Potential distribution and connectivity in two plethodontid salamanders: conservation areas and landscape corridors for two endemic species

doi:10.21203/rs.3.rs-1605028/v1

Download PDF

Research Article

Potential distribution and connectivity in two plethodontid salamanders: conservation areas and landscape corridors for two endemic species

https://doi.org/10.21203/rs.3.rs-1605028/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Habitat loss is one of the most important threats to biodiversity; it alters the habitat connectivity of species, and is among the main causes of the global amphibian extinction crisis. Identifying the potential areas of distribution and connectivity of species is of the utmost importance so that informed decisions can be made for the conservation of vulnerable amphibian populations. In this study, we performed species distribution models and used circuit theory to model omnidirectional connectivity for two plethodontid salamanders of conservation concern distributed in the forests of Chiapas, Mexico, and Guatemala (Bolitoglossa franklini and Bolitoglossa lincolni). Potential distribution maps show an affinity for well-preserved montane forests for both species. Likewise, we found that the niches of the species are not similar. The connectivity models show that the main areas of connectivity are in the Meseta Central de Chiapas, Sierra Madre de Chiapas, and the Cordillera Volcánica Guatemalense, in this last range, important areas of connectivity were located, as well as least-cost paths and barriers to the movement of both species. We identified that important areas of connectivity are not within the protected natural areas and may be threatened by the increasing influence of anthropogenic activities. The results of our study show the importance of preserving the regional forests to ensure the persistence of species with arboreal habits and high sensitivity to habitat transformation, as well as to recognize and prioritize potential areas for management and protection in both southern Mexico and Guatemala.

Plethodontidae

Bolitoglossa franklini

Bolitoglossa lincolni

forest cover loss

connectivity

Sierra Madre de Chiapas

Guatemala Mountain Range

Ecoregions

The world's forests are highly diverse in terms of species, but it is estimated that at least 46% of them have disappeared due to human activities (Crowther et al. 2015), and only 40% remain free of modification (Grantham et al. 2021). The forests in tropical regions have a third of the global biodiversity, However, there is a great loss of vertebrates, mainly due to habitat loss and land change use, affecting the diversity, distribution, and species interaction, and, hence, the ecosystem functioning (Jetz et al. 2007; Thompson et al. 2017). Currently, amphibians are facing an environmental crisis due to their population reduction and the extinction of different species in several regions of the planet (Stuart et al. 2004). For instance, in Mesoamerica, threats to amphibians are much higher than in all the rest of the American continents (Alroy et al. 2015; Scheele et al. 2019) due to habitat modification (Whitfield 2016). In the south of Mexico, a collapse and reduction of amphibian populations, particularly of amphibian species that can be found in cloud forests and within mountain ranges has been occurring for years (Lips et al. 2008). In Guatemala, local extirpations of plethodontid salamanders have been happening even in protected natural areas (Rovito et al. 2009). Among the plethodontid salamanders, the genus Bolitoglossa is widely distributed from southwestern North America to northern Brazil and the central mountains of Bolivia; the greatest diversity and endemism of the genus is found in the geological complex known as Nuclear Central America (Wake and Lynch 1976; Rovito and Parra-Olea 2016). Bolitoglossa franklini and B. lincolni are phylogenetically closely related and considered sister species (Wake and Lynch 1982; Parra-Olea et al. 2004; Vásquez-Almazán and Rovito 2014). They have an allopatric distribution and are found in pine, pine-oak, and cloud forest areas within Chiapas and Guatemala’s occidental Central Mountains (Wake et al. 1980).

Bolitoglossa franklini is a complex species, regarded as vulnerable (IUCN SSC Amphibian Specialist Group 2020a) and distributed in the Sierra Madre de Chiapas until the southwest Pacific slopes in western Guatemala (Wake et al. 1980), within pine forests and pine-oak forest between 1500 to 3200 m a.s.l. They are mainly arboreal and use bromeliads as their main microhabitat in mature and conserved forests. Bolitoglossa lincolni, however, is a near-threatened species (IUCN SSC Amphibian Specialist Group, 2020b) distributed mainly in the pine and pine-oak forests in the Central Plateau of Chiapas (CCP), the mountain range of Cuchumatanes, Cuilco Sierra, and the San Marcos region in the Cordillera Volcánica Guatemalense (GVC) (Wake and Lynch 1976; Wake et al. 1980), and it is found in disturbed areas, exhibiting a generalist habit. Despite the differences observed in both species, the populations share threats to their conservation such as the loss and transformation of habitat due to forest exploitation, agricultural activities, and the growth of human settlements in their distribution areas (IUCN SSC Amphibian Specialist Group 2020a, b).

It is crucial to delineate the species distribution area to establish conservation areas for preserving these species and habitat conservation. The species distribution models (SDM) estimate the environmental suitability and potential distribution areas using the species presence records and the geographic and environmental conditions (Peterson 2006; Elith and Leathwick 2009; Warren 2012). These predictions are valuable for assessing the potential distribution and, therefore, the conservation status of rare, endemic, or poorly studied species (Guisan et al. 2013). Such SDMs are also useful in determining suitable habitats for populations of undiscovered species, for detecting climatic barriers to dispersal, and for exploring ecological divergence between closely related taxa when species share distribution ranges (Cunningham et al. 2008; Glor and Warren et al. 2011; Hu and Jiang 2018).

Landscape connectivity is the degree to which the environment simplifies or prevents an organism’s movement between different locations (Taylor et al. 1993; Tischendorf and Fahrig 2000). Habitat fragmentation can modify connectivity, imposing barriers to organism movement and affecting the individuals’ ability to disperse and reach habitable places (Wiens et al. 2009). These factors are especially important for amphibians; since they are very susceptible to habitat transformation due to their limited vagility and high sensitivity to microclimatic variation, which affect their dispersal and increase their migrant mortality (Nowakowsky et al. 2015). A practice that can guarantee biodiversity conservation, along with the functioning and stability of the ecosystem, is the identification and design of corridors, which will facilitate the appropriate dispersion of the species in the landscape (Thompson et al. 2017). The connectivity between habitats can be modeled as a cost measure where the landscape elements are allocated resistance values according to the degree to which they allow or hinder the movement of individuals. One of the most used methodologies for this purpose is the circuit theory; this analysis uses resistance and flow values of movement and, therefore, can be applied to predict the movement and the probability of dispersal of organisms while generating measures of connectivity or isolation between populations, patches of habitats, or protected natural areas (McRae et al. 2008).

The previously described approaches mentioned are rarely used among Mexican amphibians; however, their application in the study of plethodontid salamanders in central Mexico has been useful in identifying priority areas and corridors (Vargas-Jaimes et al. 2021; Sunny et al. 2022), along with the identification of barriers and their effect on the population isolation (Velo-Antón et al. 2013; Sunny et al. 2022).

