Introduction
Habitat degradation and fragmentation are two of the strongest threats to global biodiversity (Taubert et al. 2018), which coupled with overexploitation and other major anthropogenic disturbances and human-induced environmental changes, may risk the possibility for thousands of species to survive and reproduce (Andronache et al. 2019). Habitat fragmentation imposes dramatic structural changes within native landscapes (e.g., loss of physical and functional connectivity) that directly affects the ability of species to disperse across the landscape, posing a barrier to individual (and thus genetic) admixture. Thus, for forest-dwelling species, fragmentation restricts movements such as seasonal migration or individual dispersal, particularly for those taxa that are adapted to arboreality (Cushman et al. 2013).
Anthropogenic activities such as mining and oil extraction, cattle ranching, commercial and subsistence agriculture, wood extraction, hunting, and oil palm plantations are significantly altering tropical forests worldwide (Wright 2010; Laso Bayas et al. 2022), with biodiversity loss being a major consequence of these changes. Currently, the global biodiversity crisis is being referred to as the sixth mass extinction, given the current loss of species is occurring at an exceptionally rapid rate (Ceballos et al. 2015). Several groups of vertebrates – e.g., those with large body size – are being strongly affected (Leung et al. 2020). For example, Betts et al. (2017) assessed the relationship between forest loss and extinction risk of vertebrate species at a global scale and found that deforestation increases the likelihood of species being classified as threatened with extinction. Also, they identified high-risk hotspots in southeast Asia, the central-western Amazon, and the Congo Basin, suggesting the number of threatened species is predicted to increase in the coming decades.
Advertisement
Although Central and South America still hold a significant proportion of tropical forests, compared to other regions in the world, vertebrate population declines in these ecosystems have reached dramatical high levels (Benitez-Lopez et al. 2017; Almond et al. 2020; Leung et al. 2020; Brodie et al. 2021). According to the Living Planet Index, vertebrate populations in the Americas have had an average loss > 90% in the last 50 years (Almond et al. 2020). Particularly, the lowland tropical ecosystems of North-western South America (including tropical dry and humid forests, wetlands, and natural savannas) have been pervasively affected by extensive livestock farming and large-scale agricultural monocultures (Ayram et al. 2020). For example, in the middle Magdalena River valley, in between the eastern and Central Andes mountains of Colombia, monocultures and extensive cattle ranching have transformed more than 85% of native forests and wetlands, and those remaining are highly fragmented (Ayram et al. 2020). The Magdalena River’s Valley is within one of the Global Biodiversity Hotspots: the Tumbes, Choco, Magdalena biodiversity hotspot (Myers et al. 2000) and holds a great number of native and endemic species, and many of them are currently facing threats to their existence, partly due to the isolation of their populations in small forest fragments (de Luna and Link 2018).
As most tropical ecosystems become pervasively fragmented (Taubert et al. 2018), restoration through biological corridors is beginning to play a cornerstone role in wildlife conservation strategies by recovering the connectivity of previously isolated metapopulations (Chetkiewicz et al. 2006). Initiatives implementing large regional corridors aimed to protect charismatic species (e.g., bears, jaguars, primates, amongst other taxa) are underway in many countries. For example, the iconic recovery of the golden lion tamarins (Leontopithecus chrysomelas) conservation status in eastern Brazil, has included the protection of the remaining forests within its geographic distribution, the active reintroduction of individuals and the establishment of corridors that reconnect previously isolated populations (Dosen et al. 2017). Also, transnational corridors have been planned to protect the largest neotropical felid, the jaguar (Panthera onca), including landscapes within and outside protected areas, and across regions with different levels of human disturbance (Zeller et al. 2013). Other practical methods to recover wildlife connectivity have been implemented at a very fine scale, for example with arboreal canopy bridges across highways o electric lines (Gregory et al. 2017). Currently, less studies have focused on the role of corridors in the reconnection of wildlife populations in heavily fragmented landscapes, where the native ecosystem matrix has been pervasively transformed into pastures for cattle ranching and large-scale monocultures.
Landscape connectivity is important for the maintenance of ecological dynamics and the reliable provisioning of ecosystem services. Thus, corridors between isolated forest fragments might mitigate the potentially negative effects of habitat fragmentation by maintaining landscape connectivity (Mullu 2016). Restoration corridors are broadly defined as a structurally continuous vegetation that connects previously isolated forest fragments. Corridors may help provide resilience to wild populations through promoting movement of individuals across the corridors (Rosenberg et al. 1997). It is estimated that restoration corridors allow movement in fragmented landscapes, achieving proper responses of animals to climate change to increase the probability of persistence as populations (Cushman et al. 2013). Corridors may promote the movement of wildlife between previously isolated habitat fragment increasing the probability of wild populations to persist by inhabiting a larger interconnected area and may recover species richness (lost in open pastures) and subsequently functional diversity and ecosystem dynamics (Suárez-Castro et al. 2022).
