Can bear corridors support mammalian biodiversity? A case study on Central Italian Apennines

doi:10.21203/rs.3.rs-5287788/v1

Download PDF

Research Article

Can bear corridors support mammalian biodiversity? A case study on Central Italian Apennines

https://doi.org/10.21203/rs.3.rs-5287788/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Ecological corridors are essential for maintaining ecosystem functionality, allowing species movement between protected areas. In the Central Italian Apennines, five corridors have been identified to enhance habitat connectivity for the critically endangered Marsican brown bear (Ursus arctos marsicanus). This study focuses on two of these corridors to investigate their support of other mammal species populations. We collected data from camera traps over four months, and applied the Random Encounter Model to estimate the population densities of eight meso- and macro-mammal species. We compared the densities we estimated with those reported in the literature for different locations across Europe. The results indicated higher-than-average densities for several species compared to published data, especially for ungulates. These findings underscore the importance of Marsican bear corridors for a number of other mammals, as they provide important habitats for several of them. Effective management of these corridors, with a focus on reducing human disturbance and improving habitat connectivity, will be critical for the long-term survival of both the Marsican bear and its coexisting species.

Conservation Biology

Mammals

corridors

biodiversity

density

REM

Protected areas (PAs) are the cornerstone of biodiversity conservation (Watson et al. 2014), and have increased rapidly in their global extent, now covering around 16% of the Earth's land surface (UNEP-WCMC & IUCN, 2024). PAs have been effective at preventing the extinction of several species, and the populations of many threatened mammal species rely almost entirely on these sites nowadays, due to habitat loss in unprotected parts of their distribution ranges (Pacifici et al. 2020). Yet, even effectively managed PAs might not be sufficient to preserve species with high spatial requirements, and maintaining ecological connectivity between PAs is essential in this case (Hilty et al. 2020). This is the rationale behind the inclusion of ecological corridors within the 2030 EU Biodiversity Strategy “Legally protect a minimum of 30% of the EU's land areas and 30% of the EU sea area and integrate ecological corridors” (EC 2024). This requires extending conservation intervention beyond PAs, also focussing on the restoration and maintenance of habitat corridors between isolated reserves (Fahrig 2003, Pacifici et al. 2020).

Ecological corridors can increase the persistence of species with large spatial requirements, by allowing migration and dispersal, and they can also help to reduce species’ mortality by facilitating the avoidance of predation and human disturbance (Curcic and Djurdjic 2013). They became a key element of conservation and management strategies for endangered mammal species, as they can mitigate the impact of habitat loss and fragmentation and they can increase the resilience of PAs networks (Mateo-Sánchez et al. 2014, McGuire et al. 2016). However, the ecological value of a corridor is species-specific, as corridors intended for one species may not work for other species with different habitat preferences and movement patterns (Merenlender et al. 2022). The vast majority of ecological corridors are designed for charismatic and endangered focal species, which have high conservation support from stakeholders and citizens (Keeley et al. 2019). Yet, these areas might also provide important co-benefits for several other species co-occurring with the focal species for which the corridor was designed, typically having lower spatial requirements than the latter (Wang et al. 2018).

The Marsican brown bear (Ursus arctos marsicanus) is a subspecies of the brown bear (Ursus arctos) and an endemism of the Central Italian Apennines (Ciucci and Boitani 2008). This subspecies is isolated from other bear populations, and represents one of the most endangered mammals in Europe, facing severe risk of extinction due to its very small population size (Ciucci and Boitani 2008, Ciucci et al. 2015, Gervasi and Ciucci 2018). While the population is almost entirely confined within the Abruzzo, Lazio, and Molise national Park (ALMNP), five corridor areas have been identified to enhance the connectivity towards critical habitat in nearby protected areas (Ciucci et al. 2016, Maiorano et al. 2019). In addition to the Marsican brown bear, Central Apennines host other charismatic mammal species, including endemic subspecies such as the Apennine chamois (Rupicapra pyrenaica ornata) and the Apennine wolf (Canis lupus italicus), and many other species such as the red deer (Cervus elaphus), the roe deer (Capreolus capreolus) and the porcupine (Hystrix cristata) (Loy et al. 2019). As all these species co-occur with the Marsican brown bear, they can in principle benefit from the establishment of bear ecological corridors. Indeed, it is possible that these corridors act simultaneously as connectivity areas for facilitating bear movement and as suitable habitats able to support meta-populations of other mammal species with lower spatial requirements (Thornton et al. 2016).

Here, we aim to understand if corridors delineated for enhancing the movement of the Marsican bear population actually provide suitable habitat for other mammal species, thereby having a high conservation value for them. We deployed camera traps in two Marsican bear corridors in the Central Apennines to detect the images of eight meso- and macro-mammal species with smaller spatial requirements than the brown bear: European badger (Meles meles), hare (Lepus spp.), porcupine (Hystrix cristata), red deer (Cervus elaphus), red fox (Vulpes vulpes), roe deer (Capreolus capreolus), wild boar (Sus scrofa), and wildcat (Felis silvestris silvestris). To estimate population densities for each species, we used the Random Encounter Model (REM; Rowcliffe et al. 2008).

