Connecting Colombia’s protected areas: Using a functional approach for tapir species

Abstract

Colombia is the world’s fourth most biodiverse country for mammal species. This condition is evidenced in the high number of mammal species reported in specific groups, such as tapirs. The country is considered a hotspot for the genus Tapirus, reporting three of the four species scientifically valid (Tapirus bairdii, T. pinchaque, and T. terrestris). Approximately ~ 49% of Colombia’s natural ecosystems have been transformed by human activities, and  ~  16% of the national territory is designated protected areas (PAs). In this context, the ecological connectivity between PAs is essential to improve the conservation of threatened large mammals such as tapirs and to contribute to the effectiveness of management of these areas in the current scenario of global change. We developed connectivity models for the tapir species and identified critical areas to conserve and improve ecological connections between PAs in Colombia. To this end, we constructed (i) distribution models for tapir species, (ii) movement resistance surfaces, and (iii) mapped least-cost corridors (LCCs). We also used the circuit and least-cost models to locate conservation priorities and restoration opportunities, estimating the equivalent connected area (ECA) index. Our results provide a national-level assessment of functional connectivity priorities for tapir species. This assessment could be considered as an input to guide efforts related to conservation, restoration, and implementation of management tools that facilitate the movement of tapirs through transformed landscapes. Implementing of these actions could contribute to meeting the goals of the post-2020 global biodiversity framework, which aims to achieve effective, ecologically representative, well-connected, and equitably managed PAs.

Introduction

The first strategy for conserving biodiversity in the face of the accelerated transformation of natural ecosystems over the last century is based on the designation and management of protected areas (PAs), which are intended to contribute to the long-term maintenance of the ecological functions of natural environments [1,2]. However, the role of PAs is limited by their size and the broad biological requirements of some species (e.g., large mammals with extensive migratory movements). Some areas have low levels of connectivity because they are isolated in a matrix of regions altered by human activities [2]. Reduced levels of connectivity between PAs limit several biological processes of species, such as dispersal, migration, and gene flow between populations [3]. To improve connectivity levels between PAs, international policies such as the convention on biological diversity (CBD) and the subsequent Kunming-Montreal global biodiversity framework (GBF) call on local governments to implement policies for effective and efficient management of PAs that contribute to improving ecological connectivity levels between PAs and actively involving communities in the planning, management, and climate change adaptation measures of these areas [4].

Colombia is the second most biodiverse country globally, with an estimated 10% of global biodiversity represented by more than 67.000 described species [5]. However, only 310.000 km², or  ~  16% of the country’s surface area, represents the 1.443 national, regional, and local PAs [6]. In recent decades, this country has experienced a high transformation of its natural ecosystems reaching 49% of the national continental area [7–9]. The main drivers of transformation reported are (i) deforestation for the establishment of extensive livestock productions ( ~ 420.000 km²), agro-industrial productions of oil palm (Elaeis guineensis ~ 5.000 km²), sugar cane (Saccharum officinarum ~ 2.400 km²), rice (Oryza spp.  ~  5.848 km²), banana (Musa spp.  ~  2.500 km²), coffee (Coffea spp.  ~  9.000 km²), forest plantations (Acacia magium, Eucalyptus spp., and Pinus spp.  ~  12.000 km²), illicit crops (Erythroxylum coca and Cannabis spp.  ~  2.000 km²);(ii) illegal gold mining ( ~ 2.000 km²), and (iii) road infrastructure construction ( ~ 2.000 km²) [10,11]. Drivers of the transformation of natural ecosystems in Colombia include unsustainable practices such as deforestation and inappropriate use of fire [12–15]. These actions reduce the supply of resources for large herbivorous mammal species such as wild ungulates (e.g., tapirs, peccaries, and deers species), which in turn increases the vulnerability of these species to processes such as local extinction of their populations [16–18]. These human activities have drastically shaped Colombia’s land surface, causing fluctuations in landscape connectivity and leading to a decrease in biodiversity [19,20], as well as negatively affecting the ecological, genotypic, specific, and functional characteristics of landscape units [21–23].

Tapir species in Colombia are distributed over a 631.068 km² approximately, representing 55% of the country’s continental surface [24]. The distribution of tapirs shows a remarkable altitudinal variation, ranging from sea level to 4.300 m elevation. This remarkable altitudinal range of tapir habitats allows to establish to the presence of these ungulate species in more than 50 types of natural ecosystems of the five geographical regions of Colombia. Twenty-two percent (139.008 km²) of the tapir’s range is within 26 of the country’s 58 national parks [24]. Furthermore, tapir species have been reported to require large habitat areas, with home ranges ranging from 2 km² to 40 km², and dispersal abilities greater than 10 km. The gestation period of tapirs varies between 11 and 13 months, and they typically have only one offspring per reproductive event, with parental care lasting up to 2 years [25–27]. These ecological characteristics make tapirs particularly vulnerable to habitat degradation and alteration caused by human activities, which could lead to local extinctions. In this context, both the Colombian National List of Threatened Species and the International Union for Conservation of Nature (IUCN) place tapir species in the following threat categories Tapirus bairdii and T. pinchaque in the Endangered (EN) category, and T. terrestris in the Vulnerable (VU) category [24]. Given the ecological requirements described above, tapirs are excellent biological models for assessing connectivity between protected areas. The broad ecological requirements of these species allow the identification of key areas for intervention with participatory restoration strategies. These ecological restoration actions could contribute to improving the connectivity of landscapes transformed by deforestation, such as the Pacific, Inter-Andean Valleys, and the Amazon ecoregions in Colombia.

The application of minimum cost spatial analysis and circuit theory to identify dispersal corridors of large mammals under threat, such as tapir species, are methodological approaches that could generate relevant scientific information that would contribute to the protection and improvement of connectivity between protected areas and the conservation of the biodiversity contained therein. This is even more important in a country like Colombia, which has the fourth-highest mammal diversity in the world and the third-highest in the Neotropical region, with 551 species [28]. Additionally, ensuring the movement of tapir species over a large area of the country contributes to enhancing adaptive and mitigation capacity for the adverse effects of global change [29]. In this paper, we developed connectivity models for the three tapir species reported in Colombia; these spatial analyses allowed us to identify areas within the landscape matrix that may be critical for conserving and restoring connectivity between national protected areas. Tapir species occupy predominantly forested areas but have different home range sizes, dispersal abilities, and altitudinal ranges. We modeled least-cost corridors for each species based on expert criteria, species distribution models, and climate and land cover information. We then used circuit theory models to identify bottlenecks as priority conservation areas within the corridors and a least-cost approach to identify forest restoration opportunities that could improve connectivity between protected areas. For each species, we produced maps highlighting potential distribution patterns, suitable areas, least-cost corridors, and priority conservation areas. Finally, we estimated the potential connectivity gain from forest restoration using the equivalent connected area ECA index.

Materials and methods

Study area

Colombia is located at the northwestern tip of the South American subcontinent and covers an area of  ~  1.1 million km² [30]. The country’s high spatial heterogeneity, represented by different biogeographic elements that integrate the Caribbean, Pacific, Andes, Orinoquia, Guiana, and Amazon ecoregions, contributes to the generation of speciation and endemism processes, represented by more than 56.343 species, representing 10% of the world’s diversity, placing Colombia among the 14 most megadiverse countries on the planet [31]. Colombia’s natural ecosystems cover approximately 50% to 52% of the country’s territory, ranging from the Amazon and Pacific rainforest to the high Andean [32]. The main drivers of the degradation of the natural ecosystems in Colombia over the last decades have been deforestation for established activities such as extensive cattle ranching, agro-industrial expansion, timber extraction, illegal crops, and gold mining [12,33]. The main strategy to protect natural ecosystems from environmental degradation caused by the drivers of transformation is the national system of protected areas (SINAP). This system includes 1.443 terrestrial protected areas in 15 different categories, ranging from strict protection (e.g., national parks) to private conservation areas with sustainable management practices (e.g., civil society natural reserves, see Fig 1 [6]).

[Figure omitted. See PDF.]

6]. In addition, the different types of land cover classes for 2018 [32] and departmental boundaries are shown in different colors.

Focal tapir species

The species of the genus Tapirus reported for Colombia, Tapirus bairdii (Baird’s tapir), T. pinchaque (mountain tapir), and T. terrestris (lowland tapir), are considered excellent biological models for identifying connectivity priorities within the Colombian protected area system (S1 Table). The main ecological characteristics of tapir species include (i) medium to high values of dispersal capacity [25,34–36], (ii) habitat use associated with forest land cover [35–37], (iii) tapir species are considered landscape species for high Andean forests, Orinoquia savannas and Amazonrain forests, (iv) conservation values of some protected areas [29,38], (v) tapir species in Colombia have conservation strategies at the national level [39,40], and (vi) management plans under the jurisdiction of regional environmental authorities [41–43].

Species distribution models

We developed species distribution models (SDMs) for the focal tapir species to identify their potential habitat in Colombia. First, we obtained geolocated occurrences across the full distribution of each species from the Colombian Biodiversity Information System platform (https://biodiversidad.co/), scientific articles, and national and international biological collections. A total of 3.709 records were consolidated, and distributed as follows Baird’s tapir n = 78 (2%); mountain tapir n = 481 (13%), and lowland tapir n = 3.150 (85%). Additionally, we used the R package “CoordinateCleaner” to exclude occurrence records with common spatial errors and to reduce the effect of sampling bias, occurrences were spatially reduced using the R package spThin [2,44,45] so that there were no records within 1 km of each other.

To construct the SDMs of the tapir species in Colombia, we initially considered 15 predictors before performing the variance inflation factor (VIF) analyses. These predictors have been previously used by Norris [46], Ortega-Andrade et al. [47], Cordeiro et al. [48], and Mosquera-Guerra et al. [29]. Among the predictors considered is listed: Climatic predictors were (i) annual mean temperature (BIO 01), (ii) mean diurnal range (BIO O2), (iii) isothermality (BIO 03), (iv) temperature seasonality (BIO 04), (v) temperature annual range (BIO 07), (vi) annual precipitation (BIO 12), (vii) precipitation of driest month (BIO 14), (viii) precipitation of driest quarter (BIO 17), and (ix) precipitation of coldest quarter (BIO 19). Habitat quality predictors were (i) average rainy season NDVI of the last five years, (ii) average rainy season NDVI of the last 10 years, (iii) average dry season NDVI of the last five years, and (iv) average dry season NDVI of the last 10 years, (v) elevation, and (vi) global human modification dataset (gHM). Collinearity between variables was assessed using the variance inflation factor (VIF) analysis in the open-source software R, version 4.0.3 [49–51].

The spatial resolution of the bioclimatic predictors was 500 m, and they were obtained from the WorldClim dataset [52], habitat quality predictors [53–56], and the global human modification dataset, which provides a cumulative measure of human land modification worldwide at 1 km² resolution [57]. The “gHM” values range from 0 to 1 and were calculated by estimating the proportion of a given location (pixel) that is modified, and the estimated intensity of modification associated with a given type of human modification or “stressor”. The main anthropogenic activities that transform tapir habitats in Colombia are (i) human settlements, (ii) agriculture, (iii) transportation, (iv) mining and energy production, and (v) electrical infrastructures (i) human settlement, (ii) agriculture, (iii) transportation, (iv) mining and energy production, and (v) electrical infrastructure.