In this study, we utilized SDMs to describe the potential distribution of and evaluated the similarity of the climatic niches of two sister species of the genus Bolitoglossa, (B. lincolni and B. franklini). Additionally, we applied a circuit theory approach to find the corridors between core areas, least-cost path corridors, and bottlenecks for the movement of plethodontid salamanders in the mountains of southern Mexico and Guatemala. Therefore, this study aimed to determine the potential distribution of the two salamander species, identify corridors and pinch points (bottlenecks) in the study area, and analyze the corridor network to identify important core areas and corridors to keep the network connection and, consequently, to enhance information within the conservation actions for the salamanders of southern Mexico and Central America.

Study area

The study area includes southern Mexico and western Guatemala. Specifically in the Sierra Madre de Chiapas (SMCH), Meseta Central de Chiapas (CCP), and the highlands of Guatemala (Cordillera de Los Cuchumatanes (CC), the relief of this area forms a mountainous landscape with an abrupt topography that allows the presence of different climates and ecosystems that have evolved many species and endemics (CONANP 2003). In the SMCH, there is the largest proportion of conserved forests areas under protection, like the Sepultura, El Triunfo, Tacaná Volcano, and the Frailescana, one of the most important cloud forests in the region. In Guatemala, several protected natural areas occur in the study area, like Volcán Tacaná, Tajumulco, and Volcán Lacandón, as well as the Visis-Cabá (CONAP 2021). In the study area, the predominant vegetation is cloud forests, pine forests, pine-oak forests, and agricultural areas, mainly coffee plantations, corn crops, and pastures for cattle.

Species occurrence and environmental variables

Occurrence records of Bolitoglossa franklini and B. lincolni with geographical coordinates were obtained from fieldwork from 2013 to 2021 and online databases such as the Global Biodiversity Information Facility (GBIF; https://www.gbif.org), Consejo Nacional de la Biodiversidad (CONABIO; https://www.gob.mx/conabio), and the Collection of Herpetology from the Southeast of Mexico from El Colegio de la Frontera Sur (ECOSUR-SC), Colección Zoológica Regional Secretaría de Medio Ambiente e Historia Natural. Cleaning of the occurrence data was carried out with the thin command of the library spThin (Aiello Lammens et al. 2015) for R software (4.00; R. Development Core Team 2019) as follows: 1) occurrence data was previsualized to remove erroneous records or records outside the potential geographic range, 2) duplicate records were eliminated, and 3) to avoid pseudoreplication, the species’ distribution data were subsampled considering only records that were spaced at least 1 km from each other, as this method has been shown to substantially reduce the degree of overfitting in the model (Segurado et al. 2006; Boria et al. 2014) and according to the dispersion capacity observed in Plethodontidae (Lowe et al. 2008). After filtering the data, 69 records were obtained for B. franklini and 63 for B. lincolni (Fig. 1).

Species distribution models

To delimit the distribution of a species, it is necessary to determine a calibration space for the models, or M according to the BAM diagram, which represents the accessible area for a species and is compatible with the biogeographic history of the species. (Soberón and Peterson 2005), For this purpose, we selected and grouped the Central American pine-oak forests (CAPOF), Central American montane forests (CAMF), Chiapas montane forests (CMF), and Sierra Madre de Chiapas moist forests (SMF) ecoregions according to Olson et al. 2001 (Fig. 1). For the environmental niche modeling, we use the bioclimatic layers available in Worldclim Project 2 (Fick and Hijmans 2017) and the ENVIREM climate layers (Title and Bemmels 2018). To prevent the models from having an effect caused by the over-parameterization and collinearity between the layers, we discarded some bioclimatic layers based on the variance inflation factor (VIF), which quantifies the multicollinearity between the layers, indicating the degree that a variable increases the standard error of a regression. The VIF was calculated with the package usdm 1.1.18, (Naimi et al. 2014). The final set of variables was: annual mean diurnal range (Bio 2), temperature seasonality (Bio 4), mean temperature of the wettest quarter (Bio 9), precipitation seasonality (Bio 15), precipitation of the wettest quarter (Bio 16), precipitation of the warmest quarter (Bio 18), precipitation of the coldest quarter (Bio 19), climatic moisture index, potential evapotranspiration (PET) of the driest quarter, and PET seasonality. For the environmental niche modeling, we used the maximum entropy analysis implemented in Maxent 3.4.1 (Phillips et al. 2021). Before the elaboration of the SDM, using independent evaluation metrics of the threshold, we looked for the optimal set of parameters to obtain the most parsimonious model with the ENMeval package (Muscarella et al. 2014). One test considered for the evaluation of the model was the AUC test, which measures the discriminatory capacity, where the highest values indicate that the model has a reliable capacity to distinguish between the test locations and the background points (Peterson et al. 2011). The difference in the AUC (AUC DIFF) is the difference between the AUC calculated from the training locations and the AUC test. This parameter quantifies the overfitting of the models, high values indicate excessive overfitting of the model (Warren and Seifert 2011), and the Akaike information criterion corrected for small samples (AICc) shows the goodness of fit and the complexity of the model. The models with the lowest AICc value are considered the best of the set of models obtained (Warren and Seifert 2011). To obtain the evaluation metrics, we generated 10,000 random points that were used as background points. We partitioned the number of records by the “Checkerboard1” method with an aggregation factor of 5; this method generates a grid over the study area as a Chessboard, which partitions the localities into cells (bins), although it does not offer a balanced number of presences in each cell, and it performs an equal sampling of the geographical and environmental space (Muscarella et al. 2014). Subsequently, the complexity of the models was estimated based on the regularization multiplier (RM) and the Maxent features classes. The RM determines how concentrated or adjusted the distribution is; values less than 1.0 will give rise to a more localized distribution and adjusted to the presence registers, which can lead to overfitting in the models. High values of the RM will give a more dispersed and less localized prediction. For our study, we obtained the most parsimonious model by testing all possible combinations between the RMs and Maxent's feature classes; we tested 10 RMs in a range from 1 to 5, with increments of 0.5 and six feature classes: linear (L), quadratic (Q), product (P), threshold (T), and hinge (H). With the information obtained, we built the model in Maxent, implemented in Dismo 1.1-4 (Hijmans et al. 2017) for R. Additionally, to evaluate the precision of the Maxent model, we transformed the area under the curve (AUC) by calculating the partial ROC (partial receiver operating characteristic curve graphics, Peterson et al. 2008), which was obtained by evaluating the statistical significance through null distributions; we calculated the Partial ROC in the package ntbox (Osorio-Olvera 2020) in R, applying 1000 iterations and an error of 5%. The significance of the AUC was estimated with a bootstrapping of 1000 and using 50% of the presences as training locations.

