Agroecosystems and primate conservation: Shade coffee as potentialhabitat for the conservation of Andean night monkeys in the northernAndes

Adriana Guzmána,b, Andrés Linkc,d,*, Jaime A. Castilloa,e, Jorge E. BoteroaaCentro Nacional de Investigaciones de Café, Cenicafé, Km. 4 vía antigua Chinchiná-Manizales, Caldas, ColombiabUniversidad Nacional de Colombia, sede Amazonia, Km 2 antigua vía Leticia-Tarapacá, Leticia, Amazonas, ColombiacDepartamento de Ciencias Biológicas, Universidad de Los Andes, Bogotá, Colombiad Facultad de Administración, Universidad de Los Andes, Bogotá, ColombiaeGrupo de investigación “Biología para la Conservación” Universidad Pedagógica y Tecnológica de Colombia, Avenida Central del Norte 39-115, Tunja,Colombia

A R T I C L E I N F O

Article history:Received 3 November 2014Received in revised form 3 September 2015Accepted 7 September 2015Available online xxx

Keywords:Aotus lemurinusMontane forestsHabitat useSustainable agriculture

A B S T R A C T

The global change in land use toward agroecosystems is one of the major drivers of habitat loss andfragmentation. Although simpli!ed agroecosystems such as pastures generally exclude most nativebiodiversity, other more complex systems such as agroforestry can support certain levels of biodiversityand serve as habitat, connecting corridors or stepping stones for many species. For example, traditionalshade coffee plantations can become key refuge areas for biodiversity, especially compared to moreintensive agricultural practices.The principal aim of this study was to evaluate the potential role of shade coffee plantations on the

conservation of Andean night monkeys (Aotus lemurinus) in the Colombian highlands. Two groups ofnight monkeys were followed during 16 and 7 months, and data on their ranging patterns, habitat use,activity budget and diet were recorded using focal animal sampling. We also estimated habitat wide fruitavailability at natural forests and shade coffee. Overall, both study groups preferred natural forestsrelative to their spatial cover, spending 70.3% and 94.4% of their time in natural forests, even thoughshade coffee had a yearly 10-fold total productivity advantage over natural habitats (4045 and 416 kg/h/year, respectively). Nonetheless, night monkeys used shade coffee areas throughout the year, speciallyduring the peak periods of fruit production of Prunus integrifolia (43% of time during June–July 2009; 64%of time during March 2010), a tree that dominated the shade coffee areas. The activity budget, homerange size and nightly travelled distances of Andean night monkeys in this study fall within the range ofvalues reported in previous studies on night monkeys on disturbed and undisturbed habitats. The resultsof this study provide support for the use of shade coffee plantations as complementary areas for theconservation of Andean primate populations while providing local farmers with economic alternatives.Although the conservation of montane primate populations in Colombia will largely depend on theprotection of the remaining natural habitats, agroecosystems such as shade coffee plantations mightprovide additive opportunities to increase the carrying capacity of mixed habitats and may become asuccessful conservation strategy in the pervasively transformed habitats of the northern Andes.

ã 2015 Elsevier B.V. All rights reserved.

1. Introduction

One of the major threats to wildlife and their habitats comesfrom the rapid loss of natural ecosystems as they are transformedinto agricultural !elds and pastures for livestock (Donald, 2004).

Agroecosystems, de!ned as large areas where the native fauna and"ora have been completely or partially replaced by agriculture andlivestock, cover well over 50 million km2, or roughly 25% of theglobal land surface (Altieri, 1999). Intensive agroecosystems thatcompletely replace native biodiversity leave almost no viability forlocal fauna and "ora. Nonetheless, intermediate agriculturalinterventions, such as agroforestry, have been shown to play animportant role for the conservation of birds and mammals whilesimultaneously providing farmers with economic alternatives(e.g., food or fuel) and ecological services (Caudill et al., 2014; Jha

* Corresponding author at: Departamento de Ciencias Biológicas y Facultad deAdministración, Universidad de los Andes, Bogotá, Colombia.

E-mail address: a.link74@uniandes.edu.co (A. Link).

http://dx.doi.org/10.1016/j.agee.2015.09.0020167-8809/ã 2015 Elsevier B.V. All rights reserved.

Agriculture, Ecosystems and Environment 215 (2016) 57–67

Contents lists available at ScienceDirect

Agriculture, Ecosystems and Environment

journal homepage: www.elsev ier .com/locate /agee

et al., 2011; Perfecto and Vandermeer, 2008; Pimentel et al., 1992;Vandermeer, 2003; Zarate et al., 2014). Although there is an urgentneed to preserve and protect the remaining natural ecosystems,agroecosystems may provide a complementary and successfulconservation alternative to be implemented in buffer areas nearpristine habitats or within previously intervened habitats (Schrothet al., 2004).

There is growing evidence supporting the idea that agro-ecosystems can (1) provide temporal habitat for migrating taxa(e.g., birds) and/or (2) extend the available habitat and increaseresource availability for resident species (Estrada et al., 1999, 2012;Perfecto and Ambrecht, 2003; Perfecto and Vandermeer, 2008;Rice and Greenberg, 2000). Only recently, studies have begun toassess the ability of primates to survive within agroecosystems(Estrada et al., 2012). For example, early studies on golden-headedlion tamarins (Leontopithecus chrysonelis) suggested that shadecacao plantations were successfully used by these CriticallyEndangered primates, and that they offered a conservationalternative within the heavily disrupted ecosystems in the Atlanticforest of Brazil (Alves, 1990; Raboy et al., 2004). Other Neotropicalprimates such as howler monkeys (Alouatta palliata) and spidermonkeys (Ateles geoffroyi) have been studied at agroecosystems inMexico and Costa Rica where they have been found in similarpopulation densities and with similar group size and socialstructure compared to populations living in natural forests(Estrada et al., 2006 and Refs. therein). Similar conclusions havebeen drawn from studies on old world primates and their ability tocope with agroforestry systems (Bali et al., 2007; Michon and deForesta, 1995).

A recent review on the use of agroecosystems by primates hasshown that several Neotropical primates use these transformedecosystems as they offer connectivity between native forestpatches (e.g., live fences) (Harvey et al., 2004) and increasehabitat-wide resource availability (Estrada et al., 2012). Severalgenera have been observed using agroecosystems in a permanentor temporal way, according to their biology and dietary "exibility.For example, Estrada et al. (2012) proposed that given the highecological demands of spider monkeys and their reliance on ripefruit in their diets (Di Fiore et al., 2008), they tend to useagroecosystems in a more temporal way than howler monkeys thatcan rely more heavily on leaves, which are more readily available inthese habitats. These large ateline primates have been studied insome of the most widely distributed agroecosystems: shade coffeeand cacao plantations (McCann et al., 2003; Muñoz et al., 2006;Zarate et al., 2014). Many other platyrhines that have also beenrecorded using agroecosystems include Lagothrix, Cebus, Sapajus,Saimiri and Leontopithecus (Table 1 in Estrada et al., 2012). To thebest of our knowledge no study has documented the ability of nightmonkeys (genus Aotus) to use agroecosystems.