Study area

The study area is located in the Central Apennines (Italy) and includes two ecological corridors designed to enhance the connectivity for the Marsican bear population (Ursus arctos marsicanus; Fig. 1). The two corridors connect the Abruzzo, Lazio and Molise National Park (ALMNP) with the Sirente Velino Regional Park (Corridor 1) and with the Majella National Park (Corridor 2). These are part of a set of five corridors identified in the "Piano d'azione nazionale per la tutela dell'orso bruno marsicano" (PATOM; Ciucci et al. 2016). The two corridors together cover an area of approximately 265 km², with Corridor 1 extending for around 75 km² and Corridor 2 extending for around 190 km². The elevation ranges between 579 m and 1805 m asl, and the area is mostly covered by mixed deciduous forests, grasslands, and croplands. The climate is temperate Mediterranean continental, with frequent snowfalls, cold winters and hot summers, with a temperature ranging between 24°C – 35°C during the sampling period (Fratianni and Acquaotta 2017).

Data collection

We placed 34 camera traps from 19/04/2023 to 20/11/2023, for a total of 215 days. We deployed 14 camera traps in Corridor 1 and 20 in Corridor 2, using a random sampling and following a grid with cells of 2.5 km per side. The camera traps were deployed as close as possible to the centroid of each cell, with a maximum distance from the centroid of 200 metres. Camera traps were placed on trees at a height ranging from 0.5 m to 1.7 m, and facing north orientation, to avoid direct exposure to the sun. We used Browning Patriot (BTC-PATRIOT-FHD) camera-traps with the following settings: trail mode (i.e., picture mode); capture delay of 1 seconds; multi shot set on 3 shots; image resolution of 3840 x 2160 (Ultra HD); night exposure on “mid-range” for confined spaces and “long-range” for open spaces. Checks were made regularly every three weeks for each camera trap, to control the status of the batteries and SD cards. During the sampling period, two camera traps were stolen and replacement cameras were positioned approximately 100 metres away from their original locations.

Based on a minimum number of encounters threshold > = 20 (Rowcliffe et al. 2008) and thus on the possibility to derive the needed parameters for the application of REM, we decided to estimate the density of 8 species of meso- and macro-mammals: European badger (Meles meles), hare (Lepus spp.), porcupine (Hystrix cristata), red deer (Cervus elaphus), red fox (Vulpes vulpes), roe deer (Capreolus capreolus), wild boar (Sus scrofa), wildcat (Felis silvestris silvestris). Since it was not possible to distinguish the European hare (Lepus europaeus) from the Apennine hare (Lepus corsicanus) in the images, we refer to the hare as Lepus spp. Even if we obtained many pictures of wolf (Canis lupus italicus), we decided to not consider it in our analysis because the home range of a wolf pack is much larger than our study area and a different sampling design would have been needed to estimate wolf density (Ciucci et al. 1997, Mancinelli et al. 2018).

Data processing

We processed camera trap data to calculate density estimates via the Random Encounter Model (REM; Rowcliffe et al. 2008), a model that operates with unmarked animals and estimates their population density in a given area (Rowcliffe et al. 2008). The model treats animals like ideal gas particles and estimates density within the collective detection viewsheds of a camera array. REM has been proven to be a reliable method for estimating wildlife population density in a wide range of situations and scenarios, when using appropriate methods to estimate parameters and appropriate sampling designs (Palencia et al. 2022). It is also effective for monitoring more than one species using the same survey design, because for its application it is not needed for the animals to have a high detection probability (Rowcliffe et al. 2008). We estimated densities for each corridor separately, and for the entire study area (Corridor 1 + Corridor 2).

The densities are estimated based on different parameters (Fig. 2), which can be measured directly from the camera trap pictures, without the need of auxiliary data (Caravaggi et al. 2016, Hofmeester et al. 2017). Parameters estimated include: day range (i.e., the average distance travelled by an individual during the day, estimated as the product of speed (average travel speed while active) and activity rate (proportion of day that the population spent active)), the camera traps’ effective detection zone (EDZ; i.e., the area effectively monitored by cameras, defined by the effective detection radius (EDR) and the effective detection angle (EDA)), and trapping rates (i.e., the number of independent encounters per unit time, in particular the number of encounters occurred at least 30 minutes from each other over the collective time in which the camera traps were operative) (Rowcliffe et al. 2008). All the parameters were measured by using data from Corridor 1 and Corridor 2 separately when estimating densities of each corridor, while we considered all the images together in calculating densities of the entire study area (Corridor 1 + Corridor 2) (Appendix SI, Table S2).

Random Encounter Model application

We applied REM following Palencia (2022) to estimate the density of each species (D), using the formula:

\(\:D=\frac{y}{t}\frac{\pi\:}{vr(2+a)}\) [1]

where y is the number of independent encounters, t is total camera survey effort (in days), v is the average distance travelled by an individual during a day (i.e., day range, expressed in km/day), and r and α are the radius (metres) and angle (degrees) of the camera traps detection zone, respectively. The density of the wild boar was calculated by multiplying the density of the groups by the average group size (group size = 3.31 individuals).

We chose to report the trapping rates estimated to compare the presence of the species in the individual cells of our study area (Fig. 3). We also reported the standard deviation of the trapping rates across all cells for each species (Table S3), to understand the variability of species presence in our study area.

We gathered available information on species density from the literature, to compare the densities obtained for each species with densities found elsewhere. We selected references published from the year 2000 onwards and referring to population densities estimated in Europe, giving priority to those estimated in Italy whenever available. Since it was not possible to consider only densities estimated with the camera trap method, we took into consideration studies that used different methodologies. Regarding the hare, from the literature review of Smith et al. (2005) we decided to consider only the estimates published from the year 2000 onwards.