In ecological studies and conservation implementations, the relation between species diversity and ecosystem functioning has become a central topic (Loreau et al. 2001; Cardinale et al. 2002; Tilman et al. 2014; Turnbull et al. 2016; Suárez-Castro et al. 2022). There is evidence showing that species richness alone, may not entirely reflect how species assemblages support ecosystem functions, particularly in disturbed landscapes (Mayfield et al. 2010). Overall, functional trait diversity (variation in species functional traits in a community) is considered a reliable proxy for assessing ecosystem functioning (Petchey and Gaston 2002, 2006; Mason et al. 2005). Thus, by studying the relationship between species richness and functional diversity we can reach a better understanding of how to preserve biodiversity and ecosystem functions in changing landscapes (e.g., restoration corridors).
Advertisement
The reduction of ecological complexity that takes place as undisturbed natural forests are transformed into simplified pastures and monocultures has been often related to a decline in species richness and functional diversity (Tscharntke et al. 2005). Support for the negative effect of human-induced land transformation on both species richness and functional diversity has been particularly evidenced for birds and mammals (Flynn et al. 2009). There is growing concern on the potential effects of functional diversity loss on the ability of ecosystems to provide services directly or indirectly associated to human wellbeing (Suarez-Castro et al. 2022).
In the heavily fragmented landscapes of the middle Magdalena River Valley, Fundación Proyecto Primates and Proyecto Vida Silvestre have implemented restoration corridors – thin strips of native trees (5–30 m wide) that are planted between forest fragments aiming to reconnect forest fragments structurally and functionally and the currently isolated wildlife metapopulations that are found within them. Here, our goal is to evaluate the effectiveness of restoration corridors in reconnecting terrestrial and arboreal wildlife populations (medium and large mammals and birds) in a fragmented landscape using a taxonomic and functional diversity approach. We expect that taxonomic and functional diversity will be higher in restoration corridors – compared to open pastures – providing evidence of habitat suitability for native species. We also expect the species composition of corridors to be similar to that of native forests, meaning that a large proportion of terrestrial and arboreal species use corridors to move between previously isolated forest fragments. Finally, we discuss the results of this study and highlight the opportunities and potential risks of implementing connectivity corridors to influence the successful planning and implementation of corridor-oriented conservation initiatives for large vertebrates in living in the remaining tropical forests of the world.
Methodology
Study area
This study took place at Hacienda Lusitania Private Natural Reserve, located in the lowland rainforests in between the Eastern and Central Andes cordilleras and within the middle Magdalena River basin in the department of Santander in Colombia (Fig. 1). At Hacienda Lusitania, wildlife conservation has been promoted for decades, and both hunting and logging are prohibited. The study area is made up of fragments of tropical rainforests immersed in a matrix of pastures for cattle ranching. Since 2016, Fundación Proyecto Primates and Wildlife Conservation Society Colombia have implemented 10 restoration corridors that aim to reconnect approximately 1200-ha of forest fragments (Fig. 1). Thus, restoration corridors have between one and six years of being established and have been planted with native tree species. The youngest corridors (1–2 years) have trees between one and three meters high with some isolated emergent trees that were present in the corridors before the implementations. The oldest corridors (4–6 years) have trees that range between 5 m and 15 m high, beginning to evidence a continuous canopy and forest-like structure. This scenario opens an ideal opportunity to evaluate the use (and effectivity) of corridors in order to reconnect isolated populations of medium and large mammal and birds in a heavily fragmented landscape. Our long-term terrestrial and arboreal mammal surveys in the study´s broader region (the Carare-Opon interfluvium) have provided recent evidence of the presence of medium and large terrestrial mammals with a historical distribution in this area (Link et al., unpublished data). Thus, we aim to assess whether corridors are indeed reconnecting previously isolated wildlife populations, including some of the most threatened mammals and birds such as the critically endangered brown spider monkey (Ateles hybridus) and the blue-billed curassow (Crax alberti), by quantifying taxonomic and functional diversity and comparing it among the four land cover categories studied (forest, natural corridors, restoration corridors, and pastures).