The global coffee economy is a major agricultural activity withover 10 million hectares of land devoted to coffee plantations (FAO,2008), employing over 25 million people in their supply chain(from the !elds to the consumer) and providing over 400 billioncups of coffee per year (Donald, 2004; Illy, 2002). Coffeemanagement can oscillate from “rustic” coffee which has >90%of shade and thus resembles the traditional crops that wereoriginally established under the forest’s canopy, to “sun” coffee inwhich plantations can have no shade at all (Jha et al., 2011). Severalstudies have highlighted the importance of shade coffee planta-tions as agroecosystems that favor biodiversity, by serving as bufferzones around protected areas and as dispersal corridors betweenforest patches (Dietsch et al., 2004; Jha et al., 2011; Moguel andToledo, 1999; Raman, 2006; Tejada-Cruz and Sutherland, 2004). Infact, some studies have demonstrated how primate species adaptto shade coffee and other agroecosystems (Bicca-Marques, 2003;Clarke et al., 2002; Estrada et al., 1999, 2012; Estrada and Coates-

Estrada, 1996; Harvey et al., 2004), using them in some cases as anew habitat (Ganzhorn, 1985, 1987; McCann et al., 2003; Singhet al., 2001) or for dispersal purposes.

Shade coffee also provides several additional ecosystemservices to local communities that may increase their annualyields and overall wellbeing (Jha et al., 2011). Shade trees canprovide provisioning services (MEA, 2005) through timber andnon-timber products that can be used or commercialized byfarmers, and may represent an important source of income inmonths when coffee berries are not ready to be harvested(Somarriba et al., 2004). Shade coffee can also protect the naturalpopulations of key pollinators, thus providing “regulating” servicesfor local farmers as coffee plantations have been shown to havehigher yields when additionally pollinized by natural pollinators(Klein et al., 2003). For example, in Costa Rica, coffee plantationsnear forest fragments had larger yields (up to 20% higher) thanthose further and thus not visited as intensively by naturalpollinators, with and estimated additional income of $62,000 perfarm (Ricketts et al., 2004). Finally, shade coffee can provide“supporting” services by preventing erosion, carbon capture andthrough nitrogen !xation of shade trees (e.g., Inga spp.), amongstother ecosystem services (Jha et al., 2011). In summary, there aredirect bene!ts for local households in the implementation of shadecoffee, specially in areas where shade is a required condition forcoffee management (see below). In fact, in Colombia, Laderachet al. (2009) have found that coffee plantations with higher levelsof shade produce better quality coffee when compared to thosewith lower shade coverage.

Night monkeys (genera Aotus) are nocturnal primates with a dietbased on fruits and insects and live in socially monogamous groupsthat actively defend their territories (Fernández-Duque, 2007). TheAndean night monkey (Aotus lemurinus I. Geoffroy 1843) iscategorized as a Vulnerable species by the International Unionfor the Conservation of Nature (De"er et al., 2003; IUCN, 2010),principally due to the transformation of its natural habitat foragricultural activities and urban areas (Kattán and Álvarez-López,1996). In Colombia, Andean night monkeys are restricted to theAndean forests above 1.000 m.a.s.l., which overlaps with thecountry’s most densely populated and more intensively trans-formed areas for agriculture (Etter and van Wyngaarden, 2000;Etteret al., 2006). Highland ecosystems arealso the most intensivelyused by the coffee industry in Colombia (Guhl, 2008). The NorthernAndes are considered one of the 25 Biodiversity Hotspots due to itshigh species richness and endemism (Myers et al., 2000) and aconservation priority in Colombia. However, in the Northern Andesnatural forests within protected areas are mostly restricted tohigher elevations and to some remote regions (Kattán and Álvarez-López, 1996). Thus, one of the main challenges for the conservationof Andean biota, including the Andean night monkey, is to increasethe connectivity between isolated forest patches and increase theirpotential habitat. If Andean night monkeys can adapt to these novelconditions by using the remaining habitat patches and movingthrough the new surrounding matrix, as it has been found for otherarboreal mammals (Brearley et al., 2009), their populations willmost probably be able to cope with the pervasive process of habitatloss and fragmentation.

The main objective of this study was to determine the ecologicalvalue of shade coffee for Andean night monkeys. Speci!cally, weaddressed three questions: (1) Do Andean night monkeys useshade coffee plantations? (2) Do night monkeys have a preferencefor their natural habitats or for shade coffee plantations? (3) Arethere behavioral differences in Andean night monkeys when theyare in their natural habitat versus when they are in shade coffeeplantations? We expected that Andean night monkeys would usethe shade coffee plantations if they indeed offered complementaryresources to those present in the natural forests, or else ignore

58 A. Guzmán et al. / Agriculture, Ecosystems and Environment 215 (2016) 57–67

them entirely as neither group was obliged to move through theshade coffee plantations to pass between forest fragments in theirhome ranges.

2. Methods

2.1. Study area and shade coffee plantations

Data were collected between March 2008 and June 2010 atcoffee farms located in the municipality of San Vicente del Chucurí,Santander, on the western slope of Serranía de Los Yariguies in thenorth-eastern Andean cordillera of Colombia (6!53017.4 N,73!22012.5 W), at an elevation that ranged between 1200 and1500 m.a.s.l. Mean annual temperature is 18.9 !C, and mean annualprecipitation is 1900 mm (FNC-Cenicafé, 2013). Based on informa-tion from Colombia’s National Coffee Federation (FNC), at SanVicente de Chucuri about 70% of coffee farms (2475 ha) arecerti!ed by Rainforest Alliance; therefore, at least this area shouldbe cultivated under a minimum shade of 40% cover, have twocanopy strata and a minimum of 70 shade trees per hectarebelonging to 12 different tree species. Other important socioeco-nomic activities in this municipality include cocoa and bananaplantations, and grasslands for cattle ranching.

At San Vicente de Chucuri, shade coffee has not only spreadgiven the economic incentives behind international certi!cations(e.g., Rainforest Alliance, 4C, Nespresso AAA, Coffee Practices) butis actually required in order to avoid foliar damage in coffee plants.Based on traditional knowledge and common agricultural practicesin these highlands, coffee plantations were initiated in areas whereother agricultural products (e.g., cocoa) were grown under shade inprevious decades. Shade is a key factor for agricultural practices atSan Vicente given that (1) there is a high solar irradiance that candirectly affect coffee plants, and (2) there are pronounced periodsof drought in which shade provides the appropriate conditions forreducing evaporation (thus increasing relative humidity in theunderstory and soil) and slowing the water drainage dynamics(Beer et al., 1998).

Within the study area, shade coffee has been denominated as“regulated shade coffee” or “semi-shade” coffee given that optimalshade for coffee productivity should oscillate between 40% and50%, less than in nearby highland coffee production areas(Sanchez-Clavijo et al., 2007). A key aspect of shade coffeeplantations at San Vicente is that shade trees need to be “managed”and more speci!cally they need to be trimmed annually in order tocontrol the relative shade over the crops and regulate thereproductive biology of coffee plants. Shade trees are trimmedin January (corresponding to the beginning of the dry season) inorder to exacerbate water stress on coffee plants and induce the"owering processes, and also in order to allow the regrowth ofbranches and foliage of shade trees needed during the followingcoffee fruiting periods. Thus, the selection of shade trees includeseveral legumes (e.g., Inga spp.) which are easily managed, andother trees used for timber such as Cedrela odorata (Meliaceae) andCordia alliodora (Boraginaceae), as well as a large diversity ofcanopy and subcanopy trees that had been left as shadeintentionally from the former forests of the region, such as Prunusintegrifolia (Rosaceae) which provides feeding resources for nightmonkeys and other frugivores.