Covariates extraction and comparison between corridors

We extracted the values of different environmental covariates within each corridor (mean, median, standard deviation), to make a comparison between the two areas and therefore better explain the differences in the density results we obtained. We considered the following covariates: land-cover (i.e., tree cover, shrubland, grassland, cropland, built up), fractal dimension index, elevation, slope, distance from primary roads, distance from settlements.

We extracted land-cover variables from the European Space Agency (ESA) World Cover portal (Zanaga et al. 2022). The original raster has a resolution of 10 m, and we resampled it at a resolution of 30 m to be consistent with other covariates. Fractal dimension index is a fragmentation metric that is based on the area and perimeter of the patch and describes its complexity. It was extracted using the LecoS plugin on Qgis (3.22.12-Białowieża, 2022; Jung 2016) from the tree cover raster, thus quantifying the fragmentation of forest habitats. To calculate the statistics for the elevation and the slope, we downloaded the Digital Elevation Model DEM GLO-30 (Fahrland et al. 2022) for central Italy from the Nasa Earth Observing System Data and Information System (EOSDIS) and we then calculated the zonal statistics using Qgis Processing tools. Using Qgis, we calculated the Euclidean distance of each camera trap to the nearest village. To calculate the distances of each camera trap from the primary roads, we used the roads’ shapefile downloaded from “OpenstreetMap” (Curran et al. 2013).

Estimating model parameters

We collected a total of 3,942 pictures of 15 species of wild mammals, including 3,379 pictures of our 8 study species (Table S1). We found high variability in the estimate of day range among species – i.e., the average distance that animals travel in a day – with values ranging from 6.07 km/day (SE = 1.27) for the European hare to 15.46 km/day (SE = 1.46) for the red fox (Table S2). The trapping rate – i.e., the number of independent encounters over time – varied widely in the study area, both among species and between the two corridors (Fig. 3, Table S3). The porcupine and the wildcat were not detected in several cells and showed the lowest trapping rates (mean = 0.02, sd = 0.05 encounters per unit time for porcupine; mean = 0.01, sd = 0.02 for wildcat). Conversely, the red deer and roe deer were detected in all cells and showed the highest trapping rates (mean = 0.21, sd = 0.22 and mean = 0.21, sd = 0.16, respectively). The wild boar was also detected in the entire study area except two cells of corridor 2, and showed a high trapping rate in two cells (i.e.,., 0.50 and 0.47; mean = 0.08, sd = 0.12). The red fox showed trapping rates greater than 0 in all the cells of corridor 1 but not in all cells of corridor 2 (mean = 0.10, sd = 0.11). Notably, the porcupine was only detected in three locations of Corridor 2: "Anversa" (0.27), "Lago di San Domenico'' (0.05) and "Valle Santa Margherita" (0.04).

Species density estimates

The wild boar showed the highest density in the overall study area (7.22\(\:\:\pm\:\) 1.75), followed by the roe deer (3.41 \(\:\pm\:\) 0.70 ind/km²) and the badger (1.83 \(\:\pm\:\) 0.82 ind/km²; Table 1). Instead, the wildcat and the red fox were the two species with the lowest densities (0.43 \(\:\pm\:\) 0.17 ind/km² and 1.27 \(\:\pm\:\) 0.27 ind/km², respectively).

To aid interpretation of our results, we compared the population densities estimated in our study area (i.e., the two bear corridors) with those found in the literature (Table 1, Table S4). We collected 29 studies that reported population estimates in Europe for recent years (year 2000 and onwards). Among these 29 studies, five were literature reviews that reported several density estimates (Smith et al. 2005, Melis et al. 2009, Lara-Romero et al. 2012, Mattioli et al. 2014). We found a total of 9 studies that used camera traps to estimate densities, two of which used REM. One of them is a collection of 19 works using REM to calculate the density of wild boars in different European areas (ENETWILD-consortium et al. 2022). Except for deer and roe deer, for which we have estimates within the Abruzzo Lazio and Molise national park (PNALM) using the pellet count method, we have not identified any estimates within the nearby protected areas.

We found that all the densities we estimated were comparable with the data collected from the literature. While important for results interpretation, it is important to clarify that this is not a formal comparison of densities inside vs outside corridors, as the study design behind different estimates was not the same (see Discussion).

Comparison between the two corridors and covariates extraction

The analysis of environmental covariates revealed that Corridor 1 is, on average, more cultivated, more built and less forested than Corridor 2. The fractal dimension index (i.e., patch complexity) was also higher in Corridor 1. Corridor 2, on average, showed a higher altitude and slope. Additionally, the camera traps in Corridor 1 were generally closer to primary roads compared to those in Corridor 2, but almost at the same distance from the nearest village (Fig. S2).

The environmental differences between the two corridors are also reflected in different estimates of population density for several species (Table 1, Fig. S3). In Corridor 1, species with the highest density included the wild boar (14.21 \(\:\pm\:\) 4.64 ind/km²), badger (4.73 \(\:\pm\:\) 2.05 ind/km²) and roe deer (4.67 \(\:\pm\:\) 1.47 ind/km²). In Corridor 2, species with the highest density included the hare (3.60 \(\:\pm\:\) 2.96 ind/km²), red deer (3.36\(\:\pm\:\) 1.50 ind/km²) and roe deer (2.94 \(\:\pm\:\) 0.68 ind/km²). We found Corridor 1 had higher densities than Corridor 2 for all species except the hare (where estimates in the two corridors were very similar). Some of the most notable examples were the wild boar (14.21 \(\:\pm\:\) 4.64 ind/km² in Corridor 1 vs 1.69 \(\:\pm\:\) 0.71 ind/km² in Corridor 2), European badger (4.73 \(\:\pm\:\) 2.05 ind/km² in Corridor 1 vs 1.03 \(\:\pm\:\) 0.34 ind/km² in Corridor 2) and red fox (2.65 \(\:\pm\:\) 0.63 ind/km² in Corridor 1 vs 0.49 \(\:\pm\:\) 0.14 ind/km² in Corridor 2). For the wildcat, data were not sufficient to estimate the densities for the two corridors separately.

