Costa Rica is one of the most biodiverse countries in the world, hosting almost 4% of all known mammals. Although the local mammals have undoubtedly been more investigated than those of other Central American countries, some regions still appear poorly known to date, such as the Nicoya Peninsula, NW Costa Rica. Our research focuses on the Karen Mogensen Wildlife Refuge, an important component of the Nicoya Peninsula Biological Corridor. This investigation represents the first comprehensive study of mammals of this area, with the aims to provide baseline information on the mammal community and tourism management. The investigation during a 6-year period (2018–2023) revealed the presence of 60 species, belonging to 50 genera, 23 families, and 8 orders, indicating a high species richness of the area. Among these, we highlight the occurrence of the globally Vulnerable Alouatta palliata and Cebus imitator and the Near Threatened Leopardus wiedii and Vampyrum spectrum, along with the recently recognised Lontra annectens, whose status is still Not Evaluated. The influence of tourism on activity patterns of mammals was also investigated. Using camera trapping, 13 species were detected in 537 trap-days and divided based on functional traits. No evidence of impacts caused by the presence of tourists was found on these species. This suggests that the current conservation policies of the Wildlife Refuge appear to be incorporating ecotourism in a wise and thoughtful manner. The information gathered will be a useful tool for the development and implementation of conservation strategies and actions at the local scale in the near future.

Activity patterns, camera trapping, checklist, conservation, mammal community, temporal overlap

The Central American Costa Rica, one of the 36 world biodiversity hotspots (Mittermeier et al. 2005), represents only 0.03% of the world’s land surface (Rodríguez and Chinchilla 1996; Reid 2009) but hosts almost 4% of all mammals’ species in the World (Wainwright 2007; Reid and Gόmez Zamora 2022; Mammal Diversity Database 2025). Several factors contribute to this diversity, from the tropical climate, which provides conditions that allow the growth of many autotrophic and heterotrophic organisms, to the four mountain ranges as well as the weather systems blowing in from both oceans, creating a dense and varied mosaic of habitats (Huettmann 2015). Costa Rica’s diversity is also influenced by its very position right between North and South America, connected to each other on account of the formation of the Panamanian land bridge approximately 2.8 million years ago (Woodburne 2010).

Today, the mammal conservation in Costa Rica is very delicate on account of two fundamental aspects. On the one hand, the second half of the last century was a dismal period for many Costa Rican mammals as a consequence of habitat loss caused by deforestation and pollution coming from the industrial, domestic, and agricultural sectors (Evans 1999). On the other hand, in 1970 a national park system was created and 28% of the country, in one form or another, was brought under its protection (Evans 1999). Consequently, ecotourism has grown enormously during the last decade and for many landowners, protecting forests and opening them up to tourism has become the most lucrative option. Several examples, from Costa Rica (Fletcher 2012; Hunt et al. 2015; Lopez Gutierrez et al. 2020) to many other regions of the globe (Tisdell and Wilson 2002; McConnell and Kull 2014; Buckley et al. 2016), show that ecotourism provides socio-economic and environmental sustainability.

Species coexistence is ruled by several and complex mechanisms (Chesson 2000) and is based on the partitioning of ecological niches (Amarasekare 2003). Despite being a complex concept to define, a niche is generally described by three dimensions: temporal, resource, and spatial (Case and Gilpin 1974; Albrecht and Gotelli 2001; Amarasekare 2003). Species that coexist in an area must differ in at least one of these dimensions (Gause 1934). Despite being regarded as the least important, the temporal dimension often plays a key role in species’ behaviours and interactions, and their associated consequences for community structure (Monterroso et al. 2014; Frey et al. 2017). There is a strong dependence between the species diel activity patterns and their internal “clocks” (Aschoff 1966) that is species-, sex-, and age-dependent, but several factors can affect this “clock” with consequent changes in phenotypic plasticity (Monterroso et al. 2014; Martín-Díaz et al. 2018). How species use time and distribute their activity within the diel cycle is an important aspect of animal ecology and behaviour that provides valuable information about their ecological niche (Schoener 1974; Frey et al. 2017). At the community level, understanding how species partition time contribute to explain the mechanisms facilitating stable coexistence (Carothers and Jaksić 1984; Kronfeld-Schor and Dayan 2003; Frey et al. 2017). Indeed, patterns of daily activity of the species determine the biotic (e.g., predation, competition), abiotic (e.g., light level, temperature), and anthropogenic factors to which they are exposed (Schoener 1974; Hut et al. 2012; Cox et al. 2021). For example, predators adapt their activity to maximise the probability of prey encounter and, contrarily, preys reduce their activity when predators are active (Foster et al. 2013; Monterroso et al. 2013). In addition, species with similar ecology and morphology minimise interspecific competition through temporal segregation (Lucherini et al. 2009; Di Bitetti et al. 2010; Monterroso et al. 2014).

