Mountains as key areas for Carnivore Connectivity in Neotropical Grasslands

doi:10.21203/rs.3.rs-4958975/v1

Download PDF

Research Article

Mountains as key areas for Carnivore Connectivity in Neotropical Grasslands

https://doi.org/10.21203/rs.3.rs-4958975/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Habitat loss and fragmentation threaten biodiversity, particularly for carnivores whose dispersion and population viability are compromised by reduced available habitat and anthropic elements in the landscape, such as roads and crops. The world's grasslands have experienced considerable degradation and replacement by crops and are currently limited to natural patches scattered throughout the region. In Argentina, the last remaining grassland remnants that persist are associated within the Pampean mountain systems, which act as crucial refuges for the species that inhabit them in the region. Our objectives were to identify and map priority sites and areas that can act as ecological corridors between highland grassland patches of high conservation priority. We performed connectivity analyses using Least-cost Path Models and Circuit Theory. To do this, we generated habitat suitability models by combining environmental and anthropic variables, from which resistance surfaces were generated. We highlight areas of high habitat suitability for carnivores in the Pampas ecoregion, with particular emphasis on highland grassland. We identified potential corridors and least-cost paths for five carnivore species, obtaining multi-species corridors which highlight the importance of landscape connectivity for maintaining healthy populations. Centrality analyses revealed crucial grassland remnants and valuable corridors. These findings address the challenges associated with habitat fragmentation in the Pampas ecoregion and provide guidelines for future research and carnivore conservation actions.

connectivity analysis

grasslands

carnivores

corridor design

Pampas

Habitat loss and ecosystem fragmentation are considered the main threats to biodiversity (Pereira 2010; Rands et al. 2010; Yuan et al. 2024). As a result of these processes, native habitats are restricted to patches surrounded by a matrix of agricultural lands or other land uses that, to a greater or lesser extent, cause changes in both the physical environment and the populations that inhabit them (Saunders et al. 1991). In response to these threats, various strategies and tools have been proposed to help curb the loss and fragmentation of natural habitats, such as the creation of extensive protected natural areas or the implementation of policy mechanisms based on economic incentives (Sims 2014; Batáry et al. 2015; de Vries and Hanley 2016; Timmers et al. 2022). However, at present, a large portion of terrestrial ecosystems have already been degraded and fragmented by human activity (Ellis et al. 2010; Venter et al. 2016), and as such, strategies aimed at maintaining, strengthening, or restoring connectivity in already affected ecosystems have become increasingly relevant in recent years (Grass et al. 2019; Hilty et al. 2021).

Carnivores play a crucial role in ecosystem functioning by regulating community structure and dynamics through predation pressure and competition (Terborgh and James 2010). In fragmented environments, the configuration and characteristics of natural habitat patches determine the persistence of populations within them, and the effects this has on species can vary depending on the studied group of species. For carnivores, the size of natural patches may sometimes be too small to support stable populations with large spatial requirements (Buskirk and Gittleman 1990), and the degree of isolation within patches and the quality of the surrounding matrix may also have effects on their occupancy (Prugh et al. 2008). On the other hand, landscape alterations generated by human action create linear elements such as transition zones between natural habitats and crops, fences, and roads that can negatively affect carnivore species. Fences and roads can act as barriers to individual movement (Fahrig and Rytwinski 2009; Cozzi et al. 2013), affecting dispersal and gene flow within populations (Vaeokhaw et al. 2020), and roads, in particular, increase mortality through collisions and can generate stress to species due to noise pollution and visual stimuli (Červinka et al. 2015). The connectivity analysis and corridor design between natural habitat areas have become increasingly important in recent years, primarily in addressing conservation issues affecting carnivores (Castilho et al. 2015; Rio-Maior et al. 2019).

In Argentina, Neotropical grasslands comprise approximately 398,966 km² of what is known as the Pampas ecoregion (Matteucci, 2012). Over the last century, this region has witnessed significant agricultural development, leading to a simplified and homogenized landscape, with natural grasslands being reduced to scattered patches within a matrix consisting of various land uses (Soriano, 1992). In this ecoregion, Bilenca and Miñarro (2004) have identified around 30 Valuable Grassland Areas, which are large natural grasslands considered to be in good conservation status. Several of these Valuable Grassland Areas are located towards the southeast of the ecoregion, within the Tandilia mountain system, where grasslands persist as authentic relics associated with peri-mountain plains, where rocky outcrops and shallow soils prevent cultivation (Herrera and Laterra 2011). These grasslands now cover only around 5% of their original surface area due to the high degree of agricultural transformation and serve as a last refuge for the different species that inhabit the area (Bilenca and Miñarro 2004; Valicenti et al. 2010; Velasco et al. 2013; Vera et al. 2021; Aranguren et al. 2023; Trofino-Falasco et al. 2023)

Five species of carnivores are relatively common in the Pampas ecoregion: Pampas fox (Lycalopex gymnocercus), Geoffroy's cat (Leopardus geoffroyi), Molina's Hog-nosed Skunk (Conepatus chinga), Lesser grison (Galictis cuja), and Puma (Puma concolor) (Trofino-Falasco et al. 2023). All these species are categorized as Least Concern, both nationally (Secretaría de Ambiente y Desarrollo Sustentable de la Nación y Sociedad Argentina para el Estudio de los Mamíferos 2019) and internationally (IUCN, 2024), and their distribution ranges are wide. However, within the Tandilia mountain system, they could be experiencing different pressures at the local level that could endanger their persistence. Given the region´s fragmentation and degradation level and the relevance of grassland patches as persisting refuges for species, it is necessary to consider strategies that help conserve and promote connectivity between these natural areas. In this frame, our objectives were to identify and map priority sites and areas that can act as ecological corridors between grassland patches to contribute to the connectivity knowledge of the Tandilia mountain system.

1. Study Area

The Tandilia mountain system is located in the southeastern Pampas ecoregion. It runs along a northwest-southeast diagonal, covering an extension of 350 km and a maximum width of 50–60 km. The region is characterized by a subhumid to humid temperate climate, with a marked seasonality in temperature and an average annual precipitation that ranges from 800 to 850 mm (Burgos and Vidal 1951; Valicenti et al. 2010; Falasca et al. 2000; Echeverría et al. 2017).

In this area, highland grasslands remnants are mainly associated with hills and hillocks that rise above 50 and 500 m over the Pampas plains (Cingolani 2011), and exist in areas where soil characteristics such as rocky outcrops and shallow substrate depth have hindered the advancement of agriculture (Herrera and Laterra 2007), allowing the persistence of native grassland. These highland grasslands are composed of several species of forage value that form tussocks and grasslands (Paspalum, Festuca, Poa, Stipa, among others), thistle species (Eryngium), as well as shrub species (Colletia paradoxa, Baccharis spp) (Herrera et al. 2019). Initially, grassland habitat also extended into the mountain valley areas, which currently have been replaced by pastures, forestation, and crops, forming a matrix that covers approximately 71% of the system's surface area, with soybeans, corn, sunflower, and winter cereals being the predominant crops (de Abelleyra 2023). Associated with this matrix, we also find a network of elements, such as roads and urban and rural centers, that help define the landscape along the mountain system. The entire area is connected by 297 km northeast-southeast route, passes through the main urban centers and, along with other provincial routes, forms a network of paved roads located within the mountain system (Fig. 1)

2. Data analysis

We utilized a methodology consisting of several stages of analysis (summarized in Fig. 2) to map the corridors and identify the contribution of landscape elements to connectivity. In the first stage, we selected and analyzed environmental variables (carnivore occurrence data, alongside bioclimatic variables, land cover) and anthropogenic variables (roads and urban centers) to construct the ambients factors, which resulted in the base inputs for developing suitability models to the Pampas ecoregion. Then, we obtained the resistance surfaces of the Tandilia mountain system, and finally, based on the obtained models, connectivity analyses were conducted using least-cost path models and circuit theory.

2.1 Variable analysis and environmental factor construction

We can define the habitat of a species based on the biotic and abiotic elements of the landscape that individuals use (e.g., food, cover, refuge, etc.), assuming that these elements are what they require to survive, reproduce, and move through the landscape matrix (Beier et al. 2008). Using GIS and other tools, habitat suitability models relate suitability to raster layers representing elements of the available environment in this format (land cover, distance to roads, elevation, e.g.), and refer to these layers as environmental factors (Beier et al. 2007).