Similarity, equivalence, and niche overlap

In species with shared ancestry such as Bolitoglossa franklini and B. lincolni, niche similarity may be more likely but rarely identical (Warren and Seifferd 2008). To identify environmental differences, we compare the ecological niches of the species and determine if they are significantly different from each other, we contrasted the niches of both species according to the method of Broenimann et al. (2012). The bioclimatic variables used in the modeling were reduced using an ordering technique with a principal component analysis (PCA), and then we delimited the environmental space using the first two axes of the PCA. Species presence records were plotted within the environmental space using a Kernel density function, which corrects the potential bias of sampling errors (Broenniman et al. 2012) (Fig. 2d). The niche overlap was calculated with the Schöener D metric (Schoener 1970; Rödder and Engler 2011), which calculates the habitable space based on its abundance of the species and the Hellinger distance I, which quantifies the probability of distribution of the species to inhabit a given area, calculating the overlap of the niches (Warren et al. 2008). Both metrics have ranges from 0 to 1, where 1 indicates that the niches are identical. The estimates of presence density and the similarity of the niche between species in the environmental space were repeated 100 times to generate a null distribution comparable with the observed D value; non-significant results indicate that the niches are not identical. The analyses were carried out with the ecospat 3.1 package (Di Cola et al. 2017), and the presence density for the two species was plotted with ggplot2 (Wickham 2016) for R.

Connectivity models

To evaluate the connectivity patterns between the B. franklini and B. lincolni localities, we calculated landscape resistance using circuit theory, which models the movement patterns of species in a complex landscape. The geographic area used to model connectivity was delimited by a minimum convex polygon made from the sample points, with a buffer of 20 km to minimize the increase in resistance values (Koen et al. 2010). The models were elaborated with the presence of localities and a resistance surface layer, which was elaborated from the information obtained from the land cover of Copernicus Global Land Service for the year 2019 at a resolution of 100 m (Buchhorn et al. 2020). The layers contain spatial information about the different types of coverage and the influence that humans have on them and have continuous field values expressed in percentage values (Buchhorn et al. 2020). We resampled the layers and assigned low resistance values to pixels that represented high coverage percentages, and we included a layer of roads, infrastructure, and cities obtained from OpenStreetMap (http://www.openstreetmap.org/). The layer was rasterized and rescaled using the Mosaic to Raster tool of ArcMap 10.8. Finally, the resistance values of the landscape were scaled to 5–100.

The connectivity patterns were obtained with the CIRCUITSCAPE 4.0.5 software, which calculates the effective resistance to movement and all possible paths between pairs of locations or focal nodes, where a node is arbitrarily connected to a 1-amp current source, while the other node is connected to ground (McRae 2006). The process is carried out through iterations between pairs of focal nodes, and they are expressed in current values on a map. Additionally, the least cost path (LCP) was determined with the Linkage Mapper 3.0.0 extension for ArcMap (McRae and Kavanagh 2011). The software calculates all possible routes and their cost in the landscape, showing the lowest cost routes based on the Cost Distance algorithm. This is a calculation of the minimum accumulated cost distance between two nodes. The result is a raster layer with cell values that represent the accumulated cost from a nearby source cell. From this raster, the LCPs are calculated in a vector layer of lines that establish the optimal routes for the establishment of corridors (Adriaensen et al. 2003). To assess the relative importance of core areas, we use current flow centrality analysis. In this approach, core areas are treated as nodes and the current flow through the network is calculated, thus determining the importance of each node and corridor in maintaining connectivity between habitat patches (Carroll et al. 2012; Dutta et al. 2016).

Finally, we calculated the bottlenecks or constriction zones for the movement of the species (pinch points) in the study area. These are used to identify important areas for connectivity due to a high flow of individuals, where high values indicate that corridors are vulnerable to unfavorable conditions (McRae et al. 2008).

Species distribution models

We obtained 215 records for B. franklini and 80 for B. lincolni. After applying spatial filtering, we retained 69 and 63 records, respectively. After calculating the correlation coefficient with the VIF, we selected 10 variables (Table A1) with VIF values less than 10. For B. franklini, we obtained 61 models, of which the least complex model (AICc = 1296.5) had the following parameters: LQH with a regularization multiplier (rm) of 2.5 and an AUC = 0.94. For B. lincolni, we obtained 61 models, two of them being equally parsimonious (ΔAICc ≤ 2). For this species, niche modeling was performed with the following parameters: AICc = 1219.0, LQH with a rm = 3.5, and an AUC = 0.93. The performance of the models was adequate; according to the calculation of the partial ROC bootstrap tests, 100% of the models were statistically significant with AUC ratio values of 1.89 for B. franklini and 1.86 for B. lincolni, (p < 0.001). The potential distribution of both species shows an area of suitability in the mountains of Chiapas and Guatemala (Fig. 1). For B. franklini, a potential distribution area of 8121.8 km² was obtained, and for B. lincolni, a distribution area of 9411.9 km² was determined; these areas are approximately 13.7% and 15.9% of the studied area, respectively. For B. franklini, the areas of greatest suitability were located mainly in the Cordillera Volcánica Guatemalense (GVC) and in the Sierra Madre de Chiapas (SMCH) in the Istmo de Tehuantepec. Its distribution is reduced, and only the tops of the mountains of the region have suitability. In this area, there are at least three important protected natural areas, the Biosphere Reserves of El Triunfo, La Sepultura, and Volcán Tacaná in México. Bolitoglossa lincolni, in Mexico, is distributed mainly in the Meseta Central de Chiapas (CCP); the analysis showed a region of low probability of presence for the species located in lowlands located between the CCP and the Sierra de los Cuchumatanes. In the GVC area, both species overlap in their distribution. In Mexico, the distribution of B. lincolni is not documented within federally protected areas, and in Guatemala, its distribution occurs in at least four important ANPs: Visis Cabá, Covirey, Volcán Tajumulco, and Volcán Lacandón. According to the jackknife analysis carried out in Maxent, the variable with the greatest relative contribution to the distribution models was the mean temperature of the driest quarter for both B. franklini (Bio9 = 24.1%) and B. lincolni (Bio9 = 72.8%).

Similarity, equivalence, and niche overlap

Niche overlap analysis by PCA (Fig. 1c and 1d) showed that the contributions of the first two axes to PCA were 33.3% and 23%, respectively, and the precipitation of the wettest quarter (Bio16 = 24.5%) was the variable that explains the greater variation in the PC1 component, while the variation in the PC2 component indicated that the niche overlap was mainly explained by the seasonality of PET (37.5%). Schoener's D and Warren's I tests indicated non-significant values for the equivalence and similarity of the niches between B. franklini and B. lincolni (D = 0.40, P > 0.001, I = 0.58, and P > 0.001).

