/s13364-012-0098-5.

Daghela, H. B., Fotio, D., Missoup, A. D. & Vidal, S. (2013). Shade Tree

Diversity, Cocoa Pest Damage, Yield Compensating Inputs and Farmers’

Net Returns in West Africa. Plos One, 8(3), 1-9. https://doi.org/10.1371/

journal.pone.0056115.

Engelhardt S. C. and Weladji, R. B. (2011). Effect of levels of human exposure on

ight initiation distance and distance to refuge in foraging eastern gray

squirrels (Sciurus carolinensis). Canadian Journal of Zoology, 89, 823-

830. https://doi.org/10.1139/Z11-054.

Engel, L. D., Carlen, E. J., Losos, J. B. & Winchell, K. M. (2020). Eastern gray

squirrels (Sciurus carolinensis) differ in abundance and response to

humans across urban habitats of St. Louis. Urban Naturalist, 33, 1-16.

https://www.eaglehill.us/urna-pdfs-regular/urna-033-Winchell.pdf

Ferreira, T. and Dasband, W. (2012). Imagej user guide. Revised edition. 187 p.

Flaherty, S., Patenaude, G., Close, A. & Lurz, P. W. (2012). The impact of forest

stand structure on red squirrel habitat use. Forestry, 85(3), 437-444.

https://doi.org/10.1093/forestry/cps042.

Garcés-Restrepo, M. F. and Saavedra-Rodríguez C. A. (2013). Densidad de ardilla

roja (Sciurus granatensis) en hábitats con diferentes coberturas vegetales

en los Andes de Colombia. Mastozoología Neotropical, 20, 381-386.

https://www.redalyc.org/pdf/457/45729294011.pdf

Gidoin C., Babin R., Bagny Beilhé L., Barbu C., Gosme M., Jeuffroy M-H.,

Ngo Bieng M. A., Valantin-Morison M. and Martijn ten Hoopen,

G. (7-10 September 2015). Multi-scale studies of the relationships

between cropping structure and pest and disease regulation services.

International Symposium for Farming Systems Design. Montpellier,

France.

Gough, C. M., Atkins, J. W., Fahey, R. T., & Hardiman, B. S. (2019). High rates of

primary production in structurally complex forests. Ecology. 100(10), 1-6.

https://doi.org/10.1002/ecy.2864

Holloway G. L., Smith W. P., Halpern C. B., Gitzen R. A., Maguire C. C. & West

S. D. (2012). Inuence of forest structure and experimental green-tree

retention on northern ying squirrel (Glaucomys sabrinus) abundance.

Forest Ecology and Management, 285, 187-194. https://doi.org/10.1016/j.

foreco.2013.10.026.

Holloway, G. and Malcolm, J. (2007). Nest-tree used by Northern and Southern

ying squirrels in Central Ontario. Journal of Mammalogy, 88(1), 226-

233. https://doi.org/10.1644/05-MAM-A-368R2.1.

Huber, O., Oliveira-Miranda, M., Rodríguez, J. P., Rojas-Suárez, F. & Giraldo-

Hernández, D. (2010). Libro rojo de los ecosistemas terrestres de

Venezuela. Provita. https://ecosistemasamenazados.org/les/libro_rojo_

ecosistemas_terrestres_Venezuela.pdf

Huy, B., Poudel, K., Kralicek, K., Dinh Hung, N., Van Khoa, P., Tan Phương,

V. & Temesgen, H. (2016). Allometric Equations for Estimating Tree

Aboveground Biomass in Tropical Dipterocarp Forests of Vietnam.

Forests, 7, 1-19. https://doi.org/10.1016/j.foreco.2016.10.021.

IBM Corp. (2011). IBM SPSS Statistics for windows, Version 20.0. Armonk, NY:

IBM Corp.

Kittendorf, A. and Danzer, B. (2021). Urban fox squirrels exhibit habituation to

humans but respond to stimuli from natural predators. Ethology, 9,697-

709. https://doi.org/doi:10.1111/eth.13206.

López N., Flores E., Castillo J., & Montalvan O. (2014). Plagas en Cacaotales.

Municipio Siuna. Ciencia e Interculturalidad. 14, 106-114. https://doi.

org/10.5377/rci.v14i1.1502

Mäkeläinen, S., de Knegt, H., Ovaskainen, O. & Hanski, I. (2016). Home-range

patterns and movements of the Syverian ying squirrel in urban forests:

effects of the habitat composition and connectivity. Movement Ecology,

14, 4-13.

Milosavljevic I., Esser A. D. & Cowder J. W. (2016). Effects of environmental

and agronomic factors on soil-dwelling pest communities in cereal crops.

Agriculture, Ecosystems and Environment, 225, 192-198. https://doi.

org/10.1016/j.agee.2016.04.006.