We modeled the potential habitat for tapir species using MaxEnt software v.3.4.4. implemented in the “dismo” R package with its predefined parameters [58–60]. Models were run with 10.000 randomly sampled background points and based on the number of tapir species occurrences, we partitioned the data using the “checkerboard1” (>30 records) and “checkerboard2” (>100 records) methods [61]. In addition, we tested a Target-Group Background selection (TGB) approach to account for sampling bias in the models [62]. We evaluated the optimal parameters of the resulting models based on the lowest mean test omission rate (10th percentile), followed by the highest mean test AUC [2,63].

The final predictions for tapir species were converted into binary maps of suitable and unsuitable habitats using the 10th percentile presence threshold for training. Model selection between random background points and TGB approaches was performed by FMG by visual comparison with field knowledge of tapir species distributions in Colombia [16–20]. We used the 2018 national land cover map [32] to reclassify cells with less than 50% natural forest cover in 1 km² as unsuitable. Additionally, sensitivity analyses were performed to evaluate how different thresholds affect the delimitation of habitat and non-habitat in the Maxent model results for each species. A list of thresholds was created: (i) T. bairdii (maximum value: 0.015, mean value: 0.615, and increment: 0.05), (ii) T. pinchaque (maximum value: 0.14, mean value: 0.74, and increment: 0.05), and (iii) T. terrestris (maximum value: 0.2, mean value: 0.8, and increment: 0.05). For each threshold value, the MaxEnt model raster was converted into a binary prediction map (habitat/ non-habitat). In the process of evaluating the impact of thresholds on the delimitation of habitat and non-habitat of tapir species, we used the spatial metrics (i) available habitat area (calculating the total area classified as habitat), and (ii) concordance with occurrence points (verifying how many occurrence points are within the delimited habitat).

Protected areas and resistance surfaces

With the potential suitable distributions of habitat and home range estimates, we selected the protected areas (PAs) to include in the connectivity analysis for each tapir species. To avoid mapping errors near range limits, we buffered species distributions to achieve a  ~  20% increase in their area [2,26,64]. Then, we selected the PAs having a suitable area greater than or equal to the average home range size reported in the scientific literature for tapir species (Table 1). We considered all PAs within the national system of protected areas (SINAP), including national, subnational, public, and private PAs (Fig 1).

[Figure omitted. See PDF.]

We then constructed resistance surfaces to represent the degree to which different land cover impedes or facilitates the movement of each species throughout its distribution [2,69,70]. We consulted expert mastozoologists to estimate the resistance of different land covers. Each expert estimated the difficulty of crossing at least 1 km across the different land cover classes. Land cover classes were obtained from the Colombian land cover map for 2018 (300 m resolution [2,32]), and generalized into the following categories: forest, scrub, water, degraded, barren, agriculture, and settlement (Fig 1). In addition, roads, population density, and forest fires were considered (S2 Table). Experts ranked each class using a Likert scale ranging from absolute resistance (100) to no resistance to movement (0) [2]. Finally, we used the expert responses to create resistance surfaces for each of the tapir species [71].

Connectivity modeling

Spatial least-cost analyses and circuit theory are the most commonly used methodological approaches to identify species dispersal corridors [72]. These analyses incorporate variables such as the energetic costs and mortality risks of individuals moving across different landscape attributes and vary according to the ecological requirements of species and their dispersal capacity [69,73]. Minimum cost models use resistance surfaces to estimate the shortest and least costly dispersal routes between habitat patches [73]. Circuit theory, on the other hand, identifies all possible paths between patches, allowing the detection of dispersal routes of species with the least resistance to movement under the assumption of random walking behavior [74]. We used the polygons of the PAs where each tapir species occurs in Colombia and the resistance surfaces as inputs in Linkage Mapper [2,75], and ArcGIS toolbox for performing connectivity analysis [76]. Firstly, we used the Linkage Pathways tool to map the least cost corridors (LCCs) among PAs for each tapir species. Linkage pathways produce rasters of cost-weighted distance (CWD) values between pairs of neighboring PAs [2], highlighting the path with the lowest cumulative movement resistance or least cost path (LCP) [76–78]. It then normalizes and mosaics the results from each pair of PAs, creating a single composite showing all LCPs and LCCs [2], which are wider swathes of pixels with slightly higher resistance to movement than the optimal path [77–79]. We ran the tool by removing LCPs intersecting intermediate PAs and allowing a maximum of four connected nearest neighbors. We then used Linkage Mapper’s Pinchpoint Mapper tool to identify priority conservation sites within the identified corridors (Figs 2D–4D). Pinchpoint Mapper applies circuit theory by running Circuitscape [74] within LCCs to produce current flow maps showing the net probabilities of passage through each cell as tapirs move between PAs. Therefore, pinch points are locations with high current flow concentrations indicating a greater movement probability or the lack of alternative paths to move between PAs [74]. Loss of suitable habitat at pinch points could affect connectivity significantly [74,80]. To run the tool, we tested 1, 5, and 10 km as corridors’ width limits in the CWD units, finally selecting the 10 km width.

[Figure omitted. See PDF.]

(A) Species distribution model (B) Binarization process: suitable and unsuitable habitat for the species (C) Least-cost corridors (LCCs) for the species between protected areas (PAs), and (D) Priority movement sites within corridors for the species connecting PAs.

[Figure omitted. See PDF.]

(A) Species distribution model (B) Binarization process: suitable and unsuitable habitat for the species (C) Least-cost corridors (LCCs) for the species between protected areas (PAs), and (D) Priority movement sites within corridors for the species connecting PAs.

[Figure omitted. See PDF.]

(A) Species distribution model (B) Binarization process: suitable and unsuitable habitat for the species (C) Least-cost corridors (LCCs) for the species between protected areas (PAs), and (D) Priority movement sites within corridors for the species connecting PAs.

We estimated the potential increase in connectivity among PAs by restoring priority sites using the equivalent connected area ECA index, which integrates intra–patch and inter-patch connectivity [81,82]. The ECA index is defined as the area of a single continuous patch providing the same probability of connectivity value as the actual landscape pattern within an area of interest [81]. Using the Makurhini R package, we calculated the percentage change in ECA (dECA) and habitat area (dA) under two scenarios for the three tapir species reported for Colombia (Figs 5–7; [83]). The current scenario included the natural forest cover reported for Colombia in 2018 [33], while the second added forest cover derived from restoration.

[Figure omitted. See PDF.]

[Figure omitted. See PDF.]

[Figure omitted. See PDF.]

Results

Species distribution model of the tapir species

We generated a potential distribution model for the tapir species using the MaxEnt algorithm (Figs 2A–4A). Variance inflation factor analysis (VIF) allowed the exclusion of the following highly correlated climatic variables from the initial set considered (i) annual mean temperature (BIO 01), (ii) mean diurnal range (BIO O2), (iii) annual precipitation (BIO 12), (iv) precipitation of driest month (BIO 14), (v) average rainy season NDVI of the last five years, (vi) average rainy season NDVI of the last 10 years, and (vii) average dry season NDVI of the last five years. The tapir species models integrated 12 predictors five were climatic variables, four were habitat quality variables, and three were of human origin (S3, S5, and S7 Tables). The Baird’s tapir model showed a high accuracy, with AUC value of 0.94 (S4 Fig), the mountain tapir model, with AUC value of 0.98 (S6 Fig), and the lowland tapir model, with AUC value of 0.88 (S8 Fig). The calculated metrics were plotted to identify trends and sensitivity of changes in the thresholds of the models generated for tapir species in Colombia (S9A–F Fig), and the binarization process converted the continuous gradient maps into binary distributions of suitable and unsuitable classes for tapir species in Colombia (Figs 2B–4B).

Corridors among PAs

The expert mastozoologists agreed that forest land covers had the lowest resistance values for all tapir species, while transformed and artificial land covers had the highest resistance to movement. The extent of the LCC for Baird’s tapir included 41 PAs covering an area of 64.393 km² with 29.105 km² of forest area, representing 45% of the corridor extent (Fig 2C), the LCC mountain tapir included 183 PAs covering an area of 71.183 km², of which 32.613 km² correspond to forest land covers, equivalent to 46% of the corridor extension proposed for the species (Fig 3C). Finally, the LCC lowland tapir includes 238 PAs covering an area of 416.683 km² with 149.435 km² of forest land covers, equivalent to 36% of the corridor extent (Fig 4C). The extent of land cover converted to cropland exceeds the forested areas of the LCC in the Caribbean and inter-Andean valleys regions for the lowland tapir. Conversely, the Amazon, Andean, and Pacific regions show larger extensions of forest areas within the LCCs for tapir species.

Connectivity models

Based on the pinch-point analysis for species of the genus Tapirus in Colombia, we prioritized 14,981 km² for Baird’s tapir and identified critical areas for maintaining connectivity between protected areas (PAs). These priorities consist mainly of forested areas (9.288 km²) and are distributed in the Caribbean (5.320 km²; 57%), the Chocó (2.370 km²; 25%), and the western foothills of the Andes (1.596 km²;17%) regions. Ecologically important areas for the T. bairdii individuals’ movement were identified such as the natural corridors between the Nudo de los Paramillos and Katios national natural parks (NNPs), as well as the regional integrated management districts (IMRD), the Ayapel wetland complex, and the western Antioqueño dry forest (Fig 2D).

In the case of the mountain tapir the pinch-point analysis prioritized 21.994 km² that were identified as critical areas for maintaining connectivity between PAs, corresponding to forested areas (16.245 km²) distributed mainly in the Central Andes (12.098 km²; 74.5%), the Eastern Andes (3.122 km²; 19.2%), and the Colombian massif and Andean-Amazonian regions (1.025 km²; 6.3%). Some areas of ecological importance were identified for maintaining the dispersal movements of T. pinchaque individuals such as the natural corridors between the Cocha lagoon-Cerro Patascoy national protected forest reserve (NPFR), and the Doña Juana Cascabel, Puracé, Nevado del Huila, the Hermosas, and Nevados PNNs, located in the Colombian massif and the Central Andes, and the corridor between the Serranía de los Churubelos, the Alto Fragua Indi Wasi PNNs, and the regional natural park (RNP) Miraflores-Picachos in the Andean-Amazonian region of the Eastern Andes (Fig 3D).