We defined four environmental factors (Fig. 2) to develop habitat models for each species, partially following the methodology proposed by Gonzales Saucedo et al. (2011): 1) A climate suitability model obtained from a Maximum Entropy (MaxEnt) analysis with 19 bioclimatic variables; 2) A habitat use model, which analyzes the use versus availability of land cover; 3) Distance to roads and urban centers, two variables associated with anthropogenic disturbances that are important in the case of carnivores (Ordeñana et al. 2010).

2.1.1 Occurrence data of focal species

When modeling corridors, it is essential to consider several focal species, especially those that ensure the functioning of ecological processes, within patches of natural areas and the landscape matrix. Species that can act as umbrella species and safeguard many other species are also ideal for designing corridors (Beier et al. 2008). Taking this into account, we worked with the five native carnivore species for corridor modeling in the Tandilia mountain system: Lycalopex gymnocercus, Leopardus geoffroyi, Conepatus chinga, Galictis cuja, and Puma concolor. These species have a wide distribution and occur throughout the Pampas ecoregion, so we decided to work with occurrence data from the entire ecoregion to generate the modeling inputs. Therefore, we constructed an occurrence database for each species based on records obtained from camera trap surveys in the Tandilia mountain system and with data available in online databases for the rest of the Pampas ecoregion.

We obtained camera trap records from a monitoring project for vertebrate species associated with remnants of highland grasslands. These surveys were conducted between September 2016 and May 2022 in 20 rural properties on remnants of grassland and surrounding areas, covering 50,738 ha (31 remnants). We installed 248 camera trap stations, which remained active for an average period of 21 days, accumulating a sampling effort of 5,308 trap days.

Regarding the data for the rest of the Pampas ecoregion, we used each species´ available records present in the Global Biodiversity Information Facility (GBIF) for the Pampas ecoregion. We obtained these records using the GBIF Occurrences plugin for QGIS software (https://doi.org/10.15468/dd.p8neyx). This tool allows downloading all available records in the online database with geographic coordinates, generating a point layer with the occurrences. This layer was subsequently processed, removing duplicate, invalid, and records whose coordinates coincide with urban centers.

2.1.2. Climate Suitability Models

We constructed climate suitability models based on the Ecological Niche Modeling (ENM) approach offered by MaxEnt, a maximum entropy modeling tool. Based on presence data, this method estimates functions that relate environmental variables and habitat suitability to approximate each species’ niche and potential geographic distribution (Phillips et al. 2006). We worked with 19 raster layers corresponding to bioclimatic variables (Table 1, Supplementary Information 1), with a resolution of 30 seconds (∽ 1 km²), obtained from the WorldClim platform (Fick and Hijmans et al. 2017), which represent annual trends, seasonality, and extreme or limiting environmental factors. To avoid correlation between variables, we performed a Principal Component Analysis (PCA), thus reducing their dimensionality. We worked with the Principal Components (PC), which accumulated over 90% of the data variability. Since these are orthogonal, meaning independent of each other, they are not correlated, making them an excellent alternative to use as predictor variables in the ENM (Cruz-Cárdenas et al. 2014).

Furthermore, to avoid spatial autocorrelation among occurrence records, we spatially filtered the data corresponding to each species based on their home range, using the spThin package in R (Aiello-Lammens et al. 2015). We filtered small and medium-sized species records at a distance of 5 km, while we filtered puma data at 15 km.

We performed niche models using the kuenm package in R (Cobos et al. 2019), which utilizes MaxEnt (maximum entropy) as the modeling algorithm. The package allows the creation of several candidate models by testing different combinations of parameters and evaluating them. We combined six regularization multipliers (0.1, 0.5, 1, 2, 3, 4) and all possible combinations of the five features available in MaxEnt (linear, quadratic, product, threshold, and hinge). The models obtained were evaluated based on their statistical significance (partial ROC), prediction capacity (omission rate), and model complexity (AICc). The final model was obtained by averaging 30 bootstrap models, and habitat suitability was evaluated from the raw output values of ROR. This MaxEnt modeling output directly measures habitat suitability without making assumptions about species prevalence or detection probabilities (Merow et al. 2013).

Since the niche models obtained include all sites with the same bioclimatic conditions under which each species has been recorded, the geographic distribution of the species may be overrepresented (Illoldi-Rangel and Escalante 2008). To avoid this, we applied a threshold at the 10th percentile to omit all regions with habitat suitability lower than the suitability values for the lowest 10% of occurrence records. It assumes that 10% of occurrence records within the least suitable habitats do not occur in sites that represent the species' general habitat and, therefore, should be omitted (Morrow 2019). Thus, we obtained a binary map for each species, representing their potential distribution within the Pampas ecoregion.

From these binary maps, we constructed the final climate suitability maps using the distance to centroid niche (DCN) approach (González Saucedo 2011). Under this approach, it is assumed that the ideal ecological conditions for a species are found at the centroid of its multivariate niche in environmental space and that as conditions move away from this centroid, environmental suitability decreases (Yañez et al. 2020). We calculated the centroid of the climate niche as the mean of each of the four PCs used for modeling (Martínez-Meyer et al. 2013). Then, using R software, we calculated the Euclidean distance between the niche centroid and the value of each pixel within the potential distribution area. The distance raster layers obtained were linearly rescaled between zero and one, with zero assigned to the furthest value and one to the closest value from the centroid of the climatic niche (representing the lowest and highest suitability, respectively).

2.1.3. Analysis of land cover use versus availability

To include the land cover characteristics that contribute to the presence of each species in the landscape, we conducted a habitat use versus availability analysis. For this, we obtained land cover information for the Pampas ecoregion from the MapBiomas Pampa Trinacional Collection 2.0 (Baeza et al. 2022), with a resolution of 30x30 m, which includes annual data on land cover and land use for the period from 1985 to 2021. We reclassified the original raster layer summarizing into seven categories by using the QGIS raster calculator (Supplementary Information 1, Table 2). On the other hand, since the land cover category "Areas without vegetation" includes both urban areas and other vegetation-free areas (e.g., rocky areas), we decided to differentiate this category. Firstly, we obtained a vector layer containing polygons representing all urban areas in the Pampas ecoregion from the National Geographic Institute of the Argentine Republic (IGN, https://www.ign.gob.ar/). Then, we overlaid the raster layer of land cover using the Rasterize tool in QGIS, categorizing pixels coinciding with the polygons as Urban. Subsequently, we excluded this category from our analysis by assigning no data to these pixels. For corridor and connectivity analysis, we considered urban areas to be total barriers to movement.

Once the land cover layer was processed, we conducted a habitat use versus availability analysis for each species. From all occurrence data and using QGIS, we extracted the land cover information for each record and calculated the frequency of records for each type of land cover for each species. The frequencies of records in each category correspond to the habitat used by the species. Then, we made a buffer around each record based on the reported home range size for each species (Lucherini and Luengos Vidal 2008; Castillo 2010; Manfredi et al. 2011; Elbroch and Wittmer 2012; Luengos Vidal et al. 2016), clipped the land cover layer with these buffers, and calculated the proportion of each land cover type within the home ranges of each species. We calculated the expected frequency (availability) of use from these values for each category. Then, with these observed and expected frequency values, we conducted a Chi-square goodness-of-fit test to evaluate if there are significant differences between what the species use and what is available. In cases where expected frequencies were less than five, we performed a Yates' correction, and for the puma, whose records were low, we grouped the land cover type categories to perform the Chi-square test.

To assess the use that species make of different land cover types, we calculatedconfidence intervals using the Bonferroni method with the HaviStat v2.4 program (Neu et al. 1974; Montenegro et al. 2014) to identify whether species prefer or avoid any particular land cover type or whether they use land cover based on availability. Once we determined each species’ use of different land cover types, we linearly rescaled the usage values of different categories between − 1 (avoided cover) and 1 (selected cover).

2.1.4. Distance to road and urban centers

Using the Distance Accumulation tool in ArcGIS Pro 3.1, we created a layer of distance to roads from a vector layer of paved roads provided by the IGN (IGN, https://www.ign.gob.ar/) and a layer of distance to urban centers from the previously used vector layer of urban centers. Once we obtained the layers with distance values, we linearly scaled the values between 0 (corresponding to pixels with the shortest distance to roads) and 1 (corresponding to the cell with the farthest distance to roads).

2.2. Habitat suitability models for the Pampas ecoregion

Using GIS tools, we combined rescaled layers to build habitat-suitability models for each species. The values obtained for each pixel include climatic suitability, land cover preference, and anthropogenic disturbances (distance to roads and urban centers). This procedure gives us maps with values between − 1 and 4, where the high values reflect the best habitat conditions available and the low values the most unfavorable conditions for the species.