The shade coffee production at San Vicente is imbibed within acomplex mosaic of land use practices including other agriculturalpractices (e.g., cocoa, bananas), small scale cattle ranching, andareas that are covered with undergrowth or secondary vegetation.Mature forests are restricted to areas near water sources orextremely broken terrain that has not been used for agriculturalactivities. In fact, although on average farms are roughly 12 ha the

average coffee plantation is only 3 ha, and the rest of the landscapeis comprised by other agricultural and ranching practices.

Technical assistance for local coffee farms at San Vicente deCaguan is provided by the National Coffee Federation in order tocontrol the 40–50% cover, the selection of shade trees and theirmanagement, and the trimming practices in order to induce waterstress (and thus "owering) in certain periods of the year, but also toguarantee shade (and thus humidity) during the fruiting periods.Nonetheless, the implementation of these practices relies heavilyon the traditional agricultural practices of local farmers.

Finally, although certi!cation bene!ts are part of the economicincentives for shade plantations, at San Vicente de Chucurí, coffeeplantations are shade dependant. At the time of the study coffeecerti!cation by Rainforest Alliance was very successful and stilltoday approximately 150 small farms (2–20 ha) are still certi!ed. AtSan Vicente shade coffee has been largely driven by the stringentenvironmental conditions that require shade for maximizing coffeeproduction yields and certi!cation has come as a by-product toreach better markets and increase local income. Nonetheless, thesuccess of shade coffee as a product of “good agricultural practices”largely depends on an “effective” international marketing programin order to reach the public that is willing to recognize value onresponsible social and environmental practices.

2.2. Study species

The distribution of the Andean night monkey is restricted to theAndean forests of Colombia, and perhaps Venezuela (De"er et al.,2003) between 1000 and 3000 m.a.s.l. They are considered as athreatened species given the restricted forest cover and the rapidtransformation of natural habitats into agricultural areas in theColombian Andes (Etter et al., 2008). Although scant informationon the ecology and behavior of the Andean night monkey iscurrently available, recent studies suggest their behavior is similarto that described for most other species of this unique Neotropicalgenus (Castaño et al., 2010). Night monkeys (genus Aotus) are theonly nocturnal (or cathameral) anthropoid primates. They live inmonogamous groups of two to !ve individuals (Aquino andEncarnación, 1994; Fernández-Duque and Huntington, 2002;Wright, 1978; Wright, 1994) and reproductive pairs and theiroffspring can occupy areas that range between 1.0 and 17.5 ha(Castaño et al., 2010; Fernández-Duque, 2007; Wartmann et al.,2014).

Only two primate species are present in the Andean forests ofSan Vicente de Caguan: the Andean night monkey (A. lemurinus)and the red howler monkey (Alouatta seniculus). None of theseprimates feed on coffee berries or foliage and do not represent adirect threat to coffee plantations. Thus, there is no negativeperception by local farmers on the presence of monkeys within andnearby coffee plantations, as even for folivore primates like thehowler monkey, the consumption of foliage from certain shadetrees might not have a detrimental effect on the shade (protection)provided by canopy and subcanopy trees. In summary, at SanVicente de Caguan there is an indirect and “passive” relationbetween coffee plants and monkeys, and both bene!t from shadetrees for protection from solar irradiance and as potential habitatand source of feeding resources, respectively.

2.3. Animal capture

A total of 14 platforms were built in trees and baited daily withbanana fruits for periods ranging from three to eight months. Oncewe con!rmed with camera trapping that night monkeys werevisiting the platforms searching for food, we activated a Tomahawktrap for medium size mammals (80 cm " 30 cm " 25 cm) andattempted to capture night monkeys for the following nights.

A. Guzmán et al. / Agriculture, Ecosystems and Environment 215 (2016) 57–67 59

Overall, we captured three night monkeys that were immobilizedwith 1 mg/kg of Xilacin 1% and 50 mg/kg of Ketamin HCL 50%, and!tted with MOD-80 radiocollars (Telonics Inc.ã). During thesedation period, body temperature, heart frequency and respira-tory frequency were frequently monitored. Other body measure-ments (e.g., body length, tail length, neck circumference etc.) wererecorded as well.

2.4. Habitat structure and composition

In order to describe habitat structure and composition, weestablished vegetation plots in three areas: (1) the natural forest,(2) the shade coffee plantations where group G1 ranged and (3) theshade coffee plantations where G2 ranged. At each of these threeareas, we randomly distributed ten plots of 50 " 2 m for a total of0.1 ha (Gentry,1982). In each plot, all individuals with a diameter atbreast height (DBH) > 2.5 cm were marked and identi!ed to theminimum taxonomic level. For each individual tree its height and

Fig. 1. (A) Fruit production (kg/ha) between May 2009 and April 2010 in the natural forest plot (1 ha). (B) Fruit production (kg/ha) between May 2009 and April 2010 in thecoffee shade plantation plot (1 ha). The most important plants are shown in color. Note that the two graphs have different scale on the y-axis.

60 A. Guzmán et al. / Agriculture, Ecosystems and Environment 215 (2016) 57–67

DBH were recorded. A dominance index (IVI) was calculated foreach species identi!ed, by adding its relative density, relativefrequency and relative basal area using the formula:

IVI # Ni

N

! "$ Oi

O

! "$ BAi

BA

! "

where Ni = the number of individual trees from the species i,N = total number of trees in the plots, Oi = number of plots with thepresence of species i, O = total number of plots, BAi = the sum of thebasal area of all trees of the species i, BA = the sum of the basal areaof all trees in the plots. Additionally, average canopy cover wascalculated for the three areas plus for another coffee plantationwhere group G1 was occasionally observed. At each locality,canopy cover was measured at 30 random places with adensitometer facing the four cardial points. A variance analysis(ANOVA) and Tukey–Kramer post hoc test were performed in orderto determine canopy cover differences between them.

2.5. Fruit phenology and productivity

Fruit productivity was estimated for both the natural forest andshade coffee plantations. In each type of vegetation cover (shadecoffee and natural forest) we established one transect of 1 ha(500 m " 20 m) that was surveyed monthly between May 2009 andMay 2010. During each phenological survey, we recorded all treesbearing ripe fruits and estimated their productivity (total numberof fruits) using binoculars (Estrada 1984). For every plant species,we collected a minimum of !ve fruits from the forest ground,measured their length, width and height and dried them in an ovenin order to calculate their dry weight. We estimated habitat-wideproductivity by summing up the estimated total dry mass of fruits(kg/ha) produced by each tree species during every month in eachtype of habitat. Both phenological transects were located withinthe home range of G1, therefore correlations (SAS Institute)between fruit production and habitat use were only done for thisgroup.

Fig. 2. Percentage of time (A) G1 and (B) G2 spent at the natural forest and the coffee shade, during 16 and 7 months of study, respectively.