Finally, for the lowland tapir the pinch-point analysis identifies 167.903 km² considered as critical areas for maintaining connectivity between PAs and corresponds to forested areas (149.617 km²) distributed in the Amazonia (99.174; 66%), Orinoquia (35.952; 24%), inter-Andean valleys (8.798 km²; 6%), and Caribbean (5.691 km²; 4%) regions. We identified areas of ecological importance for the movement of the T. terrestris individuals, such as the corridors between the Sierra Nevada de Santa Marta NNP, Cenagoso of Zapatosa complex IMRD, and Catatumbo – Bari NNP in the Caribbean region, as well as the Serranía de los Yarigüíes IMRD, the Serranía de los Yarigüíes NNP, Serranía de las Quinchas RNP, and dry forest on the eastern slope of the Magdalena River IMRD in the inter-Andean valleys. For the region of the Colombian Orinoquia, the regional corridor made up of El Tuparro, Serranía de Manacacías, and Serranía de la Macarena NNPs is connected through 94 NRCS and the Cinaruco national integrated management district (NMID). Finally, for the Amazon region, we identified three regional natural corridors that were integrated by Tinugua, Picachos mountain range, Alto Fragua Indi-Wasi, and Serranía de los Churumbelos-Auka Wasi PAS in the Andean Amazonian biogeographic region; Serranía del Chiribiquete, the Paya NNP, and the Nukak and Puinawai natural reserves (NRs) in the transition between the Guyana shield and the Amazon rainforest, and Yaigojé-Apaporis, Puré River, Cahunarí, and Amacayacú NNPs in the Amazon floodplain biogeographic region (Fig 4D).

Restoration priorities

We identified 29.728 km² with the potential to improve connectivity between multiple PAs by restoring degraded areas to forests in areas of tapir species occurrence in Colombia (Figs 5–7). These areas comprise mainly non-native grasslands and shrublands (20.080 km²; 67.5%) and croplands (9.648 km²; 32.5%). Restoration priorities for areas connecting Baird’s tapir habitats between PAs are concentrated in 5.693 km² (19.1%) of which non-native grasslands and shrublands are 964 km² (17%), and cropland is 4.730.5 km² (83%). Degraded habitats for the T. bairdii due to cattle ranching and banana (Musa spp.) plantations are located in the buffer areas of the PAs in the northwest Pacific, Caribbean, and Western Andes foothills. The ECA analysis indicates that restoration of the identified priorities could increase forest cover by 0.19% and functional connectivity by 0.1% (Fig 5).

In the case of the mountain tapir, the restoration priorities for areas connecting the species’ habitats between PAs are concentrated on 5.749 km² (19.3%), of which non-native grasslands and shrublands cover 5.070 km² (88%) and crops 681 km² (12%). Degraded habitats for the T. pinchaque due to cattle ranching and cultivation of vegetables (Solanum spp., Allium cepa, A. fistulousm, and Daucus carota), fruits (Fragaria ananassa, Passiflora ligularis, Solanum betaceum, and S. quitoense), and coffee (Coffea spp.) crops are located in the buffer areas of the PAs in the Colombian massif, the Andean-Amazon region, and the Central and Eastern Andes. The ECA analysis indicates that restoration of the identified priorities could increase forest cover by 37% and functional connectivity by 6.8% (Fig 6).

Finally, for the lowland tapir, the restoration priorities for areas connecting the species’ habitats between PAs are concentrated on 18.286 km² (61%), of which non-native grasslands and shrublands cover 14.049 km² (77%) and crops 4.283 km² (23%). Degraded habitats for T. terrestris due to cattle ranching and agro-industrial (sugar cane (Saccharum officinarum), rice (Oryza sativa), oil palm (Elaeis guineensis), and forestry (Acacia magium, Eucalyptus spp., and Pinus spp.), and illicit (Erythroxylum coca and Cannabis spp.) crops are located in the buffer areas of the PAs in the Caribbean, Inter-Andean valleys, Orinoquia, and Amazonia. The ECA analysis indicates that restoration of the identified priorities could increase forest cover by 10% and functional connectivity by 0.08% (Fig 7).

Discussion

Species distribution model of the tapir species

The environmental predictors used in the distribution models for the tapir species generated here are related to climatic, habitat quality, and human footprint variables. These predictors coincide with the variables used in the distribution models reported for the tapir species in the Neotropical region [e.g., 29,46–48]. These results confirm the strong dependence of tapir species on forest cover, the influence of precipitation and temperature associated with the altitudinal ranges where these species occur in Colombia, as well as the impact of deforestation, forest fires and the construction of infrastructure that fragment habitats and reduce the dispersal capacity of individuals and the levels of functional connectivity of populations [20,29].

Corridors and connectivity models

The connectivity models proposed here between protected areas (PAs) in Colombia identified relevant conservation areas to ensure good levels of functional connectivity for tapir species populations in the current scenario. These areas were located in the Andean, Caribbean, Guiana, and Amazonian regions, and are mostly coincident with other proposals that evaluated the functional connectivity of PAs at large scales, using different methodological approaches [e.g., 84,86] and with methods similar to those used in this manuscript [e.g., 2,20]. The contributions of these studies to the knowledge of the functional connectivity of PAs generated from ungulates such as T. bairdii, T. pinchaque, T. terrestris, Odocoileus virginianus, Tayassu pecari, and Dicotyles tajacu, and carnivores such as Panthera onca, Puma concolor, Tremarctos ornatus should be highlighted [e.g., 20,85,87]. These results highlight the need to identify appropriate criteria for selecting the best biological models (e.g., large mammals) that will allow us to understand and evaluate the functional connectivity between the PAs, given the increase in the negative effects of global change on biodiversity. Moreover, it’s a priority to generate and consolidate existing scientific information on spatial ecology (e.g., home ranges and core areas) and movement (e.g., dispersal capacity), as well as landscape genetics for species considered to be of conservation value in the PAs (e.g., threatened large mammals, including tapirs, and other ungulate and carnivore species [20,88]). Providing scientific information to fill ecological information these gaps for the species prioritized will allow for the development of much more robust and reliable connectivity models [89,90]. This condition is currently the main limitation in conducting connectivity analyses involving large mammal species in the Neotropical region [20]. Satellite telemetry efforts for species of the genus Tapirus in Colombia have focused mainly on the mountain tapir in the Central Andes region [66] and to a lesser extent on the lowland tapir in the Amazon River basin and Bita River basin in Colombian Orinoquia [20,37], and have been absent for Baird’s tapir in the Pacific and northwestern Caribbean. Despite these shortcomings and the limitations of spatial information on tapir species in Colombia, we use the best information available in the Neotropical region [20]. In this respect, we believe that our study identifies the main functional forest corridors for tapirs in the country, considering the current data limitations for the tapir species, mainly for the Baird tapir. In this context, it’s a priority to consider the scientific knowledge generated on the spatial ecology and movements of tapir species as an essential criterion for the management and administration of the 462 PAs destined exclusively for biodiversity conservation. The future of viable tapir populations in Colombia and the ecosystem services provided by these species depend to a large extent on a well-managed national system of protected areas connected and adapted to the negative effects of global change.

Restoration opportunities

The priority areas to be intervened with ecological restoration strategies identified here coincide with those proposed to be restored by Linero-Triana [2]. The implementation of these restoration actions could contribute to reducing the dispersal costs of tapir species and could in turn contribute to improving the levels of connectivity between the PAs at different scales. In this sense, we recommend the implementation of the objectives of the national restoration plan in aspects such as (i) recovering biodiversity, (ii) mitigating climate change, (iii) strengthening community resilience, (iv) boosting the green economy, (v) strengthening the resilience of communities, (vi) promoting the development of a green economy, and (vii) prioritizing the recovery of ecosystems at high risk of degradation in Colombia [7,8,91].

Other conservation strategies that consider the importance of implementing restoration actions in suitable habitats for tapir species in Colombia are: (i) National Program for the Conservation of the Tapirus genus in Colombia [39], (ii) the management plans for the mountain tapir [43], (iii) the management plan for the lowland tapir in the Orinoquia [41], and (iv) the conservation plan of the subspecies T. terrestris colombianus in the Sierra Nevada de Santa Marta [92]. Additionally, the national environmental system - SINA promotes the conservation of habitats and populations of threatened large mammals through the designation of ecologically key areas for these species as other effective area-based conservation measures - OMEC [93]. In the OMECs, the development of productive activities is allowed under environmental sustainability schemes, while at the same time areas of ecological importance for the occurrence of these species are protected under restoration strategies. An example of this condition occurs in the department of Casanare, located in the Orinoquia region, where private conservation initiatives through civil society nature reserves and OMECs contribute greatly to the functional connectivity of the flooded savannas of the Colombian Llanos in the absence of state protected areas in these landscapes [94]. This same condition is present to a lesser extent in the central Andes and Amazonian foothills.

Conservation implications

In the last decade, the growth in the extension and representativeness of national system of protected areas in the Colombian continental territory has been an international example, with notable efforts such as the expansion of the Serranía del Chiribiquete NNP to 42.681 km², making it the largest protected area in the country and the largest biodiversity reserve in the Colombian Amazon [95]. In addition to this initiative, the Serranía del Manacacías NNP, located in the high plains native savanna ecosystems of the Orinoquia, was recently designated as a protected area [6], in addition to the active participation of the government authorities in the nomination processes of several areas throughout the country under the other effective area-based conservation measures (OMEC) [93]. These conservation efforts by the Colombian government are largely motivated by international commitments to meet the Aichi targets of the strategic plan for biodiversity 2011–2020, and the CBD-Kunming-Montreal global biodiversity framework [4]. However, despite this progress, challenges remain in terms of management effectiveness, funding, and the involvement of local communities in PAs conservation processes to reduce biodiversity loss in these areas in these areas, especially large mammals under threat, such as tapir species.

Currently, protected areas and their buffer zones in Colombia face several threats to conserving the biodiversity contained within them. The main cause is deforestation, which makes these areas increasingly degraded and vulnerable to the negative effects of extreme climatic phenomena (e.g., altered precipitation patterns, extreme droughts, and increased wildfires) [29]. These threats could affect endangered species and the ecological integrity of these areas [2], which are intended for the conservation of Colombia’s environmental heritage. It is noteworthy that in the last four decades, an increase in the rate of transformation of natural ecosystems by deforestation for land-use change, forest fires, and biological invasions of exotic species has been reported in the country, reaching ~ 49% of the Colombian national continental territory [7,8].

The drivers of the transformation of natural ecosystems in Colombia are causing the degradation of suitable habitats for many species and therefore the reduction in habitat connectivity levels for many populations of large mammals, including a large number of endangered species, such as tapirs as reported in the Red List of the International Union for Conservation of Nature (IUCN) [96], and the National List of Threatened Species [97]. The loss of ecosystem connectivity caused by the degradation of natural environments and faunal processes leads to a reduction in the supply of ecosystem services, such as the regulation of the hydrological cycle, the fixation of atmospheric carbon, and the dilution of zoonotic loads provided by species [20,29,98,99]. In the current context of global change, tapir species are essential for the construction of ecosystem-based adaptation strategies to the negative effects of climate variability [20].

We highlight the implications of extreme meteorological events on PAs connectivity. These climatic events cause an increase in the average temperature and in turn reduce the percentage of humidity in the vegetation, generating ideal conditions for the accumulation of a greater fuel load and causing an alteration in the natural fire regime that materializes in an increase in the intensity, frequency, and extent of wildfires in the PAs of the Orinoquia, Amazon, Caribbean, and Andes regions [13]. The PAs that report the highest frequency of wildfires in Colombia are El Tuparro NNP and Cinaruco IMND in the Orinoquia region; Tinigua and Serranía del Chiribiquete NNPs in the Amazon region, Sierra Nevada de Santa Marta NNP in the Caribbean region and, to a lesser extent, Los Nevados NNP and Iguaque fauna and flora sanctuary (FFS) in the Colombian central and eastern Andes [13,29].