Connectivity analysis

According to connectivity models based on circuit theory, for B. franklini, we identified a potential corridor that extends between the "La Sepultura" Biosphere Reserve in the SMCH, Mexico, to the Chicabal volcano in the GVC (Fig. 3). The connectivity of the populations of B. franklini (Fig. 3a) is lower in the western extension of the SMCH, near the Isthmus of Tehuantepec, especially in the La Sepultura and La Frailescana. The SMCH shows important areas for connectivity and LCPs within El Triunfo Biosphere Reserve (ETBR); however, two larger areas of connectivity were also observed outside of these protected areas. These are mainly found in Mexico in the Motozintla mountains, adjacent to the Pico El Loro-Paxtal state reserve, and in Guatemala in the area between the Tajumulco Volcano reserve and the Esquipulas and Astillero San Marcos Municipal reserves (Fig. 3). The least cost routes for B. franklini are found mainly in the extreme east of the SMCH and the GVC. In Mexico, the routes with medium to high costs for the movement of the species are within the El Triunfo Biosphere Reserve, CPL and Tacaná Volcano. In Guatemala, LCPs with low to moderate values are found between the Tajumulco Volcano and Chicabal Volcano. In Mexico, the western extension of the SMCH offers the most costly and resistant routes for the movement of B. franklini, within the protected areas of La Frailescana and La Sepultura, covering a length of 109.2 km, from the area corresponding to Cerro Quetzal (CQ) in the ETBR, to Cerro Tres Picos in La Sepultura. For B. lincolni, the lowest to moderate cost LCPs are in the Central Chiapas Plateau, while the high to higher cost routes for the species are in the Cuchumatanes region and Sierra de Cuilco.

Within the distribution area of the species, the connectivity analysis detected 29 zones of relative importance for B. franklini, among them the areas of higher current-flow centrality are located in the area bordering Mexico and Guatemala, of which the most extensive habitat is located in the mountains of Motozintla (between the protected areas CPL and VTBR) and has an extension of 169.4 km2, the second most central core area is located in the VTBR and has an extension of 44.7 km2. In the ETBR, La Frailescana and La Sepultura natural areas, the core area extensions are mainly isolated on the mountain tops.

The areas of greatest current-flow centrality are found for B. lincolni in the mountains of the CCP, in the region there are areas of greatest centrality located on the summits of the Central American Pine-Oak Forest, the largest central area is in Mexico and has an extension of 172 km2, the second largest area (100.2 km2) is in the area of the Tajumulco Volcano in Guatemala.

According with Pinch points analysis, constriction zones or unfavorable conditions for the that could break the corridors of both species are found at several points along the corridors, potentially due to the topography of the terrain and climatic conditions of the low-elevation lands.

Our results show a significant area of environmental suitability for B. franklini and B. lincolni in the mountains of southern Mexico and western Guatemala; the distribution of both species overlaps in the Guatemalan Volcanic Mountain range. However, these species distributions are restricted by environmental factors such as the presence of climatic barriers. Contrary to the shared evolutionary relationship between the studied species, the niches of both species are not corresponding, possibly due to the divergence time between them. The GVC, SMCH, and CCP are central areas for the climatic suitability and lower-cost paths for both species, yet there are also potential corridors outside of the protected natural area limits of the region. These results are useful for the habitat conservation of both species and other amphibians that are distributed in the same areas, and specifically, for the priority areas identification to maintain connectivity in cover and land use variations in both southern Mexico and Guatemala.

Plethodontid salamanders are lungless organisms highly reliant on cold and humid environments (Kozak 2017). They are adapted to the environmental conditions of the mountainous areas of Chiapas and Guatemala, which are characterized by their species richness (Wake and Lynch 1976), and for their significant role in the Bolitoglossa genus diversification (Rovito et al. 2015). The species distribution model shows that the ideal climatic conditions for B. franklini distribution are found mainly in the GVC, the pine-oak forests of Central America, and the montane forests of Central America. In Mexico, its distribution occurs in the humid forests of the SMCH but can be restricted to the top of the mountains, as the surroundings are characterized by suboptimal habitats that may be uninhabitable due to their poor climatic suitability. Therefore, migration may be impossible even at short geographical distances (Sexton et al. 2009). Within our results, the distribution of B. lincolni is listed mainly in the pine-oak forests of Central America in the CCP, the Central American montane forests of the Cuchumatanes, and the GVC. Its distribution overlaps with that of B. franklini in the same areas, where it has been previously proposed as a zone of secondary contact, in which the hybridization of both species is promoted (Wake et al. 1980). Furthermore, our SDM confirms the affinity of both species, which was previously discussed by Wake et al. (1980), who acknowledged a greater affinity of B. franklini for cloud forests.

The temperature often determines the geographic limits for ectothermic organisms, such as amphibians (Kearney and Porter 2009), and the distribution of plethodontid amphibians is often related to humid climates that reduce the risk of desiccation (Baken et al. 2020). While a negative relationship with temperature is expected for Mesoamerican salamanders (Kozak and Wiens 2012), the analyses show that the distributions of B. franklini and B. lincolni are, in both cases, explained by the mean temperature of the drier quarter (January–March, Magaña et al. 1999), and this indicates the species preference for temperate climates with little variation throughout the year. This outcome, an overlap distribution, is mainly explained by the precipitation in PC1 and the seasonality of PET in PC2, indicating that the distribution of both species takes place in areas that have stable climates with low precipitation levels and little variability in PET throughout the year. Closely related species are expected to share certain levels of environmental similarity since the rate of niche evolution is not fast enough to eliminate all signs of common ancestry, and such species will not be, however, identical sister species (Warren et al. 2008; Warren et al. 2014). In the present study, we observed that both species show sympatry and geographic ranges overlapping in some areas of their distribution; however, the similarity analyses and equivalence show that their niches are not equivalent or similar to each other. This can imply that their coexistence does not show competitive exclusion due to their phylogenetic position (Warren et al. 2014), and this can be a consequence of the wide number of climates and niches available in the neotropical region compared to other regions with plethodontids presence, thus easing physiological pressures and species interactions (Baken et al. 2020).

The dispersal ability of species and landscape components influence the movement resistance and the connectivity patterns of populations (Cushman et al. 2013). The limited vagility of amphibians and their high sensitivity to microclimatic variations connotes a significant barrier to their dispersal in open environments with dry or warm microclimates and the loss of vegetation (Nowakowski et al. 2015). Plethodontid salamanders are generally restricted to a narrow range of environmental conditions (Grover 1998). Our results show that the SMCH and the GVC have important corridors for B. franklini, yet the areas with greater connectivity and less resistance to movement are particularly found between the territorial limits of Mexico and Guatemala. Our centrality analysis shows that, in the GVC and between the Pico de Loro, El Paxtal, and Volcán Tacaná reserves (VTBR) in Mexico, and between the VTBR and Volcán Tajumulco in Guatemala are significant habitat areas for focal species, there are conserved forests with canopy cover > 70% (Buchhorn, 2020) where significant communities of salamanders are found (Wake and Lynch, 1976). In the case of B. lincolni, a wide-ranging corridor is observed in the region of the Chiapas montane forests, which includes the CCP, the Sierra de los Cuchumatanes, Cuilco and the GVC. These areas present a complex matrix composed of the presence of pine and pine-oak forests under different stages of succession and agricultural areas (INEGI, 2018; Buchhorn, 2020).

