Diversity and activity of medium- and large-sized mammals

in agroforestry systems, Santander, Colombia

Sergio A. García-Mieles1 , Carlos H. Cáceres-Martínez1 , Diana V. Caballero-Aldana2 , and Giovany Díaz-Peñaloza3,4,5* .

1Grupo de Investigación en Ecología y Biogeografía, Universidad de Pamplona, Norte de Santander, Colombia. E-mail: samieles16@gmail.com (SAG-M); carloscaceres@unipamplona.edu.co (CHC-M).

2ISAGEN S.A. E. S. P., Medellín, Antioquia, Colombia. E-mail: dianav.caballero@gmail.com (DVC-A).

3Maestría en Ciencias, Biología, Facultad de Ciencias, Universidad Nacional de Colombia, Bogotá D.C., Colombia.

4Colección Nacional de Anfibios, Instituto de Ciencias Naturales, Universidad Nacional de Colombia, Bogotá D.C., Colombia.

5Grupo de Investigación Morfología y Ecología Evolutiva, Instituto de Ciencias Naturales, Universidad Nacional de Colombia, Bogotá D.C., Colombia.

*Corresponding author: fgiodiaz24@gmail.com

The expansion of the agricultural frontier for food production is one of the main threats to biodiversity in Colombia. Agroforestry systems represent a productive strategy based on mixed cropping with timber, fruit, or native species that enhance production and can provide resources for diverse animal species. In Colombia, cacao is generally cultivated under agroforestry systems, with Santander being the main national producer. Because these systems harbor species sensitive to anthropogenic disturbance, such as mammals, the objective of this study was to evaluate the taxonomic diversity and activity patterns of medium- and large-sized mammals associated with cacao agroforestry systems. The study area included three cacao agroforestry systems located on different farms between San Vicente de Chucurí and Zapatoca, Santander, Colombia. Three Bushnell Essential E3 camera traps were installed on each farm and remained active from January 2022 to July 2023. Relative abundance indices, diversity based on Hill numbers, activity patterns, and overlap were estimated. Based on an effort of 1,529 camera-trap nights and 1,154 independent detections, 17 species of medium- and large-sized mammals were recorded, some of which were categorized as Near Threatened (NT) and Vulnerable (VU). The assemblage exhibited predominantly nocturnal–crepuscular activity. Possible shifts in activity patterns in response to human presence were identified, as observed in Dasyprocta punctata. Additionally, comparisons among potentially competing species revealed temporal segregation in rodents and overlap in marsupials and cingulates. These results highlight the potential of cacao agroforestry systems to provide resources such as water, food, and shelter for these mammals.

Keywords: Camera trapping, human presence, relative abundance, temporal overlap, Theobroma cacao.

La expansión de la frontera agrícola para la producción de alimentos es una de las principales amenazas para la biodiversidad en Colombia. Los sistemas agroforestales representan una estrategia productiva basada en cultivos mixtos con especies maderables, frutales o nativas que favorecen la producción y pueden proporcionar recursos a diversas especies animales. En Colombia, el cacao se desarrolla generalmente mediante sistemas agroforestales, siendo Santander el principal productor nacional. Debido a que estos sistemas albergan especies sensibles a la intervención antrópica, como los mamíferos, el objetivo del estudio fue evaluar la diversidad taxonómica y los patrones de actividad de mamíferos medianos y grandes asociados a sistemas agroforestales de cacao. El estudio se desarrolló en tres sistemas agroforestales de cacao ubicados en diferentes predios entre San Vicente de Chucurí y Zapatoca, Santander, Colombia. Se instalaron tres cámaras-trampa Bushnell Essential E3 en cada predio, las cuales permanecieron activas de enero de 2022 a julio de 2023. Con la información obtenida se estimaron índices de abundancia relativa, diversidad mediante números de Hill, patrones de actividad y solapamiento. Como resultado de 1.529 noches/trampa y 1.154 detecciones independientes, se registraron 17 especies de mamíferos medianos y grandes, algunas de ellas categorizadas como Casi Amenazadas (NT, Near Threatened) y Vulnerables (VU, Vulnerable). El ensamblaje presentó principalmente actividad nocturna-crepuscular. Se identificaron posibles ajustes en el patrón de actividad frente a la presencia humana, como en el caso de Dasyprocta punctata. Además, se comparó la actividad entre especies potencialmente competidoras observando segregación temporal en roedores y solapamiento en marsupiales y cingulados. Estos resultados destacan el potencial de los sistemas agroforestales de cacao para proveer recursos como agua, alimento y refugio a estos mamíferos.

Palabras clave: Abundancia relativa, fototrampeo, presencia humana, solapamiento temporal, Theobroma cacao.

Agriculture is one of the anthropogenic activities that sustain human populations (Godfray et al. 2010; Bula and Tessmer 2020). This activity generally depends on large land areas for food production (Dirzo and Raven 2003), leading to large-scale tree clearing that affects plant and animal biodiversity, including mammals (Haddad et al. 2015). As an alternative, agroforestry systems (AFS) have been implemented for food production (Zermeño-Hernández et al. 2016). Agroforestry systems consist of combinations of crops with timber, fruit, or native species that provide shade to cultivated plants (Moreno-Calles et al. 2013). In this context, AFS have been considered as productive systems capable of providing shelter and resources for different mammal species (Beer 1999; Bhagwat et al. 2008). One of the crops commonly grown under this modality is cacao (Jaimes-Suárez et al. 2022).

In Colombia, cacao cultivation is generally carried out under AFS, with Santander Department being the main producer (Jaimes-Suárez et al. 2022). This department has more than 59,000 hectares under cultivation and contributes 41% of the national production of dry cocoa beans (Ministerio de Agricultura y Desarrollo Rural 2021). Within this region, San Vicente de Chucurí and El Carmen de Chucurí are the municipalities with the highest cacao production, accounting for 48.7% of departmental production. Due to its importance, cacao agroforestry is considered a sustainable production system from social, economic, and biophysical perspectives (Jaimes-Suárez et al. 2022), as well as a strategy to reduce pressure on primary forests, preserve biodiversity, and contribute to food security (Palencia-Calderón and Mejía-Flórez 2004). Among the ecosystem services provided by these systems are carbon sequestration, water cycle regulation, and the provision of habitat for diverse species of birds, insects, and mammals (Jaimes-Suárez et al. 2022).