Table 1

Density estimates and standard errors (S.E.) derived from the application of Random Encounter Model (REM), for eight species of meso- and macro-mammals. Densities of all species are reported separately for Corridor 1, Corridor 2, and for the whole study area, with the exception of the wildcat (Felis silvestris silvestris) which could only be reported for the whole area. The last column refers to the other densities found from the literature.
Species	Corridor 1 (ind/km²)	Corridor 2 (ind/km²)	Corridor 1 + 2 (ind/km²)	Literature values (ind/km²)
European badger (Meles meles)	4.73 \(\:\pm\:\) 2.05	1.03 \(\:\pm\:\) 0.34	1.83 \(\:\pm\:\) 0.82	Min: 0.26 Max: 3.81 Median: 0.85
Hare (Lepus spp)	3.22 \(\:\pm\:\) 1.58	3.60 \(\:\pm\:\) 2.96	3.39 \(\:\pm\:\) 0.86	Min: 0.0023 Max: 82 Median: 9.15
Porcupine (Hystix cristata)	1.84 \(\:\pm\:\) 0.68	1.08 \(\:\pm\:\) 0.67	1.40 \(\:\pm\:\) 0.47	Min: 0.44 Max: 0.49 Median: 0.46
Red deer (Cervus elaphus)	3.36\(\:\pm\:\)1.50	2.71 \(\:\pm\:\) 0.99	3.16 \(\:\pm\:\) 0.92	Min: 1.72 Max: 8.5 Median: 2.85
Red fox (Vulpes vulpes)	2.65 \(\:\pm\:\) 0.63	0.49 \(\:\pm\:\) 0.14	1.27 \(\:\pm\:\) 0.27	Min: 0.21 Max: 4.4 Median: 0.66
Roe deer (Capreolus capreolus)	4.67 \(\:\pm\:\) 1.47	2.94 \(\:\pm\:\) 0.68	3.41 \(\:\pm\:\) 0.70	Min: 0.11 Max: 53.8 Median: 15.45
Wild boar (Sus scrofa)	14.21 \(\:\pm\:\) 4.64	1.69 \(\:\pm\:\) 0.71	7.22\(\:\:\pm\:\) 1.75	Min: 0.35 Max: 47 Median: 6.54
Wildcat (Felis silvestris silvestris)	NA	NA	0.43 \(\:\pm\:\) 0.17	Min: 0.069 Max: 1.36 Median: 0.33

We estimated the densities of eight meso- and macro-mammal species in two areas of Central Apennines managed as connectivity corridors for the Marsican bear by local administration and NGOs (Ciucci et al. 2016, Maiorano et al. 2019). We estimated densities both at the individual corridor level and for the entire study area, and compared these with density values from other European areas. We found that the densities we estimated are on average higher than those found in the literature, suggesting that bear corridors host areas of high ecological value for several other species. The comparison between the two corridor areas allowed us to understand which environmental conditions could best favour the presence of the species analysed. Corridor 1 showed higher density values than corridor 2 almost for all species. Species such as foxes and wild boars could benefit from the greater (albeit light) anthropic presence in corridor 1, while the roe deer could benefit from mosaic spaces with a high ecotone index characterised by the continuous alternation of open environments with herbaceous vegetation and broad-leaved woods, typical of corridor 1.

The three species of ungulates (i.e., roe deer, red deer, wild boar), as expected, showed the highest densities and the highest trapping rates. This is consistent with the trend of the last few years, which sees these species expanding their ranges and increasing their population numbers (Rondinini et al. 2022). Italian ungulates are mostly opportunistic and generalist species, so they can adapt to several ecological conditions, and they have exploited the massive abandonment of mountains and hills by humans, specifically the more forested environments (Rondinini et al. 2022). Our results constitute important knowledge in itself, since this information is especially scarce in our study area and, more broadly, in the central Apennines.

Density estimates for individual species

We found our results aligning with or exceeding previously reported values in the literature for almost all species (see Appendix SII for additional discussion).

For the European badger, our estimated density (1.83 ± 0.82 ind/km²) was higher than average but still comparable to other European studies that employed camera traps (Lara-Romero et al. 2012). Although Italy lacks comprehensive data on badger populations, one study in the River Po plain using camera traps reported lower densities (0.93–1.4 ind/km²) in hilly regions (Balestrieri et al. 2016).

Regarding the hare, we estimated a density of 3.39 ± 0.86 ind/km², which is on the lower end compared to other European estimates (5.6 ind/km² to 82 ind/km² ) (Smith et al. 2005). However, most of these studies used old methods, such as transect counts or spotlight surveys, which complicate direct comparisons. The only other study conducted in Italy (Genghini and Capizzi 2005) reported much lower hare densities (0.0027 ± 0.0007 ind/km²). For porcupines, our density estimate (1.40 ± 0.47 ind/km²) was higher than most reported in the literature (e.g., 0.49 ind/km² in Lombardy - Palencia et al. 2024). This was probably due to the positioning of our study area within the porcupine’s core distribution area, where habitat changes driven by global warming and agricultural abandonment have facilitated the species’ range expansion (Mori et al. 2021).