2.3. Modeling of connectivity in the Tandilia System

To analyze connectivity, we used circuit theory (McRae et al. 2008) and least-cost modeling (Adriaensen et al. 2003), as both approaches have been employed in connectivity analysis and have proven to be useful (LaRue and Nielsen 2008; Carroll et al. 2012; Etherington 2016; Belote et al. 2016). To apply them, we reduced the scale of analysis and focused on the Tandilia mountain system, where remnants of grassland persist within an agricultural matrix. We manually identified and delimited these remnants based on remote sensing imagery using GIS tools, and selected grassland remnants with areas larger than 500 ha to connect and analyze. Finally, we delimited the matrix area to be analyzed by applying a 15 km buffer around all mapped remnants.

2.3.1. Resistance surfaces

Cost-distance and circuit theory models require a resistance surface representing the relative effort required for an organism to occupy a pixel on a map (Wade et al. 2015). We understand resistance as the magnitude with which a pixel facilitates or limits the movement of organisms through it (Spear et al. 2010). We also assume that suitability is synonymous with habitat permeability, i.e., the degree to which the landscape allows the passage of organisms or ecological processes (Singleton et al. 2002) and is inversely related to resistance (Beier et al. 2007; Pullinger and Johnson 2010). These models can be thought as a set of scores where the end of the scale reflects low resistance, i.e., high habitat quality, and the other end high resistance, i.e., low habitat quality (Beier et al. 2008). With this in mind, we constructed a resistance layer by linearly scaling between 1 and 100 the suitability layer and applying an inverse linear function. This resulted in a resistance map for each species, where values of 1 in cells with lower resistance correspond to cells with higher values in the suitability layer (value of 4), and 100 in cells with the highest resistance correspond to the worst suitability values for the species (value of -1).

2.3.2 Corridors Identification

To identify and map areas that could serve as ecological corridors between grassland remnants, we assessed connectivity in the Tandilia mountains using least-cost models (Adriaensen et al. 2003). We calculated cost-weighted distances between patches (core areas) from a resistance surface and then identified and mapped the least-cost paths between areas based on these values using Linkage Mapper ver. 3.1 (McRae and Kavanagh 2011). This program allows corridor identification based on a maximum geographic distance or a cost-weighted distance, among other criteria, to include species dispersal in the modeling. We chose not to limit and overestimate corridor mapping, as our goal is not to map dispersal pathways but to identify areas that may promote connectivity. We generated an integrated corridor map for all carnivore species from the five individual corridor models. In order to do this, we reclassified the values of each species model into deciles ranging from 1 to 10 and then summed them. We obtained an integrated corridor map with values ranging from 5 (low quality and higher travel costs for all the species together) to 50 (higher quality and lower travel costs) (Belote et al. 2016).

2.3.3. Centrality Analysis

We estimated the centrality values for each species model. One way to analyze landscape connectivity is to assume that the landscape acts as an electrical circuit. Since a circuit's connectivity increases with the number of connections, distance metrics based on electrical connectivity apply to processes that respond positively to increased connections and redundancy. The relationship between current, voltage, and resistance with random walks in circuits makes it possible to link this approach to movement ecology, providing concrete ecological interpretations of parameters and predictions from circuit theory (McRae et al. 2008). Under this approach, the Centrality Mapper tool within Linkage Mapper (McRae 2012), calculates the "current flow centrality" through a network of corridors and nodes (patches). The centrality values calculated for network components indicate how important a link or node is to maintaining overall connectivity.

We compile an occurrence database (Supplementary Information 2) consisting of 391 records for L. gymnocercus, 140 for L. geoffroyi, 132 for C. chinga, 95 for G. cuja, and 30 for P. concolor. After spatial filtering, 162, 77, 66, 63, and 15 records remained, respectively, and these were used to generate the climatic niche models. As a result of the PCA performed with the bioclimatic variables, we constructed four raster layers corresponding to the first four principal components that accumulated more than 90% of the original variation of the dataset for modeling in MaxEnt.

By modeling climatic ecological niches, we obtained 930 candidate niche models (186 for each species). We selected the best model for each species, considering statistical significance, a low omission rate, and the Delta AICc (Supplementary Information 1, Table 3).

We obtained habitat suitability models for the Pampas ecoregion (Fig. 3). The values obtained in each model indicated that the southeast region for the Pampas ecoregion (Fig. 3) contains good habitat suitability for the five studied carnivore species. We elaborated five resistance surfaces from each suitability model for the Tandilia mountain system, corresponding to each species (Fig. 4). In resistance surfaces, we observed areas with high resistance values in the intermountain areas and patches with low resistance values.

The selection of areas to connect resulted in 46 highland grassland remnants covering approximately 81,962 ha of grassland surface, meaning that we worked with 68% of the total grassland area that persists in the Tandilia mountain system.

From the connectivity analysis, we obtained a cost surface showing all corridor alternatives for Tandilia mountains and the least-cost paths (LCP) for all five species: 125 paths for L. gymnocercus, 122 paths for L. geoffroyi, 125 paths for C. chinga, 127 paths for G. cuja, and 111 paths for P. concolor (Fig. 5). The sum of the corridors resulted in a cost surface considering all five species at once (Fig. 6). The centrality analysis allowed us to identify those grassland remnants in the Tandilia mountain system that contributed most to the overall landscape connectivity (Fig. 5) and those LCPs that were most valuable for maintaining high connectivity (Fig. 5).

Our work represents a first approach to designing corridors as conservation alternatives for carnivore species and all the ecosystem processes accompanying them in the Pampas grasslands. Carnivores inhabit a wide range of habitats with good tolerance to disturbed environments (Donadio et al. 2001; Yensen and Tarifa 2003; Lucherini and Luengos Vidal 2008; Tellaeche et al. 2014; Guidobono et al. 2016; De Lucca and Chimento 2020; Zanón Martínez et al. 2022) and generally show high dispersal capacity (Boitani and Powell 2012), allowing individuals to move across the landscape searching for suitable habitats. Several studies have considered these characteristics and worked with carnivores to create connectivity networks, model habitats, select conservation areas, and assess functional connectivity across large landscapes (Huck et al. 2010; González Saucedo 2011; Crooks et al. 2011; Zemanova et al. 2017). In Argentina, work on modeling and designing ecological corridors is scarce and has focused only on forest environments such as the Chaco and the Misiones jungle (Torrella et al. 2018; DeMatteo et al. 2023; Mereles et al. 2023). On the other hand, although the Pampas grassland has been the focus of studies and connectivity analysis for some authors (Di Giacomo et al. 2010; Herrera and Laterra 2011; Herrera et al. 2017; Gandini et al. 2019), to date, the design of ecological corridors has not been addressed as a conservation alternative for these highly altered and fragmented ecosystems. In this work, we obtained habitat suitability models for the Pampas ecoregion and multispecies corridors for the five carnivorous species reported in the study area (Barquez et al. 2006; Aranguren et al. 2023; Trofino-Falasco et al. 2023).

Habitat suitability models as input for the design of ecological corridors allow species-specific requirements to be incorporated into the connectivity analysis, overcoming the purely structural approach, which is why they have gained some relevance in recent years (Huck et al. 2010; Dong et al. 2021; Dutta et al. 2022). Our study selected variables relevant to the region's carnivores and generated habitat suitability models for the Pampas ecoregion, identifying areas with high and low suitability. Since previous studies have shown that the distribution of mammal species is strongly influenced by vegetation cover and topography (Holland and Bennett 2007; Heikinheimo et al. 2012; Ang’ila et al. 2023), it is understandable that, in an environment as altered as the Pampas grassland, the highest fitness values are found in areas where natural vegetation persists. Our results indicate that large areas towards the southeast of the ecoregion have high fitness values for the five studied carnivore species. The high values of habitat suitability observed here could be associated with the persistence of natural grasslands, which have not been replaced by other uses, such as crops, given their soil conditions. On the one hand, the low, flood-prone soils of the Flooding Pampas, and on the other, the rocky soils of the Tandilia and Ventania mountain system in the Southern Pampas have limited the agriculturalization process in these regions (Bilenca and Miñarro 2004; Bilenca et al. 2008). However, it is essential to understand that despite persisting, these grasslands, like all or the vast majority of Pampean habitats, are subject to the pressures and modifications that humans have exerted throughout the region in the last century (Bilenca et al. 2012). In the Tandilia mountain system, natural grasslands have been replaced by crops, pastures, and forestations, mainly within inter-mountain and lowland areas (Bilenca and Miñarro 2004; Bilenca et al. 2008; Herrera et al. 2019). Taking this into account and observing in detail the northwest-southeast diagonal that comprises the mountain system of Tandilia, the habitat suitability values are low, except for the areas that include the hills and “cerrilladas”, which are distributed along the system as small patches. These patches include relic mountain grasslands that, despite being subject to a certain degree of grazing, still conserve much of the structure and composition of the native grasslands (Bilenca and Miñarro 2004; Valicenti et al. 2010; Herrera et al. 2019), which would explain their excellent habitat suitability values. We can also observe this in the resistance surfaces generated for the Tandilia mountain system, where we observe the areas with the lowest resistance associated with the patches of highland grasslands and various edges of roads and streams. Similarly, within the Ventania mountain system, areas with good habitat suitability are present within the mountain system, which is surrounded by extensive areas of low habitat quality. The habitat suitability of these mountain areas can be attributed to the relic grasslands present there. The outcomes of our habitat suitability assessment within the Pampas ecoregion enable us to pinpoint areas for future research on a broader regional scale.