A. Guzmán et al. / Agriculture, Ecosystems and Environment 215 (2016) 57–67 61

2.6. Home range and habitat preferences

Radiotracking equipment consisted of two CommunicationSpecialist receivers R-1000 and two 2-element "exible antennasRA-14 (Telonics Incã). Behavioral follows were conducted fromdusk to dawn (18:00–06:00) during !ve complete nights eachmonth accounting for a total of 60 h of behavioral sampling permonth. During behavioral follows, the group’s location wasrecorded every 20 mins using a GPS (Garmin 60CSx). Total distancetraveled per night was calculated for both groups by calculating theEuclidean distance between sampling points. The home range ofeach group was calculated with the minimum convex polygonmethod (MCP 95%) and the Kernel density estimators (95%), andcore areas were estimated using Kernel density estimators (50%)using ArcGIS 9.3. Habitat preferences were determine using G tests,between the observed number of location points in each type ofvegetation cover and the expected number of location points basedon the proportion of shade coffee and natural habitat in each ofboth study group’s home range.

2.7. Behavior and diet

Behavioral data were collected throughout the night, and20 min focal animal samples (Altmann, 1974) were conducted torecord activity budget and social behavior every time that it waspossible to observe the individuals directly using infrared Bushnellbinoculars. Focal samples with less than 12 min of direct followswere discarded. Once the focal animal was selected, data on itsactivity and behavior were recorded using 2-minute instantaneousrecords along the 20 min of observation. Behavior was classi!ed in!ve mutually exclusive categories: resting (R), moving (M),foraging (F), social interactions (S) and other non-social behaviors(O). The foraging category included search, manipulation andingestion. When the behavior could not be classi!ed in anycategory because the individual was out of view, the behaviorfrequencies of those focal samples were adjusted to 100%. Thisaccounted for less than 5% of the focal animal samples.

Finally, we opportunistically recorded the diet of Andean nighmonkeys and divided foraging categories into the following fooditems: fruits, "owers, leaves, and insects. When possible, werecorded the items eaten by night monkeys, marked the feeding

Fig. 3. Home range and core areas of (A) G1 and (B) G2. The kernel 95% home range estimates are shown by the external polygons, while the core areas estimated at Kernel 50%are shown as the internal polygons for each group.

62 A. Guzmán et al. / Agriculture, Ecosystems and Environment 215 (2016) 57–67

trees with sequential numbers and identi!ed them later to thelower taxonomic level possible.

3. Results

3.1. Animal capture

Three groups of night monkeys were recorded visiting thebaited platforms, but only individuals belonging to two differentgroups were captured. We tagged one adult female and a juvenilemale in group one (G1), and one adult female in group two (G2)with radio collars. Group G1 was composed of three individualsthroughout the study period, however in December 2009 thetagged adult female died, and another individual joined the group.This new individual was assumed to be a female since it wasrepeatedly seen during social interactions with the tagged juvenilemale. Group G2 was composed of !ve individuals including areproductive pair, two juveniles and one dependant infant.

3.2. Habitat description

A total of 59 plant morphospecies were identi!ed in thevegetation plot located in the natural forest. This area can be best

considered as a secondary forest in an advanced stage of succession,since species with the highest IVI were mostly pioneer tree species(i.e., Acalypha spp., Vateria sp., Ochroma pyramidale and Cecropiaspp.), although it also contained trees more readily found in matureforests (i.e., Pouteria lucuma). Average canopy cover of the naturalforest was estimated to be 90.6% (SD = 3.5, N = 30). In the vegetationplot located in the shade coffee plantation where G1 was mainlyobserved, we identi!ed a total of 14 plant morphospecies. Specieswith higher IVI were coffee trees (Coffea arabica), followed bybanana plants (Musa spp.) and canopy trees P. integrifolia andErythrina poeppigiana. Average canopy cover of the natural forestwas estimated to be 80% (SD = 14.7, N = 30), however due to thelarge banana plants leaves, this value could have been over-estimated in this habitat. G1 occasionally used another coffeeplantation, which average canopy cover was estimated to be of 62%(SD = 13.2, N = 30). In the vegetation plot located in the shade coffeeplantation used by G2, we identi!ed 11 plant morphospecies. IVIvalues indicated that in this area the most important plants werecoffee trees (C. arabica), followed by the shade trees Carapa sp., E.poeppigiana and Inga sp. Average canopy cover of this plantationwas estimated to be 66% (SD = 18, N = 30).

Fig. 4. (A) Activity budget of both study groups of Andean night monkeys and (B) activity budgets of night monkeys in natural forests and shade coffee plantations.

A. Guzmán et al. / Agriculture, Ecosystems and Environment 215 (2016) 57–67 63

3.3. Fruit phenology and productivity

Fruit production in the natural forest varied monthly,producing slightly more fruits during the wet season than duringthe dry season, however fruit production was not correlated withrainfall. Total fruit production for the natural forest was estimatedin 415.5 kg/ha per year (Fig 1A). Fruit production in the shadecoffee plantation plot was bimodal, with one peak in August andthe other in March, mainly due to the fruit production of twospecies: P. integrifolia and Margaritaria sp. (Fig 1B). Total fruitproduction in coffee shade plantations was ten times greater thanthe fruit production of the natural forest (4044.8 kg/ha/yr),excluding coffee berries and bananas.

3.4. Home range and habitat use

Ranging data for G1 was collected during 16 months (March2009–June 2010) and for G2 during seven months (December2009–June 2010). There were no differences in distances travelledby night between both study groups (G1: 593 % 181 m; range: 155–1040 m; N = 61 nights; G2: 603 %156 m, range: 343–1003 m,N = 28 nights; Mann–Whitney U = 832.5, p = 0.843). Home rangesfor group G1 and G2 were estimated using MPC (95%) at 4.0 ha and4.7 ha, respectively. Using the kernel density estimator method thehome ranges of G1 and G2 at Kernel 95% were 4.0 and 6.5 ha,respectively, and both datasets reached an asymptotic accumula-tion curve after 15 nights of sampling. The study groups homeranges were in an extremely broken terrain with elevationsoscillating between 1200 m.a.s.l. to 1500 m.a.s.l. Core areasestimated through Kernel 50% were of 0.9 and 1.5 ha for G1 andG2, respectively.

The kernel 95% home range of G1 was comprised by 44% ofnatural forests and 56% of coffee-shade plantations. This groupspent on average 29.7% % 14.9 (N = 16 months) of its monthly timein coffee shade plantations and the remaining 70.3% in naturalforests (Fig. 2). Also, they spent more time in the coffee shadeduring the dry season and more time in the natural forest duringthe wet season. The home range of G2 is composed by 85.5% ofnatural forest and 14.5% of coffee-shade plantations and G2 usedcoffee shade in average 5.6% % 6.6 (N = 7 months) of their monthlytime and remaining 94.4% in the natural forest (Fig. 3). For bothgroups, the observed pattern suggested that Andean nightmonkeys preferred natural forest habitat as they use it moreoften than what was expected based on its availability (G1:p < 0.001, G2: p < 0.001). Nonetheless, the core areas used by bothstudy groups also included a large proportion of coffee-shadeplantation (G1: 67.4% of core area; G2: 15.5% of core area)suggesting they were important areas within the home ranges.Finally, habitat use by G1 was correlated with habitat-wide fruitproductivity on the coffee shade area (p = 0.001, r = 0.8), and also,with fruit productivity of P. integrifolia (p = 0.004, r = 0.7), the mostabundant coffee shade species.