Connectivity between protected areas (PAs) where tapir species occur in Colombia is essential to guarantee the long-term viability of these ungulate populations. This study allowed the identification of low-cost biological corridors, pinch points, and potential areas to restore connectivity between PAs. However, there are significant challenges to maintaining adequate levels of connectivity, related to habitat fragmentation due to the high rate of deforestation, the development of road infrastructure, the impacts of climate variability and wildfires, in addition to the difficulties in implementing effective policies for the conservation of habitats and threatened species in the country. In this context, it is essential to adopt measures that promote functional connectivity between PAs that host these large threatened mammals, through the creation of biological corridors, ecosystem restoration, and mitigation of anthropogenic threats. The implementation of actions based on scientific research is crucial to ensure the conservation of ecosystem services provided by biodiversity in PAs and to strengthen the adaptive capacity of these areas in the face of global change.

Supporting information

S1 Table. Information on the ecology, behavior, and movements of tapir species reported in the scientific literature.

https://doi.org/10.1371/journal.pone.0323175.s001

(DOCX)

S2 Table. Resistance values used for this study.

https://doi.org/10.1371/journal.pone.0323175.s002

(DOCX)

S3 Table. Percent contribution and permutation importance of environmental and human-derived predictors used in the construction of the T. bairdii distribution model.

https://doi.org/10.1371/journal.pone.0323175.s003

(DOCX)

S4 Fig. SMD response curves of the T. bairdii distribution model.

https://doi.org/10.1371/journal.pone.0323175.s004

(DOCX)

S5 Table. Percent contribution and permutation importance of environmental and human-derived predictors used in the construction of the T. pinchaque distribution model.

https://doi.org/10.1371/journal.pone.0323175.s005

(DOCX)

S6 Fig. SMD response curves of the T. pinchaque distribution model.

https://doi.org/10.1371/journal.pone.0323175.s006

(DOCX)

S7 Table. Percent contribution and permutation importance of environmental and human-derived predictors used in the construction of the T. terrestris distribution model.

https://doi.org/10.1371/journal.pone.0323175.s007

(DOCX)

S8 Fig. SMD response curves of the T. terrestris distribution model.

https://doi.org/10.1371/journal.pone.0323175.s008

(DOCX)

S9 Fig. (A-F) Plots of the metrics are calculated to identify trends and sensitivity to changes in the threshold.

(A, B) Baird’s tapir, (C, D) Mountain tapir, and (E, F) Lowland tapir.

https://doi.org/10.1371/journal.pone.0323175.s009

(DOCX)

S1 File. The spatial analyses performed for this manuscript were uploaded to the open access platform GitHub.

This information can be found at the following link.

https://doi.org/10.1371/journal.pone.0323175.s010

(DOCX)

Acknowledgments

We are grateful to the local communities and government authorities of the department of Caquetá, as well as to the Centro de Investigación de la Biodiversidad Andino Amazónica (INBIANAM), the Asociación Colombiana de Zoología (ACZ), the Fundación Puerto Rastrojo, the Fundación Naturaleza y Energía, the Instituto Amazónico de Investigaciones Científicas (SINCHI), and the Instituto de Investigaciones de Recursos Biológicos Alexander von Humboldt (IAvH), mammal reference collection of the IAvH – M, and the Universidad de la Amazonia (Uniamazonia) for providing initial raw information on the occurrence data used to construct the focal mammal species models in the Andean Amazonian and Amazon floodplain landscapes. This research was supported by the Ministry of Science, Technology and Innovation of Colombia through the Postdoctoral Fellowships Orientada por Misiones 2023 and by the Universidad de la Amazonia through the project: “Diseño participativo de paisajes multifuncionales, ambientalmente sostenibles y resilientes al cambio climático en el piedemonte Andino-Amazónico, departamento de Caquetá, BPUA - 2024-2-0003”, and the Fundación Naturaleza y Energía for supporting the implementation of conservation actions with local communities. Finally, we are grateful to the Universidad Nacional de Colombia (UNAL), and the Sistema Nacional de Regalías (SNR), through the of the project “Diseño participativo de estrategias para la reducción de incendios forestales, la conservación de la biodiversidad y el desarrollo regional en paisajes multifuncionales de Vichada (BPIN 2020000100456)”.

References

1. 1. Brennan A, Naidoo R, Greenstreet L, Mehrabi Z, Ramankutty N, Kremen C. Functional connectivity of the world’s protected areas. Science. 2022;376(6597):1101–4. pmid:35653461

* View Article

* PubMed/NCBI

* Google Scholar

2. 2. Linero-Triana D, Correa-Ayram CA, Velásquez-Tibatá J. Prioritizing ecological connectivity among protected areas in Colombia using a functional approach for birds. Global Ecology and Conservation. 2023;48:e02713.

* View Article

* Google Scholar

3. 3. Schmidt C, Domaratzki M, Kinnunen RP, Bowman J, Garroway CJ. Continent-wide effects of urbanization on bird and mammal genetic diversity. Proc Biol Sci. 2020;287(1920):20192497. pmid:32019443

* View Article

* PubMed/NCBI

* Google Scholar

4. 4. CDB. First draft of the post-2020 global biodiversity framework. 2021. https://www.cbd.int/doc/c/abb5/591f/2e46096d3f0330b08ce87a45/wg2020-03-03-en.pdf

* View Article

* Google Scholar

5. 5. SiB. Sistema de Información sobre Biodiversidad de Colombia [Internet]. 2024 [cited 2024 Nov 6. ]. https://biodiversidad.co/

* View Article

* Google Scholar

6. 6. PNN. Registro Único Nacional de Áreas Protegidas [Internet]. 2024 [cited 2024 Nov 6. ]. https://runap.parquesnacionales.gov.co/cifras

* View Article

* Google Scholar

7. 7. Etter A, Andrade A, Saavedra-Ramírez K, Arévalo P, Amaya-Valderrama P. Risk assessment of Colombian continental ecosystems: An application of the Red List of Ecosystems methodology (v. 2.0). Pontificia Universidad Javeriana, Conservación Internacional Colombia; 2017.

8. 8. Etter A, Andrade A, Nelson CR, Cortés J, Saavedra K. Assessing restoration priorities for high-risk ecosystems: An application of the IUCN Red List of Ecosystems. Land Use Policy. 2020;99:104874.

* View Article

* Google Scholar

9. 9. Etter A, Andrade A, Saavedra K, Amaya P, Cortés J, Arévalo P. Ecosistemas colombianos: una aplicación de la lista roja de ecosistemas a los ecosistemas terrestres continentales. Bogotá D.C., Colombia: Centro Editorial Javeriano; 2020a.

10. 10. Murillo-Sandoval PJ, Clerici N, Correa-Ayram C. Rapid loss in landscape connectivity after the peace agreement in the Andes-Amazon region. Global Ecology and Conservation. 2022;38:e02205.

* View Article

* Google Scholar

11. 11. Murillo-Sandoval PJ, Kilbride J, Tellman E, Wrathall D, Van Den Hoek J, Kennedy RE. The post-conflict expansion of coca farming and illicit cattle ranching in Colombia. Sci Rep. 2023;13(1):1965. pmid:36737650

* View Article

* PubMed/NCBI

* Google Scholar

12. 12. Armenteras D, Cabrera E, Rodríguez N, Retana J. National and regional determinants of tropical deforestation in Colombia. Reg Environ Change. 2013;13(6):1181–93.

* View Article

* Google Scholar

13. 13. Armenteras D. Cambios en los patrones espaciales de área quemada en Colombia, ¿qué ha pasado en las dos primeras décadas del siglo XXI? Rev Acad Colomb Cienc Ex Fis Nat. 2022.

* View Article

* Google Scholar

14. 14. González TM, González-Trujillo JD, Muñoz A, Armenteras D. Differential effects of fire on the occupancy of small mammals in neotropical savanna-gallery forests. Perspectives in Ecology and Conservation. 2021;19(2):179–88.

* View Article

* Google Scholar

15. 15. Mosquera-Guerra F, Moreno-Niño N, Garcia-Suabita W, Armenteras-Pascual D. Impact of wildfires on the suitability habitat of the lowland tapir in a native savanna landscape of northern South America. Discov Conserv. 2024;1(1).

* View Article

* Google Scholar

16. 16. García M, Jordan C, O’Farril G, Poot C, Meyer N, Estrada N, et al. Tapirus bairdii. The IUCN Red List of Threatened Species 2016: e.T21471A45173340. 2024. https://dx.doi.org/10.2305/IUCN.UK.2016-1.RLTS.T21471A45173340.en

* View Article

* Google Scholar

17. 17. Lizcano DJ, Amanzo J, Castellanos A, Tapia A, Lopez-Malaga CM. Tapirus pinchaque. The IUCN Red List of Threatened Species 2016: e.T21473A45173922. 2024. https://dx.doi.org/10.2305/IUCN.UK.2016-1.RLTS.T21473A45173922.en

* View Article

* Google Scholar

18. 18. Varela D, Flesher K, Cartes JL, de Bustos S, Chalukian S, Ayala G, et al. Tapirus terrestris. The IUCN Red List of Threatened Species 2019: e.T21474A45174127. 2024. https://dx.doi.org/10.2305/IUCN.UK.2019-1.RLTS.T21474A45174127.en

* View Article

* Google Scholar

19. 19. Marco Palamara G, Rozenfeld A, de Santana CN, Klecka J, Riera R, Eguíluz VM, et al. Biodiversity dynamics in landscapes with fluctuating connectivity. Ecography. 2023;2023(7).

* View Article

* Google Scholar

20. 20. Mosquera-Guerra F, Barreto S, Moreno-Niño N, González-Delgado TM, Armenteras-Pascual D. Habitat connectivity of threatened ungulate species in a native savanna landscape of northern South America. Mamm Biol. 2024;104(3):259–75.

* View Article

* Google Scholar

21. 21. de Chazal J, Rounsevell MDA. Land-use and climate change within assessments of biodiversity change: A review. Global Environmental Change. 2009;19(2):306–15.

* View Article

* Google Scholar

22. 22. Haddad NM, Brudvig LA, Clobert J, Davies KF, Gonzalez A, Holt RD, et al. Habitat fragmentation and its lasting impact on Earth’s ecosystems. Sci Adv. 2015;1(2):e1500052. pmid:26601154

* View Article

* PubMed/NCBI

* Google Scholar

23. 23. IPBES. Global assessment report on biodiversity and ecosystem services of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services. Final report. IPBES secretariat. Bonn, Germany [Internet]. 2019. https://doi.org/10.5281/zenodo.3831673

24. 24. Mosquera-Guerra F, González-Delgado T, Mantilla-Meluk H, Jiménez-Ortega AM, Velásquez-Valencia A, Armenteras-Pascual D. Colombia País de Tapires. Florencia, Colombia: Universidad de la Amazonia; 2024b.

25. 25. Reyna-Hurtado R, Sanvicente-López M, Pérez-Flores J, Carrillo-Reyna N, Calmé S. Insights into the multiannual home range of a Baird’s tapir (Tapirus bairdii) in the Maya Forest. Therya. 2016;7(2):271–6.