Fig. 1
Study area (Hacienda Lusitania Natural Reserve) showing camera trap locations and the main land cover types
×
![]()
Camera trap sampling
Overall, we set up 98 camera-trap sites (each containing a single camera), 19 were arboreal and 79 were terrestrial, covering four land cover categories. 37 cameras were placed in forest fragments that ranged from 20 to 600 ha in size, 28 cameras were placed in restoration corridors (corresponding to 5–30 m wide strips of restoration areas planted with native species), 17 cameras were placed in natural riparian corridors (corresponding to 10–30 m wide strips of riparian forests, generally adjacent to major creeks), and 16 cameras were placed in open pastures used for cattle ranching, these cameras were attached to fences or isolated trees within the pastures (Fig. 1). We used two digital camera trap models: [1] Bushnell TrophyCam HD and [2] Browning Strike Force throughout the study. Cameras were set to work 24 h per day and were active during periods of 16 to 104 days. Terrestrial cameras were placed on suitable trees at an approximate height of 40 cm. Arboreal cameras were placed on trees at an average height of 12 m, targeting large branches that appeared to be important travel pathways for animals while moving between trees. We did not install arboreal cameras in the pastures. All cameras were set up to take photos with a 30 s delay between consecutive events and to take three sequential photos per event. The spatial location of each camera trap (site) was recorded using a handheld GPS unit (Garmin GPSMAP 64s). Images obtained from camera traps were identified and organized using the open-access software Wild.ID (Fegraus and MacCarthy 2016) and metadata was exported as a CSV file. We used a time interval of 30 min to determine the independence between records, and the package camtrapR – in the statistical software R – was used to generate the record tables for all analyses (Niedballa et al. 2016; R Core Team 2022).
Diversity metrics
Initially, we calculated a relative abundance index (RAI) for each species and for each camera trap location. The index was calculated as the number of independent events divided by the sampling effort (number of activity days of each camera) and multiplied by 100 camera trap-days. Then we estimated metrics of taxonomic and functional diversity for each camera trap location. We used both abundance weighted metrics and metrics based on presence-absence data. For taxonomic diversity, we first constructed species accumulation curves per land cover category using the BiodoversityR package (Kindt and Kindt 2019) to ensure that sampling was representative for each category, and then we estimated species richness and Shannon index (Magurran and McGill 2010) using the RAI. For functional diversity, we used the Petchey and Gaston’s dendrogram index for multiple traits (FDpg) and the functional evenness (Petchey and Gaston 2002). For functional metrics we used the fundiv package (Gagic et al. 2015) in the statistical software R (R Core Team 2022). We used six ecological traits: body mass, diet breath, home range, maximum longevity, activity (diurnal, nocturnal), and locomotion (arboreal, terrestrial) (S2). Traits were extracted from the PanTHERIA (Jones et al. 2009) and EltonTraits (Wilman et al. 2014) databases; if a trait was not available for a species, we used the value from the closest related species (Chen et al. 2022).
Data analysis
Based on the relative abundance index (RAI), we performed a non-metric multidimensional scaling (NMDS) using Bray Curtis distances in the vegan package in R (R Core Team 2022; Oksanen et al.35). This ordination analysis allows us to organize camera trap locations by their similarity in species composition. Then, we performed a permutational multivariate analysis of variance (PERMANOVA) using the adonis2 function in vegan package (Oksanen et al.35) to evaluate differences in composition depending on the four categories of land cover included in this study (forest, natural corridor, corridor, and pasture), pairwise comparisons were performed with the pairwiseAdonis package (Martinez 2020). In order to identify the species that contributed most to variation in community structure between camera-trap sites, we applied the envfit function (vegan package) to the species RAI and ordination axis scores. We used Kruskal-Wallis Rank Sum Test, and a Pairwise Wilcoxon Rank Sum Test to compare the distributions of both taxonomic and functional diversity metrics between the four categories of land cover described above. Finally, we evaluated the relationship between species richness and functional diversity (FDpg) using a general linear model (GLM). All statistical analyses were performed in the software R (R Core Team 2022).
Results
Total sampling effort was 8497 camera trap nights, with an average of 86.7 days of activity per camera trap. Overall, we recorded 36 species, 28 corresponding to terrestrial and arboreal mammals and 8 of ground-dwelling birds. Species RAI values were segregated differentially between the four different land covers where cameras were installed, and they provide evidence about species uniquely (or mainly) present inside forest fragments, species that use both forest fragments and natural corridors, species present in all four land covers including restoration corridors, and species mainly using open pastures and grasslands (Fig. 2).