Mammals play a fundamental role in ecosystems (Racero-Casarrubia and Reyes-Cogollo 2022), fulfilling important ecological functions such as seed dispersal, soil aeration and water infiltration, and the regulation of plant, animal, and insect populations (Martín-Regalado and Briones-Salas 2024). They are generally considered indicators of ecosystem health because they can be sensitive to anthropogenic disturbances such as habitat loss and fragmentation, which may lead to changes in their activity, structure, composition, and distribution (Rumiz 2010; Buchmann et al. 2013). In agricultural landscapes, their presence can provide valuable information about the sustainability of management practices and the capacity of production systems to conserve biodiversity.

Studies conducted in agricultural systems have shown that plantations can affect mammal species diversity in surrounding areas and agricultural landscapes (Navarro and Gómez 2015; De Giorgi Peirano 2016). Conversely, research carried out in Costa Rica, Brazil, and Mexico suggests that cacao cultivation can provide shelter and food resources for a wide variety of animal and plant species (Guiracocha et al. 2001; Cassano et al. 2012; Sánchez-Soto et al. 2025). In Colombia, studies by Peralta-Zapata et al. (2019) and López-Ramírez et al. (2020) have reported similar findings. Therefore, the objectives of this study were to: 1) estimate taxonomic diversity, 2) describe activity patterns, and 3) determine temporal overlap in the activity patterns of medium- and large-sized mammals in cacao agroforestry systems in San Vicente de Chucurí and Zapatoca, Santander, Colombia.

Materials and methods

Study area. The study was conducted in three cacao agroforestry farms located in the municipalities of San Vicente de Chucurí and Zapatoca, Santander Department, northeastern Colombia (Figure 1). The sites correspond to: A) Montegrande Farm, located at 6.9252° N, −73.3997° W, and 480 m above sea level (masl); B) Villa del Rosario Farm, located at 6.9177° N, −73.4086° W, and 428 masl; and C) Miraflores farm, located at 6.9355° N, −73.3611° W, and 995 masl (Figure 1).

Montegrande Farm (23 ha) and Villa del Rosario Farm (43 ha), both located in the municipality of San Vicente de Chucurí, are cacao farms managed under agroforestry systems integrated with cattle production. In these systems, cacao is cultivated in association with avocado and citrus trees, while cattle graze within the agroforestry areas. In contrast, Miraflores Farm (20 ha), located in the municipality of Zapatoca, consists of a cacao-based agroforestry system in which cacao is cultivated in association with orange trees and plantain. Together, the three farms exhibit a vegetation structure comprising tree, shrub, and herbaceous strata, which enhances habitat heterogeneity, provides resources and shelter for wildlife, and promotes landscape connectivity (Guzmán et al. 2016).

The region exhibits low annual thermal variation, with temperatures ranging from 20 to 28 °C and a bimodal rainfall pattern, with rainy seasons during March–May and September–November, ensuring relatively constant water availability throughout the year (Instituto Geográfico Agustín Codazzi 2025).

Sampling design. A camera-trapping survey using single-station setups was conducted between January 2022 and July 2023 in cacao agroforestry systems. Nine Bushnell Essential E3 (Bushnell, USA) camera traps were installed, with three cameras deployed per farm (sampling unit). Camera-trapping stations were established opportunistically (ad libitum) within the cacao plantations, considering internal footpaths, sites with evidence of mammal activity (e.g., tracks and feces), and recommendations from the landowners. A minimum distance of 100 m was maintained between stations, following the recommendation of Moreno et al. (2025).

Cameras were mounted on tree trunks approximately 30 cm above ground level. In one case, a camera was installed at a height greater than 1 m due to terrain conditions. Ca-mera traps were programmed to record 30-s videos with a 15-s interval between consecutive trigger events and ope-rated continuously over a 24-h period (Díaz-Pulido and Payán-Garrido 2012). Video recordings facilitated species identification by allowing the observation of morphological and behavioral characteristics. Each camera-trapping station was georeferenced using a Garmin GPS 60CSx and checked monthly to ensure proper functioning and to retrieve recorded data (Ministerio de Medio Ambiente et al. 2021).

Collected data were organized in a database including farm name, camera identification, record code, coordinates, date (dd/mm/yyyy), time, species, common name, number of individuals, sex (when identifiable), conservation status, and general observations. Medium-sized (>101 g and <10 kg) and large (>10 kg) mammals were defined following Hernández-Hernández et al. (2018). Specialized literature was used for species identification (Pérez-Gómez et al. 2025; Ramírez-Chaves et al. 2025). Trophic guilds were classified according to López-Ramírez et al. (2020) and Pérez-Gómez et al. (2025). Conservation status followed Resolution 0126 issued by the Colombian Ministry of Environment and Sustainable Development (2024) and the International Union for Conservation of Nature (IUCN) Red List (IUCN 2025). Human activity was recorded incidentally by the camera traps and corresponded primarily to farmers carrying out cacao crop management. Records were identified through the visual recognition of complete individuals or visible body parts, during both diurnal and nocturnal periods.

Data analysis. Sampling effort (camera-trap nights) was calculated as the total number of nights that each camera remained active (O’Brien et al. 2003).

Relative abundance. Relative abundance was estimated using the relative abundance index (RAI), calculated as the number of independent detections divided by sampling effort and multiplied by 100 (Pardo-Vargas and Payán-Garrido 2015). Independent detections were defined as consecutive records of different species, or consecutive records of the same species separated by more than 60 min (Tobler et al. 2008). In a capture or record the total number of individuals was counted when they could be reliably distinguished (Rowcliffe et al. 2008).

Taxonomic diversity. Diversity was assessed using Hill numbers with the iNEXT package in R version 4.4.1, with a minimum of 100 permutations, 1,000 bootstrap replications, and 95% confidence intervals (Hill 1973; Jost 2006; Hsieh et al. 2016). Hill numbers partition diversity into three orders: Order 0 diversity (q0) corresponds to species richness, order 1 diversity (q1) represents common species according to their abundance, and order 2 diversity (q2) accounts for dominance (Chao and Jost 2012).