The presence of humans in wilderness and protected areas can disturb wildlife, inducing behavioural changes and shifts in diel activity patterns, even in response to non-lethal human activities such as hiking and recreational trekking (Taylor and Knight 2003). For example, in the absence of tourists, leopards (Panthera pardus) exhibit increased diurnal activity compared to periods when tourists are present (Ngoprasert et al. 2017). Similarly, other large carnivores such as jaguars (Panthera onca) and pumas (Puma concolor) have shown increased nocturnality in response to human presence (Ouboter et al. 2021). This shift to nocturnal behaviour has been extensively documented in a global meta-analysis by Gaynor et al. (2018), which demonstrated that, worldwide, mammals are increasingly becoming nocturnal as a strategy to avoid human disturbance. Mammals of all trophic levels, including carnivores, omnivores, and herbivores, have exhibited increased nocturnality in response to human presence, with diurnal species becoming more nocturnal and crepuscular or nocturnal species further intensifying their nocturnal activity (Gaynor et al. 2018).

Most studies on temporal patterns in species and community ecology have used camera trapping, a method that has proven to be very efficient for this kind of research (Frey et al. 2017). Temporal camera trapping data offer the opportunity to investigate diel activity patterns and temporal partitioning in sympatric mammals (e.g., Di Bitetti et al. 2010; Monterroso et al. 2014; Sunarto et al. 2015), as well as evaluating prey-predator and predator-predator interactions (e.g., Botts et al. 2020; Hernández-Sánchez and Santos-Moreno 2020; Marinho et al. 2020), and effects of human-driven disturbances on wildlife (Wang et al. 2015; Ngoprasert et al. 2017). Camera trapping has also proven to be particularly suitable to assess species richness of an area (Rovero et al. 2014; Rovero and Spitale 2016) and to study rare and elusive animals (Rovero et al. 2010), common characteristics of Neotropical mammals (Thompson 2004; Sanderson and Trolle 2005; Botts et al. 2020).

In this study we investigated the mammal community of the Karen Mogensen Wildlife Refuge, located in the southern part of Nicoya Peninsula, northwestern Costa Rica. This region is one of the most poorly studied in all of Costa Rica (Timm et al. 2009), although it hosts the most threatened forest ecosystem of Central America (Janzen 1988; Huettmann 2015). Indeed, for the Karen Mogensen Wildlife Refuge, as for most of Costa Rican protected areas, ecotourism represents a fundamental income for its maintenance. Our study, which represents the first comprehensive study of the mammals of this protected area, intends to report baseline information on mammal community, species temporal interactions, and human influences. Specifically, we aimed to: (1) create a complete checklist of mammals present in the entire area; (2) examine the diel activity patterns of mammals in the Karen Mogensen Wildlife Refuge; (3) assess the effectiveness of tourism management in the protected area by evaluating variations in the activity rhythm of species and functional groups in relation to the presence or absence of tourists. The documentation of mammalian species richness, along with exploratory analyses aimed at assessing potential impacts of tourism, provides important insights into wildlife behavioural responses to human disturbance within the Wildlife Refuge.

Overall, our research represents the first comprehensive investigation of the mammals of the Wildlife Refuge, revealing a high species richness that represents more than 25% of all mammal species documented for Costa Rica. We highlight that the species richness of non-volant mammals of the Wildlife Refuge is substantially in line with other neotropical areas of recent investigation (e.g., Da Silva et al. 2018; Hoskins et al. 2018; Pommer-Barbosa et al. 2023), or even higher if we consider that most of these studies covered larger areas. Furthermore, our findings suggest that the current tourism management in the Karen Mogensen Wildlife Refuge seems to be effective, as we did not detect significant shifts in species’ diel activity between days with and without tourist presence. This may indicate that, currently, human disturbance associated with tourism does not appear to substantially affect the short-term temporal activity of mammal species.