Despite the dimensions of the species’ potential corridors, favorable conditions for their movement are not found all over their extent. According to our circuit analyses, the routes with the lowest cost and the greatest connectivity for the dispersal of B. franklini are found in the GVC in western Guatemala, between the Volcán Tacaná and Volcán Tajumulco reserves. This area has mixed forests and conserved mature forests, with canopy cover greater than 70% (Buchhorn et al. 2020) and is surrounded by forest areas with integrity rates of less than 50% (Grantham et al. 2020).

The areas with lower connectivity are found in the SMCH between the protected natural areas of El Triunfo and La Sepultura. In contrast to the GVC, this area corresponds to a mountain chain that descends in altitude from 3,000 to 1,500 m a.s.l., and it is characterized by having a rugged relief in the Isthmus of Tehuantepec (Wake and Lynch 1976). The cloud forest habitats within these areas are restricted to the top of mountains, above 1800 m a.s.l.; therefore, the habitable areas for B. franklini and other plethodontid species are reduced to isolated habitat patches (sky islands), causing allopatric distributions because of historical climatic fluctuations (Wake and Lynch 1982).

Salamanders are organisms with low vagility rates, with less than 1 km movement ranges (Smith and Green 2005). For such species with limited dispersal capacity, their habitats are expected to offer suitable conditions for their maintenance, reproduction, and to preserve their dispersion routes as well as the evolutionary process at different spatial and temporal scales (Pelletier et al. 2014; Hilty et al. 2019). Consequently, the loss of these areas due to habitat destruction and changes in land use can modify the migration routes or affect their general movement (Ray et al. 2002). According to the bottlenecks (pinch points) for B. franklini, the corridors most susceptible to losing connectivity are in Guatemala, within the Tacaná and Tajumulco volcanos. There are potential dispersal routes between these protected natural areas and the surrounding local reserves; nonetheless, these areas are immersed in a matrix with low rates of landscape connectivity (Grantham et al. 2020) since these areas are between the cities of the central region of Guatemala and the Pacific coastal region.

In Mexico, important low-cost paths are found in the SMCH, between the Tacaná Volcano and the Cordón Pico de Loro Paxtal reserve. This area represents a connection point between the SMCH and the GVC; however, at the same time, it has high anthropogenic disturbances such as high urban areas and mining activity (Godínez-Gómez et al. 2020). For B. lincolni, we identified different bottlenecks in the corridors, and the largest ones are found in the CCP. This area is within a matrix of predominantly secondary pine remnants, pine-oak forests, and agricultural areas (INEGI 2018). Herein are the main urbanized areas of the region and traditional agriculture practices, as well as low-intensity forestry practices (González-Espinosa et al. 2007). Even though the region has potential connectivity routes for B. lincolni, these areas are not under any federal conservation criteria.

Implications for conservation

Our study suggests that the preservation of the region’s forests is a conservation priority for the habitat and connectivity areas of the salamander populations found in southern Mexico and Central America, specifically those found in the GVC and SMCH and the CCP, given that in these areas, climatic suitability and potential corridors for the species studied are present. It is essential to know that the areas identified in this study face several challenges for their conservation. The main challenge is to prevent the loss of tropical forests, especially for species with arboreal habits that depend on the vertical structure and microhabitats they provide. This is extremely important since, for example in Mexico, 594 kha of humid primary forest were lost between 2002 and 2019, which places Mexico as the ninth tropical country in primary forest loss according to the Global Forest Watch (Hansen et al. 2013). Ecological studies looking at the importance of forest and habitat structure conservation of plethodontids have examined all the negative consequences and the pressure that changes in land use and climate change exert on the salamander habitats (Díaz-Garcia et al. 2019; Vargas-Jaimes et al. 2021) that can drive salamanders to decline in their populations or even to extinction. The SMCH is a valuable corridor for several vertebrate species (Ocampo et al. 2020; Ceballos et al. 2021). In this region, it is estimated that at least 12% of the forest cover was lost between 1970 and 2000 because of agricultural activities and forest extraction (Cortina-Villar et al. 2012). The conversion of forests to agricultural areas can mean additional pressure for the populations and potential corridors of B. franklini and other salamanders found in the SMCH, which is also considered one of the most significant coffee-growing areas in Mexico (Schroth et al. 2009).

In Guatemala, the expansion of coffee plantations represents the greatest threat to the populations of several salamander species (IUCN SSC Amphibian Specialist Group 2020). The pine and oak forests of the CCP, the main habitat of B. lincolni, lost about 50% of their cover during the last 50 years (Cayuela et al. 2006), although the decline tendency has reduced to 14% of the forest during the last 20 years (Global Forest Watch 2021). For the Guatemalan volcanic mountain range, the corridors in which both species distributions overlap, the greatest threats are erosion and natural and human-caused fires for anthropogenic actives (Bullock 2020). This area is rich in approximately 15 different salamander species (Wake et al. 1987) and we consider that it is necessary to preserve the secondary contact zones to preserve the salamander biodiversity and genetic diversity. Our results show that, due to possible competition events between B. lincolni and B. franklini, B. lincolni can displace B. franklini, which is more sensitive to habitat changes, whereas B. lincolni can survive in moderately disturbed areas (Wake et al. 1980; IUCN SSC Amphibian Specialist Group 2020a).

Connectivity preservation is one of the expected benefits of protected natural areas; nevertheless, most of those protected areas are isolated by eroded habitat, which breaks with the connectivity of the landscape (Ward et al. 2020). In the Sierra Madre de Chiapas, most of the forest cover that surrounds the reserve core areas corresponds to traditional shade-grown coffee plantations (Schrot et al. 2009).

Such agroforestry systems maintain intermediate levels of connectivity for some vertebrate species (Ocampo et al. 2020); however, amphibians are more susceptible to habitat change. Likewise, low diversity of amphibian species has been reported in cultivated areas compared to primary forests (Whitfield 2016). The plethodontid salamanders have little resilience to the transformation and loss of forest cover (Díaz-García et al. 2019), although most of the agriculture takes place in the lowlands, we suppose that with global change these may increase in the highlands of the mountains, reducing or even displacing the areas inhabited by these salamanders. (Schrot et al. 2009).

Currently, the coverage of the protected natural areas is insufficient on a global scale, as 25% of amphibians are not distributed within the limits of protected natural areas (Nori et al. 2015). In Mexico, at least two out of five threatened species of salamanders do not inhabit protected natural areas (García-Bañuelos et al. 2019). Our results reinforce the above and show that a large portion of the habitat and potential corridors of both species is found outside the protected natural areas, mainly for B. lincolni. The connectivity corridors and current-flow centrality areas for this plethodontid in México are found in private natural areas, as there are no nature reserves under federal protection or large-scale conservation actors (WPDA, 2021), local communities’ actions could be effective for protecting significant habitats (Stachowiak et al. 2021). Our study shows the importance of forests for the salamanders’ distributions, and it identifies key areas for conservation through the biological corridor identification and potential distribution areas for the species studied. This data can be useful for reinforcement and/or conservation protection measures, such as the expansion of protected natural areas and restoration activities for amphibian conservation in areas with anthropogenic activities.