Mollineau F., Bekele F. & García G. (2008). The Neo-tropical red squirrel (Sciurus

granatensis) as a pest of cacao (Theobroma cacao L.) in the International

Cacao Genebank, Trinidad. Tropical Agriculture Trinidad. 85, 1-12.

Monge, J. and Hilje, L. (2006). Hábitos alimenticios de la ardilla (Sciurus

variegatoides), (Rodentia: Sciuridae). Nicoya, Costa Rica. Revista de

Biología Tropical, 54(2), 681-686. https://www.scielo.sa.cr/pdf/rbt/

v54n2/3855.pdf

Portillo, E., Martínez, E., Araujo, F., Parra, R. y Esparza, D. (1995). Diagnóstico

técnico-agronómico para el cultivo de cacao (Theobroma cacao L.) en

el Sur del Lago de Maracaibo. Revista de la Facultad de Agronomía

Universidad del Zulia. 12, 151-166. https://produccioncienticaluz.org/

index.php/agronomia/article/view/26016

Potash, A., Conner, M. & McClleery, R. (2019). Vertical and horizontal vegetation

cover synergistically shape prey behavior. Animal Behaviour, 152, 19-44.

https://doi.org/10.1016./j.anbehav.2019.04.007.

Reher, S., Dausmann, K., Warnecke, L. & Turner, J. (2016). Food availability

affects habitat use of Eurasian squirrels (Sciurus vulgaris) in a semi-

urban environment. Journal of Mammalogy, 97(6), 1543-1554. https://

doi.org/10.1093/jmammal/gyw105.

Sagorski, M. E. and Swihart, R. K. (2021). Raptor resource use in agroecosystems:

cover crops and denitions of availability matter. Avian Conservation and

Ecology, 16, 1-14. https://doi.org/10.5751/ACE01719-160101.

Silva-Ferreira, A., Peres, C., Dodonov, P., & Righetto-Cassano, C. (2020). Multi-

scale mammal responses to agroforestry landscapes in the Brazilian

Atlantic Forest: the conservation value of forest and traditional shade

plantations. Agroforestry Systems. 94: 2331-2341. https://doi.org/10.1007/

s10457-020-00553-y.

Smith, R. H. and Nott H. M. (1988). Rodent damage to cocoa in Equatorial

Guinea. FAO Plant Protection Bulletin. 36, 19-124.

Sonwa D. J., Weise, S. F., Schroth, G., Janssens, M. J. & Shapiro, H.Y. (2018).

Structure of cocoa farming systems in west and central Africa: a review.

Agroforestry Systems, 93(5), 2009-2025. https://doi.org/10.1007/s10457-

018-0306-7.

Tinajero, R., Barragán, F. & Chapa-Vargas, L. (2017). Raptor functional diversity

in scrubland-agricultural landscapes of Northern-Central-Mexican

dryland environments. Tropical Conservation Science, 10: 1-18. https://

doi.org/10.1177/1940082917712426.

Uchida, K. (2019). The town squirrel and the country squirrel: multiple behavioral

comparisons I Eurasian red squirrel between two environments. Ph. D.

Thesis. Scholl of Environmental Science. University of Hokaido, Japan.

https://eprints.lib.hokudai.ac.jp/dspace/bitstream/2115/83720/1/Kenta_

UCHIDA.pdf. https://doi.org/10.3106/041.042.0104

Uchida, K., Susuki, K. K., Shimamoto, T., Yanagawa, H. & Koizumi, I. (2017).

Escaping height in a tree represents a potential indicator of fearfulness in

arboreal squirrels. Mammal Study, 42, 39-43.

Vivas L. 1992. Los Andes Venezolanos. Academia Nacional de La Historia.

Italgraca S.R.L.

Verbeeck, H., Bauters, M., Toby, J., Schenkin, A., Disney, M.I., & Calders, K.

(2019). Time for a Plant Structural Economics Spectrum. Frontiers

in Forests and Global Change. 2, 1-5. https://doi.org/10.3389/

ffgc.2019.00043.

Walter, J. Stovall, A., & Atkins, J. (2021). Vegetation structural complexity and

biodiversity in the Great Smoky Mountains. Ecosphere. 12(3), 1-15.

https://doi.org/10.1002/ecs2.3390.

Warren J. and Emmandie, D. (1993). Rodent resistance in cacao, Theobroma

cacao L. Tropical Agriculture, 70, 286-288.

Wilson, J.A., Kelt, D. & Van Vuren, D. (2008). Home range and activity of the

northern ying squirrels (Glaucomys sabrinus) in the Sierra Nevada. The

Southwestern Naturalist, 53(1), 21-28. "https://doi.org/10.1894/0038-

4909(2008)53%5b21:HRAAON%5d2.0.CO;2" https://doi.

org/10.1894/0038-4909(2008)53[21:HRAAON]2.0.CO;2