Several authors agree that generating multi-species corridors is essential when planning strategies that promote landscape connectivity (Liu et al. 2018; Ersoy et al. 2019); in this way, it is guaranteed that the corridors not only function as areas that favor the movement of focal species but also allow the flow of the ecosystem processes that accompany them. In addition, the design of multi-species corridors can be thought of in terms of umbrella species, which have ample space and resource requirements, such as carnivores, and by protecting them or designing corridors for them, other species are indirectly favored (Caro 2010). Brodie et al. (2015) found that when seeking to generate multi-species corridors, selecting species whose ecological requirements are similar increases their efficiency and reduces costs compared to corridors designed for different ecological groups.

A fairly common approach to corridor design is multispecies modeling based on large herbivores (Crego et al. 2021; Riggio et al. 2022). Many herbivores are considered landscape species, given their great dispersal capacity and low population density, which makes them good focal species (Mosquera-Guerra et al. 2024). Unfortunately, these species do not represent a good alternative for the corridor design in the Pampas grassland. Currently, the community of large herbivorous mammals native to the Pampas grassland is reduced to small relictual populations of pampas deer (Ozotoceros bezoarticus) and guanacos (Lama guanicoe), and are restricted to a small area within the ecoregion (Merino et al. 2019; Carmanchahi et al. 2019). These populations have almost disappeared from the region due to replacing natural cover with crops, introducing competing species, and hunting (Bilenca et al., 2008; Vila 2006). Contrary to this, carnivore species have been able to adapt and use the new environment that the Pampas ecoregion represents today. Although some local populations may be more affected than others, some of the carnivore species reported for the region (fox, cat, skunk) are common and frequent (Caruso et al. 2017; Trofino-Falasco et al. 2023). As for the Puma, this species suffered a significant population decline during the 20th century, to the point of being considered extinct in the region (Cabrera and Yepes, 1940). Despite this, in recent decades, it has been colonizing the Pampas grassland, reestablishing itself as the top predator within the system (De Lucca and Chimento 2020).

We were able to generate a multi-species corridor map for the Tandilia system. Identifying common corridors for carnivore species gives us not only information on valuable areas for their movement and dispersal but also allows us to identify priority sites where we can focus specific conservation actions for these particular species, such as the mitigation of conflict with livestock producers, given that in the Pampas ecoregion, species such as the puma and the gray fox are considered harmful to livestock activities (Caruso et al. 2017), and in the Tandilia system the situation is similar (Aranguren in prep.). Despite the advantages of thinking about multi-species corridors designed for carnivores, we must not ignore the fact that many of these are habitat generalists, with good plasticity and tolerance to degraded environments (Buskirk and Gittleman 1990), therefore generating corridors designed only for them can lead to a “negative umbrella effect” affecting those secondary, specialist species with more limited environmental requirements, which are expected to be protected under the umbrella (Beier et al. 2008).

In our results, we observed that some main corridors (those that coincide with the lowest-cost roads) run either entirely or in some sections along paved roads or routes. These results could be attributed to broad shoulders and abundant vegetation, coupled with the fact that roads offer the shortest distance between two areas. In landscapes dominated by agricultural matrix, road edges or shoulders can serve as habitat for species, both for plant communities and for arthropods and small mammals, including some carnivores (Tikka et al. 2001; Ruiz-Capillas et al. 2013; Rotholz and Mandelik 2013; Carmona et al. 2024). Depending on their characteristics, road edges, and shoulders can also act as corridors that allow for the dispersal of some species between patches of natural habitat (Eversham and Telfer 1994; Vermeulen and Opdam 1995; Le Viol et al. 2008; Redon et al. 2015; Galantinho et al. 2020). Considering this and the results obtained, some of the shoulders of the roads that run through the Tandilia system could act as corridors for the studied species, with many of them being associated with roads in various sections (Fig. 7).

Nonetheless, the disturbances generated by these routes usually awaken behavioral changes in species, such as avoidance behaviors (McClure et al. 2013; Navarro-Castilla et al. 2014; Grilo et al. 2015), which lead carnivore species to not take advantage of these environments for dispersal. However, the proximity to the routes exposes the carnivores to the risk of being struck by vehicles. Death due to being run over on roads is one of the threats that local populations face (Schwab and Zandbergen 2011; Gregory et al. 2021). This forces us to reevaluate these environments as potential corridors and invest in mitigation or management efforts. In any case, since they can act as available habitats and corridors for other groups of species in a landscape as intervened and fragmented as the Pampas, it is crucial to aim for their adequate management, such as implementing regulations that control agricultural activities on shoulders (e.g., Rimoldi and Chimento, 2020), a common practice in the Pampas ecoregion.

The centrality analysis of the grassland relics allowed us to identify that the patches that contribute most to the general connectivity are located in the central zone of the system (area surrounding the city of Tandil). Taking into account the spatial configuration of the mountain range on a diagonal and that centrality reflects the contribution of the patches to the general connectivity of the network (McRae et al. 2008), these central habitat patches would be acting as a bridge between those that are distributed at the northwestern and southeastern borders of the system. On the other hand, we can also find numerous patches of various sizes in this central region, which have many connections to nearby patches. For Jaimes et al. (2019), this central area of the mountain system counts as an essential landscape unit worth conserving, given that mountainous environments predominate, and are present as large patches, and the anthropic elements within the matrix are few. On the other hand, their results also highlight the importance of the patch system within the southeastern area of the system. In our analysis, these groups of patches presented average centrality values. That is, they contribute less to the general connectivity of the system than others. However, these patches that stand out in the study by Jaimes et al. (2019) make up groups of large patches, with smaller patches as satellites, which can be of great value at a local scale within the mountain system. Although the analysis approach in both studies differed, both represent sufficient evidence for conserving these areas with important average centrality values.

Some of the grassland patches that showed high centrality values and that form important landscape units for connectivity (Jaimes et al. 2019), were among the most extensive patches in the system. The fact that these structurally critical patches for connectivity are also large gives them additional value as priority areas for conservation. According to the theory of island biogeography, larger patches or islands tend to provide more refuge and resources for species, favoring greater biodiversity (Laurance 2000; MacArthur and Wilson 2001). In fragmented environments, habitat patches are vital refuges for species (Fahrig 2003; Holland and Bennett 2007; Anand et al. 2010; Herrera and Laterra 2011), and while the surface area of the studied patches might not be sufficient to satisfy the needs of carnivore species with large spatial requirements, it could be adequate to sustain small populations of mesocarnivores, whose home ranges are less extensive (Donadio et al. 2001; Manfredi et al. 2006; Lucherini and Luengos Vidal 2008). On the other hand, for the puma, a species that travels great distances and whose area of action can exceed the size of grassland patches (Elbroch et al. 2009), the availability of large patches in a fragmented landscape guarantees access to shelter and prey, promoting their occupation (Stoner et al. 2023), which would facilitate the movement of this species through the landscape.

We were able to identify priority corridors and areas to maintain the connectivity of the focal species. Looking at the mapped corridors, we can find several smaller pass-over natural grassland patches that were not included in the analysis. These results could be due to small habitat patches that act as stepping stones between larger patches. Herrera et al. (2017) found that small habitat patches for the Tandilia system function as stepping stones, particularly with regards to the connectivity of species that move long distances. Considering this evidence, searching for conservation and management strategies for these areas is advisable.