3.5. Behavior

Group G1 spent resting (34.4% % SD = 27.1), followed by foraging(27.1%, SD = 26.6), moving (26.6%, SD = 19.8), socializing (5.6%,SD = 15.0), other (3.1%, SD = 8.5) and out of view (3.1%, SD = 7.2). Onthe other hand, group G2 principal activity was resting (43.0 %,SD = 24.1), followed by moving (30.9%, SD = 16.3), foraging (16.4%,SD = 18.7), socializing (6.3%, SD = 16.2), other (1.8%, SD = 4.7) andout of view (1.5%, SD = 4.8) (Fig. 4).

Table 1Plant species consumed by group G1 and group G2. Fruits (FR), "owers (FL), natural forest (NF), coffee plantation (C).

Family Species Item Habitat Consumed by G1 Consumed by G2 IVI (NF) IVI (C)

Cecropiaceae Cecropia sp. FR NF * 0.12

Clusiaceaea Clusia sp. FR NF *

Euphorbiaceae Margaritaria sp. FR C * 0.05

Fabaceae Erythrina poeppigiana FL C, NF * * 0.09 0.14

Indeterminado Sp. 1 FR NF * 0.02

Meliaceae Carapa sp. FR C *

Mimosaceae Inga sp. FR C * * 0.04 0.07

Moraceaeb Ficus sp. 1 FR NF *Ficus sp. 2 FR NF *Ficus sp. 3 FR NF *

Myrsinaceae Myrsine sp. FR C * 0.02

Myrtaceae Psidium gujava FR C *Myrcia sp. 1 FR NF * 0.02Myrcia sp. 2 FR, FL C *

Rosaceae Prunus integrifolia FR C * 0.22

Rutaceae Citrus sp. FR C * 0.14

Sapotaceae Pouteria sp. FR NF * 0.07

Solanaceae Solanum sp. FR NF *

Sapindaceaea Serjania sp. FR NF *

a Lianas.b Hemiepiphytes.

64 A. Guzmán et al. / Agriculture, Ecosystems and Environment 215 (2016) 57–67

Analysis of variance (ANOVA) revealed differences in thenightly behavior of both study groups in their time spent foraging(p = 0.003, F = 92, df = 1) and moving (p = 0.041, F = 42, df = 1). G2employed more time in moving but less in foraging than G1(Fig. 4A). There were also differences in the time spent resting(p = 0.011, F = 66, df = 1) and moving (p = 0.014, F = 62, df = 1) both inthe natural forests and the coffee shade plantations. Both groupsrested more when they were in the coffee shade and moved morewhen they were in the natural forest (Fig. 4B). We did not foundstatistical differences or interactions between groups and habitattype.

3.6. Diet and important shade species

A total of 19 plant morphoespecies, belonging to 15 genusand 14 families were identi!ed in both groups’ diet. Both groupsbased their diet principally on fruits; 86.6% for group G1 and63.6% for group G2; followed by "owers (10.1% and 31.3%,respectively), leaves (2.3% and 6.0%, respectively) and insects(1% and 0% respectively). However, the low percentage of insectsin the diet might be an artifact of the limited visibility to detectshort insect-eating bouts at night. At the study area, a few treespecies used in coffee shade plantations provided food resourcesfor G1 (e.g., fruits from P. integrifolia and "owers from Erythrinapoepiggiana) and for G2 (e.g., fruits from Inga sp. and "owersfrom E. poepiggiana.) Other important food sources for nightmonkeys were fruits from Cecropia sp., Citrus sp. and Myrcia sp.(Table 1).

4. Discussion

The results of this study provide evidence that shade coffeeplantations can provide additional habitat and important foodresources for the Andean night monkeys, and thus open a novelopportunity to implement conservation initiatives that bene!tboth, the local biota and local communities. Andean night monkeysat San Vicente de Chucuri use the shade coffee plantationsspecially during the fruiting season of P. integrifolia, con!rming thepotential ecological value of these agroecosystems. Our resultsbuild on the growing evidence that, if correctly implemented,agroecosystems can provide additional opportunities for theconservation of primates and other native wildlife (Cruz-Laraet al., 2004; Perfecto and Armbrecht, 2003; Estrada et al., 2012).

The use of shade coffee plantations as a biodiversity conserva-tion strategy has been controversial. Some authors have suggestedthat shade cover negatively in"uences coffee production (Perfectoet al., 2005), reducing farmers total income or revenue. Also,wildlife in agroecosystems or biological corridors may be morereadily exposed to hunting pressure or pathogens (Chapman et al.,2005). In general terms, shade reduces the incoming solarradiation reaching the coffee trees, resulting in a lower productiv-ity, when compared with full exposure plantations (Farfán, 2014).However, shade, or at least some degree of shade, is required inregions like San Vicente de Chucuri where weather conditions,especially rainfall and cloud cover patterns include dry periodsthat would stress the plantation and affect its long termproductivity, and in highly erosion prone areas (Farfán andJaramillo, 2009). Given this climatological scenario (which is wellunderstood by local farmers), the decision to include shade in thedevelopment of coffee plantations has not been driven fromscienti!c data, but rather, from the traditional system local farmershave used for decades. Local knowledge has played a key role indetermining the characteristics of coffee shade at San Vicente, asmost of the adjacent highland regions require of a very dense shade(Sanchez-Clavijo et al., 2007) as climatological variables (e.g.,irradiance and drought) are more severe.

Shade has other possible economic bene!ts: it reduces weedgrowth and therefore weed control costs, it provides additionalorganic matter from falling leaves, improving soil qualities andreducing fertilizing costs (Farfán, 2014), and could also provideadditional bene!ts such as timber or other products (Jha et al.,2011; Rice, 2008), that could generate income for the farmers.Shade plantations also contribute to regional environmentalsustainability through carbon sequestration, watershed protection,soil conservation, climate regulation, and of course biodiversityconservation (Jha et al., 2011; Tscharntke et al., 2011). Thoseecosystem services need to be further evaluated and valued.Programs that compensate farmers for the environmental servicesprovided by their shade coffee plantations, would contribute to theeconomic viability of shade as an agricultural production systemwith potential as a conservation strategy. Furthermore, since thesurrounding environmental conditions in"uence the agroecosys-tem fauna and the vice versa, conservation efforts should beenhanced at a regional scale (Altieri, 1999; López-Barrera andLandgrave, 2008).

The results of this study also build on the scant information onthe ecology of Andean night monkeys. To the best of ourknowledge, only one !eld study has been recently made on theirbehavior and ecology in natural but heavily fragmented habitat(Castaño et al., 2010). Consequently this study contributes to theunderstanding of the ecology of one of the few Neotropicalprimates living in montane forests. The results of our study on theecology and behavior of Andean night monkeys, such as group size,home range size, and total distance travel each night, fall withinthe range of values reported in other studies of Aotus spp. (Aquinoand Encarnación, 1994; Fernández-Duque, 2007; Fernández-Duque and Huntington, 2002; Fernández-Duque et al., 2008;Wartmann et al., 2014; Wright, 1978; Wright, 1994). However,activity budgets we measured were different from those observedby Castaño et al. (2010). For example, in our study, Andean nightmonkeys spent nearly 40% of their time resting, while in Castañoet al. (2010) resting only accounted on average for 5.3% % 5.5%(range 0–15.5%, N = 6 groups). This difference could be a result ofdifferences in data collection methods (focal animal samplingversus scan sampling), the level of habituation of the study groups,and/or differences in the environmental conditions of both studies;monkeys in this study using continuous forest surrounded by highcoffee-shade cover (60–80%) probably had more food resourcesavailable than monkeys using very small isolated forest fragmentsin the study by Castaño et al. (2010).