* View Article

* Google Scholar

26. 26. Wade AA, McKelvey KS, Schwartz MK. Resistance-surface-based wildlife conservation connectivity modeling: Summary of efforts in the United States and guide for practitioners. Rocky Mountain Research Station: U.S. Department of Agriculture, Forest Service; 2015. https://doi.org/10.2737/rmrs-gtr-333

27. 27. Foerster CR, Vaughan C. Home Range, Habitat Use, and Activity of Baird’s Tapir in Costa Rica1. Biotropica. 2002;34(3):423–37.

* View Article

* Google Scholar

28. 28. Ramírez-Chaves HE, Leuro Robles NG, Castaño RA, Morales-Martínez DM, Suárez Castro AF, Rodríguez-Posada ME, et al. Mamíferos de Colombia. v1.14. SCMas. Dataset/Checklist. 2024.

* View Article

* Google Scholar

29. 29. Mosquera-Guerra F, Moreno-Niño N, Barreto S, Armenteras-Pascual D. Habitat suitability and representation of the potential distribution of the lowland tapir (Tapirus terrestris) within land cover types and protected areas of eastern Colombia. Mamm Res. 2024;69(4):507–18.

* View Article

* Google Scholar

30. 30. Sánchez-Cuervo AM, Aide TM, Clark ML, Etter A. Land cover change in Colombia: surprising forest recovery trends between 2001 and 2010. PLoS One. 2012;7(8):e43943. pmid:22952816

* View Article

* PubMed/NCBI

* Google Scholar

31. 31. IAvH. Biodiversidad colombiana: números para tener en cuenta; 2024.

32. 32. IDEAM. Mapa de Coberturas de la tierra Metodología Corine Land Cover Escala 1:100.000 Periodo 2018. 2024.

33. 33. Dávalos LM, Sanchez KM, Armenteras D. Deforestation and Coca Cultivation Rooted in Twentieth-Century Development Projects. BioScience. 2016;66(11):974–82.

* View Article

* Google Scholar

34. 34. Castellanos AX. Iridum/GPS telemetry to study home range and population density of mountain tapirs in the río Papallacta watershed, Ecuador. The Newsletter of the IUCN/SSC TSG. 2013;22:20–26.

* View Article

* Google Scholar

35. 35. Medici EP. Assessing the viability of lowland tapir populations in a fragmented landscape. Tesis de Doctoral. 2010.

36. 36. Medici EP, Mezzini S, Fleming CH, Calabrese JM, Noonan MJ. Movement ecology of vulnerable lowland tapirs between areas of varying human disturbance. Mov Ecol. 2022;10(1):14. pmid:35287742

* View Article

* PubMed/NCBI

* Google Scholar

37. 37. González TM, González-Trujillo JD, Palmer JRB, Pino J, Armenteras D. Movement behavior of a tropical mammal: The case of Tapirus terrestris. Ecological Modelling. 2017;360:223–9.

* View Article

* Google Scholar

38. 38. Montenegro OL, López-Arévalo H, Mora-Beltrán C, Lizcano DJ, Serrano H, Mesa E, et al. Tropical ungulates of Colombia. Gallina-Tessaro S. Ecology and conservation of tropical ungulates in Latin America. Cham, Germany: Springer; 2019. p. 159.

39. 39. Montenegro OL. Programa nacional para la conservación del género Tapirus en Colombia. Bogotá D.C., Colombia: Dirección de Ecosistemas, Ministerio de Ambiente, Vivienda y Desarrollo Territorial; 2005. p. 100.

40. 40. Parra-Romero A, García Martínez RY, Escobar-Lasso S, Lizcano DJ, Aconcha-Abril I, Vélez L, et al. Programa de Conservación de la Danta de Montaña (Tapirus pinchaque) en Parques Nacionales Naturales de Colombia. Bogotá, D.C. Colombia: Parques Nacionales Naturales de Colombia, Iniciativa de Conservación de Tapires de Colombia (CTC)/ IUCN SSC Tapir Specialist Group y Proyecto de Conservación de Aguas y Tierras – ProCAT Colombia; 2021. p. 78.

41. 41. Álviz A, Pérez AK. Plan para la conservación de la danta de tierras bajas (Tapirus terrestris) en los departamentos de Casanare, Arauca y Vichada. Yopal, Casanare: Corporinoquia-Fundación Orinoquia Biodiversa; 2015. p. 60.

42. 42. Restrepo H, Gómez E.I. Plan de Conservación para oso andino (Tremarctos ornatus) y danta de montaña (Tapirus pinchaque) en el departamento de Putumayo. Mocoa, Putumayo: WWF – Corpoamazonia; 2017. p. 54.

43. 43. Mantilla-Meluk H, Mosquera-Guerra F, Trujillo F, Lizcano D, Sandoval S, González TM, Londoño MC. Plan de manejo para la conservación de la danta de montaña (Tapirus pinchaque) en el departamento del Quindío. Armenia, Quindío: Corporación Autónoma Regional del Quindío (CRQ); 2017. p. 60.

44. 44. Zizka A, Silvestro D, Andermann T, Azevedo J, Duarte Ritter C, Edler D, et al. CoordinateCleaner: Standardized cleaning of occurrence records from biological collection databases. Methods Ecol Evol. 2019;10(5):744–51.

* View Article

* Google Scholar

45. 45. Aiello‐Lammens ME, Boria RA, Radosavljevic A, Vilela B, Anderson RP. spThin: an R package for spatial thinning of species occurrence records for use in ecological niche models. Ecography. 2015;38(5):541–5.

* View Article

* Google Scholar

46. 46. Norris D. Model Thresholds are More Important than Presence Location Type: Understanding the Distribution of Lowland tapir (Tapirus Terrestris) in a Continuous Atlantic Forest of Southeast Brazil. Tropical Conservation Science. 2014;7(3):529–47.

* View Article

* Google Scholar

47. 47. Ortega-Andrade HM, Prieto-Torres DA, Gómez-Lora I, Lizcano DJ. Ecological and geographical analysis of the distribution of the mountain tapir (Tapirus pinchaque) in Ecuador: importance of protected areas in future scenarios of global warming. PLoS One. 2015;10(3):e0121137. pmid:25798851

* View Article

* PubMed/NCBI

* Google Scholar

48. 48. Cordeiro PJL, Fragoso JMV, Crawshaw D, Oliveira LFB. Lowland tapir distribution and habitat loss in South America. PeerJ. 2016; 4:e2456.

* View Article

* Google Scholar

49. 49. Naimi B, Hamm NAS, Groen TA, Skidmore AK, Toxopeus AG. Where is positional uncertainty a problem for species distribution modelling? Ecography. 2013;37(2):191–203.

* View Article

* Google Scholar

50. 50. Akinwande MO, Dikko HG, Samson A. Variance Inflation Factor: As a Condition for the Inclusion of Suppressor Variable(s) in Regression Analysis. OJS. 2015;05(07):754–67.

* View Article

* Google Scholar

51. 51. R Development Core Team. R: a language and environment for statistical computing. R Foundation for statistical computing. Version 3.1.1. Vienna, Austria: R Foundation for Statistical Computing; 2024.

52. 52. Hijmans RJ, Cameron SE, Parra JL, Jones PG, Jarvis A. Very high resolution interpolated climate surfaces for global land areas. Int J Climatol. 2005;25(15):1965–78.

* View Article

* Google Scholar

53. 53. Farr TG, Rosen PA, Caro E, Crippen R, Duren R, Hensley S, et al. The Shuttle Radar Topography Mission. Reviews of Geophysics. 2007;45(2).

* View Article

* Google Scholar

54. 54. Didan K. MOD13A1 MODIS/Terra vegetation indices 16-day L3 global 500m SIN grid V006 [data set]. NASA EOSDIS land processes DAAC. 2024. https://lpdaac.usgs.gov/products/mod13a1v006/

* View Article

* Google Scholar

55. 55. IDEAM. Actualización de cifras de monitoreo de la superficie de bosque y la deforestación – Año 2021; 2022.

56. 56. IGAC. Cartografía base para Colombia (1:100.000). 2024.

57. 57. Kennedy CM, Oakleaf JR, Theobald DM, Baruch-Mordo S, Kiesecker J. Managing the middle: A shift in conservation priorities based on the global human modification gradient. Glob Chang Biol. 2019;25(3):811–26. pmid:30629311

* View Article

* PubMed/NCBI

* Google Scholar

58. 58. Phillips SJ, Anderson RP, Schapire RE. Maximum entropy modeling of species geographic distributions. Ecological Modelling. 2006;190(3–4):231–59.

* View Article

* Google Scholar

59. 59. Phillips SJ, Dudík M. Modeling of species distributions with Maxent: new extensions and a comprehensive evaluation. Ecography. 2008;31(2):161–75.

* View Article

* Google Scholar

60. 60. Hijmans RJ, Phillips S, Leathwick J, Elith J. dismo: Species Distribution Modeling. R package version 1. 2023;3–14. https://github.com/rspatial/dismo

* View Article

* Google Scholar

61. 61. Muscarella R, Galante PJ, Soley‐Guardia M, Boria RA, Kass JM, Uriarte M, et al. ENMeval: An R package for conducting spatially independent evaluations and estimating optimal model complexity for Maxent ecological niche models. Methods Ecol Evol. 2014;5(11):1198–205.

* View Article

* Google Scholar

62. 62. Phillips SJ, Dudík M, Elith J, Graham CH, Lehmann A, Leathwick J, et al. Sample selection bias and presence-only distribution models: implications for background and pseudo-absence data. Ecol Appl. 2009;19(1):181–97. pmid:19323182

* View Article

* PubMed/NCBI

* Google Scholar

63. 63. Shcheglovitova M, Anderson RP. Estimating optimal complexity for ecological niche models: A jackknife approach for species with small sample sizes. Ecological Modelling. 2013;269:9–17.

* View Article

* Google Scholar

64. 64. Koen EL, Bowman J, Sadowski C, Walpole AA. Landscape connectivity for wildlife: development and validation of multispecies linkage maps. Methods Ecol Evol. 2014;5(7):626–33.

* View Article

* Google Scholar

65. 65. Foerster C. Results of a five-year telemetry study of Baird’s tapir in Costa Rica. P. 9. Proceedings of the First International Tapir Symposium. San Jose, Costa Rica. 2001.

66. 66. Lizcano D, Cavelier J. Using GPS Collars to Study Mountain Tapirs (Tapirus pinchaque) in the Central Andes of Colombia. Tapir Conservation. 2004;13:18–23.

* View Article

* Google Scholar

67. 67. Downer CC. The mountain tapir, endangered ‘flagship’ species of the high Andes. Oryx. 1996;30(1):45–58.

* View Article

* Google Scholar

68. 68. Noss A, Cuellar RL, Barrientos J, Maffei L, Cuellar E, Arispe R, et al. A camera trapping and radiotelemetry study of lowland tapir (Tapirus terrestris) in Bolivian dry forest. Tapir Conserv. 2003;12:24–32.