We also identified several corridors that maintain a considerable width along their route and connect valuable areas for landscape connectivity (greater centrality), which we consider a priority in the mountain system. However, some of these corridors are interrupted by barriers such as paved routes, which could be an important barrier to the movement of species, not only because of the risk of mortality but because some species avoid them, thus promoting the fragmentation of their populations (Grilo et al. 2015). Therefore, it would be necessary to mitigate these effects by resorting to underground passageways, which have proven helpful in facilitating the flow of small and medium-sized carnivores through these types of routes (Grilo et al. 2008).

In conclusion, our study shows us that despite the Pampas ecoregion's history of transforming and replacing grasslands, there are still areas with good environmental suitability for carnivore species. Mountain systems would be essential conservation sites towards the south of the region. We were also able to pinpoint several priority areas for conserving connectivity within the Tandilia system for the studied carnivore species, as well as the corridors that could keep them connected. A study like this in the region can be used as a frame of reference for future studies or the implementation of conservation actions in the area that favor the use of space by carnivore species.

Competing Interests

The authors have no relevant financial or non-financial interests to disclose.

Funding

This study was partially supported by Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET PIP. 11220220100207CO), Agencia Nacional de Promoción Científica y Tecnológica (ANPCYT, PICT 2015–2281), Neotropical Grassland Conservancy (Student Grant Program 2021), Universidad Nacional del Centro de la Provincia de Buenos Aires (UNICEN − 03-PIO-103C), IDEA WILD (IDEA WILD Equipment Support 2021). M.F. Aranguren, M.G. Pizzarello, V. Leber, D. Franzoia Moss, and J. Dopazo were supported by fellowships from the Consejo Nacional de Investigaciones Científicas y Técnicas de Argentina (CONICET). C. Trofino-Falasco is a support technician of Comisión de Investigaciones Cientificas de la Provincia de Buenos Aires (CICPBA), V. Simoy, M.A. Velasco, and I. Berkunsky are CONICET Research Fellows.

Author Contributions

All authors contributed to the study's conception and design. M. Florencia Aranguren and M. Veronica Simoy performed material preparation, data collection, and analysis. Clara Trofino Falasco and Gimena Pizzarello collaborated on the camera trap data collection. M. Florencia Aranguren wrote the manuscript's first draft, and all authors commented on previous versions. All authors read and approved the final manuscript.

Acknowledgments

We want to thank field assistants and collaborators (Carmela Marin, Manuela Santiago, Celena Sarasola, Estefania Paz, Maximiliano Calcagno, Ailen Chuchuy, M. Eva Cabanellas, Claudio Santiago, Estefania Marisol Avalo, Felisa La Pescadora) for their kind assistance and support during fieldwork activities with trap cameras. We also appreciate the collaboration of owners and managers of highland grassland remnants for allowing us to visit their properties: Federico Juana (Estancia Las Mercedes), Paulo Mosca (Estancia Nilonil), Reina Feldman (Estancia Sanmalucon), Mario Bustillo (Estancia La Asunción), Alfonso (Valle de Los Ciervos), Emilio Milanessi (Estancia Chapaleofu), Manuel Castelar (La Argentina), Raul Eyheramendy (Sierra Alta) and Tomas Pérez Marino (Estancia El Bonete). We thank Claudio Barletta (Reserva Natural Sierra del Tigre) for his attention and collaboration. Also, we would like to thank Estefanía Paz for her support and contributions during the development of this work. We want to thank the assistance of the Scouts de la Ciencia program for their cooperation in fieldwork with camera traps. Also, we would like to thank M. Ignacio Simoy, Santiago Linares, and David Vera for their collaboration with the analysis in R and GIS.

Data Availability

A part of the data set of occurrences filtered and generated during the current study is available in the GBIF repository: https://doi.org/10.15468/dd.p8neyx. The other part corresponds to camera trap data, available in Supplementary Information 2.