Activity patterns. Only species (including humans) with more than 20 independent records were included (Pardo et al. 2021). Daily activity patterns were estimated using the overlap package in R version 4.4.1, which fits kernel density functions to the temporal distribution of species activity records (Meredith and Ridout 2017). Temporal distribution was evaluated using the Rayleigh test implemented in the circular package over a 24-h period for each species; values close to 1 indicate a stronger temporal preference, whereas values close to 0 indicate a more uniform temporal distribution (Agostinelli and Lund 2017). Species were classified according to their activity pattern as diurnal (08:00–18:00 h), nocturnal (20:00–06:00 h), morning crepuscular (06:00–08:00 h) and evening crepuscular (18:00–20:00 h) (Monroy-Vilchis et al. 2011).

Activity overlap. Temporal overlap between species was estimated using the coefficient Δ (delta), a non-parametric measure that quantifies the degree of coincidence between activity patterns, taking values close to 0 when overlap is low and close to 1 when it is high (Ridout and Linkie 2009; Mandujano and Pérez-Solano 2019). This coefficient is derived from the area of intersection between the activity density curves of two species (Schmid and Schmidt 2006). Given the available sample size, the Δ1 estimator was used, as it is recommended for datasets with fewer than 50 records and provides more robust estimates for small sample sizes than Δ4 and Δ5, which require more than 70 records (Mandujano and Pérez-Solano 2019). Overlap analysis was conducted under two comparison criteria: (1) between diurnal wild mammal species and human activity, represented by records of farmers carrying out cacao crop management; and (2) among potentially competing species, belonging to the same taxonomic order and trophic guild. Finally, Δ values were interpreted following the classification proposed by Monterroso et al. (2014), which defines low overlap (Δ ≤ 0.5), moderate overlap (0.5 < Δ ≤ 0.75), and high overlap (Δ > 0.75).

Results

With a sampling effort of 1,529 camera-trap nights, a total of 1,154 independent records of 17 species of medium- and large-sized mammals were obtained, distributed across six orders, 13 families, and 16 genera (Table 1). The most representative order was Carnivora, represented by five families, followed by Rodentia and Cingulata, with three and two families, respectively. Artiodactyla, Didelphimorphia, and Pilosa were each represented by a single family. Among families, Felidae contained the highest number of species (3), followed by Didelphidae and Mustelidae (2 each).

Relative abundance. The species with the highest relative abundance was A. rufinus (RAI = 25.64; 392 independent records), clearly exceeding all other species (Table 1; Figure 2). This was followed by D. marsupialis (15.63), C. paca (10.66), and D. punctata (10.46), all showing RAI values considerably higher than the overall average. Species with intermediate RAI values included S. granatensis (2.81), P. cancrivorus (1.90), D. fenestratus (1.77), M. myosuros (1.64), and C. centralis (1.44) (Table 1; Figure 2). In contrast, the lowest relative abundances corresponded to species with RAI values < 1, including E. barbara (0.92), and C. thous (0.52) (Table 1; Figure 2). Finally, the lowest number of independent records were observed for L. wiedii, L. pardalis, and G. vittata, whereas C. semistriatus and H. yagouaroundi, were each represented by a single independent record (Table 1; Figure 2).

Taxonomic diversity. Hill number analyses indicated that estimated richness was equivalent to observed richness (17 species), suggesting that no additional species were detected beyond those recorded during sampling (Figure 3). Likewise, sample coverage values were high (SC = 0.9983), indicating that the sampled assemblage was nearly completely represented (Figure 3). Although asymptotic estimators showed a slight tendency toward higher values (estimated richness ≈ 18 species), they did not differ significantly from the observed richness (p-value = 0.723). This suggests that sampling reached an asymptote and that the sampling effort was sufficient to characterize the taxonomic diversity of the assemblage (Figure 3).

In terms of diversity in the cacao agroforestry system, 17 species were recorded (q0; s.e. = 17.99 ± 0.98; 95% CI = 17.00–19.93), of which approximately 7 correspond to common species (q1; s.e. = 6.95 ± 0.19; 95% CI = 6.47–7.32) and 5 to dominant species (q2; s.e. = 4.99 ± 0.16; 95% CI = 4.61–5.33). These results indicate that, despite the high species richness, most records were concentrated on a relatively small number of species, reflecting a community with marked dominance and an uneven distribution of abundances (Figure 3).

Activity patterns. Activity patterns were described for nine of the 17 recorded mammal species. The assemblage exhibited predominantly nocturnal–crepuscular activity. Five species showed nocturnal and crepuscular activity, two showed diurnal and crepuscular activity, and two were strictly nocturnal (Table 1; Figure 4). Species with nocturnal and evening crepuscular activity included D. fenestratus, which was active from 19:00 to 00:00 h, with activity peaks between 19:00–21:00 h and 22:00–23:00 h; A. rufinus, active between 18:00 and 05:00 h, with a peak between 01:00–02:00 h; M. myosuros, active from 19:00 to 04:00 h, with greater activity between 19:00–21:00 h; D. marsupialis, active from 19:00 to 05:00 h, with peaks between 01:00–03:00 h and 19:00–20:00 h; and P. cancrivorus, active from 18:00 to 07:00 h, with higher activity between 01:00–03:00 h and 19:00–21:00 h (Figure 4).

The species with diurnal and evening crepuscular activity was D. punctata, whose activity was recorded from 05:00 to 19:00 h, with higher activity between 17:00–19:00 h, whereas S. granatensis exhibited diurnal and morning crepuscular activity between 06:00 and 17:00 h, with a peak between 06:00–10:00 h. In contrast, the species with strictly nocturnal activity included C. centralis, active from 19:00 to 06:00 h, with a peak between 04:00–05:00 h, and C. paca, active between 20:00 and 06:00 h, with greater activity between 01:00–02:00 h. Additionally, human activity was recorded between 07:00 and 18:00 h, with a peak between 08:00–10:00 h (Figure 4).

Activity overlap and comparisons. Temporal overlap was evaluated between human activity and species exhibiting diurnal activity (D. punctata and S. granatensis; Figure 5). D. punctata and humans showed low temporal overlap (Δ = 0.42; Figure 5A), whereas S. granatensis and humans showed high temporal overlap (Δ = 0.79; Figure 5B).

Comparisons among species with similar ecological traits revealed different levels of overlap (Figure 5). Among armadillos, C. centralis and D. fenestratus showed moderate temporal overlap (Δ = 0.60), indicating partial overlap in their activity patterns. In contrast, the opossums D. marsupialis and M. myosuros exhibited high temporal overlap (Δ = 0.86), suggesting very similar activity patterns. Finally, among large rodents, C. paca and D. punctata showed low temporal overlap (Δ = 0.10), reflecting marked temporal segregation in their activity (Figure 5E).