Fig. 2
Relative abundance index (RAI) of ground dwelling species (mammals and birds) through different land covers at Hacienda Lusitania Natural Reserve
×
![]()
Our results evidence how the largest primates (Ateles hybridus and Alouatta seniculus), other strictly arboreal taxa (Coendou prehensilis and Potos flavus) and forest-dwelling rodents (Dasyprocta punctata and Cuniculus paca) are mostly found in forest fragments and riparian corridors with similar tree structure (to forest fragments). Many medium and small primates (Cebus versicolor and Aotus griseimembra) and arboreal taxa, large omnivores such as the white-collared pecari (Pecari tajacu) and the largest terrestrial birds including the CR blue-billed curassow (Crax alberti) use restoration corridors and tend to avoid moving through open pastures. Other species, mainly including large predators (Puma concolor) and mesopredators (e.g., Leopardus pardalis, Eira Barbara and Didelphis marsupialis) moved with less restrictions across the landscape and used all four land covers, including a relative high activity recorded in restoration corridors (compared to pastures). Finally, three taxa (Colinus cristatus, Cerdocyon thous and Hydrochoerus isthmius) were mostly detected in open grasslands (Fig. 2).
The NMDS ordination analysis had statistical support (stress value of 0.17) and evidenced that camera-trap locations formed different groups within the ordination space, highlighting the strong influence of land cover types on the vertebrate community structure (PERMANOVA, R2 = 0.13, p = < 0.001) (Fig. 3). Pairwise comparisons evidenced that, in terms of composition, forests are different from all other land cover categories: Forest-Natural corridor (R2 = 0.08, p = 0.001), Forest-Corridor (R2 = 0.11, p = 0.001), and Forest-Pasture (R2 = 0.12, p = 0.001). Comparisons between the other land cover categories showed more subtle differences: Corridor-Natural corridor (R2 = 0.05, p = 0.004), Corridor-Pasture (R2 = 0.04, p = 0.055) (not statistically different), Natural corridor-Pasture (R2 = 0.09, p = 0.001). Using the envfit function with a p-value ≤ 0.01 we identified nine key species influencing the differences between camera-trap sites: The night monkey (Aotus griseimembra), the brown spider monkey (Ateles hybridus), the brown-eared wooly opossum (Caluromys lanatus), the varied white-fronted capuchin (Cebus versicolor) and the red-tailed squirrel (Sciurus granatensis) were arboreal species most often recorded in forests; the lowland paca (Cuniculus paca), the collared peccary (Pecari tajacu) and the brown four-eyed opossum (Metachirus nudicaudatus) were terrestrial species frequently recorded in forests, and the crab-eating fox (Cerdocyon thous) and the lesser capybara (Hydrochoerus isthmius) were more often associated to open pastures.
Fig. 3
Non-metric multidimensional scaling (NMDS) showing the species composition structure depending on the different land cover types (forest, natural corridors, corridors, and pasture) and the species that contributed the most to the differences between camera-trap sites
×
![]()
We found significant differences in diversity metrics between land cover categories from the Kruskal Wallis and Wilcoxon tests. In terms of taxonomic richness, we found a pattern of higher species richness in the forest, followed by the corridors and finally the pasture (Fig. 4). There were significant differences between forest and corridor (p = 0.010), forest and pasture (p = < 0.001) and natural corridor and pasture (p = 0.015). For functional diversity, the Petchey and Gaston’s dendrogram index showed a similar pattern to that obtained with species richness, with higher values of functional diversity in forest, followed by corridors and finally lower values in open pastures. There were significant differences between forest and corridor (p = 0.01), and forest and pastures (p = 0.007). For evenness we found higher values in pastures, followed by corridors, and then forest. There were significant differences between forest and corridor (p = 0.031). Finally, we found a positive relationship between species richness and functional diversity measured as the Petchey and Gaston’s dendrogram index (FDpg) (p = < 0.001, R2 = 0.87, Estimate = 0.16) (Fig. 5).