Adriaensen F, Chardon JP, De Blust G et al (2003) The application of ‘least-cost’ modelling as a functional landscape model. Landsc Urban Plan 64:233–247. https://doi.org/10.1016/S0169–2046(02)00242–6
Aiello-Lammens ME, Boria RA, Radosavljevic A et al (2015) spThin: an R package for spatial thinning of species occurrence records for use in ecological niche models. Ecography 38:541–545. https://doi.org/10.1111/ecog.01132
Anand MO, Krishnaswamy J, Kumar A, Bali A (2010) Sustaining biodiversity conservation in human-modified landscapes in the Western Ghats: Remnant forests matter. Biol Conserv 143:2363–2374. https://doi.org/10.1016/j.biocon.2010.01.013
Ang’ila RO, Kimuyu DM, Wambugu GM et al (2023) Fine-scale variation in soil and topography influences herbaceous vegetation and the distribution of large mammalian herbivores. Afr J Ecol 61:706–716. https://doi.org/10.1111/aje.13166
Aranguren MF, Velasco MA, Trofino-falasco C et al (2023) Mammals of the Tandilia Mountain system, current species inhabiting Pampean highland grasslands. Neotropical Biol Conserv 18:13–29. https://doi.org/10.3897/neotropical.18.e98374
Baeza S, Vélez-Martin E, De Abelleyra D et al (2022) Two decades of land cover mapping in the Río de la Plata grassland region: The MapBiomas Pampa initiative. Remote Sens Appl Soc Environ 28:100834. https://doi.org/10.1016/j.rsase.2022.100834
Barquez RM, Mónica Díaz M, Ojeda RA (2006) MAMÍFEROS DE ARGENTINA Sistemática y Distribución Editores 2006. SAREM Sociedad Argentina para el Estudio de los Mamíferos
Batáry P, Dicks LV, Kleijn D, Sutherland WJ (2015) The role of agri-environment schemes in conservation and environmental management. Conserv Biol 29:1006–1016. https://doi.org/10.1111/cobi.12536
Beier P, Majka D, Jenness J (2007) Conceptual steps for designing wildlife corridors. Corridor Design, Arizona, USA
Beier P, Majka DR, Spencer WD (2008) Forks in the Road: Choices in Procedures for Designing Wildland Linkages. Conserv Biol 22:836–851. https://doi.org/10.1111/j.1523–1739.2008.00942.x
Belote RT, Dietz MS, McRae BH et al (2016) Identifying Corridors among Large Protected Areas in the United States. PLoS ONE 11:e0154223. https://doi.org/10.1371/journal.pone.0154223
Bilenca D, Codesido M, Gonzáles Fischer C et al (2012) Impactos de la transformación agropecuaria sobre la biodiversidad en la provincia de Buenos Aires. Rev Mus Argent Cienc Nat
Bilenca D, Codesido M, Maria C, Fischer G (2008) Cambios en la fauna pampeana. Cienc Hoy
Bilenca D, Miñarro F (2004) Identificación de Áreas Valiosas de Pastizal (AVPs) en las Pampas y Campos de Argentina, Uruguay y sur de Brasil, 1a edn. Fundación Vida Silvestre Argentina, Buenos Aires, Argentina
Boitani L, Powell RA (2012) Carnivore Ecology and Conservation: A Handbook of Techniques. OUP Oxford
Brodie JF, Giordano AJ, Dickson B et al (2015) Evaluating multispecies landscape connectivity in a threatened tropical mammal community. Conserv Biol 29:122–132. https://doi.org/10.1111/cobi.12337
Burgos JJ, Vidal AL (1951) Los climas de la República Argentina según la nueva clasificación de Thornthwaite. Meteoros 1:3–32
Buskirk SM, Gittleman JL (1990) Carnivore Behavior, Ecology, and Evolution
Cabrera AL, Yepes J (1940) Mamíferos Sudamericanos (vida, costumbres y descripción). Historia Natural Ediar, Cía. Argentina de Editores, Buenos Aires, p 370
Carmanchahi PD, Panebianco A, Leggieri L et al (2019) Lama guanicoe. In: SAyDS–SAREM (eds.) Categorización 2019 de los mamíferos de Argentina según su riesgo de extinción. Lista Roja de los mamíferos de Argentina. Digital version: http://cma.sarem.org.ar
Carmona G, Virgós E, Burgos T, Barrientos R (2024) Factors determining roadkills in a mammal carnivore are road-type specific. Mamm Biol 104:175–183. https://doi.org/10.1007/s42991-024-00400-y
Caro T (2010) Conservation by Proxy: Indicator, Umbrella, Keystone, Flagship, and Other Surrogate Species. Island
Carroll C, McRAE BH, Brookes A (2012) Use of Linkage Mapping and Centrality Analysis Across Habitat Gradients to Conserve Connectivity of Gray Wolf Populations in Western North America. Conserv Biol 26:78–87. https://doi.org/10.1111/j.1523–1739.2011.01753.x
Caruso N, Vidal L, Lucherini; et al (2017) Carnivoros en el sudoeste de la provincia de Buenos Aires: ecología y conflictos con los ganaderos
Castilho CS, Hackbart VCS, Pivello VR, dos Santos RF (2015) Evaluating Landscape Connectivity for Puma concolor and Panthera onca Among Atlantic Forest Protected Areas. Environ Manage 55:1377–1389. https://doi.org/10.1007/s00267-015-0463–7
Castillo DF (2010) ECOLOGÍA ESPACIAL, TEMPORAL Y TRÓFICA DEL ZORRINO (Conepatus chinga) EN. UN ÁREA NATURAL Y UN ÁREA DE USO AGRÍCOLA
Červinka J, Riegert J, Grill S, Šálek M (2015) Large-scale evaluation of carnivore road mortality: the effect of landscape and local scale characteristics. Mammal Res 60:233–243. https://doi.org/10.1007/s13364-015-0226–0
Cingolani CA (2011) The Tandilia system of Argentina as a southern extension of the Río de la Plata craton: An overview. Int J Earth Sci 100:221–242. https://doi.org/10.1007/s00531-010-0611-5
Cobos ME, Peterson AT, Osorio-Olvera L, Jiménez-García D (2019) An exhaustive analysis of heuristic methods for variable selection in ecological niche modeling and species distribution modeling. Ecol Inf 53:100983. https://doi.org/10.1016/j.ecoinf.2019.100983
Cozzi G, Broekhuis F, McNutt JW, Schmid B (2013) Comparison of the effects of artificial and natural barriers on large African carnivores: Implications for interspecific relationships and connectivity. J Anim Ecol 82:707–715. https://doi.org/10.1111/1365–2656.12039
Crego RD, Wells HBM, Ndung’u KS et al (2021) Moving through the mosaic: identifying critical linkage zones for large herbivores across a multiple-use African landscape. Landsc Ecol 36:1325–1340. https://doi.org/10.1007/s10980-021-01232–8
Crooks KR, Burdett CL, Theobald DM et al (2011) Global patterns of fragmentation and connectivity of mammalian carnivore habitat. Philos Trans R Soc B Biol Sci 366:2642–2651. https://doi.org/10.1098/rstb.2011.0120
Cruz-Cárdenas G, López-Mata L, Villaseñor JL, Ortiz E (2014) Potential species distribution modeling and the use of principal component analysis as predictor variables. Rev Mex Biodivers 85:189–199. https://doi.org/10.7550/rmb.36723
De Lucca ER, Chimento NR (2020) El Puma (Puma concolor) en las Pampas de la provincia de Buenos Aires: Una actualización sobre distribución geografica y conflicto con el hombre. Hist Nat 10:53–79. https://doi.org/10.1157/13097569
de Vries FP, Hanley N (2016) Incentive-Based Policy Design for Pollution Control and Biodiversity Conservation: A Review. Environ Resour Econ 63:687–702. https://doi.org/10.1007/s10640-015-9996–8
DeMatteo KE, Escalante OM, Alegre DMI et al (2023) A multispecies corridor in a fragmented landscape: Evaluating effectiveness and identifying high-priority target areas. PLoS ONE 18:e0283258. https://doi.org/10.1371/journal.pone.0283258
Di Giacomo ASD, Vickery PD, Casañas H et al (2010) Landscape associations of globally threatened grassland birds in the Aguapey river Important Bird Area, Corrientes, Argentina. Bird Conserv Int 20:62–73. https://doi.org/10.1017/S0959270909990177
Donadio E, Di Martino S, Aubone M, Novaro AJ (2001) Activity patterns, home-range. and habitat selection of the common hog-nosed skunk, Conepatus chinga (Mammalia, mustelidae), in the northwestern Patagonia. Mammalia 65:49–54
Dong X, Zhang J, Gu X et al (2021) Evaluating habitat suitability and potential dispersal corridors across the distribution landscape of the Chinese red panda (Ailurus styani) in Sichuan, China. Glob Ecol Conserv 28:e01705. https://doi.org/10.1016/j.gecco.2021.e01705
Dutta T, Sharma S, Meyer NFV et al (2022) An overview of computational tools for preparing, constructing and using resistance surfaces in connectivity research. Landsc Ecol 37:2195–2224. https://doi.org/10.1007/s10980-022-01469-x
Echeverría ML, Alonso SI, Comparatore VM (2017) Survey of the vascular plants of Sierra Chica, the untouched area of the Paititi Natural Reserve (southeastern Tandiliamountain range, Buenos Aires province, Argentina). Check List 13:1003–1036. https://doi.org/10.15560/13.6.1003
Elbroch LM, Wittmer HU (2012) Puma spatial ecology in open habitats with aggregate prey. Mamm Biol 77:377–384. https://doi.org/10.1016/j.mambio.2012.02.010
Elbroch M, Wittmer HU, Saucedo C, Corti P (2009) Dispersión de larga distancia de un puma (Puma concolor puma) macho en Patagonia. Rev Chil Hist Nat 82:459–461. https://doi.org/10.4067/S0716–078X2009000300011
Ellis EC, Klein Goldewijk K, Siebert S et al (2010) Anthropogenic transformation of the biomes, 1700 to 2000. Glob Ecol Biogeogr 19:589–606. https://doi.org/10.1111/j.1466–8238.2010.00540.x