Discussion

The richness recorded in cacao agroforestry systems in San Vicente de Chucurí and Zapatoca constitutes an important proportion of the mammals reported for Colombia and the Santander Department (Vélez et al. 2018; Ramírez-Chaves et al. 2025), including species with different conservation statuses such as L. wiedii (NT) and A. rufinus (VU), as well as species with limited information such as C. centralis (DD) (IUCN 2026). The 17 documented species are consistent with previous studies conducted in cacao agroforestry systems and other transformed landscapes, where similar or higher values have been reported, supporting ecological value of these systems compared to more homogeneous matrices such as oil palm plantations (Pardo-Vargas and Payán-Garrido 2015; Peralta-Zapata et al. 2019; López-Ramírez et al. 2020). Comparable species richness has also been reported in natural ecosystems (Cáceres-Martínez et al. 2016; Hernández-Hernández et al. 2018; Mosquera-Guerra et al. 2018) suggesting that cacao agroforestry systems may contribute to landscape connectivity and the maintenance of mammal diversity in human-modified landscapes (Ferreira et al. 2020; Bedoya-Durán et al. 2023).

In studies monitoring medium- and large-sized mammals with high dispersal capacity, methodological aspects such as the independence of records for estimating relative abundance require careful consideration (Tobler et al. 2008). This is particularly relevant in local cacao agroforestry systems, where the sampling area is constrained by the size of individual farms and camera traps are typically installed 100–300 m apart (López-Ramírez et al. 2020). Under these conditions, standardization criteria based on the temporal separation of records are valuable, whereby records are considered independent after a predefined time interval has elapsed (Tobler et al. 2008). An important consideration of these approaches is the difficulty of drawing causal inferences about behavior and intraspecific interactions, focusing instead on describing activity patterns and the degree of temporal overlap among the species studied (Tobler et al. 2008).

Species with higher relative abundance, such as A. rufinus and D. marsupialis, may be favored by resource availability and their capacity to adapt to disturbed environments, allowing them to perform key ecological roles such as seed dispersal and population regulation (Ceballos and Oliva 2005; Franco-Quimbay and Rojas-Robles 2015; Saldaña-Garro et al. 2019). Similarly, species such as C. paca and D. punctata reflect tolerance to disturbance and fulfill important functions in ecosystem dynamics, particularly through herbivory and seed dispersal processes (Wright and Duber 2001; Naughton-Treves et al. 2003). In contrast, the low number of records of carnivores such as H. yagouaroundi, L. pardalis, and L. wiedii may be associated with their large home ranges, low population densities, and the limitations inherent to the sampling design; however, their presence supports the availability of prey and the ecological functionality of these agroforestry systems (Dillon and Kelly 2008; Payán-Garrido and Soto-Vargas 2012; Pinilla-Buitrago et al. 2015).

Species exhibiting nocturnal and crepuscular activity generally showed patterns consistent with previously reported for these taxa (Blake et al. 2012; Mosquera-Guerra et al. 2018). However, some differences were observed, as in A. rufinus, whose activity has been described as both diurnal and nocturnal (Blake et al. 2012), but which in this study showed a predominantly nocturnal pattern. Similarly, the nocturnal–crepuscular activity of D. marsupialis, D. fenestratus, and P. cancrivorus was consistent with previous studies (Lira-Torres and Briones-Salas 2012; Fonseca-Prada et al. 2023), whereas the diurnal patterns of D. punctata and S. granatensis also agreed with reports from other contexts (Cáceres-Martínez et al. 2016; Mosquera-Guerra et al. 2018; Fonseca-Prada et al. 2023). However, D. punctata has been reported to exhibit variable activity patterns, ranging from diurnal to crepuscular or nocturnal depending on factors such as resource availability, predation risk, and human activity (Hernández-Pérez et al. 2015; Albanesi et al. 2016; Encalada 2018).

Overlap with human activity revealed species-specific responses. S. granatensis showed high temporal overlap, suggesting greater temporal coincidence with human activity, consistent with reports describing variable behavioral responses to humans (Nitikman 1985) and its frequent occurrence in cultivated areas, where it may directly interact with cacao resources (Monje 2018; Peralta-Zapata et al. 2019; López-Ramírez et al. 2020). In contrast, D. punctata exhibited lower temporal overlap, indicating reduced temporal coincidence with human activity recorded within the cacao agroforestry system (Gálvez et al. 2024). This difference may be associated with their contrasting ecological habits: squirrels are arboreal and generally remain above human activity, whereas D. punctata, being terrestrial, may be more exposed to human presence and hunting pressure (Torres et al. 2018).

Among species with similar ecological characteristics, the degree of overlap varied according to their activity strategies. In armadillos, partial coincidence in activity periods was observed, suggesting a shared use of the temporal niche with some degree of differentiation (Fonseca-Prada et al. 2023). In contrast, opossums exhibited highly overlapping activity patterns, indicating greater similarity in their temporal organization (Moreno et al. 2025). Likewise, large rodents, showed marked temporal segregation, which may reflect temporal niche partitioning that could reduce competition for resources (Lira-Torres and Briones-Salas 2012).

In summary, these results suggest that cacao agroforestry systems may promote the coexistence of medium- and large-sized mammals with different ecological requirements by providing conditions that facilitate temporal segregation among species, a mechanism that may help reduce the potential for competition (Guiracocha et al. 2001; Guzmán et al. 2016). Furthermore, the differences observed in the degree of temporal overlap with human activity indicate that some species exhibit greater tolerance to productive environments, whereas others may adjust their activity patterns in response to human presence (Torres et al. 2018). In this context, cacao agroforestry systems may play an important role in biodiversity conservation by providing complementary habitats that support the presence of wildlife in transformed landscapes and contribute to landscape connectivity (Guzmán et al. 2016).