Undoubtedly, the mammal community of the area consists of various small-sized terrestrial species, but during the investigation only three species belonging to two families (one Heteromyidae and two Cricetidae) have been occasionally recorded and identified. Dedicated investigations are needed for this component of the local mammalian diversity. On the contrary, regarding the chiropterans, although the results may be rather provisional, the record of 30 species indicates a high species richness, especially in relation to the small extension of the Wildlife Refuge. This number corresponds approximately to ca 25% of the Costa Rican species of bats known to date, with representatives of five out of nine families present in the country as a whole (York et al. 2019).

Some significant considerations regarding conservation issues can be drawn by our records. Two of the species reported, the mantled howler and the Panamanian white-faced capuchin, are globally Vulnerable and decreasing population trend (IUCN 2025). As for these primates, we highlight that the populations inhabiting the Wildlife Refuge show a good conservation status with more groups inhabiting the forested environment. Furthermore, one non-volant recorded species, i.e., margay, is globally Near Threatened and with decreasing populations. Besides, the status of the recently recognised northern neotropical river otter L. annectens is still Not Evaluated, even though until 2024 this species was considered L. longicaudis, whose status is Near Threatened (IUCN 2025). In addition, other three species are not threated but are characterised by globally decreasing populations, namely jaguarundi, kinkajou, and Derby’s woolly opossum (IUCN 2025). Another interesting information is the record of southern spotted skunk, as this species was not known throughout most of Costa Rica, and its southernmost global range borders correspond to the inner part of northern Costa Rica (IUCN 2025; GBIF 2025). Thus, our data extend the global range of this Mephitidae to northwestern Costa Rica, specifically to southern Nicoya Peninsula.

Further remarks concern the chiropterans. The spectral bat Vampyrum spectrum (Fig. 5) is rare in Costa Rica, globally Near Threatened and with decreasing populations (IUCN 2025). A nursing female was recorded in December 2022, highlighting that this species is not a sporadic occurrence in the area (authors pers. obs.). In addition, the northern ghost bat Diclidurus albus (Fig. 4) is an uncommon species throughout the whole country (Wainwright 2007; Loza et al. 2018; Morazán-Fernández et al. 2023). This bat was observed regularly over the years, but with no more than 3–4 individuals. The chiropterans detected include frugivorous (e.g., Artibeus spp., Dermanura spp.), insectivorous (e.g., Myotis spp.), carnivorous (e.g., V. spectrum), and haematophagous (the vampire bats Desmodus rotundus and Diphylla ecaudata) taxa. The high diversity along with the variety of feeding habits of these animals, known to be bioindicators (Wainwright 2007 and literature therein), emphasise the remarkable level of ecosystem preservation of the Wildlife Refuge.

The camera trapping led for a 4-month period allowed us to register 13 species in 537 trap-days. Taking into consideration the short period and the presence of four almost strictly arboreal species among the 30 non-volant included into the final checklist, we underscore the usefulness of this data collection method for stablishing baseline information on the mammal community of an area.