Our red deer density (3.16 ± 0.92 ind/km²) closely matched previous estimates for central Italy and aligned with estimates from the PNALM, where a pellet count survey recorded a density of 3.8 ind/km² (Latini 2019). In contrast, roe deer densities in our study (3.41 ± 0.70 ind/km²) fell within the medium-to-low range of European estimates (0.11–53.80 ind/km²) (Melis et al. 2009). In regions with higher predator presence, such as ours, lower roe deer densities are expected, as predators like wolves and bears tend to limit their populations. For instance, in areas with no predators, such as Ticino National Park, roe deer density of 30.7 ind/km² was found (De Pasquale et al. 2019).

Our density estimate of red fox (1.27 ± 0.27 ind/km²) was higher than other European studies (e.g., 0.23–1.62 ind/km² in the Mediterranean area - Jimenez et al. 2019). We found Corridor 1 having more than twice the fox density of Corridor 2, likely due to its greater degree of human presence and lower elevation. Red foxes, being highly adaptable and able to leave in human-dominated landscapes, benefit from such conditions (Alexandre et al. 2020).

For wild boar, our estimated density (7.34 ± 1.78 ind/km²) was consistent with other studies that adopted REM (range values of 0.35 ind/km² in Croatia to 15.25 ind/km² in Italy - ENETWILD-consortium et al. 2022). Due to their high reproductive rate, migratory behavior, and adaptability to various habitats, wild boar densities are notoriously difficult to estimate accurately (ENETWILD consortium et al. 2018). However, the high trapping rate of wild boars in our study suggested a genuinely high density in the area, in line with broader trends of wild boar colonization of urban and peri-urban environments, due to readily available food sources and low hunting pressure (Amendolia et al. 2019).

For wildcat, we estimated a density of 0.43 ± 0.17 ind/km², consistent with previous European studies that used camera traps. For example, Anile et al. (2014) found similar densities (0.32–1.36 ind/km²) in Sicily using different methods, including REM. Other studies in mountainous regions (Maronde et al. 2020, Fonda et al. 2022) reported comparable densities (0.26 and 0.35 ind/km²) using camera traps.

Management implications and Research implications

Our work demonstrated that ecological corridors defined for the Marsican bear host high densities of several other mammal species, highlighting the crucial role that these areas play in supporting mammalian biodiversity in the Central Apennines. Albeit not formally comparable, due to different analytical protocols, the densities of the ungulates (roe deer, red deer and wild boar) were similar with the ones found in the literature in the nearby PAs. Unfortunately, we were unable to conduct a formal comparative study in nearby protected areas using the same methods we deployed in corridor areas, due to strict park protection policies and the significantly larger sampling effort required. Future studies should focus on estimating densities within PAs, where population assessments are missing for many species, or are only reported in internal reports.

We found that other species (e.g., the porcupine) are also present in high densities in the bear corridors. Human-wildlife conflict, particularly between large ungulates or porcupines and local farmers, represents a significant challenge in these areas. These species frequently use agricultural lands for foraging, which increases the likelihood of crop damage and subsequent tensions with local communities. In this sense, it is essential to monitor the population development of those species over an extended time (White and Ward 2010). The installation of electric fences, already recommended for mitigating bear-human conflicts in the LIFE project “Bear-Smart Corridors” (Cipollone et al. 2024), could be effective in preventing damage caused by other species. Moreover, compensation schemes and community engagement initiatives could foster coexistence, reducing the negative impacts of wildlife on human livelihoods while promoting the ecological benefits of maintaining healthy mammal populations. These activities however might fall outside the jurisdiction of park authorities, and will likely require the involvement of governmental and NGO actors.

Since these areas are of great importance for the entire community of meso- and macro-mammals, management strategies should focus on maintaining and enhancing their ecological connectivity, facilitating species movement and dispersal beyond protected areas (Fahrig 2003, Pacifici et al. 2020). This is in line with the Kunming-Montreal Global Biodiversity Framework, which sets a target to protect 30% of the Earth’s land and sea areas by 2030, emphasizing the need for ecological corridors to connect fragmented habitats (CBD 2021). Likewise, this goal is in line with the 2030 EU Biodiversity Strategy and the Italian National Strategy for Biodiversity 2030, which stress the importance of integrating ecological corridors to link isolated PAs (EC 2020, MASE 2023).

Thus, the management of these corridors must reduce human-induced pressures, for instance by mitigating road impact with strict regulation of vehicular access in dirt roads and critical areas during sensitive periods such as mating (Ciucci et al. 2016). Projects like LIFE "Strade" have also demonstrated the effectiveness of measures like road signals and awareness campaigns in reducing wildlife mortality due to road accidents (Valfrè and Cipollone 2016).

The management of ecological corridors in the Central Apennines must focus on enhancing connectivity, reducing human-induced pressures, and promoting coexistence between wildlife and local communities. These actions will ensure the long-term survival of the Marsican bear and other mammal species that rely on these critical habitats (either for movement or for survival). Future research should continue monitoring the population dynamics within these corridors, as well as assessing the broader ecological impacts of ongoing conservation initiatives.