Ersoy E, Jorgensen A, Warren PH (2019) Identifying multispecies connectivity corridors and the spatial pattern of the landscape. Urban Urban Green 40:308–322. https://doi.org/10.1016/j.ufug.2018.08.001
Etherington TR (2016) Least-Cost Modelling and Landscape Ecology: Concepts, Applications, and Opportunities. Curr Landsc Ecol Rep 1:40–53. https://doi.org/10.1007/s40823-016-0006–9
Eversham BC, Telfer MG (1994) Conservation value of roadside verges for stenotopic heathland Carabidae: corridors or refugia? Biodivers Conserv 3:538–545. https://doi.org/10.1007/BF00115159
Fahrig L (2003) Effects of Habitat Fragmentation on Biodiversity. Annu Rev Ecol Evol Syst 34:487–515. https://doi.org/10.1146/annurev.ecolsys.34.011802.132419
Fahrig L, Rytwinski T (2009) Effects of Roads on Animal Abundance: an Empirical Review and Synthesis. Ecol Soc 14
Falasca S, Ulberich A, Bernabé M, Mordenti S (2000) Principales características agroclimáticas del sudeste bonaerense, República Argentina. Revija za Geografijo 127:91–102
Fick SE, Hijmans RJ (2017) WorldClim 2: new 1-km spatial resolution climate surfaces for global land areas. Int J Climatol. https://doi.org/10.1002/joc.5086
Galantinho A, Herrera JM, Eufrázio S et al (2020) Los arcenes de las carreteras proporcionan conectividad a los pequeños mamíferos: un estudio de caso con ratones de campo (Apodemus sylvaticus) en un sistema agrosilvopastoril. J Environ Manage 258:110033. https://doi.org/10.1016/j.jenvman.2019.110033
Gandini ML, Lara BD, Moreno LB et al (2019) Trends in fragmentation and connectivity of Paspalum quadrifarium grasslands in the Buenos Aires province. Argentina PeerJ 7:e6450. https://doi.org/10.7717/peerj.6450
González Saucedo ZY (2011) Conectividad funcional para el puma (Puma concolor) en el centro de México. 58
Grass I, Loos J, Baensch S et al (2019) Land-sharing/-sparing connectivity landscapes for ecosystem services and biodiversity conservation. People Nat 1:262–272. https://doi.org/10.1002/pan3.21
Gregory A, Spence E, Beier P, Garding E (2021) Toward Best Management Practices for Ecological Corridors. Land 10:140. https://doi.org/10.3390/land10020140
Grilo C, Bissonette JA, Santos-Reis M (2008) Response of carnivores to existing highway culverts and underpasses: implications for road planning and mitigation. Biodivers Conserv 17:1685–1699. https://doi.org/10.1007/s10531-008-9374–8
Grilo C, Smith DJ, Klar N (2015) Carnivores. In: Handbook of Road Ecology. pp 300–312
Guidobono JS, Muñoz J, Muschetto E et al (2016) Food habits of Geoffroy ’ s cat (Leopardus geoffroyi) in agroecosystem habitats of Buenos Aires, Argentina. Ecol Austral 26:40–50
Heikinheimo H, Eronen JT, Sennikov A et al (2012) Convergence in the distribution patterns of Europe’s plants and mammals is due to environmental forcing. J Biogeogr 39:1633–1644. https://doi.org/10.1111/j.1365–2699.2012.02723.x
Herrera L, Montti L, Sabatino M, De Rito M (2019) El paisaje serrano de Tandilia: un tesoro geológico, ecológico y cultural. Cienc Hoy 28:44–50
Herrera L, Laterra P (2007) Relaciones entre la riqueza y la composición florística con el tamaño de fragmentos de pastizales en la Pampa Austral, Argentina. Panorama de la Ecología de Paisajes en Argentina y en Países de Sudamericanos. Ediciones INTA. Buenos Aires, pp 387–396
Herrera LP, Laterra P (2011) Relative influence of size, connectivity and disturbance history on plant species richness and assemblages in fragmented grasslands. Appl Veg Sci 14:181–188. https://doi.org/10.1111/j.1654–109X.2010.01117.x
Herrera LP, Sabatino MC, Jaimes FR, Saura S (2017) Landscape connectivity and the role of small habitat patches as stepping stones: an assessment of the grassland biome in South America. Biodivers Conserv 26:3465–3479. https://doi.org/10.1007/s10531-017-1416–7
Hilty J, Worboys GL, Keeley A et al (2021) Lineamientos para la conservación de la conectividad a través de redes y corredores ecológicos. IUCN, International Union for Conservation of Nature
Holland GJ, Bennett AF (2007) Occurrence of small mammals in a fragmented landscape: the role of vegetation heterogeneity. Wildl Res 34:387–397. https://doi.org/10.1071/WR07061
Huck M, Jędrzejewski W, Borowik T et al (2010) Habitat suitability, corridors and dispersal barriers for large carnivores in Poland. Acta Theriol (Warsz) 55:177–192. https://doi.org/10.4098/j.at.0001–7051.114.2009
Illoldi-Rangel P, Escalante T (2008) De los modelos de nicho ecológico a las áreas de distribución geográfica. Biogeografía 3:7–12
Jaimes FR, Sabatino M, Herrera LP (2019) Caracterización del paisaje serrano de Tandilia (Buenos Aires, Argentina) utilizando SIG. Una aproximación para definir áreas prioritarias para la conservación.: Landscape characterization Tandilia´s system (Buenos Aires, Argentina) with GIS. An approach to define conservation areas priority. Rev Estud Ambient - Environ Stud J 7:39–65
LaRue MA, Nielsen CK (2008) Modelling potential dispersal corridors for cougars in midwestern North America using least-cost path methods. Ecol Model 212:372–381. https://doi.org/10.1016/j.ecolmodel.2007.10.036
Laurance WF (2000) Do edge effects occur over large spatial scales? Trends Ecol Evol 15:134–135. https://doi.org/10.1016/S0169–5347(00)01838–3
Le Viol I, Julliard R, Kerbiriou C et al (2008) Plant and spider communities benefit differently from the presence of planted hedgerows in highway verges. Biol Conserv 141:1581–1590. https://doi.org/10.1016/j.biocon.2008.04.003
Liu C, Newell G, White M, Bennett AF (2018) Identifying wildlife corridors for the restoration of regional habitat connectivity: A multispecies approach and comparison of resistance surfaces. PLoS ONE 13:e0206071. https://doi.org/10.1371/journal.pone.0206071
Lucherini M, Luengos Vidal EM (2008) Lycalopex Gymnocercus (Carnivora: Canidae). Mamm Species 820:1–9. https://doi.org/10.1644/820.1
Luengos Vidal EM, Castillo DF, Caruso NC et al (2016) Field capture, chemical immobilization, and morphometrics of a little-studied South American carnivore, the lesser grison. Wildl Soc Bull 40:400–405. https://doi.org/10.1002/wsb.654
MacArthur RH, Wilson EO (2001) The Theory of Island Biogeography. Princeton University Press
Manfredi C, Lucherini M, Soler L et al (2011) Activity and movement patterns of Geoffroy’s cat in the grasslands of Argentina. Mamm Biol 76:313–319. https://doi.org/10.1016/j.mambio.2011.01.009
Manfredi C, Soler L, Lucherini M, Casanave EB (2006) Home range and habitat use by Geoffroy’s cat (Oncifelis geoffroyi) in a wet grassland in Argentina. J Zool 268:381–387. https://doi.org/10.1111/j.1469–7998.2005.00033.x
Martínez-Meyer E, Díaz-Porras D, Peterson AT, Yáñez-Arenas C (2013) Ecological niche structure and rangewide abundance patterns of species. Biol Lett 9:20120637. https://doi.org/10.1098/rsbl.2012.0637
Matteucci SD (2012) Ecorregión Pampa. In: Morello J, Matteucci SD, Rodriguez AF, (Eds) Ecorregiones y complejos Ecosistémicos de Argentina. Orientación Gráfica Editora, Buenos Aires, 391–446
McClure CJW, Ware HE, Carlisle J et al (2013) An experimental investigation into the effects of traffic noise on distributions of birds: avoiding the phantom road. Proc R Soc B Biol Sci 280:20132290. https://doi.org/10.1098/rspb.2013.2290
McRae B (2012) Centrality Mapper Connectivity Analysis Software. Nat, Conserv
McRae BH, Dickson BG, Keitt TH, Shah VB (2008) Using Circuit Theory to Model Connectivity in Ecology, Evolution, and Conservation. Ecology 89:2712–2724. https://doi.org/10.1890/07–1861.1
Mereles F, Nazaro G, Villalba L et al (2023) Designing biological corridors for the Pilcomayo River region (Gran Chaco) of Argentina, Bolivia and Paraguay. Rev Soc Científica Parag 28:220–249. https://doi.org/10.32480/rscp.2023.28.2.220
Merino ML, Cirignoli S, Perez Carusi L et al (2019) Ozotoceros bezoarticus. In: SAyDS–SAREM (eds.) Categorización 2019 de los mamíferos de Argentina según su riesgo de extinción. Lista Roja de los mamíferos de Argentina. Versión digital: http://cma.sarem.org.ar
Merow C, Smith MJ, Silander JA (2013) A practical guide to MaxEnt for modeling species’ distributions: What it does, and why inputs and settings matter. Ecography 36:1058–1069. https://doi.org/10.1111/j.1600–0587.2013.07872.x
Montenegro JA, Acosta A, Reimer JD (2014) HaviStat© v2.2: Application to estimate preference for habitat and resources. Univ Sci 19:333–337. https://doi.org/10.11144/Javeriana.SC19–3.haea
Morrow CB (2019) Thresholding species distribution models. In: Cecina Babich Morrow. https://babichmorrowc.github.io/post/2019-04-12-sdm-threshold/. Accessed 19 Aug 2024
Mosquera-Guerra F, Barreto S, Moreno-Niño N et al (2024) Habitat connectivity of threatened ungulate species in a native savanna landscape of northern South America. Mamm Biol 104:259–275. https://doi.org/10.1007/s42991-024-00404–8