* View Article

* Google Scholar

69. 69. Zeller KA, McGarigal K, Whiteley AR. Estimating landscape resistance to movement: a review. Landscape Ecol. 2012;27(6):777–97.

* View Article

* Google Scholar

70. 70. Taylor PD, Fahrig L, Henein K, Merriam G. Connectivity Is a Vital Element of Landscape Structure. Oikos. 1993;68(3):571.

* View Article

* Google Scholar

71. 71. Ersoy E, Jorgensen A, Warren PH. Identifying multispecies connectivity corridors and the spatial pattern of the landscape. Urban Forestry & Urban Greening. 2019;40:308–22.

* View Article

* Google Scholar

72. 72. Anantharaman R, Hall K, Shah V, Edelman A. Circuitscape in Julia: high performance connectivity modelling to support conservation decisions. ArXiv. 2019.

* View Article

* Google Scholar

73. 73. McClure ML, Hansen AJ, Inman RM. Connecting models to movements: testing connectivity model predictions against empirical migration and dispersal data. Landscape Ecol. 2016;31(7):1419–32.

* View Article

* Google Scholar

74. 74. McRae BH, Dickson BG, Keitt TH, Shah VB. Using circuit theory to model connectivity in ecology, evolution, and conservation. Ecology. 2008;89(10):2712–24. pmid:18959309

* View Article

* PubMed/NCBI

* Google Scholar

75. 75. McRae BH, Kavanagh DM. Linkage Mapper Connectivity Analysis Software [Internet]. 2011 [cited 2024 Nov 6. ]. https://linkagemapper.org/

* View Article

* Google Scholar

76. 76. ESRI. ArcGIS desk- top: release 10.8.1; 2024.

77. 77. Adriaensen F, Chardon JP, de Blust G, Swinnen E, Villalba S, Gulinck H, et al. The application of ‘least-cost’ modelling as a functional landscape model. Landsc. Urban Plan. 2003;64(4):233–247.

* View Article

* Google Scholar

78. 78. McRae B, Kavanagh D. Linkage Mapper Toolbox: Linkage Pathways Tool User Guide. The Nature Conservancy, Adze Informatics; 2021.

* View Article

* Google Scholar

79. 79. Schrott G, Shinn J. A landscape connectivity analysis for the Coastal Marten (Martes caurina humboldtensis). Arcata, CA: U.S. Fish and Wildlife Service; 2020. p. 123.

80. 80. Castilho CS, Hackbart VCS, Pivello VR, dos Santos RF. Evaluating Landscape Connectivity for Puma concolor and Panthera onca Among Atlantic Forest Protected Areas. Environ Manage. 2015;55(6):1377–89. pmid:25860593

* View Article

* PubMed/NCBI

* Google Scholar

81. 81. Saura S, Estreguil C, Mouton C, Rodríguez-Freire M. Network analysis to assess landscape connectivity trends: Application to European forests (1990–2000). Ecological Indicators. 2011;11(2):407–16.

* View Article

* Google Scholar

82. 82. Saura S, Pascual-Hortal L. A new habitat availability index to integrate connectivity in landscape conservation planning: Comparison with existing indices and application to a case study. Landscape and Urban Planning. 2007;83(2–3):91–103.

* View Article

* Google Scholar

83. 83. Godínez-Gómez O, Urquiza-Haas T, Koleff P, Correa Ayram C, Castillo L. ¿Qué tan conectados están los sistemas nacionales de áreas protegidas terrestres en Latinoamérica y el Caribe? In: Álvarez Malvido M, Lázaro C, de Lamo X, Juffe-Bignoli D, Cao R, Bueno P, et al. Informe Planeta Protegido 2020: Latinoamérica y el Caribe. Ciudad de México, México; Cambridge, UK; Gland, Switzerland; Bogotá D.C., Colombia: RedParques, UNEP-WCMC, CMAP-UICN, WWF, CONANP y Proyecto IAPA. 2021. 148 p.

84. 84. Pineda-Zapata S, González-Ávila S, Armenteras D, González-Delgado TM, Morán-Ordoñez A. Mapping the way: identifying priority potential corridors for protected areas connectivity in Colombia. Perspectives in Ecology and Conservation. 2024;22(2):156–66.

* View Article

* Google Scholar

85. 85. Rabinowitz A, Zeller KA. A range-wide model of landscape connectivity and conservation for the jaguar, Panthera onca. Biological Conservation. 2010;143(4):939–45.

* View Article

* Google Scholar

86. 86. Zárrate Charry DA, González-Maya JF, Arias-Alzate A, Jiménez-Alvarado JS, Reyes Arias JD, Armenteras D, et al. Connectivity conservation at the crossroads: protected areas versus payments for ecosystem services in conserving connectivity for Colombian carnivores. R Soc Open Sci. 2022;9(1):201154. pmid:35242340

* View Article

* PubMed/NCBI

* Google Scholar

87. 87. Payán-Garrido E, von Hildebrand M. La conectividad a gran escala como herramienta clave para la conservación en Colombia. In: Gómez MF, Moreno LA, Andrade GI, Rueda C. Biodiversidad 2015. Estado y Tendencias de la Biodiversidad Continental de Colombia. Instituto Alexander von Humboldt; 2016. p. 167.

88. 88. Zeller KA, Jennings MK, Vickers TW, Ernest HB, Cushman SA, Boyce WM. Are all data types and connectivity models created equal? Validating common connectivity approaches with dispersal data. Diversity and Distributions. 2018;24(7):868–79.

* View Article

* Google Scholar

89. 89. Pullinger MG, Johnson CJ. Maintaining or restoring connectivity of modified landscapes: evaluating the least-cost path model with multiple sources of ecological information. Landscape Ecol. 2010;25(10):1547–60.

* View Article

* Google Scholar

90. 90. Sawyer SC, Epps CW, Brashares JS. Placing linkages among fragmented habitats: do least-cost models reflect how animals use landscapes? Journal of Applied Ecology. 2011;48(3):668–78.

* View Article

* Google Scholar

91. 91. Vanegas Pinzón S, Ospina Arango OL, Escobar Niño GA, Ramírez W, Sánchez JJ. Plan Nacional de Restauración ecológica, rehabilitación y recuperación de áreas disturbadas. Bogotá D.C., Colombia: Ministerio de Ambiente y Desarrollo Sostenible; 2015. p. 50.

92. 92. Corpoguajira, Fundación Nativa. Programa de implementación de la estrategia de conservación de la danta colombiana (Tapirus terrestris colombianus) en la Guajira. 2 Rioacha, Guajira: Corporación Autónoma Reginal de la Guajira; 2008. p. 15.

93. 93. UNEP-WCMC. Integrated Biodiversity Assessment Tool. 2024. https://www.unep-wcmc.org/en

* View Article

* Google Scholar

94. 94. López-Arévalo HF, Liévano-Latorre LF, Montenegro Díaz OL. The role of small reserves on mammal conservation in Colombia. Caldasia. 2021; 43:354–65.

* View Article

* Google Scholar

95. 95. Mantilla-Meluk H, Mosquera-Guerra F, Trujillo F, Pérez N, Velásquez-Valencia A, Vargas Pérez A. Mamíferos del sector norte del Parque Nacional Natural Serranía de Chiribiquete. Colombia Amaz. 2017;10:99–134.

* View Article

* Google Scholar

96. 96. IUCN. The IUCN Red List of Threatened Species. 2024. https://www.iucnredlist.org/

* View Article

* Google Scholar

97. 97. MinAmbiente. Resolución 0126 de 2024. https://www.minambiente.gov.co/wp-content/uploads/2024/02/Resolucion-0126-de-2024.pdf; 2024.

* View Article

* Google Scholar

98. 98. Dirzo R, Young HS, Galetti M, Ceballos G, Isaac NJB, Collen B. Defaunation in the Anthropocene. Science. 2014;345(6195):401–6. pmid:25061202

* View Article

* PubMed/NCBI

* Google Scholar

99. 99. Galetti M, Gonçalves F, Villar N, Zipparro VB, Paz C, Mendes C, et al. Causes and Consequences of Large-Scale Defaunation in the Atlantic Forest. Marques MCM, Grelle CEV. The Atlantic forest history, biodiversity, threats and opportunities of the mega-diverse. New York, USA: Springer; 2021. p. 27.

Citation: Mosquera-Guerra F, Barreto S, Palencia-Rivera JD, Velásquez-Valencia A, Mantilla-Meluk H, Bruges-Morales GA, et al. (2025) Connecting Colombia’s protected areas: Using a functional approach for tapir species. PLoS One 20(5): e0323175. https://doi.org/10.1371/journal.pone.0323175

About the Authors:

Federico Mosquera-Guerra

Roles: Writing – original draft

E-mail: [email protected]

Affiliations: Centro de Investigación de la Biodiversidad Andino Amazónica (INBIANAM), Universidad de Amazonia (UA), Florencia, Colombia, Grupo de Ecología del Paisaje y Modelación de Ecosistemas (ECOLMOD), Universidad Nacional de Colombia (UNAL), Bogotá D.C., Colombia, Fundación Omacha (FO), Bogotá D.C., Colombia

ORICD: https://orcid.org/0000-0002-6915-8602

Sebastian Barreto

Roles: Investigation

Affiliation: Grupo de Ecología del Paisaje y Modelación de Ecosistemas (ECOLMOD), Universidad Nacional de Colombia (UNAL), Bogotá D.C., Colombia

Juan D. Palencia-Rivera

Roles: Investigation

Affiliation: Fundación Omacha (FO), Bogotá D.C., Colombia

Alexander Velásquez-Valencia

Roles: Investigation

Affiliation: Centro de Investigación de la Biodiversidad Andino Amazónica (INBIANAM), Universidad de Amazonia (UA), Florencia, Colombia

Hugo Mantilla-Meluk

Roles: Investigation

Affiliation: Centro de Estudios de Alta Montaña (CEAM), Universidad del Quindío (UQ), Armenia, Quindío, Colombia

Gustavo A. Bruges-Morales

Roles: Methodology

Affiliation: Grupo de Investigación en Ciencias Aplicadas (GINCAP), Universidad Autónoma de Bucaramanga (UAB), Bucaramanga, Colombia

Alex M. Jiménez-Ortega

Roles: Methodology

Affiliation: Programa de Biología, Universidad Tecnológica del Chocó (UTCH), Quibdó, Colombia

Fernando Trujillo

Roles: Methodology

Affiliation: Fundación Omacha (FO), Bogotá D.C., Colombia

Dolors Armenteras-Pascual

Roles: Methodology

Affiliation: Grupo de Ecología del Paisaje y Modelación de Ecosistemas (ECOLMOD), Universidad Nacional de Colombia (UNAL), Bogotá D.C., Colombia

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

[/RAW_REF_TEXT]

References

1. Brennan A, Naidoo R, Greenstreet L, Mehrabi Z, Ramankutty N, Kremen C. Functional connectivity of the world’s protected areas. Science. 2022;376(6597):1101–4. pmid:35653461

2. Linero-Triana D, Correa-Ayram CA, Velásquez-Tibatá J. Prioritizing ecological connectivity among protected areas in Colombia using a functional approach for birds. Global Ecology and Conservation. 2023;48:e02713.