Authors' contributions

Chiara Dragonetti, Moreno Di Marco, Jan-Niklas Trej and Piero Visconti conceived the ideas and designed the methodology; Chiara Dragonetti, Jan-Niklas Trej and Niccolò Ceci collected the data; Niccolò Ceci, Chiara Dragonetti, Jan-Niklas Trej and Stefan Von Kempis analysed the data and performed the Random Encounter Model; all authors contributed to the writing, reviewing, editing and finalizing of the manuscript.

Acknowledgments

We thank all the Rewilding Apennines team and all the volunteers of the 2023 summer season for contributing substantially to the data collection.

We thank the Genzana and Alto Gizio reserve and in particular Antonio di Croce and Antonio Monaco for the exchange of ideas and for intellectually supporting this research. MDM acknowledges support from The European Union–NextGenerationEU as part of the National Biodiversity Future Center, Italian National Recovery and Resilience Plan (NRRP) Mission 4 Component 2 Investment 1.4 (CUP: B83C22002950007).

UNEP-WCMC & IUCN. (2024) Protected Planet: the World Database of Protected Areas (WDPA).
Alexandre, M., D. Hipólito, E. Ferreira, C. Fonseca, and L. M. Rosalino. 2020. Humans do matter: determinants of red fox (Vulpes vulpes) presence in a western Mediterranean landscape. Mammal Research 65:203–214.
Amendolia, S., M. Lombardini, P. Pierucci, and A. Meriggi. 2019. Seasonal spatial ecology of the wild boar in a peri-urban area. Mammal Research 64:387–396.
Anile, S., B. Ragni, E. Randi, F. Mattucci, and F. Rovero. 2014. Wildcat population density on the Etna volcano, Italy: a comparison of density estimation methods. Journal of Zoology 293:252–261.
Balestrieri, A., E. Cardarelli, M. Pandini, L. Remonti, N. Saino, and C. Prigioni. 2016. Spatial organisation of european badger (Meles meles) in Northern Italy as assessed by camera-trapping. European Journal of Wildlife Research 62:219–226.
Biodiversity strategy for 2030 - European Commission. 2024. <https://environment.ec.europa.eu/strategy/biodiversity-strategy-2030_en>. Accessed 1 Jul 2024.
Caravaggi, A., M. Zaccaroni, F. Riga, S. C. Schai-Braun, J. T. A. Dick, W. I. Montgomery, and N. Reid. 2016. An invasive-native mammalian species replacement process captured by camera trap survey random encounter models. Remote Sensing in Ecology and Conservation 2:45–58.
CBD. 2021. First Draft of the Post-2020 Global Biodiversity Framework. CBD/WG2020/3/3.
Cipollone, M., D. Bormpoudakis, A. Carrara, G. Chatzinakos, A. Chovardas, D. D’Amico, Antonio Di Croce, Antonio Di Nunzio, S. Frau, A. Karamanlidis, V. Koutis, P. Leone, Y. Mertzanis, E. Olivieri, S. Psaroudas, A. Tavone, and D. Vavylis. 2024. GUIDELINES FOR A BEAR-SMART COMMUNITY. <https://rewilding-apennines.com/wp-content/uploads/sites/9/2024/06/GUIDELINES-FOR-BSC.pdf>. Accessed 1 Oct 2024.
Ciucci, P., and L. Boitani. 2008. The Apennine Brown Bear: A Critical Review of Its Status and Conservation Problems. Volume 19. <https://about.jstor.org/terms>.
Ciucci, P., L. Boitani, F. Francisci, and G. Andreoli. 1997. Home range, activity and movements of a wolf pack in central Italy. Journal of Zoology 243:803–819.
Ciucci, P., V. Gervasi, L. Boitani, J. Boulanger, D. Paetkau, R. Prive, and E. Tosoni. 2015. Estimating abundance of the remnant Apennine brown bear population using multiple noninvasive genetic data sources. Journal of Mammalogy 96:206–220.
Ciucci, P., L. L. Maiorano, M. Chiaverini, and Falco. 2016. Aggiornamento della cartografia di riferimento del PATOM su presenza e distribuzione potenziale dell’orso bruno marsicano nell’Appennino centrale. Azione A2: Relazione tecnica finale. Ministero dell’Ambiente e della Tutela del Territorio e del Mare e Unione Zoologica Italiana, Roma. 84 pagg.
Curcic, N., and S. Djurdjic. 2013. The actual relevance of ecological corridors in nature conservation. Journal of the Geographical Institute Jovan Cvijic, SASA 63:21–34.
Curran, K., J. Crumlish, and G. Fisher. 2013. OpenStreetMap. Pages 540–549 in. Geographic Information Systems: Concepts, Methodologies, Tools, and Applications. IGI Global. <https://www.igi-global.com/chapter/content/www.igi-global.com/chapter/content/70461>. Accessed 30 Jun 2024.
De Pasquale, D. D., O. Dondina, E. Scancarello, and A. Meriggi. 2019. Long-term viability of a reintroduced population of roe deer Capreolus capreolus, in a lowland area of northern Italy. Folia Zoologica 68:9–20.
EC. 2020. Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions — EU Biodiversity Strategy for 2030 — Bringing nature back into our lives (COM(2020) 380 final, 20.5.2020).