Navarro-Castilla Á, Mata C, Ruiz-Capillas P et al (2014) Are Motorways Potential Stressors of Roadside Wood Mice (Apodemus sylvaticus) Populations? PLoS ONE 9:e91942. https://doi.org/10.1371/journal.pone.0091942
Neu CW, Byers CR, Peek JM (1974) A Technique for Analysis of Utilization-Availability Data. J Wildl Manag 38:541. https://doi.org/10.2307/3800887
Pereira JA (2010) Activity pattern of Geoffroy’s cats (Leopardus geoffroyi) during a period of food shortage. J Arid Environ 74:1106–1109. https://doi.org/10.1016/j.jaridenv.2010.03.017
Phillips SJ, Anderson RP, Schapire RE (2006) Maximum entropy modeling of species geographic distributions. Ecol Model 190:231–259. https://doi.org/10.1016/j.ecolmodel.2005.03.026
Prugh LR, Hodges KE, Sinclair ARE, Brashares JS (2008) Effect of habitat area and isolation on fragmented animal populations. Proc Natl Acad Sci 105:20770–20775. https://doi.org/10.1073/pnas.0806080105
Pullinger MG, Johnson CJ (2010) Maintaining or restoring connectivity of modified landscapes: evaluating the least-cost path model with multiple sources of ecological information. Landscape Ecol 25:1547–1560. https://doi.org/10.1007/s10980-010-9526-6
Rands MRW, Adams WM, Bennun L et al (2010) Biodiversity Conservation: Challenges Beyond 2010. Science 329:1298–1303. https://doi.org/10.1126/science.1189138
Redon L, Le Viol I, Jiguet F et al (2015) Road network in an agrarian landscape: Potential habitat, corridor or barrier for small mammals? Acta Oecol 62:58–65. https://doi.org/10.1016/j.actao.2014.12.003
Riggio J, Foreman K, Freedman E et al (2022) Predicting wildlife corridors for multiple species in an East African ungulate community. PLoS ONE 17:e0265136. https://doi.org/10.1371/journal.pone.0265136
Rimoldi PG, Chimento NR (2020) Nuevo registro de Herpailurus yagouaroundi y actualización de su distribución en el sur de Santa Fe, República Argentina. Notas sobre Mamíferos Sudamericanos, 2
Rio-Maior H, Nakamura M, Álvares F, Beja P (2019) Designing the landscape of coexistence: Integrating risk avoidance, habitat selection and functional connectivity to inform large carnivore conservation. Biol Conserv 235:178–188. https://doi.org/10.1016/j.biocon.2019.04.021
Rotholz E, Mandelik Y (2013) Roadside habitats: effects on diversity and composition of plant, arthropod, and small mammal communities. Biodivers Conserv 22:1017–1031. https://doi.org/10.1007/s10531-013-0465–9
Ruiz-Capillas P, Mata C, Malo JE (2013) Road verges are refuges for small mammal populations in extensively managed Mediterranean landscapes. Biol Conserv 158:223–229. https://doi.org/10.1016/j.biocon.2012.09.025
Saunders DA, Hobbs RJ, Margules CR (1991) Biological Consequences of Ecosystem Fragmentation: A Review. Conserv Biol 5:18–32. https://doi.org/10.1111/j.1523–1739.1991.tb00384.x
Schwab AC, Zandbergen PA (2011) Vehicle-related mortality and road crossing behavior of the Florida panther. Appl Geogr 31:859–870. https://doi.org/10.1016/j.apgeog.2010.10.015
Sims KRE (2014) Do Protected Areas Reduce Forest Fragmentation? A Microlandscapes Approach. Environ Resour Econ 58:303–333. https://doi.org/10.1007/s10640-013-9707-2
Singleton PH, Gaines WL, Lehmkuhl JF (2002) Landscape Permeability for Large Carnivores in Washington: A Geographic Information System Weighted-Distance and Least-Cost Corridor Assessment. US Department of Agriculture, Forest Service, Pacific Northwest Research Station
Soriano A, León RJC, Sala OE et al (1992) Río de la Plata grasslands. In: Coupland RT (ed) Ecosystems of the world: Natural grasslands. Elsevier, New York, pp 367–407
Spear SF, Balkenhol N, Fortin M et al (2010) Use of resistance surfaces for landscape genetic studies: considerations for parameterization and analysis. Mol Ecol 19:3576–3591. https://doi.org/10.1111/j.1365–294X.2010.04657.x
Stoner DC, McDonald Z, Coon CAC (2023) Stepping stones to extirpation: Puma patch occupancy thresholds in an urban-wildland matrix. Ecol Evol 13:e10381. https://doi.org/10.1002/ece3.10381
Tellaeche CG, Reppucci JI, Vidal EML, Lucherini M (2014) New data on the distribution and natural history of the lesser grison (Galictis cuja), hog-nosed skunk (Conepatus chinga), and culpeo (Lycalopex culpaeus) in northwestern Argentina. Mammalia 78:261–266. https://doi.org/10.1515/MAMMALIA–2013–0005
Terborgh J, James AE (eds) (2010) Trophic Cascades: Predators, Prey and the Changing Dynaics of Nature. Island
Tikka PM, Högmander H, Koski PS (2001) Road and railway verges serve as dispersal corridors for grassland plants. Landsc Ecol 16:659–666. https://doi.org/10.1023/A:1013120529382
Timmers R, van Kuijk M, Verweij PA et al (2022) Conservation of birds in fragmented landscapes requires protected areas. Front Ecol Environ 20:361–369. https://doi.org/10.1002/fee.2485
Torrella SA, Piquer-Rodríguez M, Levers C et al (2018) Multiscale spatial planning to maintain forest connectivity in the Argentine Chaco in the face of deforestation. Ecol Soc 23
Trofino-Falasco C, Simoy MV, Aranguren MF et al (2023) How effective is camera trapping in monitoring grassland species in the southern Pampas ecoregion? Rev Mex Biodivers 94:e945243. https://doi.org/10.22201/ib.20078706e.2023.94.5243
Vaeokhaw S, Ngoprasert D, Swatdipong A et al (2020) Effects of a highway on the genetic diversity of Asiatic black bears. Ursus 2020:1–15. https://doi.org/10.2192/URSUS-D–18–00013.2
Valicenti R, Farina E, Scaramuzzino R, DAlfonso C (2010) Ordenación de la vegetación en el paisaje Boca de la Sierras (Azul, Sistema de Tandilia). Rasadep 1:111–122
Velasco MA, Lutz MA, Igor B et al (2013) Mammals of protected area La Poligonal and neighborhood areas in Tandilia hills, Buenos Aires, Argentina. Check List 9:1510–1513
Venter O, Sanderson EW, Magrach A et al (2016) Sixteen years of change in the global terrestrial human footprint and implications for biodiversity conservation. Nat Commun 7:12558. https://doi.org/10.1038/ncomms12558
Vera DG, Di Pietro DO, Tettamanti G et al (2021) An annotated list of the reptiles of the highland grassland of tandilia mountains, Argentina. Neotropical Biol Conserv 16:185–204. https://doi.org/10.3897/neotropical.16.e60629
Vermeulen HJW, Opdam PFM (1995) Effectiveness of roadside verges as dispersal corridors for small ground-dwelling animals: A simulation study. Landsc Urban Plan 31:233–248. https://doi.org/10.1016/0169–2046(94)01050-I
Vila AR (2006) Ecología y conservación del venado de las pampas (Ozotoceros bezoarticus celer. Cabrera 1943) en la Bahía Samborombón. Provincia de Buenos Aires
Wade AA, McKelvey KS, Schwartz MK (2015) Resistance-surface-based wildlife conservation connectivity modeling: Summary of efforts in the United States and guide for practitioners. United States Department of Agriculture, Forest Service, Rocky Mountain Research Station
Yañez C, Martín G, Osorio-Olvera L et al (2020) The Abundant Niche-centroid Hypothesis: Key Points About Unfilled Niches and the Potential Use of Supraspecfic Modeling Units. Biodivers Inf 15:92–102. https://doi.org/10.17161/bi.v15i2.13218
Yensen E, Tarifa T (2003) Galictis cuja. Mamm Species 1–8. https://doi.org/10.1644/728
Yuan R, Zhang N, Zhang Q (2024) The impact of habitat loss and fragmentation on biodiversity in global protected areas. Sci Total Environ 931:173004. https://doi.org/10.1016/j.scitotenv.2024.173004
Zanón Martínez JI, Seoane J, Kelly MJ et al (2022) Assessing carnivore spatial co-occurrence and temporal overlap in the face of human interference in a semiarid forest. Ecol Appl 32. https://doi.org/10.1002/EAP.2482
Zemanova MA, Perotto-Baldivieso HL, Dickins EL et al (2017) Impact of deforestation on habitat connectivity thresholds for large carnivores in tropical forests. Ecol Process 6:1–11. https://doi.org/10.1186/s13717-017-0089–1

Download PDF

Reviewers agreed at journal
04 Sep, 2024
Reviewers invited by journal
28 Aug, 2024
Editor assigned by journal
24 Aug, 2024
First submitted to journal
22 Aug, 2024

You are reading this latest preprint version

Mountains as key areas for Carnivore Connectivity in Neotropical Grasslands

Status:

Version 1

Abstract

Figures

Introducción

Methodology

1. Study Area

2. Data analysis

2.1 Variable analysis and environmental factor construction

2.1.1 Occurrence data of focal species

2.1.2. Climate Suitability Models

2.1.3. Analysis of land cover use versus availability

2.1.4. Distance to road and urban centers

2.2. Habitat suitability models for the Pampas ecoregion

2.3. Modeling of connectivity in the Tandilia System

2.3.1. Resistance surfaces

2.3.2 Corridors Identification

2.3.3. Centrality Analysis

Results

Discussion

Declarations

Competing Interests

Funding

Author Contributions

Acknowledgments

Data Availability

References

Supplementary Files

Status:

Version 1