The conservation value of cacao agroforestry systems may also be influenced by the surrounding landscape configuration. In the municipalities of San Vicente de Chucurí and Zapatoca, cacao plantations are embedded within a heterogeneous matrix composed of forest remnants, riparian vegetation, secondary forests, and other productive systems, which promotes connectivity with less transformed areas (Solarte-Soto et al. 2025). This landscape configuration may facilitate the movement of medium and large-sized mammals among habitat fragments, allowing cacao agroforestry systems to function as complementary habitats for foraging, shelter, and movement rather than as isolated environments (López-Ramírez et al. 2020). The presence of species with large home ranges and those sensitive to habitat transformation suggests that these systems maintain a certain degree of connectivity for wildlife (Higuera-Mora et al. 2026). The contribution of cacao agroforestry systems to mammal conservation depends not only on their structural complexity but also on the quantity, quality, and connectivity of the surrounding natural vegetation, factors that help mitigate the effects of habitat fragmentation and promote biodiversity persistence in productive landscapes (Higuera-Mora et al. 2026)

Conclusions

Cacao agroforestry systems in the municipalities of San Vicente de Chucurí and Zapatoca, Santander, northeastern Colombia, harbor a diversity of 17 species of medium- and large-sized mammals, with a marked number of common and dominant species within the assemblage. Activity patterns showed a predominance of nocturnal and crepuscular habits, with five nocturnal–crepuscular species (A. rufinus, D. fenestratus, D. marsupialis, M. myosuros, P. cancrivorus), two strictly nocturnal species (C. centralis, C. paca), and two diurnal–crepuscular species (D. punctata, S. granatensis), in addition to human activity concentrated during morning hours. Activity pattern analyses revealed different levels of temporal overlap, reflecting reduced temporal coincidence of D. punctata and greater temporal coincidence of S. granatensis with human presence. Based on taxonomic group and trophic guild, greater overlap was observed among ecologically similar species such as D. marsupialis–M. myosuros, and stronger temporal segregation among species with different strategies such as C. centralis–D. fenestratus and C. paca–D. punctata. These results indicate that the variety of resources provided by cacao agroforestry systems, such as water, food, and shelter, can be utilized by a community of medium- and large-sized terrestrial mammals, highlighting their potential contribution to biodiversity conservation and landscape connectivity in productive landscapes.

Acknowledgments

The authors express their gratitude to ISAGEN S.A. E.S.P. for providing the resources and equipment necessary for the execution of this research. This study would not have been possible without the support and participation of the owners and managers of Montegrande, Villa del Rosario, and Miraflores farms, to whom we are grateful for granting access to their properties and for their time during working days.

Declaration of artificial intelligence use

During the preparation of this manuscript, the authors used GPT-5.3 to improve grammar and obtain a preliminary translation of the text. This tool was not used for the creation of figures or tables, data generation, statistical analysis, or the scientific interpretation of the results. The authors critically reviewed and edited all AI-assisted content and hold full responsibility for the originality, integrity, and conclusions presented in this work.

Author contributions

Sergio A. García-Mieles and Giovany Díaz-Peñaloza concei-ved and designed the study. Both authors conducted fieldwork, collected and curated the data, performed data processing and analyses, interpreted the results, and prepared figures and tables. Sergio A. García-Mieles and Giovany Díaz-Peñaloza also wrote the first draft of the manuscript, critically revised subsequent versions, and approved the final version for publication. Carlos H. Cáceres-Martínez and Diana V. Caballero-Aldana contributed logistical support, institutional agreements, and financial resources that enabled the execution of field activities and data collection in the study localities. They also reviewed the manuscript and approved its final version.

Literature cited

Agostinelli C, and Lund U. 2017. circular: circular statistics. R package version 0.4-93. Department of Environmental Sciences, Informatics and Statistics, Ca’ Foscari University and California Polytechnic State University.

Albanesi S, Jayat JP, and Brown A. 2016. Patrones de actividad de mamíferos de medio y gran porte en el piedemonte de Yungas del noroeste argentino. Mastozoología Neotropical 23:335–358.

Bedoya‐Durán MJ, Jones HH, Malone KM, and Branch LC. 2023. Continuous forest at higher elevation plays a key role in maintaining bird and mammal diversity across an Andean coffee‐growing landscape. Animal Conservation 26:714–728. https://doi.org/10.1111/acv.12857

Beer J. 1999. Theobroma cacao: un cultivo “agroforestal”. Agroforestería en las Américas 6:4.

Bhagwat SA, Willis KJ, Birks HJB, and Whittaker RJ. 2008. Agroforestry: a refuge for tropical biodiversity? Trends in Ecology & Evolution 23:261–267.

Blake JG, Mosquera D, Loiselle BA, Swing K, Guerra J, and Romo D. 2012. Temporal activity patterns of terrestrial mammals in lowland rainforest of eastern Ecuador. Ecotropica 18:137–146.

Buchmann CM, Schurr FM, Nathan R, and Jeltsch F. 2013. Habitat loss and fragmentation affecting mammal and bird communities: the role of interspecific competition and individual space use. Ecological Informatics 14:90–98.

Bula A, and Tessmer GA. 2020. Importancia de la agricultura en el desarrollo socioeconómico. Informe del Observatorio UNR No. 50, Puente Académico No. 16. Rosario (ARG): Universidad Nacional de Rosario.

Cáceres-Martínez CH, Acevedo-Rincón AA, and González-Maya JF. 2016. Terrestrial medium and large sized-mammals diversity and activity patterns from Tamá National Natural Park and buffer zone, Colombia. Therya 7:285–298. https://doi.org/10.12933/therya-16-397

Cassano CR, Barlow J, and Pardini R. 2012. Large mammals in an agroforestry mosaic in the Brazilian Atlantic Forest. Biotropica 44:818–825.

Ceballos G, and Oliva G. 2005. Los mamíferos silvestres de México. Mexico City (MEX): Fondo de Cultura Económica, Comisión Nacional para el Conocimiento y Uso de la Biodiversidad.

Chao A, and Jost L. 2012. Coverage-based rarefaction and extrapolation: standardizing samples by completeness rather than size. Ecology 93:2533–2547.

De Giorgi Peirano A. 2016. Diversidad de aves y mamíferos presentes en tres campos de soja en el departamento de Colonia y su relación con el daño a dicho cultivo [Bachelor’s Thesis]. [Montevideo (URY)]: Universidad de la República.

Díaz-Pulido A, and Payán-Garrido E. 2012. Manual de fototrampeo: una herramienta de investigación para la conservación de la biodiversidad en Colombia. Bogotá (COL): Instituto de Investigación de Recursos Biológicos Alexander von Humboldt, Panthera Colombia.