3. Schmidt C, Domaratzki M, Kinnunen RP, Bowman J, Garroway CJ. Continent-wide effects of urbanization on bird and mammal genetic diversity. Proc Biol Sci. 2020;287(1920):20192497. pmid:32019443

4. CDB. First draft of the post-2020 global biodiversity framework. 2021. https://www.cbd.int/doc/c/abb5/591f/2e46096d3f0330b08ce87a45/wg2020-03-03-en.pdf

5. SiB. Sistema de Información sobre Biodiversidad de Colombia [Internet]. 2024 [cited 2024 Nov 6. ]. https://biodiversidad.co/

6. PNN. Registro Único Nacional de Áreas Protegidas [Internet]. 2024 [cited 2024 Nov 6. ]. https://runap.parquesnacionales.gov.co/cifras

7. Etter A, Andrade A, Saavedra-Ramírez K, Arévalo P, Amaya-Valderrama P. Risk assessment of Colombian continental ecosystems: An application of the Red List of Ecosystems methodology (v. 2.0). Pontificia Universidad Javeriana, Conservación Internacional Colombia; 2017.

8. Etter A, Andrade A, Nelson CR, Cortés J, Saavedra K. Assessing restoration priorities for high-risk ecosystems: An application of the IUCN Red List of Ecosystems. Land Use Policy. 2020;99:104874.

9. Etter A, Andrade A, Saavedra K, Amaya P, Cortés J, Arévalo P. Ecosistemas colombianos: una aplicación de la lista roja de ecosistemas a los ecosistemas terrestres continentales. Bogotá D.C., Colombia: Centro Editorial Javeriano; 2020a.

10. Murillo-Sandoval PJ, Clerici N, Correa-Ayram C. Rapid loss in landscape connectivity after the peace agreement in the Andes-Amazon region. Global Ecology and Conservation. 2022;38:e02205.

11. Murillo-Sandoval PJ, Kilbride J, Tellman E, Wrathall D, Van Den Hoek J, Kennedy RE. The post-conflict expansion of coca farming and illicit cattle ranching in Colombia. Sci Rep. 2023;13(1):1965. pmid:36737650

12. Armenteras D, Cabrera E, Rodríguez N, Retana J. National and regional determinants of tropical deforestation in Colombia. Reg Environ Change. 2013;13(6):1181–93.

13. Armenteras D. Cambios en los patrones espaciales de área quemada en Colombia, ¿qué ha pasado en las dos primeras décadas del siglo XXI? Rev Acad Colomb Cienc Ex Fis Nat. 2022.

14. González TM, González-Trujillo JD, Muñoz A, Armenteras D. Differential effects of fire on the occupancy of small mammals in neotropical savanna-gallery forests. Perspectives in Ecology and Conservation. 2021;19(2):179–88.

15. Mosquera-Guerra F, Moreno-Niño N, Garcia-Suabita W, Armenteras-Pascual D. Impact of wildfires on the suitability habitat of the lowland tapir in a native savanna landscape of northern South America. Discov Conserv. 2024;1(1).

16. García M, Jordan C, O’Farril G, Poot C, Meyer N, Estrada N, et al. Tapirus bairdii. The IUCN Red List of Threatened Species 2016: e.T21471A45173340. 2024. https://dx.doi.org/10.2305/IUCN.UK.2016-1.RLTS.T21471A45173340.en

17. Lizcano DJ, Amanzo J, Castellanos A, Tapia A, Lopez-Malaga CM. Tapirus pinchaque. The IUCN Red List of Threatened Species 2016: e.T21473A45173922. 2024. https://dx.doi.org/10.2305/IUCN.UK.2016-1.RLTS.T21473A45173922.en

18. Varela D, Flesher K, Cartes JL, de Bustos S, Chalukian S, Ayala G, et al. Tapirus terrestris. The IUCN Red List of Threatened Species 2019: e.T21474A45174127. 2024. https://dx.doi.org/10.2305/IUCN.UK.2019-1.RLTS.T21474A45174127.en

19. Marco Palamara G, Rozenfeld A, de Santana CN, Klecka J, Riera R, Eguíluz VM, et al. Biodiversity dynamics in landscapes with fluctuating connectivity. Ecography. 2023;2023(7).

20. Mosquera-Guerra F, Barreto S, Moreno-Niño N, González-Delgado TM, Armenteras-Pascual D. Habitat connectivity of threatened ungulate species in a native savanna landscape of northern South America. Mamm Biol. 2024;104(3):259–75.

21. de Chazal J, Rounsevell MDA. Land-use and climate change within assessments of biodiversity change: A review. Global Environmental Change. 2009;19(2):306–15.

22. Haddad NM, Brudvig LA, Clobert J, Davies KF, Gonzalez A, Holt RD, et al. Habitat fragmentation and its lasting impact on Earth’s ecosystems. Sci Adv. 2015;1(2):e1500052. pmid:26601154

23. IPBES. Global assessment report on biodiversity and ecosystem services of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services. Final report. IPBES secretariat. Bonn, Germany [Internet]. 2019. https://doi.org/10.5281/zenodo.3831673

24. Mosquera-Guerra F, González-Delgado T, Mantilla-Meluk H, Jiménez-Ortega AM, Velásquez-Valencia A, Armenteras-Pascual D. Colombia País de Tapires. Florencia, Colombia: Universidad de la Amazonia; 2024b.

25. Reyna-Hurtado R, Sanvicente-López M, Pérez-Flores J, Carrillo-Reyna N, Calmé S. Insights into the multiannual home range of a Baird’s tapir (Tapirus bairdii) in the Maya Forest. Therya. 2016;7(2):271–6.

26. Wade AA, McKelvey KS, Schwartz MK. Resistance-surface-based wildlife conservation connectivity modeling: Summary of efforts in the United States and guide for practitioners. Rocky Mountain Research Station: U.S. Department of Agriculture, Forest Service; 2015. https://doi.org/10.2737/rmrs-gtr-333

27. Foerster CR, Vaughan C. Home Range, Habitat Use, and Activity of Baird’s Tapir in Costa Rica1. Biotropica. 2002;34(3):423–37.

28. Ramírez-Chaves HE, Leuro Robles NG, Castaño RA, Morales-Martínez DM, Suárez Castro AF, Rodríguez-Posada ME, et al. Mamíferos de Colombia. v1.14. SCMas. Dataset/Checklist. 2024.

29. Mosquera-Guerra F, Moreno-Niño N, Barreto S, Armenteras-Pascual D. Habitat suitability and representation of the potential distribution of the lowland tapir (Tapirus terrestris) within land cover types and protected areas of eastern Colombia. Mamm Res. 2024;69(4):507–18.

30. Sánchez-Cuervo AM, Aide TM, Clark ML, Etter A. Land cover change in Colombia: surprising forest recovery trends between 2001 and 2010. PLoS One. 2012;7(8):e43943. pmid:22952816

31. IAvH. Biodiversidad colombiana: números para tener en cuenta; 2024.

32. IDEAM. Mapa de Coberturas de la tierra Metodología Corine Land Cover Escala 1:100.000 Periodo 2018. 2024.

33. Dávalos LM, Sanchez KM, Armenteras D. Deforestation and Coca Cultivation Rooted in Twentieth-Century Development Projects. BioScience. 2016;66(11):974–82.

34. Castellanos AX. Iridum/GPS telemetry to study home range and population density of mountain tapirs in the río Papallacta watershed, Ecuador. The Newsletter of the IUCN/SSC TSG. 2013;22:20–26.

35. Medici EP. Assessing the viability of lowland tapir populations in a fragmented landscape. Tesis de Doctoral. 2010.

36. Medici EP, Mezzini S, Fleming CH, Calabrese JM, Noonan MJ. Movement ecology of vulnerable lowland tapirs between areas of varying human disturbance. Mov Ecol. 2022;10(1):14. pmid:35287742

37. González TM, González-Trujillo JD, Palmer JRB, Pino J, Armenteras D. Movement behavior of a tropical mammal: The case of Tapirus terrestris. Ecological Modelling. 2017;360:223–9.

38. Montenegro OL, López-Arévalo H, Mora-Beltrán C, Lizcano DJ, Serrano H, Mesa E, et al. Tropical ungulates of Colombia. Gallina-Tessaro S. Ecology and conservation of tropical ungulates in Latin America. Cham, Germany: Springer; 2019. p. 159.

39. Montenegro OL. Programa nacional para la conservación del género Tapirus en Colombia. Bogotá D.C., Colombia: Dirección de Ecosistemas, Ministerio de Ambiente, Vivienda y Desarrollo Territorial; 2005. p. 100.

40. Parra-Romero A, García Martínez RY, Escobar-Lasso S, Lizcano DJ, Aconcha-Abril I, Vélez L, et al. Programa de Conservación de la Danta de Montaña (Tapirus pinchaque) en Parques Nacionales Naturales de Colombia. Bogotá, D.C. Colombia: Parques Nacionales Naturales de Colombia, Iniciativa de Conservación de Tapires de Colombia (CTC)/ IUCN SSC Tapir Specialist Group y Proyecto de Conservación de Aguas y Tierras – ProCAT Colombia; 2021. p. 78.

41. Álviz A, Pérez AK. Plan para la conservación de la danta de tierras bajas (Tapirus terrestris) en los departamentos de Casanare, Arauca y Vichada. Yopal, Casanare: Corporinoquia-Fundación Orinoquia Biodiversa; 2015. p. 60.

42. Restrepo H, Gómez E.I. Plan de Conservación para oso andino (Tremarctos ornatus) y danta de montaña (Tapirus pinchaque) en el departamento de Putumayo. Mocoa, Putumayo: WWF – Corpoamazonia; 2017. p. 54.

43. Mantilla-Meluk H, Mosquera-Guerra F, Trujillo F, Lizcano D, Sandoval S, González TM, Londoño MC. Plan de manejo para la conservación de la danta de montaña (Tapirus pinchaque) en el departamento del Quindío. Armenia, Quindío: Corporación Autónoma Regional del Quindío (CRQ); 2017. p. 60.

44. Zizka A, Silvestro D, Andermann T, Azevedo J, Duarte Ritter C, Edler D, et al. CoordinateCleaner: Standardized cleaning of occurrence records from biological collection databases. Methods Ecol Evol. 2019;10(5):744–51.

45. Aiello‐Lammens ME, Boria RA, Radosavljevic A, Vilela B, Anderson RP. spThin: an R package for spatial thinning of species occurrence records for use in ecological niche models. Ecography. 2015;38(5):541–5.

46. Norris D. Model Thresholds are More Important than Presence Location Type: Understanding the Distribution of Lowland tapir (Tapirus Terrestris) in a Continuous Atlantic Forest of Southeast Brazil. Tropical Conservation Science. 2014;7(3):529–47.

47. Ortega-Andrade HM, Prieto-Torres DA, Gómez-Lora I, Lizcano DJ. Ecological and geographical analysis of the distribution of the mountain tapir (Tapirus pinchaque) in Ecuador: importance of protected areas in future scenarios of global warming. PLoS One. 2015;10(3):e0121137. pmid:25798851

48. Cordeiro PJL, Fragoso JMV, Crawshaw D, Oliveira LFB. Lowland tapir distribution and habitat loss in South America. PeerJ. 2016; 4:e2456.