A number of studies based on or at least including this kind of data collection have been conducted in several forested neotropical locations in recent years, focusing on medium and large-sized mammals. Specifically for Costa Rica, a recent study by Beal et al. (2020) revealed the presence of 19 species for a 2-year period (1440 trap-nights) in one of the most diverse areas of the country: Piedras Blancas National Park. Similarly, Botts et al. (2020) recorded 24 species during a 10-year period (59,919 trap-days) and at 12 different sites in several locations of the country. As for other neotropical countries, for instance Hoskins et al. (2018) recorded 24 species in the Cusuco National Park, Honduras, based on different methods and during a 10-year study period (ca 900 trap-days). Da Silva et al. (2018) reported 26 species in the Iguaçu National Park, Brazil, based on camera traps at 37 sites and for an 8-month period in two different years (over 9,000 trap-days). An investigation of patches of semi-deciduous Atlantic Forest in Minas Gerais, Brazil, revealed the presence of 16 species of mammals; the study was carried out with 16 camera traps for a 2-month period (more than 900 trap-days; Costa et al. 2019). Another study in a Brazilian area, the Cazumbá-Iracema Reserve, revealed the presence of 15 species by means of 5 camera traps utilised for one year (240 trap-days; Oliveira and Calouro 2020). Jansen et al. (2020) reported 24 species in an area within the Chiquitano Dry Forest, Bolivia, using 26 camera-traps for an 8-month period (1,700 trap-days). A total of 20 species was detected in more than 2,000 ha of forest remnants in Conceição dos Ouros, Brazil (6,120 trap-days; Vilas Boas et al. 2022). Generally, as reported above, among recent studies carried out in different forested neotropical contexts, the species richness revealed is higher on average (sometimes more than twice) than the diversity that came to light in our camera trapping survey. Nevertheless, it must be noted that all these studies covered broader areas and/or longer timespans, and most of them used a higher number of camera traps, which meant a higher number of trap-days (often at least twice the trap-days in our study). If we only consider research in areas analogous to the Karen Mogensen Wildlife Refuge in terms of both size and habitat typology, and based on the use of a similar number of camera traps, the species richness recorded is largely comparable to the diversity revealed by the camera trapping activity of the present study (e.g., Oliveira and Calouro 2020: 15 species vs 13 species recorded in the Karen Mogensen Wildlife Refuge). A general consideration related to our camera trapping survey is that, as our data collection was derived from the use of camera traps placed 40–50 cm immediately above ground level, information regarding mammals with mostly arboreal activity is sparse (i.e., Panamanian white-faced capuchin) or totally lacking, as in the case of kinkajou, Derby’s woolly opossum, mantled howler, and Mexican hairy porcupine. Despite this, it must be noted that two partially arboreal species (northern tamandua and white-nosed coati) were included in the analyses because of the relatively high number of records.

The camera trapping survey allowed to gather information for studying the impact of tourism on the local mammal community. Most literature studies demonstrated that tourist activities alter the wildlife activity patterns. In a global review, Gaynor et al. (2018) revealed a strong effect of humans on daily patterns, increasing wildlife nocturnality. This shift from natural patterns has implications on fitness population persistence, community interactions, and evolution but may facilitate human-wildlife coexistence (Gaynor et al. 2018). Oberosler et al. (2017) in a protected area in the Italian Alps, found direct impact of humans on the mammals’ activity pattern: all species significantly decreased their activity during the hours most frequented by tourists. Matthews et al. (2006) and Belotti et al. (2012) demonstrated that tourist activities altered the activity patterns of black bear (Ursus americanus) and Eurasian lynx (Lynx lynx), respectively. Díaz-Ruiz et al. (2016) argued that human disturbance is a factor that affected the activity pattern of red fox (Vulpes vulpes), known to exhibit a behavioural plasticity, decreasing its diurnality in areas with high human presence. Conversely, in the Lapa Rios Ecolodge Natural Reserve in Costa Rica, a recent study highlighted no indication that visitation and tourism activity have negative impacts on wildlife (Lopez Gutierrez et al. 2020). The authors, comparing the hours of activity between tourists and wildlife, declared that species patterns seem to be driven by specific ecological requirement and behaviour of animals rather that by human activity. The study area and the management strategies described by Lopez Gutierrez et al. (2020) are different from the studies discussed above, but very similar to those of the Karen Mogensen Wildlife Refuge. In particular, both areas present human activities related to ecotourism in the broadest sense (Fletcher 2012; Hunt et al. 2015; Lopez Gutierrez et al. 2020). A small part of the Karen Mogensen Wildlife Refuge is crossed by a single narrow road that is not accessible to standard motor vehicles. Furthermore, the number of tourists is limited, and they are always accompanied by guides. Acknowledging that some species are under-represented in the overall pool of observations, our results show, in agreement with Lopez Gutierrez et al. (2020), that diurnal activity patterns of mammals and species groups that share similar traits, i.e., the activity cycle and the diet, do not appear to be affected by human presence. However, our study focused specifically on diel activity patterns to explore potential short-term behavioural responses of mammals to tourism. While this approach offers valuable insight into temporal shifts in activity between day with and without tourists, it represents only one aspect of species’ ecological responses. Therefore, we cannot exclude the possibility that tourism may influence other aspects of mammal ecology, such as spatial distribution and habitat use. Further research with broader methodological approaches and larger sample sizes, with larger temporal scale, would be necessary to comprehensively evaluate the potential effects of tourism on wildlife. Nevertheless, our current findings did not reveal significant short-term behavioural changes associated with tourist presence, and the Wildlife Refuge supports a high level of mammalian diversity, suggesting that current tourism levels may not be exerting strong negative impacts. More generally, the information gained highlights the importance of the investigated area for conservation. Despite its relatively limited extension, it represents an area of high biodiversity and an essential part of the Biological Corridor of the Nicoya Peninsula. A further relevant reason for the present study stems from the increased knowledge of the mammal community in one of the least studied regions of Costa Rica, though characterised by a globally high conservation interesting environment, i.e., the threatened tropical dry forest ecosystem of Mesoamerica (Janzen 1988; Huettmann 2015).