The ecological "exibility of some primates has shown thateven at a local scale, social groups (and even individuals) maybehave differently to better cope with immediate ecologicalchallenges and opportunities (Chapman, 1987; González andStevenson, 2009; Stevenson et al., 1998). In this study, differencesin behavior between the study groups could be a result ofdifferences in group size and/or habitat quality within their homeranges. At the time of the study, group G2 was composed of moreindividuals than G1; therefore, they possibly needed to movemore and use more area in order to ful!ll their energetic andnutritional needs. On the other hand, coffee shade used by G1 wasmore diverse and offered more resources than coffee shade usedby G2. Also, the fact that night monkeys might be spending largerproportion of time moving and travelling in natural forestscompared to shade coffee plantations might be due to thedifferences in fruit availability between these habitats and thefact that P. integrifolia trees were extremely common andproductive in the shade coffee. Thus, night monkeys might haveneeded to spend less time moving between feeding sources inshade coffee compared to more diverse areas in the natural forest.Most studies on night monkeys (Aotus spp.) have shown that theyfeed principally on fruits, although their diet can vary signi!cantly

A. Guzmán et al. / Agriculture, Ecosystems and Environment 215 (2016) 57–67 65

between species and localities (Fernández-Duque, 2007; Fernán-dez-Duque et al., 2008). At San Vicente, night monkeys mostly atefruits and "ower consumption was high in group G2 (31.3%).Further systematic studies on their diet should be conducted inorder to quantify "ower production in coffee shade trees anddetermine its importance for night monkey's behavior and habitatuse. Finally, although there were slight differences in the activitybudgets of both study groups, they had very similar rangingpatterns and nightly travel distances.

If shade coffee or cocoa plantations are an alternative strategytoward the conservation of local wildlife, the selection of shadetree species, their diversity and structure might be of crucialimportance in order to provide a functional shade that providesadditional habitat and increases temporal and spatial resourceavailability for wildlife living within agroecosystems. For primatesand other frugivores, the selection of trees for shade that produce"eshy fruits might increase habitat-wide resource availability ormay be chosen to include species that fruit during periods of fruitscarcity. For local farmers, the selection of shade trees might allowthem to count with timber and non-timber products which canprovide additional economic income to local farmers. For example,the tree P. integrifolia that offered important fruit resources forAndean night monkeys at San Vicente de Chucuri also has timber-yielding properties, and it's widely used for construction, industry"oors, lathery, wagon bodywork and construction beams (Acero-Duarte, 1985). Also, farmer may select species with high nitrogen!xation rates (e.g., Inga spp.) and reduce fertilization costs,amongst many other direct bene!ts from shade trees (see reviewin Jha et al., 2011).

Although in recent decades wildlife conservation has beenfocused on the establishment of protected areas, increasing habitatquality and connectivity in human dominated landscapes canfurther expand the integrity of ecosystems and the goods andservices they provide to humankind (Bhagwat et al., 2008). As wecontinue to witness the high rates of habitat loss in tropical forests,the sustainable management of socio-ecological systems becomesa complementary alternative to conservation.

5. Conclusions

In conclusion, this study provides a notable example of thepotential bene!ts of agroecosystems for the Andean night monkey,and in particular for shade coffee plantations adjacent to naturalhabitats. Andean night monkeys might not be able to survivewithout their natural forests, specially if they mainly rely on asingle trees species of the agroecosystems that do not fruit year-round, as observed at San Vicente de Chucuri. We point out that themost successful conservation initiative for the Andean nightmonkey might be associated with the conservation of naturalhabitats. Nonetheless, given that A. lemurinus might be able toadapt and live in transformed habitats (Castaño et al., 2010), theuse of coffee shade plantations in altered habitats or surroundingprotected areas (e.g., buffer zones) can act as a complementaryconservation initiative for increasing ecosystem connectivity forthese primates and other local wildlife that heavily rely on Andeanmountain forests.

Acknowledgments

The following institutions supported this study logistically and!nancially: Rainforest Alliance, Cenicafé, San Vicente de ChucuríFNC Extension Service and Fundacion Natura. Our special thanks tothe Vijagual coffee farm owners where the study was conducted.We also thank the comments of Sian Evans, Robert Cooper, Pablo R.Stevenson, Gabriela de Luna and Camila Pizano on earlier versions

of the manuscript. Eduardo Fernández-Duque help us during theinitial phase of project planning. Oscar Castillo and NathaliaBustamante helped us with spatial analyses. We also thank, theveterinary Nestor Varela from Matecaña Zoo of Pereira who helpedus on the immobilization procedures.

References

Acero-Duarte, L.E., 1985. Árboles de la zona cafetera Colombiana. Ediciones FondoCultural Cafetero, Bogotá, Colombia, pp. 303.

Altmann, J., 1974. Observational study of behavior: sampling methods. Behaviour48, 1–27.

Alves, M.C., 1990. The role of cacao plantations in the conservation of the Atlanticforest of Southern Bahia, Brazil. MSc. Thesis. University of Florida, Gainesville.

Altieri, M.A., 1999. The ecological role of biodiversity in agroecosystems. Agric.Ecosyst. Environ. 74, 19–31.

Aquino, R., Encarnación, F., 1994. Owl monkey populations in Latin America: !eldwork and conservation. In: Baer, J., Weller, R., Kakoma, I. (Eds.), Aotus: The OwlMonkey. Academic Press, United States, pp. 59–95.

Bali, A., Kumarb, A., Krishnaswamy, J., 2007. The mammalian communities in coffeeplantations around a protected area in the Western Ghats, India. Biol. Conserv.139, 93–102.

Beer, J., Muschler, R., Kass, D., Somarriba, E., 1998. Shade management in coffee andcacao plantations. Directions in Tropical Agroforestry Research. Springer,Netherlands, pp. 139–164.

Bhagwat, S.A., Willis, K.J., Birks, H.J.B., Whittaker, R.J., 2008. Agroforestry: a refugefor tropical biodiversity? Trends Ecol. Evol. 23, 261–267.

Bicca-Marques, J.C., 2003. How do howler monkeys cope with habitatfragmentation? In: Marsh, L.K. (Ed.), Primates in Fragments: Ecology andConservation. Kluwer Academic/ Plenum Publishers, New York, pp. 283–303.

Brearley, G., Bradley, A., Bell, S., MacAlpine, C., 2009. In"uence of contrasting urbanedges on the abundance of arboreal mammals: a study of squirrel gliders(Petaururus norfolcensis) in Southeast Queensland, Australia. Biol. Conserv. 141,60–71.

Castaño, J.H., Cardona, D.M., Botero, J.E., 2010. Ecología del mono nocturno andino(Aotus lemurinus), en fragmentos de bosque subandinos de Colombia. In: PereiraBengoa, V., Stevenson, P.R., Bueno, M.L., Nassar-Montoya, F. (Eds.), Primatologíaen Colombia: avances al principio del milenio. Asociación Colombiana dePrimatología, Bogotá, pp. 69–90.

Caudill, S.A., DeClerck, F.A.J., Husband, T.P., 2014. Connecting sustainable agricultureand wildlife conservation: does shade coffe provide habitat for mammals?Agric. Ecosyst. Environ. 199, 85–93.