Dillon A, and Kelly MJ. 2008. Ocelot home range, overlap and density: comparing radio telemetry with camera trapping. Journal of Zoology 275:391–398.

Dirzo R, and Raven PH. 2003. Global state of biodiversity and loss. Annual Review of Environment and Resources 28:137–167.

Encalada L. 2018. Patrones de actividad diaria de mamíferos medianos y grandes de la Reserva Biológica Uyuca mediante fototrampeo [Master’s Thesis]. [Zamorano (HND)]: Escuela Agrícola Panamericana.

Ferreira AS, Peres CA, Dodonov P, and Cassano CR. 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

Fonseca-Prada K, Botero-Henao N, Mendoza-Mora A, and Tunarrosa-Echeverría E. 2023. Patrones de actividad de mamíferos medianos en fragmentos de bosque de Marquetalia (Caldas, Colombia). Revista Mutis 13:1–13. https://doi.org/10.21789/22561498.1852

Franco-Quimbay J, and Rojas-Robles R. 2015. Frugivoría y dispersión de semillas de la palma Oenocarpus bataua en dos regiones con diferente estado de conservación. Actualidades Biológicas 37:45–60.

Gálvez D, Romero E, Murcia-Moreno D, Bonilla B, and Valdés R. 2024. Daily activity patterns in agoutis (Dasyprocta spp) in response to relaxed predation. Heliyon 10:e39986. https://doi.org/10.1016/j.heliyon.2024.e39986

Godfray HCJ, Beddington JR, Crute IR, Haddad L, Lawrence D, Muir JF, et al. 2010. Food security: the challenge of feeding 9 billion people. Science 327:812–818. https://doi.org/10.1126/science.1185383

Guiracocha G, Harvey CA, Somarriba E, Krauss U, and Carrillo E. 2001. Conservación de la biodiversidad en sistemas agroforestales con cacao y banano en Talamanca, Costa Rica. Agroforestería en las Américas 8:7–11.

Guzmán A, Link A, Castillo JA, and Botero JE. 2016. Agroecosystems and primate conservation: shade coffee as potential habitat for the conservation of Andean night monkeys in the northern Andes. Agriculture, Ecosystems & Environment 215:57–67.

Haddad NM, Brudvig LA, Clobert J, Davies KF, Gonzalez A, Holt RD, et al. 2015. Habitat fragmentation and its lasting impact on Earth’s ecosystems. Science Advances 1:e1500052.

Hernández-Hernández JC, Chávez C, and List R. 2018. Diversidad y patrones de actividad de mamíferos medianos y grandes en la Reserva de la Biosfera La Encrucijada, Chiapas, México. Revista de Biología Tropical 66:634–646.

Hernández-Pérez E, Reyna-Hurtado R, Castillo-Vela G, Sanvicente-López M, and Moreira-Ramírez J. 2015. Fototrampeo de mamíferos terrestres de talla mediana y grande asociados a petenes del noroeste de la península de Yucatán, México. Therya 6:559–574. https://doi.org/10.12933/therya-15-290

Higuera-Mora NC, Clavijo-Ponce NL, Posada-Almanza RH, León-Sicard TE, and Roa-Fuentes LL. 2026. The multi-scale role of biodiversity in shaping cocoa agroforestry system performance. Agroforestry Systems 100:161. https://doi.org/10.1007/s10457-026-01501-y

Hill MO. 1973. Diversity and evenness: a unifying notation and its consequences. Ecology 54:427–432.

Hsieh TC, Ma KH, and Chao A. 2016. iNEXT: an R package for rarefaction and extrapolation of species diversity (Hill numbers). Methods in Ecology and Evolution 7:1451–1456.

Instituto Geográfico Agustín Codazzi [IGAC]. 2025. Modelo Digital de Terreno del municipio de San Vicente de Chucuri (Santander). Santander (COL): Instituto Geográfico Agustín Codazzi https://datos.icde.gov.co/datasets/e82a7ebe958340779124973c0bc16ada

IUCN. 2025. The IUCN Red List of Threatened Species. Version 2025-2. [Accessed January 21st, 2026]. https://www.iucnredlist.org

Jaimes-Suárez YY, Agudelo-Castañeda GA, Báez-Daza EY, Montealegre-Bustos F, Rengifo-Estrada GA, and Rojas-Molina J. 2022. Modelo productivo para el cultivo de cacao (Theobroma cacao L.) en el departamento de Santander (2a edición). Mosquera (COL): Corporación Colombiana de Investigación Agropecuaria (AGROSAVIA).

Jost L. 2006. Entropy and diversity. Oikos 113:363–375.

Lira-Torres I, and Briones-Salas M. 2012. Abundancia relativa y patrones de actividad de los mamíferos de los Chimalapas, Oaxaca, México. Acta Zoológica Mexicana 28:566–585.

López-Ramírez C, Restrepo-Quiroz T, and Solari S. 2020. Diversidad y ecología de mamíferos no voladores asociados a un sistema agro-productivo de cacao, Granja Yariguíes, Santander, Colombia. Actualidades Biológicas 42:1–13.

Mandujano S, and Pérez-Solano LA. 2019. Fototrampeo en R: organización y análisis de datos. Vol. 1. Xalapa (MEX): Instituto de Ecología A.C.

Martín-Regalado N, and Briones-Salas M. 2024. ¿Qué hacen los mamíferos en los ecosistemas? Therya ixmana 3:47–48.

Meredith M, and Ridout M. 2017. overlap: estimates of coefficient of overlapping for animal activity patterns. R package version 0.3.0.

Ministerio de Agricultura y Desarrollo Rural. 2021. Cadena del cacao: cifras sectoriales. Bogotá (COL): Ministerio de Agricultura y Desarrollo Rural.

Ministerio de Ambiente y Desarrollo Sostenible. 2024. Listado oficial de las especies silvestres amenazadas de la diversidad biológica colombiana continental y marino costera. Bogota (COL): Ministerio de Ambiente y Desarrollo Sostenible.

Ministerio de Medio Ambiente, Organización de las Naciones Unidas Medio Ambiente, and Corporación Nacional Forestal. 2021. Manual de uso de trampas cámaras para el monitoreo de carnívoros nativos y exóticos. Santiago (CHL): Ministerio de Medio Ambiente, Organización de las Naciones Unidas Medio Ambiente.