Our investigation underscores that the wilderness of the Wildlife Refuge is undergoing a period of renaissance since its creation in 1996, marked by the presence of a rich animal diversity with a diversified vegetation, similarly to other well-known and more popular parks and natural reserves in Costa Rica, all of which have greater extensions. As reported by a former investigation on the avifauna, several parts of the area have shown a clear regeneration process, with the growth of secondary forest surrounding the patches of primary forest already present before the birth of the Wildlife Refuge (Dal Zotto et al. 2017). The species richness reported here proportionately matches recent ornithological studies conducted in the area, which proved the presence of more than 220 bird species (Dal Zotto et al. 2017; Lopez et al. 2023; authors pers. obs.). Undeniably, there is clear evidence that the area is recovering from historical anthropogenic disturbances, as the mammal community shows signs of ecological regeneration.

The present study allowed us to define and understand the current dynamics occurring between and among the mammal species of the Wildlife Refuge. This information is fundamental for future conservation plans, in order to permit a positive coexistence between the local human communities and the autochthonous fauna. Wildlife conservation can be effective only if innovative strategies at a local scale take into account the coupled nature of social and ecological systems (Guerrero and Wilson 2016). This may seem easy, but, for instance, we must know that the jaguar (Panthera onca), once living in the area, disappeared during the 1970s precisely because they were killed by local farmers who responded to the predation of their cattle by this felid. Our investigation showed that the puma, which had also disappeared from the area for several years, is now back in the Wildlife Refuge. In fact, the puma was detected in 2018, after years without any indications of its presence (L. A. Mena Aguilar pers. obs.) and our study confirmed its first documented reproduction within the area, thanks to some videos portraying females with two cubs. Its permanent presence must be taken into consideration in order to avoid repeating the fate suffered by the jaguar. We must also consider that this last felid is recolonising the north-west of Costa Rica, on the border with Nicaragua, so we cannot exclude that, in the next few years, it will come to populate again the area that today corresponds to the Refuge. More generally, in Costa Rica, areas such as the targeted one, although protected by national laws as Wildlife Refuges, need more custody from poaching, which unfortunately has become increasingly frequent in recent years (authors pers. obs.), as well as from the presence of free-ranging dogs, which have been shown to have deleterious effects on wildlife, acting as predators or competitors to native species, especially in fragmented habitats or well-preserved areas of tropical environments (Carvalho et al. 2019). These issues are not easy to solve since they require long-term projects to help the local people understand the importance of the fauna and flora that surround them, both in terms of ecosystemic concerns and of economic returns. It is our hope that, with the additional knowledge gained from surveys such as this one, local governors and stakeholders will make prudent decisions to preserve and defend these areas and allow the development or the continuation of conservation projects. To this end, a positive scenario results from the recent conversion of the private Karen Mogensen Reserve into the Wildlife Refuge (Refugio de Vida Silvestre), which is included in the network of the public protected areas of Costa Rica.

We want to thank Arnulfo and Mery Quiros Salazar, the president Patricia Slump, and all the ASEPALECO (Asociación Ecológica Paquera, Lepanto y Cóbano) team, all the members of the association Foreste per Sempre OdV, Davide Iamonte and Esther Marneffe. Thanks to Luca Lombroso (Geophysical Observatory, University of Modena and Reggio Emilia) for providing us with the meteorological data. We express our gratitude to Dr. Bernal Rodríguez (Universidad de Costa Rica), Diana Pereira Brenes, Naidely Vidaurre Quesada, R. Andres Hernández Jimenez, Keilyn Castro Avendaño, Thomas McFarland, and all the researchers involved in the monitoring of chiropterans for sharing the data collected. Finally, we want to thank two anonymous reviewers who allowed us to substantially improve this manuscript, and Nathalie Yonow for the final editing.