ENETWILD consortium, O. Keuling, M. Sange, P. Acevedo, T. Podgorski, G. Smith, M. Scandura, M. Apollonio, E. Ferroglio, and J. Vicente. 2018. Guidance on estimation of wild boar population abundance and density: methods, challenges, possibilities. EFSA Supporting Publications 15:1449E.
ENETWILD-consortium, P. Acevedo, V. Aleksovski, M. Apollonio, O. Berdión, J. Blanco-Aguiar, L. del Rio, A. Ertürk, L. Fajdiga, F. Escribano, E. Ferroglio, G. Gruychev, I. Gutiérrez, V. Häberlein, B. Hoxha, K. Kavčić, O. Keuling, C. Martínez-Carrasco, P. Palencia, P. Pereira, R. Plhal, K. Plis, T. Podgórski, C. Ruiz, M. Scandura, J. Santos, J. Sereno, A. Sergeyev, V. Shakun, R. Soriguer, A. Soyumert, N. Sprem, S. Stoyanov, G. Smith, A. Trajçe, N. Urbani, S. Zanet, and J. Vicente. 2022. Wild boar density data generated by camera trapping in nineteen European areas. EFSA Supporting Publications 19:7214E.
Explore the World’s Protected Areas. n.d. <https://www.protectedplanet.net/en>. Accessed 5 Oct 2024.
Fahrig, L. 2003. Effects of Habitat Fragmentation on Biodiversity. Annual Review of Ecology, Evolution, and Systematics. Volume 34. Annual Reviews Inc.
Fahrland, E., H. Paschko, P. Jacob, and H. Kahabka. 2022. Copernicus DEM Delivery Review Organisation Note (DEL-06).
Fonda, F., G. Bacaro, S. Battistella, G. Chiatante, S. Pecorella, and M. Pavanello. 2022. Population density of European wildcats in a pre-alpine area (northeast Italy) and an assessment of estimate robustness. Mammal Research 67:9–20.
Fratianni, S., and F. Acquaotta. 2017. The Climate of Italy. Pages 29–38 in M. Soldati and M. Marchetti, editors. Landscapes and Landforms of Italy. Springer International Publishing, Cham. <https://doi.org/10.1007/978-3-319-26194-2_4>. Accessed 13 Mar 2024.
Genghini, M., and D. Capizzi. 2005. Habitat improvement and effects on brown hare Lepus europaeus and roe deer Capreolus capreolus: a case study in northern Italy. Wildlife Biology 11:319–329.
Gervasi, V., and P. Ciucci. 2018. Demographic projections of the Apennine brown bear population Ursus arctos marsicanus (Mammalia: Ursidae) under alternative management scenarios. European Zoological Journal 85:243–253.
Hilty, J., G. L. Worboys, A. Keeley, S. Woodley, B. J. Lausche, H. Locke, M. Carr, I. Pulsford, J. Pittock, J. W. White, D. M. Theobald, J. Levine, M. Reuling, J. E. M. Watson, R. Ament, and G. M. Tabor. 2020. Guidelines for conserving connectivity through ecological networks and corridors. IUCN. <https://portals.iucn.org/library/node/49061>. Accessed 8 Oct 2024.
Hofmeester, T. R., J. M. Rowcliffe, and P. A. Jansen. 2017. A simple method for estimating the effective detection distance of camera traps. Remote Sensing in Ecology and Conservation 3:81–89.
Jimenez, J., R. Chandler, J. Tobajas, E. Descalzo, R. Mateo, and P. Ferreras. 2019. Generalized spatial mark–resight models with incomplete identification: An application to red fox density estimates. Ecology and Evolution 9:4739–4748.
Jung, M. 2016. LecoS - A python plugin for automated landscape ecology analysis. Ecological Informatics 31:18–21.
Keeley, A. T. H., P. Beier, T. Creech, K. Jones, R. H. G. Jongman, G. Stonecipher, and G. M. Tabor. 2019. Thirty years of connectivity conservation planning: An assessment of factors influencing plan implementation. Environmental Research Letters 14.
Lara-Romero, C., E. Virgós, and E. Revilla. 2012. Sett density as an estimator of population density in the European badger Meles meles. Mammal Review 42:78–84.
Latini, R. 2019. STIMA DELLA DENSITÀ DELLA POPOLAZIONE DI CERVO E DI CAPRIOLO NEL PARCO NAZIONALE D’ABRUZZO, LAZIO E MOLISE.
Loy, A., G. Aloise, L. Ancillotto, F. M. Angelici, S. Bertolino, D. Capizzi, R. Castiglia, P. Colangelo, L. Contoli, B. Cozzi, D. Fontaneto, L. Lapini, N. Maio, A. Monaco, E. Mori, A. Nappi, M. Podestà, D. Russo, M. Sarà, M. Scandura, and G. Amori. 2019. Mammals of Italy: An annotated checklist. Hystrix 30.
Maiorano, L., L. Chiaverini, M. Falco, and P. Ciucci. 2019. Combining multi-state species distribution models, mortality estimates, and landscape connectivity to model potential species distribution for endangered species in human dominated landscapes. Biological Conservation 237:19–27.
Mancinelli, S., L. Boitani, and P. Ciucci. 2018. Determinants of home range size and space use patterns in a protected wolf (Canis lupus) population in the central apennines, Italy. Canadian Journal of Zoology 96:828–838.
Maronde, L., B. T. McClintock, U. Breitenmoser, and F. Zimmermann. 2020. Spatial capture–recapture with multiple noninvasive marks: An application to camera-trapping data of the European wildcat (Felis silvestris) using R package multimark. Ecology and Evolution 10:13968–13979.