Chapman, C.A., 1987. Flexibility in diets of three species of Costa Rican primates.Folia Primatol. 49, 90–105.

Chapman, C.A., Gillespie, T.R., Goldberg, T.L., 2005. Primates and the ecology or theirinfectious diseases: how will anthropogenic change affect host-parasiteinteractions? Evvol. Anthropol. 14, 134–144.

Clarke, M.R., Collins, D.A., Zucker, E.L., 2002. Responses to deforestation in a group ofmantled howlers (Alouatta palliata) in Costa Rica. Int. J. Primatol. 23, 365–381.

Cruz-Lara, L.E., Lorenzo, C., Soto, L., Naranjo, E., Ramírez-Marcial, N., 2004.Diversidad de mamíferos en cafetales y selva mediana de las cañadas de la selvaLacandona, Chiapas, México. Acta Zool. Mex. 20, 63–81.

De"er, T.R., Rodríguez-Mahecha, J.V., Hernández-Camacho, J.I., 2003. Conservationpriorities for Colombian primates. Primate Conserv. 19, 10–18.

Dietsch, T.V., Philpott, S.M., Rice, R.A., Greenberg, R., Bichier, P., 2004. Conservationpolicy in coffee landscapes. Science 303, 625–626.

Di Fiore, A., Link, A., Dew, J.L., 2008. Diets of wild spider monkeys. In: Campbell, C.J.(Ed.), Spider Monkeys: Behavior, Ecology and Evolution of the Genus Ateles.Cambridge University Press, Cambridge, pp. 81–137.

Donald, P.F., 2004. Biodiversity impacts of some agricultural commodity productionsystems. Conserv. Biol. 18, 17–37.

Estrada, A., 1984. Resource use by howler monkeys in the rain forest of Los Tuxtlas,Veracruz, Mexico. Int. J. Primatol. 5, 105–131.

Estrada, A., Coates-Estrada, R., 1996. Tropical rain rainforest fragmentation and wildpopulations of primates at Los Tuxtlas. Int. J. Primatol. 5, 759–783.

Estrada, A., Juan-Solano, S., Ortiz-Martínez, T., Coates-Estrada, R., 1999. Feeding andgeneral activity patterns of a howler monkey (Alouatta palliata) troop living in arainforest fragment at Los Tuxtlas, Mexico. Am. J. Primatol. 48, 167–183.

Estrada, A., Saenz, J., Harvey, C.A., et al., 2006. Primates in agroecosystems:conservation value of agricultural practices in Mesoamerican landscapes. In:Estrada, A., Garber, P.A., Pavelka, M.S.M., Luecke, L.G. (Eds.), New Perspectives inthe Study of Mesoamerican Primates: Distribution, Ecology, Behavior andConservation. Springer Press, New York, pp. 437–470.

Estrada, A., Raboy, B., Olivera, L.C., 2012. Agroecosystems and primate conservationin the tropics: a review. Am. J. Primatol. 74, 696–711.

Etter, A., van Wyngaarden, W., 2000. Patterns of landscape transformation inColombia, with emphasis in the Andean Region. Ambio 29, 412–439.

Etter, A., McAlpine, C., Wilson, K., Phinn, S., Possingham, H., 2006. Regional patternsof agricultural land use and deforestation in Colombia. Agric. Ecosyst. Environ.114, 369–386.

Etter, A., McAlpine, C., Possingham, H., 2008. Historical patterns and drivers oflandscape change in Colombia since 1500: a regionalized spatial approach. Ann.Assoc. Am. Geogr. 98 (1), 2–23.

66 A. Guzmán et al. / Agriculture, Ecosystems and Environment 215 (2016) 57–67

FAO, 2008. Food and Agriculture Organization. Diet, Nutrition, and the Prevention ofChronic Disease. WHO, Geneva.

Farfán, F., Jaramillo, A., 2009. Sombrío para el cultivo de café según la nubosidad dela región. Cenicafé Avance Técnico No. 379.

Farfán, F., 2014. Agroforestería y sistemas agroforestales con café. Cenicafé,Manizales, Caldas, Colombia.

Federación Nacional de Cafeteros de Colombia, Centro Nacional de Investigacionesde Café, Cenicafé, Chinchiná (Colombia), 2013. Anuario Meteorológico Cafetero2012. Cenicafé, Chinchiná, Colombia.

Fernández-Duque, E., 2007. The Aotinae: social monogamy in the only nocturnalhaplorhines. In: Campbell, C.J., Fuentes, A., MacKinnon, K.C., Panger, M., Bearder,S. (Eds.), Primates in Perspective. Oxford University Press, Oxford, pp. 139–154.

Fernández-Duque, E., Huntington, C., 2002. Disappearances of individuals fromsocial groups have implications for understanding natal dispersal inmonogamous owl monkeys (Aotus azarai). Am. J. Primatol. 57, 219–225.

Fernández-Duque, E., Di Fiore, A., Carrillo-Bilbao, G., 2008. Behavior, ecology anddemography of Aotus vociferans in Yasuní National Park, Ecuador. Int. J. Primatol.29, 421–431.

Ganzhorn, J.U., 1985. Utilization of eucalyptus and pine plantations by brownlemurs in the eastern rainforest of Madagascar. Primate Conserv. 6, 34–35.

Ganzhorn, J.U., 1987. A possible role of plantations for primate conservation inMadagascar. Am. J. Primatol. 12, 205–215.

Gentry, A.H., 1982. Patterns of Neotropical plant diversity. Evol. Biol. 15, 1–84.González, M., Stevenson, P.R., 2009. Patterns of daily movement, activities and diet

in Woolly monkeys (Genus Lagothrix): a comparison between sites andmethodology. In: Potoki, E., Krasinki, J. (Eds.), Primatology: Theories, Methodsand Research. Nova Science Publishers Inc., New York, pp. 171–186.

Guhl, A., 2008. Coffee production intensi!cation and landscape change in Colombia,1970–2002. In: Millington, A., Jepson, W. (Eds.), Land-Change Science in theTropics: Changing Agricultural Landscapes. Springer, New York, pp. 93–116.

Harvey, C., Tucker, N., Estrada, A., 2004. Can live fences, isolated trees andwindbreaks help conserve biodiversity within fragmented tropical landscapes?In: Schroth, G., Fonseca, G., Gascon, C., Vasconcelos, H., Izac, A.H., Harvey, C.(Eds.), Agroforestry and Conservation of Biodiversity in Tropical Landscapes.Island Press Inc., New York, pp. 261–289.

Illy, E., 2002. The complexity of coffee. Sci. Am. 6, 86–91.IUCN, 2010. Red List of Threatened Species. IUCN, Gland, Switzerland and

Cambridge, UK. www.iucnredlist.org.Jha, S., Bacon, C.M., Philpott, S.M., Rice, R.A., Méndez, V.E., Läderach, P., 2011. A

review of ecosystem services, farmer livelihoods, and value chains in shadecoffee agroecosystems. In: Campbell, W.B., López-Ortiz, S. (Eds.), Integratin GAgriculture, Conservation and Ecotourism: Examples from the Field. Springer,Netherlands, pp. 141–208.

Kattán, G.H., Álvarez-López, H., 1996. Preservation and management of biodiversityin fragmented landscapes in the Colombian Andes. In: Schelhas, J., Greenberg, R.(Eds.), Forest Patches in Tropical Landscapes. Island Press, Washington, D.C, pp.3–18.