Fig. 4
Taxonomic and functional diversity between land covers at Hacienda Lusitania. The first panel shows taxonomic richness, the second panel shows the Petchey and Gaston’s dendrogram index for multiple traits, and the third panel shows the functional evenness
×
![]()
Fig. 5
Relationship between the functional diversity index (FDpg) and richness, discriminated by land cover categories and showing the Petchey and Gaston’s functional dendrogram
×
![]()
Discussion
The results of this study provide evidence supporting that restoration corridors can have a positive impact in reconnecting vertebrate populations in the pervasively fragmented landscapes of the Middle Magdalena Valley, in Colombia. Although in our study a few species were naturally well adapted to use forest edges, wetlands and human-made pastures (e.g., crab-eating fox, capybaras), most medium and large mammals and terrestrial birds seemed to have strong dispersal limitations in areas without forest cover which may be reduced by using vegetated corridors. Our results also highlight the different response of wildlife to use the corridors. For example, while large arboreal mammals and some strictly forest dwelling species rarely used the restoration corridors, most of terrestrial mammals and birds included in this study did use corridors to move between forest fragments (Fig. 2). However, we note that some cameras had records of brown spider monkeys using them, and although we did not include the age of the corridors in the analysis, we know that these are older corridors (5–6 years old).Although sampling was not evenly distributed across the four different land cover categories, it is evident that most species tend to be recorded more frequently in the corridors compared to open pastures (Fig. 2). Interestingly, corridors seem to be used by terrestrial birds, including the Critically Endangered blue-billed curassow, which may increase the resilience of this threatened population, given that previously isolated individuals are now able to move across fragments and increase their population-wide reproductive connectivity. The relative low use of corridors by large primates, such as spider monkeys and howler monkeys, may also be influenced by their territorial behaviour. These primates have a high degree of site fidelity to their home ranges and explore new areas slowly and progressively (Ramos-Fernandez et al. 2013; Asensio et al. 2018) but may act as key pathways for natal dispersal of individuals (which are generally rare but important events for maintaining the genetic integrity of primate populations (Di Fiore et al. 2009).
In the heterogeneous landscape of the middle Magdalena basin, species composition in forests tends to be relatively similar and this may be largely influenced by the presence of several taxa that are mainly present if areas with complex and robust vegetation structure (Fig. 3). The results of our NMDS analysis also show that the two more distinct habitats are pastures and forest fragments, and that corridors are habitats that are used by most terrestrial mammals and birds in the study area, as corridors are starting to be used by forest taxa not commonly present in pastures (Fig. 2). Additionally, based on the multidimensional analysis graphic (Fig. 3) we consider species composition variation in the corridors is driven by its use by both forest-dwelling species and other species more adapted to open habitats, Thus, species richness in the corridors has intermediate values between pastures and forests, thus suggesting corridors are used more often by a large array of mammals and birds that may seldom use pastures to move between forest fragments.
The PERMANOVA, pairwise comparisons and the envfit analyses show a differentiation between three groups of animals on how they used the different land covers/habitats: A group of species mainly using forested areas, a small group of species using pastures (and seldom other land covers) and finally a large group of animals using most vegetation covers available in the study region (Figs. 2 and 3). While arboreal taxa have a strong tendency to rely on forests, pastures were mainly used by generalist taxa (such as capybaras) which are largely aquatic and the pastures in the farm have several artificial lagoons made for cattle, and by crab-eating foxes.
At the community level, we recorded higher species richness in restoration corridors compared to open pastures. This may reflect the preference for many mammals to use corridors, if available, and avoid crossing between fragment through open grasslands. Although species richness in restoration corridors is not as high as that recorded for riparian corridors and forest fragments, it does seem that it might mitigate the potential risks associated to using open habitats (Fig. 4).
The Petchey and Gaston’s dendrogram index (FDpg) provide a reliable measure of species trait’ values complementarity (greater differences in species’ trait values are greater trait complementarity and greater FDpg values.) in the communities. Based on that, we found that functional diversity (FDpg) is higher on forest fragments and natural corridors (Fig. 4), reflecting major complementarity between species traits within these structurally more complex land covers, that is, higher values of FDpg representing greater differentiation between species trait values (Petchey and Gaston 2002, 2006). Conversely, in pastures, lower values of functional diversity reflect a higher functional redundancy that may be explained by the dominance of generalist species in pastures communities because they are adapted to the environmental conditions of opened areas (Petchey and Gaston 2002) (Fig. 4). Corridors FDpg values were intermediate between pastures and both, forests and natural corridors, suggesting that communities in corridors had a lower functional diversity compared to communities in forests, but more functional diversity than those communities in open pastures. This result supports the role of restoration corridors in promoting the recovery of functional diversity of communities in the pervasively transformed landscape of the Middle Magdalena Valley in Northern South America. The lower values of functional evenness in forests, that increases in corridors and reaches highest values in pastures, might be reflecting that the larger number of species present in the forest have uneven abundances, contrastingly, the few species recorded in pastures have more even and similar abundances (Fig. 2). These results demonstrate the importance of corridors in recovering functional space in the previously isolated mammal and terrestrial bird communities of the forests of the Middle Magdalena River Basin. The positive relationship between species richness and functional diversity in this study (Fig. 5) possibly is reflecting that in this heterogeneous local area the resources productivity that increases with the establishment of corridors may promote trait divergence and increase the species richness in the landscape (Le Provost et al. 2020; van’t Veen et al. 2020; Suárez-Castro et al. 2022).