Monje J. 2018. Pasado, presente y futuro del manejo de vertebrados plaga en Costa Rica. Revista de Ciencias Ambientales 52:221–238.

Monterroso P, Alves PC, and Ferreras P. 2014. Plasticity in circadian activity patterns of mesocarnivores in Southwestern Europe: implications for species coexistence. Behavioral Ecology and Sociobiology 68:1403–1417. https://doi.org/10.1007/s00265-014-1748-1

Monroy-Vilchis O, Zarco-González MM, Rodríguez-Soto C, Soria-Díaz L, and Urios V. 2011. Fototrampeo de mamíferos en la Sierra Nanchititla, México: abundancia relativa y patrón de actividad. Revista de Biología Tropical 59:373–383.

Moreno G, Lizcano D, and Sánchez F. 2025. Activity patterns and habitat selection by terrestrial mammals in a mosaic landscape of silvopastoral system and forest fragments in the Andean piedmont in Colombia. Folia Oecologica 52:162–173.

Moreno-Calles AI, Toledo VM, and Casas A. 2013. Los sistemas agroforestales tradicionales de México: una aproximación biocultural. Botanical Sciences 91:375–398. .

Mosquera-Guerra F, Trujillo F, Díaz-Pulido A, and Mantilla-Meluk H. 2018. Diversidad, abundancia relativa y patrones de actividad de los mamíferos medianos y grandes asociados a los bosques riparios del río Bita, Vichada, Colombia. Biota Colombiana 19:202–218.

Naughton-Treves L, Mena JL, Treves A, Alvarez N, and Radeloff VC. 2003. Wildlife survival beyond park boundaries: the impact of slash-and-burn agriculture and hunting on mammals in Tambopata, Peru. Conservation Biology 17:1106–1117. https://doi.org/10.1046/j.1523-1739.2003.02045.x

Navarro JF, and Muñoz J. 2000. Manual de huellas de algunos mamíferos terrestres de Colombia. Medellín (COL): Multimpresos.

Navarro JF, Arias-Alzate A, and Delgado CA. 2024. Manual de huellas de algunos mamíferos terrestres de Colombia. Segunda Edición. Medellín (COL): Universidad CES.

Navarro PJ, and Gómez LA. 2015. Terrestrial mammal diversity in forests close to pineapple crops, Cutris San Carlos, Costa Rica. Cuadernos de Investigación UNED 7:59–65.

Nitikman LZ. 1985. Sciurus granatensis. Mammalian Species 246:1–8.

O’Brien TG, Kinnaird MF, and Wibisono HT. 2003. Crouching tigers, hidden prey: Sumatran tiger and prey populations in a tropical forest landscape. Animal Conservation 6:131–139.

Palencia-Calderón GE, and Mejía-Flórez LA. 2004. Una aproximación a los sistemas agroforestales con cacao. Bucaramanga (COL): Corporación Colombiana de Investigación Agropecuaria (AGROSAVIA).

Pardo LE, Edwards W, Campbell MJ, Gómez-Valencia B, Clements GR, and Laurance WF. 2021. Effects of oil palm and human presence on activity patterns of terrestrial mammals in the Colombian Llanos. Mammalian Biology 101:775–789.

Pardo-Vargas LE, and Payán-Garrido E. 2015. Mamíferos de un agropaisaje de palma de aceite en las sabanas inundables de Orocué, Casanare, Colombia. Biota Colombiana 16:54–66.

Payán-Garrido E, and Soto-Vargas C. 2012. Los felinos de Colombia. Bogotá (COL): Ministerio de Ambiente y Desarrollo Sostenible, Instituto de Investigación de Recursos Biológicos Alexander von Humboldt, Panthera Colombia.

Peralta-Zapata N, Sanchez-Ardila J, Sáenz-Jiménez FJ, Serrano-Sanchez S, Ardila-Rueda CE, Garces-Quiroz E, et al. 2019. Fauna y cultivos: interacciones y propuestas de manejo. Bogota (COL): ISAGEN – Fundación Natura.

Pérez-Gómez K, Fernández-Rodríguez RC, and Moreno-Niño N. 2025. Guía ilustrada de los mamíferos de Colombia. Cali (COL): McMullan Birding & Publishers S.A.S.

Pinilla-Buitrago GE, Rojas-Rojas AV, Liévano-Latorre LF, and Ramírez-Chaves HE. 2015. Familia Felidae. In: Suárez-Castro AF, and Ramírez-Chaves HE, editors. Los carnívoros terrestres y semiacuáticos continentales de Colombia. Guía de campo. Bogota (COL): Universidad Nacional de Colombia; p. 78–103.

Racero-Casarrubia J, and Reyes-Cogollo KI. 2022. Mamíferos carismáticos emblemas de la conservación de Colombia. Therya ixmana 1:111–113.

Ramírez-Chaves HE, Morales-Martínez DM, Cardona-Giraldo A, Castellanos O, Ospina O, Ossa-López PA, et al. 2025. Systematics of the dwarf red brocket, Mazama rufina (Pucheran, 1851) (Mammalia: Artiodactyla: Cervidae), with the description of a new genus. Zootaxa 5711:223–240.

Ridout MS, and Linkie M. 2009. Estimating overlap of daily activity patterns from camera trap data. Journal of Agricultural, Biological, and Environmental Statistics 14:322–337.

Rowcliffe JM, Field J, Turvey ST, and Carbone C. 2008. Estimating animal density using camera traps without the need for individual recognition. Journal of Applied Ecology 45:1228–1236.

Rumiz DI. 2010. Roles ecológicos de los mamíferos medianos y grandes. In: Wallace RB, editor. Distribución, ecología y conservación de los mamíferos medianos y grandes de Bolivia. Santa Cruz de la Sierra (BOL): Centro de Ecología Difusión, Fundación Simón I. Patiño; p. 53–73.

Saldaña-Garro I, Cadavid-Ramírez AC, and Gómez-Ruiz DA. 2019. Abundancia relativa y patrones de actividad de Didelphis marsupialis en un área periurbana de Medellín, Colombia. Revista MVZ Córdoba 24:7366–7371.

Sánchez-Soto S, Guzmán-Canul AR, Domínguez-Angulo S, and Mendoza-Romero A. 2025. Registro de mamíferos no voladores en plantaciones de cacao en el Corredor Biológico Mesoamericano, México. Mammalogy Notes 11:481.