Acknowledgments

We are grateful to the editor and two anonymous reviewers for their comments. We thank Olivia Davis for the English review. R.B.H was on his Postdoctoral stay at UAEMex (CONACYT: I1200/94/2020). A.S received financial support from the Secretary of Research and Advanced Studies (SYEA) of the Universidad Autónoma del Estado de México (Grant: 4732/2019CIB).

Adriaensen F, Chardon JP, De Blust G, Swinnen E, Villalba S, Gulinck H, Matthysen E (2003) The application of ‘least-cost’modelling as a functional landscape model. Landsc Urban Plan 64(4):233–247.
Aiello-Lammens ME, Boria RA, Radosavljevic A, Vilela B, Anderson RP (2015) spThin: an R package for spatial thinning of species occurrence records for use in ecological niche models. Ecography 38(5):541–545.
Alroy J (2015) Current extinction rates of reptiles and amphibians. Proc Natl Acad Sci USA 112(42):13003–13008.
Baken EK, Mellenthin LE, Adams DC (2020) Macroevolution of desiccation-related morphology in plethodontid salamanders as inferred from a novel surface area to volume ratio estimation approach. Evolution 74(2):476–486.
Broennimann O, Fitzpatrick MC, Pearman PB, Petitpierre B, Pellissier L, Yoccoz NG, Thuiller W, Fortin MJ, Randin C, Zimmermann NE, Graham CH (2012) Measuring ecological niche overlap from occurrence and spatial environmental data. Glob Ecol Biogeogr 21(4):481–497.
Buchhorn M, Smets B, Bertels L, De Roo B, Lesiv M, Tsendbazar NE, Herold M, Fritz S (2020) Copernicus Global Land Service: Land Cover 100m: Collection 3 Epoch 2015, Globe. Version V3. 0.1)[Data set].
Ceballos G, de la Torre JA, Zarza H, Huerta M, Lazcano-Barrero MA, Barcenas H, Cassaigne I, Chávez C, Carreón G, Caso A, Carvajal S, García A, Morales JJ, Moctezuma O, Monroy-Vilchis O, Ruiz F, Torres-Romero EJ (2021) Jaguar distribution, biological corridors and protected areas in Mexico: from science to public policies. Landsc Ecol 36:3287–3309.
Crowther TW, Glick HB, Covey KR, Bettigole C, Maynard DS, Thomas SM, Smith JR, Hintler G, Duguid MC, Amatulli G, Tuanmu MN, Jetz W, Salas C, Stam C, Piotto D, Tavani R, Green S, Bruce G, Williams SJ, Wiser SK, Huber MO, Hengeveld GM, Nabuurs GJ, Tikhonova E, Borchardt P, Li CF, Powrie LW, Fischer M, Hemp A, Homeier J, Cho P, Vibrans AC, Umunay PM, Piao SL, Rowe CW, Ashton MS, Crane PR, Bradford MA (2015) Mapping tree density at a global scale. Nature 525:201–205.
Cushman SA, Landguth EL, Flather CH (2013) Evaluating population connectivity for species of conservation concern in the American Great Plains. Biodivers Conserv 22:2583–2605.
Di Cola V, Broennimann O, Petitpierre B, Breiner FT, D’Amen M, Randin C, Engler R, Pottier J, Pio D, Dubuis A, Pellissier L, Mateo RG, Hordijk W, Salamin N, Guisan A (2017) ecospat: an R package to support spatial analyses and modeling of species niches and distributions. Ecography 40:774–787.
Díaz-García JM, López-Barrera F, Toledo-Aceves T, Andresen E, Pineda E (2020) Does Forest restoration assist the recovery of threatened species? A study of cloud forest amphibian communities. Biol Conserv 242:108400.
Dirzo R, Young HS, Galetti M, Ceballos G, Isaac NJ, Collen B (2014) Defaunation in the Anthropocene. Science 345(6195):401–406.
Dutta T, Sharma S, McRae BH, et al (2016) Connecting the dots: mapping habitat connectivity for tigers in central India. Regional Environmental Change 16:53–67.
Elith J, Leathwick JR (2009) Species Distribution Models: Ecological Explanation and Prediction Across Space and Time. Annu Rev Ecol Evol Syst 40:677–697.
Fick SE, Hijmans RJ (2017) WorldClim 2: new 1-km spatial resolution climate surfaces for global land areas. Int J Climatol 37:4302–4315.
Giam X (2017) Global biodiversity loss from tropical deforestation. Proc Natl Acad Sci USA 114:5775.
Glor RE, Warren D (2011) Testing ecological explanations for biogeographic boundaries. Evolution 65:673–683.
Godínez-Gómez O, Schank C, Mas JF, Mendoza E (2020) An integrative analysis of threats affecting protected areas in a biodiversity stronghold in Southeast Mexico. Glob Ecol Conserv 24:e01297.
González-Espinosa M, Ramírez-Marcial N, Newton AC, Rey-Benayas JM, Camacho-Cruz A, Armesto JJ, Lara A, Premoli AC, Williams-Linera G, Altamirano A, Alvarez-Aquino C, Cortés M, Echeverría C, Galindo-Jaimes L, Muñiz-Castro MA, Núñez-Ávila MC, Pedraza RA, Rovere AE, Smith-Ramírez C, Thiers O, Zamorano C (2007) Restoration of Forest Ecosystems in Fragmented Landscapes of Temperate and Montane Tropical Latin America. In: Biodiversity Loss and Conservation in Fragmented Forest Landscapes: The Forests of Montane Mexico and Temperate South America. A. C. Newton eds, CABI, 416pp.
Grantham HS, Duncan A, Evans TD, Jones KR, Beyer HL, Schuster R, Walston J, Ray JC, Robinson JG, Callow M, Clements T, Costa HM, DeGemmis A, Elsen PR, Ervin J, Franco P, Goldman E, Goetz S, Hansen A, Hofsvang E, Jantz P, Jupiter S, Kang A, Langhammer P, Laurance WF, Lieberman S, Linkie M, Malhi Y, Maxwell S, Mendez M, Mittermeier R, Murray NJ, Possingham H, Radachowsky J, Saatchi S, Samper C, Silverman J, Shapiro A, Strassburg B, Stevens T, Stokes E, Taylor R, Tear T, Tizard R, Venter O, Visconti P, Wang S, Watson JEM (2020) Anthropogenic modification of forests means only 40% of remaining forests have high ecosystem integrity. Nat Commun 11:5978.
Grover MC (1998) Influence of Cover and Moisture on Abundances of the Terrestrial Salamanders Plethodon cinereus and Plethodon glutinosus. J Herpetol 32:489–497.
Guisan A, Tingley R, Baumgartner JB, Naujokaitis-Lewis I, Sutcliffe PR, Tulloch AIT, Regan TJ, Brotons L, McDonald-Madden E, Mantyka-Pringle C, Martin TG, Rhodes JR, Maggini R, Setterfield SA, Elith J, Schwartz MW, Wintle BA, Broennimann O, Austin M, Ferrier S, Kearney MR, Possingham HP, Buckley YM (2013) Predicting species distributions for conservation decisions. Ecol Lett 16:1424–1435.