Klein, A.M., Steffan-Dewenter, I., Tscharntke, T., 2003. Fruit set of highland coffeeincreases with the diversity of pollinating bees. Proc. Biol. Sci. 270, 955–961.

Laderach, P., Oberthiir, T., Pohlan, J., 2009. Management of Intrinsic QualityCharacteristics for Specialty Coffee—a Framework Developed and Tested inCoffee Growing Regions Across Latin America. Dr. Miiller Verlag, Saarbriick,Germany, pp. 157.

López-Barrera, F., Landgrave, R., 2008. Variación de la biodiversidad a nivel paisaje.In: Manson, R., Hernandez-Ortiz, V., Gallina, S., Ymehltreter, K. (Eds.),Agroecosistemas Cafetaleros De Veracruz Biodiversidad, Manejo YConservación. Instituto de Ecología A.C. INECOL. Instituto Nacional de EcologíaINE-SEMARNAT, México.

McCann, C., Williams-Guillen, K., Koontz, F., Roque Espinoza, A.A., MartínezSánchez, J.C., Koontz, C., 2003. Shade coffee plantations as wildlife refuge forMantled Howler Monkeys (Alouatta palliata) in Nicaragua. In: Marsh, L.K. (Ed.),Primates in Fragments: Ecology and Conservation. Kluwer AcademicPublishers/Plenum Press, New York, pp. 321–341.

Michon, G., de Foresta, H., 1995. The Indonesian agro-forest model. In: Halladay, P.,Gimour, D.A. (Eds.), Cons Erving Biodiversity Outside Protected Areas: the Roleof Traditional Agroecosystems. IUCN, Gland, pp. 90–106.

Millenium Ecosystem Assessment (MEA), 2005. Ecosystems and Human Well-being: Synthesis. Island Press, Washington, D.C.

Moguel, P., Toledo, V.M., 1999. Biodiversity conservation in traditional coffeesystems of Mexico. Conserv. Biol. 13, 11–21.

Muñoz, D., Estrada, A., Naranjo, E., Ochoa, S., 2006. Foraging ecology of howlermonkeys (Alouatta palliata) in a cacao (Theobroma cacao) plantation, Tabasco,Mexico. Am. J. Primatol. 68, 127–142.

Perfecto, I., Armbrecht, I., 2003. The coffee agrocosystem in the neotropics:combining ecological and economic goals. In: Vandermeer, J. (Ed.), TropicalAgroecosystems, Advances in Agroecology Series. CRC, Boca Raton, pp. 159–194.

Perfecto, I., Vandermeer, J., 2008. Biodiversity conservation in tropicalagroecosystems. Ann. N. Y. Acad. Sci. 1134, 173–200.

Perfecto, I., Vandermeer, J., Mas, A., Soto-Pinto, L., 2005. Biodiversity, yield, andcoffee certi!cation. Ecol. Econ. 54, 435–446.

Pimentel, D., Stachow, D.A., Takacs, H.W., Brubaker, H.W., Dumas, A.R., Meaney, J.J.,O’Neil, J.A.S., Onsi, D.E., Corzilius, D.B., 1992. Conserving biological diversity inagricultural/forestry systems. Bioscience 5, 354–362.

Raboy, B.E., Christman, M.C., Dietz, J.M., 2004. The use of degraded and shade cocoaforests by endangered goldenheaded lion tamarins Leontopithecus chrysomelas.Oryx 38, 75–83.

Raman, T.R.S., 2006. Effects of habitat structure and adjacent habitats on birds intropical rainforest fragments and shaded plantations in the Western Ghats,India. Biodivers. Conserv. 15, 1577–1607.

Rice, R.A., 2008. Agricultural intensi!cation with agroforestry: the case of coffee andwood products. Agric. Ecosyst. Environ. 128, 212–218.

Rice, R.A., Greenberg, R., 2000. Cacao cultivation and the conservation of biologicaldiversity. Ambio 3, 167–176.

Ricketts, T.H., Daily, G.C., Ehrlich, P.R., Michener, C.D., 2004. Economic value oftropical forest to coffee production. Proc. Natl. Acad. Sci. U. S. A. 101, 12579–12582.

Sanchez-Clavijo, L.M., Botero, J.E., Velez, J.G., 2007. Estructura, diversidad y potencialpara conservación de los sombríos en cafetales de tres localidades de Colombia.Cenicafe 58, 304–307.

Schroth, G., Fonseca, G., Gascon, C., Vasconcelos, H., Izac, A.M., Harvey, C., 2004.Agroforestry and Conservation of Biodiversity in Tropical Landscapes. IslandPress Inc., New York.

Singh, M., Kumara, H.N., Kumar, M.A., Sharma, A.K., 2001. Behavioral responses oflion-tailed macaques (Macaca silenus) to a changing habitat in a tropical rainforest fragment in the Western Ghats, India. Folia Primatol. 72, 278–291.

Somarriba, E., Harvey, C., Samper, M., Anthony, F., Gonzalez, J., Staver, C., Rice, R.,2004. Biodiversity in coffee plantations. In: Schroth, G., Fonseca, G., Harvey, C.,Gascon, C., Vasconcelos, H., Izac, A.M.N. (Eds.), Biodiversity in CoffeePlantations. Island Press, Washington, D.C, pp. 198–226.

Stevenson, P.R., Quiñones, M.J., Ahumada, J.A., 1998. Effects of fruit patch availabilityon feeding subgroup size and spacing patterns in four primate species at TiniguaNational Park, Colombia. Int. J. Primatol. 19, 313–324.

Tejada-Cruz, C., Sutherland, W.J., 2004. Bird responses to shade coffee production.Anim. Conserv. 7, 169–179.

Tscharntke, T., Clough, Y., Bhagwat, S.A., Buchori, D., Faust, H., Hertel, D., Holscher, D.,Juhrbandt, J., Kessler, M., Perfecto, I., Scherber, C., Schroth, G., Veldkamp, E.,Wanger, T.C., 2011. Multifunctional shade-tree management in tropicalagroforestry landscapes—a review. J. Appl. Ecol. 48, 619–629.

Vandermeer, J.H., 2003. Tropical Agroecosystems. CRC Press, Boca Raton, FL.Wartmann, F.M., Juarez, C.P., Fernandez-Duque, E., 2014. Size, Site !delity, and

overlap of home ranges and core areas in the socially monogamous owl monkey(Aotus azarai) of northern Argentina. Int. J. Primatol. 35, 919–939.

Wright, P., 1978. Home range, activity pattern, and agonistics encounters of a groupof night monkeys (Aotus trivirgatus) in Peru. Folia Primatol. 29, 43–55.

Wright, P., 1994. The behavior and ecology of the owl monkey. In: Baer, J., Weller, R.,Kakoma, I. (Eds.), Aotus: the Owl Monkey. Academic Press, United States, pp.97–112.

Zarate, D.A., Andresen, E., Estrada, A., Serio-Silva, J.C., 2014. Black howler monkey(Alouatta pigra) activity: foraging and seed dispersal patterns in shaded cocoaplantations versus rainforests in southern Mexico. Am. J. Primatol. 76, 890–899.

A. Guzmán et al. / Agriculture, Ecosystems and Environment 215 (2016) 57–67 67