Schmid F, and Schmidt A. 2006. Nonparametric estimation of the coefficient of overlapping–theory and empirical application. Computational Statistics & Data Analysis 50:1583–1596.

Solarte-Soto JA, Ospina-Bautista F, Fremout T, Melo-Zipacon WF, Valencia-Molina MC, and Pulleman M. 2025. Biodiversity and ecosystem services of cacao agroforestry arrangements in the Colombian Amazon. Agroforestry Systems 99:235. https://doi.org/10.1007/s10457-025-01326-1

Superina M, and Trujillo F. 2023. Armadillos de los Llanos Orientales de Colombia. Bogota (COL): Fundación Omacha y Fondation Segré.

Tobler MW, Carrillo-Percastegui SE, Leite Pitman R, Mares R, and Powell G. 2008. An evaluation of camera traps for inventorying large- and medium-sized terrestrial rainforest mammals. Animal Conservation 11:169–178.

Torres DF, Oliveira ES, and Alves RR. 2018. Conflicts between humans and terrestrial vertebrates: a global review. Tropical Conservation Science 11:940082918794084. https://doi.org/10.1177/1940082918794084

Vélez D, Rey J, Castro-Moreno C, Ortiz A, and González A. 2018. Línea base de especies en términos de su vocación de conservación y uso sostenible del departamento de Santander – Proyecto Santander BIO. Instituto de Investigación de Recursos Biológicos Alexander von Humboldt.

Wright SJ, and Duber HC. 2001. Poachers and forest fragmentation alter seed dispersal, seed survival, and seedling recruitment in the palm Attalea butyracea, with implications for tropical tree diversity. Biotropica 33:583–595.

Zermeño-Hernández I, Pingarroni A, and Martínez-Ramos M. 2016. Agricultural land-use diversity and forest regeneration potential in human-modified tropical landscapes. Agriculture, Ecosystems & Environment 230:210–220.

Associated editor: Gabriel P. Andrade Ponce

Submitted: April 8, 2026; Reviewed: June 17, 2026

Accepted: August 18, 2026; Published online: September 7, 2026

THERYA, 2027, Vol. 18(1):XXX-XXX

DOI: 10.12933/therya.2027.6309 ISSN 2007-3364

Figure 1. Location of camera-trapping stations (triangles) at Montegrande (A), Villa del Rosario (B), and Miraflores (C) farms, in the municipalities of San Vicente de Chucurí (A–B) and Zapatoca (C), Santander, Colombia.

Table 1. Medium- and large-sized mammal species recorded in cacao agroforestry systems in San Vicente de Chucurí and Zapatoca, Santander, Colombia. NI: number of independent records. RAI: relative abundance index (%). N: nocturnal, D: diurnal, EV: evening crepuscular, MC: morning crepuscular, and DD: data deficient. (*) Species recorded at all farms.

Scientific name	Common name / local name	NI	RAI	Activity pattern	Trophic guild
Artiodactyla
Cervidae
Andinocervus rufinus*	Northern Andean Red Brocket / Venado	392	25.64	N-CV	Herbivore
Carnivora
Canidae
Cerdocyon thous	Crab-eating Fox / Zorro perro	8	0.52	DD	Omnivore
Felidae
Leopardus pardalis	Ocelot / Ocelote	4	0.26	DD	Carnivore
Leopardus wiedii	Margay	6	0.39	DD	Carnivore
Herpailurus yagouaroundi	Jaguarundi / Gato pardo	1	0.07	DD	Carnivore
Mephitidae
Conepatus semistriatus	Striped Hog-nosed Skunk / Mapuro	1	0.07	DD	Omnivore
Mustelidae
Galictis vittata	Greater Grison / Hurón	4	0.26	DD	Carnivore
Eira barbara	Tayra / Zorro umba	14	0.92	DD	Omnivore
Procyonidae
Procyon cancrivorus	Crab-eating Raccoon / Zorra patona	29	1.9	N-CV	Omnivore
Cingulata
Chlamyphoridae
Cabassous centralis	Northern Naked-tailed Armadillo / Armadillo cola de trapo	22	1.44	N-CV	Insectivore
Dasypodidae
Dasypus fenestratus*	Western Andean Long-nosed Armadillo / Armadillo de nueve bandas	27	1.77	N-CV	Omnivore
Didelphimorphia
Didelphidae
Didelphis marsupialis*	Northern Black-eared Opossum / Faro or Fara	239	15.63	N-CV	Omnivore
Metachirus myosuros	Common Brown Four-eyed Opossum / Cuatrojos	25	1.64	N-CV	Omnivore
Pilosa
Myrmecophagidae
Tamandua mexicana	Northern Tamandua / Hormiguero	16	1.05	DD	Insectivore
Rodentia
Cuniculidae
Cuniculus paca*	Lowland Paca / Tinajo	163	10.66	N	Frugivore
Dasyproctidae
Dasyprocta punctata*	Central American Agouti / Ñeque	160	10.46	D-CM	Frugivore
Sciuridae
Syntheosciurus granatensis*	Red-tailed Squirrel / Ardita	43	2.81	D-CM	Frugivore

Figure 2. Relative abundance index of the species recorded in cacao agroforestry systems in San Vicente de Chucurí and Zapatoca, Santander, Colombia. Species are grouped according to their relative abundance as high (purple), intermediate (blue), and low (green). Illustrations were obtained from Navarro and Muñoz (2000); Superina and Trujillo (2023) and Navarro et al. (2024).

Figure 3. Rarefaction–extrapolation curves based on Hill numbers of orders q0, q1, and q2 in cacao agroforestry systems in San Vicente de Chucurí and Zapatoca, Santander, Colombia. The solid line indicates interpolation, the dashed line indicates extrapolation, and the shaded area represents the 95% confidence interval.

Figure 4. Kernel density of daily activity patterns of nine species of medium- and large-sized mammals recorded in cacao agroforestry systems in San Vicente de Chucurí and Zapatoca, Santander, Colombia (including humans). The x-axis represents a 24-hour period, and the y-axis indicates activity density.

Figure 5. Activity overlap between diurnal species and humans, and among ecologically similar species in San Vicente de Chucurí and Zapatoca, Santander, Colombia. Panels A and B correspond to comparisons based on the wildlife–human activity criterion. Panels C, D, and E correspond to comparisons between ecologically similar species, selected according to taxonomic order and trophic guild.