Hilty JA, Keeley ATH, Lidicker Jr. WZ, Merenlender AM (2019) Corridor Ecology: Linking Landscapes for Biodiversity Conservation and Climate Adaptation (second edition 2019. Island Press, Washington, D.C., USA. 350pp.
Hu J, Jiang J (2018) Inferring ecological explanations for biogeographic boundaries of parapatric Asian mountain frogs. BMC Ecology 18:3.
IUCN SSC Amphibian Specialist Group (2020a) Bolitoglossa franklini. The IUCN Red List of Threatened Species 2020: e.T59163A53975747. https://dx.doi.org/10.2305/IUCN.UK.2020-2.RLTS.T59163A53975747.en. Downloaded on 18 september 2021.
IUCN SSC Amphibian Specialist Group (2020b) Bolitoglossa lincolni. The IUCN Red List of Threatened Species 2020: e.T59173A53976133. https://dx.doi.org/10.2305/IUCN.UK.2020-3.RLTS.T59173A53976133.en. Downloaded on 07 October 2021.
Jetz W, Wilcove DS, Dobson AP (2007) Projected Impacts of Climate and Land-Use Change on the Global Diversity of Birds. PLOS Biol 5:e157.
Kearney M, Porter W (2009) Mechanistic niche modelling: combining physiological and spatial data to predict species’ ranges. Ecol Lett 12:334–350.
Koen EL, Garroway CJ, Wilson PJ, Bowman J (2010) The effect of map boundary on estimates of landscape resistance to animal movement. PLOS ONE 5:e11785.
Kozak KH (2017) What Drives Variation in Plethodontid Salamander Species Richness over Space and Time? Herpetologica 73:220–228.
Kozak KH, Wiens JJ (2012) Phylogeny, ecology, and the origins of climate–richness relationships. Ecology 93:S167–S181.
Lips KR (2008). Decline of a Tropical Montane Amphibian Fauna. Conservation Biology 12:106–117.
Lowe WH, Mcpeek MA, Likens GE, Cosentino BJ (2008) Linking movement behaviour to dispersal and divergence in plethodontid salamanders. Mol Ecol 17:4459–4469.
Magaña V, Amador JA, Medina S (1999) The Midsummer Drought over Mexico and Central America. J Clim 12:1577–1588.
McRae BH (2006) Isolation by resistance. Evolution 60:1551–1561.
McRae BH, Dickson BG, Keitt TH, Shah VB (2008) Using circuit theory to model connectivity in ecology, evolution, and conservation. Ecology 89:2712–2724.
McRae BH, Kavanagh DM (2011) Linkage mapper connectivity analysis software. The Nature Conservancy. https://code.google.com/p/linkage-mapper/
Muscarella R, Galante PJ, Soley-Guardia M, Boria RA, Kass JM, Uriarte M, Anderson RP (2014) ENMeval: An R package for conducting spatially independent evaluations and estimating optimal model complexity for Maxent ecological niche models. Methods Ecol Evol 5:1198–1205.
Naimi B, Hamm NAS, Groen TA, Skidmore AK, Toxopeus AG (2014) Where is positional uncertainty a problem for species distribution modelling? Ecography 37:191–203.
Nowakowski AJ, Veiman-Echeverria M, Kurz DJ, Donnelly MA (2015) Evaluating connectivity for tropical amphibians using empirically derived resistance surfaces. Ecol Appl 25:928–942.
Ocampo MCE, Castillo Santiago MÁ, Ochoa-Gaona S, Enríquez PL, Sibelet N (2019) Assessment of Habitat Quality and Landscape Connectivity for Forest-Dependent Cracids in the Sierra Madre del Sur Mesoamerican Biological Corridor, México. Trop Conserv Sci 12:1940082919878827.
Olson DM, Dinerstein E, Wikramanayake ED, Burgess ND, Powell GVN, Underwood EC, D’amico JA, Itoua I, Strand HE, Morrison JC, Loucks CJ, Allnutt TF, Ricketts TH, Kura Y, Lamoreux JF, Wettengel WW, Hedao P, Kassem KR (2001) Terrestrial Ecoregions of the World: A New Map of Life on Earth: A new global map of terrestrial ecoregions provides an innovative tool for conserving biodiversity. BioScience 51:933–938.
OpenStreetMap (2020) Disponible: (http://www.openstreetmap.org/),
Parra-Olea G, García-París M, Wake DB (2004) Molecular diversification of salamanders of the tropical American genus Bolitoglossa (Caudata: Plethodontidae) and its evolutionary and biogeographical implications. Biol J Linn Soc 81:325–346.
Pelletier D, Clark M, Anderson MG, Rayfield B, Wulder MA, Cardille JA (2014) Applying Circuit Theory for Corridor Expansion and Management at Regional Scales: Tiling, Pinch Points, and Omnidirectional Connectivity. PLOS ONE 9:e84135.
Peterson AT (2006) Ecologic Niche Modeling and Spatial Patterns of Disease Transmission. Emerg Infect Dis 12:1822–1826.
Peterson AT (2011) Ecological niche conservatism: a time-structured review of evidence. J Biogeogr 38:817–827.
Phillips SJ, Dudík M. (2008) Modeling of species distributions with Maxent: new extensions and a comprehensive evaluation. Ecography 31(2):161–175.
Raxworthy CJ, Martinez-Meyer E, Horning N, Nussbaum RA, Schneider GE, Ortega-Huerta MA, Townsend Peterson A (2003) Predicting distributions of known and unknown reptile species in Madagascar. Nature 426:837–841.
Ray N, Lehmann A, Joly P (2002) Modeling spatial distribution of amphibian populations: a GIS approach based on habitat matrix permeability. Biodivers Conserv 11:2143–2165.
Rödder D, Engler JO (2011) Quantitative metrics of overlaps in Grinnellian niches: advances and possible drawbacks. Glob Ecol Biogeogr 20:915–927.
Rovito SM, Parra-Olea G (2016) Neotropical Plethodontid Biogeography: Insights from Molecular Phylogenetics. Copeia 104:222–232.
Rovito SM, Parra-Olea G, Recuero E, Wake DB (2015) Diversification and biogeographical history of Neotropical plethodontid salamanders. Zool J Linn Soc 175:167–188.
Rovito SM, Parra-Olea G, Vásquez-Almazán CR, Papenfuss TJ, Wake DB (2009) Dramatic declines in neotropical salamander populations are an important part of the global amphibian crisis. Proc Natl Acad Sci USA 106:3231.
Scheele BC, Pasmans F, Skerratt LF, Berger L, Martel A, Beukema W, Acevedo AA, Burrowes PA, Carvalho T, Catenazzi A, De la Riva I, Fisher MC, Flechas SV, Foster CN, Frías-Álvarez P, Garner TWJ, Gratwicke B, Guayasamin JM, Hirschfeld M, Kolby JE, Kosch TA, La Marca E, Lindenmayer DB, Lips KR, Longo AV, Maneyro R, McDonald CA, Mendelson J, Palacios-Rodriguez P, Parra-Olea G, Richards-Zawacki CL, Rödel MO, Rovito SM, Soto-Azat C, Toledo LF, Voyles J, Weldon C, Whitfield SM, Wilkinson M, Zamudio KR, Canessa S (2019) Amphibian fungal panzootic causes catastrophic and ongoing loss of biodiversity. Science 363:1459.