This study provides some of the first empirical evidence on how restoration corridors that connect isolated forest fragments may be an effective conservation initiative to protect biodiversity in heavily degraded and fragmented landscapes. This coincides with the more available theoretical approaches to habitat connectivity and its potential effects on biodiversity (Correa Ayram et al. 2016). In the middle Magdalena River valley, the implementation of vegetated corridors reconnecting isolated forest fragments may be a complementary conservation strategy to some CR taxa such as the brown spider monkey (Ateles hybridus) and the blue billed curassow (Crax alberti).
Many studies related to the implementation of restoration corridors have been made, but until now, an assessment of their potentiality to reconnect isolated wildlife populations has been scant. While some of these studies have found that restoration corridors do support species conservation, the design of these studies plays an important role in their effectiveness (Beier and Noss 1998). Furthermore, there has not been a preliminarily study that shows a negative effect of restoration corridors, thus supporting current initiatives oriented to reconnecting isolated fragments within the fragmented landscapes of Middle Magdalena River Valley in Colombia (Beier and Noss 1998).
There has been a general expectation that individual movements across corridors might increase population maintenance and integrity, as well as maintain the historical community composition of native habitats, although evidence for this is still limited (Mullu 2016). This study provides some of the first empirical evidence supporting how vegetated corridors that reconnect previously isolated forest fragments may influence how species occupy their landscapes (by varying the spatial configuration and degree habitat resistance). We document higher species richness in corridors compared to pastures, and more similar community assemblages in corridors compered to forest fragments. By increasing the effective size of forest fragments through vegetated corridors we predict that variables associated to habitat structure and composition will resemble those found in continuous forests, which are currently largely absent in the middle Magdalena River basin. Thus, we expect that through the reconnection of isolated forest fragment in the pervasively fragmented landscape of the Magdalena River valley, these threatened ecosystems within the Tumbes-Choco-Magdalena biodiversity hotspot will maintain its ample and unique biodiversity, and mitigate the potential loss of species and their connectivity, compared to a scenario of forests restrained to isolated and disconnected fragments.
In conclusion, our study provides initial evidence on the positive effects of connectivity corridors to mitigate the negative effects of habitat fragmentation in the pervasively fragmented tropical rainforests in Colombia, and worldwide (Taubert et al. 2018). According to the obtained results, animals are using the stablished corridors to move between forest fragments, and species richness and composition seems to be trending towards that of the remaining native forests. Finally, we caution that the implementation of vegetated corridors may be an effective strategy to increase the resilience of wild populations, but may pose novel challenges as well, as for example increase vulnerability to human-hunting and intra-specific dynamics, which will need to be further addressed in future studies. Nonetheless, the results of our study provide evidence for the successful and viable implementation of conservation strategies in areas heavily devoted to agricultural or extensive cattle ranching activities, that in tropical countries like Colombia may represent the vast majority of area influenced by human activities.
Acknowledgements
We are extremely thankful to the Jaramillo family who have allowed us to implement a project on recovering forest connectivity, and protecting local biodiversity, within a matrix of pastures dedicated to extensive cattle ranching. We also thank Ronald Mejía, Arnulfo Montoya, Fabian Castillo and many other local community members and Colombian students that have planted and maintained our restoration corridors. We thank the generous financial support of several organizations for restorations and wildlife conservation including the World Wildlife Fund (Russel E Train Education for Nature -EFN- grants SW72, RA75, RH47 and RK60), European Outdoor Conservation Association, Mohamed bin Zayed Species Conservation Fund, International Primate Protection League, Foundation Ensemble and IUCN-Netherlands (SPN grants). We also thank the Facultad de Ciencias and Vicerrectoria de Investigación at Universidad de los Andes, Wildlife Conservation Society Colombia and Ecopetrol for their financial support.
Declarations
Competing interests
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