49. Naimi B, Hamm NAS, Groen TA, Skidmore AK, Toxopeus AG. Where is positional uncertainty a problem for species distribution modelling? Ecography. 2013;37(2):191–203.

50. Akinwande MO, Dikko HG, Samson A. Variance Inflation Factor: As a Condition for the Inclusion of Suppressor Variable(s) in Regression Analysis. OJS. 2015;05(07):754–67.

51. R Development Core Team. R: a language and environment for statistical computing. R Foundation for statistical computing. Version 3.1.1. Vienna, Austria: R Foundation for Statistical Computing; 2024.

52. Hijmans RJ, Cameron SE, Parra JL, Jones PG, Jarvis A. Very high resolution interpolated climate surfaces for global land areas. Int J Climatol. 2005;25(15):1965–78.

53. Farr TG, Rosen PA, Caro E, Crippen R, Duren R, Hensley S, et al. The Shuttle Radar Topography Mission. Reviews of Geophysics. 2007;45(2).

54. Didan K. MOD13A1 MODIS/Terra vegetation indices 16-day L3 global 500m SIN grid V006 [data set]. NASA EOSDIS land processes DAAC. 2024. https://lpdaac.usgs.gov/products/mod13a1v006/

55. IDEAM. Actualización de cifras de monitoreo de la superficie de bosque y la deforestación – Año 2021; 2022.

56. IGAC. Cartografía base para Colombia (1:100.000). 2024.

57. Kennedy CM, Oakleaf JR, Theobald DM, Baruch-Mordo S, Kiesecker J. Managing the middle: A shift in conservation priorities based on the global human modification gradient. Glob Chang Biol. 2019;25(3):811–26. pmid:30629311

58. Phillips SJ, Anderson RP, Schapire RE. Maximum entropy modeling of species geographic distributions. Ecological Modelling. 2006;190(3–4):231–59.

59. Phillips SJ, Dudík M. Modeling of species distributions with Maxent: new extensions and a comprehensive evaluation. Ecography. 2008;31(2):161–75.

60. Hijmans RJ, Phillips S, Leathwick J, Elith J. dismo: Species Distribution Modeling. R package version 1. 2023;3–14. https://github.com/rspatial/dismo

61. Muscarella R, Galante PJ, Soley‐Guardia M, Boria RA, Kass JM, Uriarte M, et al. ENMeval: An R package for conducting spatially independent evaluations and estimating optimal model complexity for Maxent ecological niche models. Methods Ecol Evol. 2014;5(11):1198–205.

62. Phillips SJ, Dudík M, Elith J, Graham CH, Lehmann A, Leathwick J, et al. Sample selection bias and presence-only distribution models: implications for background and pseudo-absence data. Ecol Appl. 2009;19(1):181–97. pmid:19323182

63. Shcheglovitova M, Anderson RP. Estimating optimal complexity for ecological niche models: A jackknife approach for species with small sample sizes. Ecological Modelling. 2013;269:9–17.

64. Koen EL, Bowman J, Sadowski C, Walpole AA. Landscape connectivity for wildlife: development and validation of multispecies linkage maps. Methods Ecol Evol. 2014;5(7):626–33.

65. Foerster C. Results of a five-year telemetry study of Baird’s tapir in Costa Rica. P. 9. Proceedings of the First International Tapir Symposium. San Jose, Costa Rica. 2001.

66. Lizcano D, Cavelier J. Using GPS Collars to Study Mountain Tapirs (Tapirus pinchaque) in the Central Andes of Colombia. Tapir Conservation. 2004;13:18–23.

67. Downer CC. The mountain tapir, endangered ‘flagship’ species of the high Andes. Oryx. 1996;30(1):45–58.

68. Noss A, Cuellar RL, Barrientos J, Maffei L, Cuellar E, Arispe R, et al. A camera trapping and radiotelemetry study of lowland tapir (Tapirus terrestris) in Bolivian dry forest. Tapir Conserv. 2003;12:24–32.

69. Zeller KA, McGarigal K, Whiteley AR. Estimating landscape resistance to movement: a review. Landscape Ecol. 2012;27(6):777–97.

70. Taylor PD, Fahrig L, Henein K, Merriam G. Connectivity Is a Vital Element of Landscape Structure. Oikos. 1993;68(3):571.

71. Ersoy E, Jorgensen A, Warren PH. Identifying multispecies connectivity corridors and the spatial pattern of the landscape. Urban Forestry & Urban Greening. 2019;40:308–22.

72. Anantharaman R, Hall K, Shah V, Edelman A. Circuitscape in Julia: high performance connectivity modelling to support conservation decisions. ArXiv. 2019.

73. McClure ML, Hansen AJ, Inman RM. Connecting models to movements: testing connectivity model predictions against empirical migration and dispersal data. Landscape Ecol. 2016;31(7):1419–32.

74. McRae BH, Dickson BG, Keitt TH, Shah VB. Using circuit theory to model connectivity in ecology, evolution, and conservation. Ecology. 2008;89(10):2712–24. pmid:18959309

75. McRae BH, Kavanagh DM. Linkage Mapper Connectivity Analysis Software [Internet]. 2011 [cited 2024 Nov 6. ]. https://linkagemapper.org/

76. ESRI. ArcGIS desk- top: release 10.8.1; 2024.

77. Adriaensen F, Chardon JP, de Blust G, Swinnen E, Villalba S, Gulinck H, et al. The application of ‘least-cost’ modelling as a functional landscape model. Landsc. Urban Plan. 2003;64(4):233–247.

78. McRae B, Kavanagh D. Linkage Mapper Toolbox: Linkage Pathways Tool User Guide. The Nature Conservancy, Adze Informatics; 2021.

79. Schrott G, Shinn J. A landscape connectivity analysis for the Coastal Marten (Martes caurina humboldtensis). Arcata, CA: U.S. Fish and Wildlife Service; 2020. p. 123.

80. Castilho CS, Hackbart VCS, Pivello VR, dos Santos RF. Evaluating Landscape Connectivity for Puma concolor and Panthera onca Among Atlantic Forest Protected Areas. Environ Manage. 2015;55(6):1377–89. pmid:25860593

81. Saura S, Estreguil C, Mouton C, Rodríguez-Freire M. Network analysis to assess landscape connectivity trends: Application to European forests (1990–2000). Ecological Indicators. 2011;11(2):407–16.

82. Saura S, Pascual-Hortal L. A new habitat availability index to integrate connectivity in landscape conservation planning: Comparison with existing indices and application to a case study. Landscape and Urban Planning. 2007;83(2–3):91–103.

83. Godínez-Gómez O, Urquiza-Haas T, Koleff P, Correa Ayram C, Castillo L. ¿Qué tan conectados están los sistemas nacionales de áreas protegidas terrestres en Latinoamérica y el Caribe? In: Álvarez Malvido M, Lázaro C, de Lamo X, Juffe-Bignoli D, Cao R, Bueno P, et al. Informe Planeta Protegido 2020: Latinoamérica y el Caribe. Ciudad de México, México; Cambridge, UK; Gland, Switzerland; Bogotá D.C., Colombia: RedParques, UNEP-WCMC, CMAP-UICN, WWF, CONANP y Proyecto IAPA. 2021. 148 p.

84. Pineda-Zapata S, González-Ávila S, Armenteras D, González-Delgado TM, Morán-Ordoñez A. Mapping the way: identifying priority potential corridors for protected areas connectivity in Colombia. Perspectives in Ecology and Conservation. 2024;22(2):156–66.

85. Rabinowitz A, Zeller KA. A range-wide model of landscape connectivity and conservation for the jaguar, Panthera onca. Biological Conservation. 2010;143(4):939–45.

86. Zárrate Charry DA, González-Maya JF, Arias-Alzate A, Jiménez-Alvarado JS, Reyes Arias JD, Armenteras D, et al. Connectivity conservation at the crossroads: protected areas versus payments for ecosystem services in conserving connectivity for Colombian carnivores. R Soc Open Sci. 2022;9(1):201154. pmid:35242340

87. Payán-Garrido E, von Hildebrand M. La conectividad a gran escala como herramienta clave para la conservación en Colombia. In: Gómez MF, Moreno LA, Andrade GI, Rueda C. Biodiversidad 2015. Estado y Tendencias de la Biodiversidad Continental de Colombia. Instituto Alexander von Humboldt; 2016. p. 167.

88. Zeller KA, Jennings MK, Vickers TW, Ernest HB, Cushman SA, Boyce WM. Are all data types and connectivity models created equal? Validating common connectivity approaches with dispersal data. Diversity and Distributions. 2018;24(7):868–79.

89. Pullinger MG, Johnson CJ. Maintaining or restoring connectivity of modified landscapes: evaluating the least-cost path model with multiple sources of ecological information. Landscape Ecol. 2010;25(10):1547–60.

90. Sawyer SC, Epps CW, Brashares JS. Placing linkages among fragmented habitats: do least-cost models reflect how animals use landscapes? Journal of Applied Ecology. 2011;48(3):668–78.

91. Vanegas Pinzón S, Ospina Arango OL, Escobar Niño GA, Ramírez W, Sánchez JJ. Plan Nacional de Restauración ecológica, rehabilitación y recuperación de áreas disturbadas. Bogotá D.C., Colombia: Ministerio de Ambiente y Desarrollo Sostenible; 2015. p. 50.

92. Corpoguajira, Fundación Nativa. Programa de implementación de la estrategia de conservación de la danta colombiana (Tapirus terrestris colombianus) en la Guajira. 2 Rioacha, Guajira: Corporación Autónoma Reginal de la Guajira; 2008. p. 15.

93. UNEP-WCMC. Integrated Biodiversity Assessment Tool. 2024. https://www.unep-wcmc.org/en

94. López-Arévalo HF, Liévano-Latorre LF, Montenegro Díaz OL. The role of small reserves on mammal conservation in Colombia. Caldasia. 2021; 43:354–65.

95. Mantilla-Meluk H, Mosquera-Guerra F, Trujillo F, Pérez N, Velásquez-Valencia A, Vargas Pérez A. Mamíferos del sector norte del Parque Nacional Natural Serranía de Chiribiquete. Colombia Amaz. 2017;10:99–134.

96. IUCN. The IUCN Red List of Threatened Species. 2024. https://www.iucnredlist.org/

97. MinAmbiente. Resolución 0126 de 2024. https://www.minambiente.gov.co/wp-content/uploads/2024/02/Resolucion-0126-de-2024.pdf; 2024.

98. Dirzo R, Young HS, Galetti M, Ceballos G, Isaac NJB, Collen B. Defaunation in the Anthropocene. Science. 2014;345(6195):401–6. pmid:25061202

99. Galetti M, Gonçalves F, Villar N, Zipparro VB, Paz C, Mendes C, et al. Causes and Consequences of Large-Scale Defaunation in the Atlantic Forest. Marques MCM, Grelle CEV. The Atlantic forest history, biodiversity, threats and opportunities of the mega-diverse. New York, USA: Springer; 2021. p. 27.