Schoener TW (1970) Nonsynchronous spatial overlap of lizards in patchy habitats. Ecology 51:408–418.
Schroth G, Laderach P, Dempewolf J, Philpott S, Haggar J, Eakin H, Castillejos T, Garcia Moreno J, Soto Pinto L, Hernandez R, Eitzinger A, Ramirez-Villegas J (2009) Towards a climate change adaptation strategy for coffee communities and ecosystems in the Sierra Madre de Chiapas, Mexico. Mitig Adapt Strateg Glob Chang 14:605–625.
Sexton JP, McIntyre PJ, Angert AL, Rice KJ (2009) Evolution and Ecology of Species Range Limits. Annu Rev Ecol Evol Syst 40:415–436.
Smith AM, Green DM (2005) Dispersal and the metapopulation paradigm in amphibian ecology and conservation: are all amphibian populations metapopulations? Ecography 28:110–128.
Stachowiak C, Crain BJ, Kroetz K, et al (2021) Protected Areas Established by Local Communities through Direct Democracy Encompass Habitat for Species as Effectively as Protected Areas Planned over Large Spatial Scales. Environmental Management 67:242–250. https://doi.org/10.1007/s00267-020-01403-4
Stuart SN, Chanson JS, Cox NA, Young BE, Rodrigues ASL, Fischman DL, Waller RW (2004) Status and Trends of Amphibian Declines and Extinctions Worldwide. Science 306:1783.
Sunny A, López-Sánchez M, Ramírez‐Corona F, Suárez‐Atilano M, González‐Fernández A (2022) Genetic diversity and functional connectivity of a critically endangered salamander. Biotropica 54(1):42–56.
Taylor PD, Fahrig L, Henein K, Merriam G (1993) Connectivity Is a Vital Element of Landscape Structure. Oikos 68:571–573.
Thompson PL, Rayfield B, Gonzalez A (2017) Loss of habitat and connectivity erodes species diversity, ecosystem functioning, and stability in metacommunity networks. Ecography 40:98–108.
Tischendorf L, Fahrig L (2000) On the Usage and Measurement of Landscape Connectivity. Oikos 90:7–19.
Title PO, Bemmels JB (2017) ENVIREM: an expanded set of bioclimatic and topographic variables increases flexibility and improves performance of ecological niche modeling. Ecography 41:291–307.
Vargas-Jaimes J, González-Fernández A, Torres-Romero EJ, Bolom-Huet R, Manjarrez J, Gopar-Merino F, Pacheco XP, Garrido-Garduño T, Chávez C, Sunny A (2021) Impact of climate and land cover changes on the potential distribution of four endemic salamanders in Mexico. Journal for Nature Conservation 64:126066.
Vásquez-Almazán CR, Rovito SM (2014) A New Species of Black Bolitoglossa (Caudata: Plethodontidae) from Guatemala. J Herpetol 48:518–524
Velo-Antón G, Parra JL, Parra-Olea G, Zamudio KR (2013) Tracking climate change in a dispersal-limited species: reduced spatial and genetic connectivity in a montane salamander. Mol Ecol 22:3261–3278.
Wake D, Lynch JF (1976) The distribution, ecology, and evolutionary history of plethodontid salamanders in tropical America. Nat Hist Mus Los Angeles County Sci Bull 25:1–65.
Wake DB, Lynch JF (1982) Evolutionary Relationships among Central American Salamanders of the Bolitoglossa franklini Group, with a Description of a New Species from Guatemala. Herpetologica 38:257–272.
Wake DB, Yang SY, Papenfuss TJ (1980) Natural Hybridization and Its Evolutionary Implications in Guatemalan Plethodontid Salamanders of the Genus Bolitoglossa. Herpetologica 36:335–345.
Ward M, Saura S, Williams B, Ramírez-Delgado JP, Arafeh-Dalmau N, Allan JR, Venter O, Dubois G, Watson JEM (2020) Just ten percent of the global terrestrial protected area network is structurally connected via intact land. Nat Commun 11:4563.
Warren DL (2012) In defense of ‘niche modeling.’ Trends Ecol Evol 27:497–500.
Warren DL, Cardillo M, Rosauer DF, Bolnick DI (2014) Mistaking geography for biology: inferring processes from species distributions. Trends Ecol Evol 29:572–580.
Warren DL, Glor RE, Turelli M (2008) Environmental niche equivalency versus conservatism: quantitative approaches to niche evolution. Evolution 62:2868–2883.
Warren DL, Seifert SN (2011) Ecological niche modeling in Maxent: the importance of model complexity and the performance of model selection criteria. Ecol Appl 21:335–342.
Whitfield SM, Lips KR, Donnelly MA (2016) Amphibian Decline and Conservation in Central America. Copeia 104:351–379.
Wickham H (2016) ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York. ISBN 978-3-319-24277-4
Wiens JA, Stralberg D, Jongsomjit D, Howell CA, Snyder MA (2009) Niches, models, and climate change: Assessing the assumptions and uncertainties. Proc Natl Acad Sci USA 106:19729.
WPDA: The World Database on Protected Areas. Available: https://www.protectedplanet.net/en/thematic-areas/wdpa?tab=WDPA

No competing interests reported.

Supportinginformation15032022.docx

Download PDF

Editorial decision: Major revision
13 Jul, 2022
Reviews received at journal
04 Jul, 2022
Reviewers agreed at journal
21 Jun, 2022
Reviewers invited by journal
24 May, 2022
Editor assigned by journal
01 May, 2022
Submission checks completed at journal
29 Apr, 2022
First submitted to journal
28 Apr, 2022

You are reading this latest preprint version

Potential distribution and connectivity in two plethodontid salamanders: conservation areas and landscape corridors for two endemic species

Status:

Version 1

Abstract

Figures

Introduction

Methods

Study area

Species occurrence and environmental variables

Species distribution models

Similarity, equivalence, and niche overlap

Connectivity models

Results

Species distribution models

Similarity, equivalence, and niche overlap

Connectivity analysis

Discussion

Implications for conservation

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1