MASE. 2023. Strategia Nazionale per la Biodiversità (SNB) 2030. Ministero dell’Ambiente e della Sicurezza Energetica Direzione Generale Patrimonio Naturalistico e Mare Divisione III – Strategie della Biodiversità. <www.mase.gov.it/>.
Mateo-Sánchez, M. C., S. A. Cushman, and S. Saura. 2014. Connecting endangered brown bear subpopulations in the Cantabrian Range (north-western Spain). Animal Conservation 17:430–440.
Mattioli, L., P. Forconi, D. Berzi, and P. Perco. 2014. WOLF POPULATION ESTIMATE IN ITALY AND MONITORING PERSPECTIVES. <http://webgis.ar-tel.it/wide>.
McGuire, J. L., J. J. Lawler, B. H. McRae, T. A. Nuñez, and D. M. Theobald. 2016. Achieving climate connectivity in a fragmented landscape. Proceedings of the National Academy of Sciences of the United States of America 113:7195–7200.
Melis, C., B. Jędrzejewska, M. Apollonio, K. A. Bartoń, W. Jędrzejewski, J. D. C. Linnell, I. Kojola, J. Kusak, M. Adamic, S. Ciuti, I. Delehan, I. Dykyy, K. Krapinec, L. Mattioli, A. Sagaydak, N. Samchuk, K. Schmidt, M. Shkvyrya, V. E. Sidorovich, B. Zawadzka, and S. Zhyla. 2009. Predation has a greater impact in less productive environments: variation in roe deer, Capreolus capreolus, population density across Europe. Global Ecology and Biogeography 18:724–734.
Merenlender, A. M., A. T. H. Keeley, and J. A. Hilty. 2022. Ecological corridors for which species? Therya 13:45–55.
Mori, E., G. F. Ficetola, R. Bartolomei, G. Capobianco, P. Varuzza, and M. Falaschi. 2021. How the South was won: current and potential range expansion of the crested porcupine in Southern Italy. Mammalian Biology 101:11–19.
Pacifici, M., M. D. Marco, and J. E. M. Watson. 2020. Protected areas are now the last strongholds for many imperiled mammal species. Conservation Letters 13.
Palencia, P. 2022. REM analysis vignette.
Palencia, P., P. Barroso, J. Vicente, T. R. Hofmeester, J. Ferreres, and P. Acevedo. 2022. Random encounter model is a reliable method for estimating population density of multiple species using camera traps. Remote Sensing in Ecology and Conservation 8:670–682.
Palencia, P., S. Zanet, P. Barroso, R. Vada, F. Benatti, F. Occhibove, F. Meriggi, and E. Ferroglio. 2024. How abundant is a species at the limit of its distribution range? Crested porcupine Hystrix cristata and its northern population. Ecology and Evolution 14:e10793.
Rondinini, C., A. Battistoni, V. Peronace, and C. Teofili. 2022. LISTA ROSSA DEI VERTEBRATI ITALIANI. <www.iucn.it>.
Rowcliffe, J. M., J. Field, S. T. Turvey, and C. Carbone. 2008. Estimating animal density using camera traps without the need for individual recognition. Journal of Applied Ecology 45:1228–1236.
Smith, R. K., N. V. Jennings, and S. Harris. 2005. A quantitative analysis of the abundance and demography of European hares Lepus europaeus in relation to habitat type, intensity of agriculture and climate. Mammal Review. Volume 35. Blackwell Publishing Ltd.
Thornton, D., K. Zeller, C. Rondinini, L. Boitani, K. Crooks, C. Burdett, A. Rabinowitz, and H. Quigley. 2016. Assessing the umbrella value of a range-wide conservation network for jaguars (Panthera onca). Ecological Applications 26:1112–1124.
Valfrè, D., and M. Cipollone. 2016. Realizzazione di opere di mitigazione per la messa in sicurezza della SR 83 “Marsicana” tra gli abitati di Gioia dei Marsi e Opi a favore dell’orso bruno marsicano (Ursus arctos marsicanus) nell’Appennino centrale (Italia). <www.salviamolorso.it>.
Wang, F., W. J. McShea, S. Li, and D. Wang. 2018. Does one size fit all? A multispecies approach to regional landscape corridor planning. Diversity and Distributions 24:415–425.
Watson, J. E. M., N. Dudley, D. B. Segan, and M. Hockings. 2014. The performance and potential of protected areas. Nature. Volume 515. Nature Publishing Group.
White, P. C. L., and A. I. Ward. 2010. Interdisciplinary approaches for the management of existing and emerging human–wildlife conflicts. Wildlife Research 37:623–629.
Zanaga, D., R. Van De Kerchove, D. Daems, W. De Keersmaecker, C. Brockmann, G. Kirches, J. Wevers, O. Cartus, M. Santoro, S. Fritz, M. Lesiv, M. Herold, N.-E. Tsendbazar, P. Xu, F. Ramoino, and O. Arino. 2022. ESA WorldCover 10 m 2021 v200. <https://zenodo.org/records/7254221>. Accessed 31 May 2024.

The authors declare no competing interests.

SupportingInformationprpr.pdf

Download PDF

Version 1

posted

You are reading this latest preprint version

Can bear corridors support mammalian biodiversity? A case study on Central Italian Apennines

Status:

Version 1

Abstract

Figures

Introduction

Material and Methods

Results

Discussion

Declarations

Authors' contributions

Acknowledgments

References

Additional Declarations

Supplementary Files

Status:

Version 1