4.2 Conservation of connectivity networks and core habitat distribution:
Effective conservation of large carnivores requires identifying predicted core habitats and corridor networks between them (Cushman et al., 2018; Khosravi et al., 2018; Shahnaseri et al., 2019). According to our connectivity analysis the southern parts of the study area were predicted to contain the largest extent of potentially suitable habitats for both target species (Figure 5). We identified four and two important core habitats for grey wolf and golden jackal in the southern part of the landscape, respectively (Figure 3 to 5). High connectivity areas in the southern parts of the study area are predicted to connect these core habitats (Figure 6).
Most important linkage for both species occurred in the South from East to West which is strong for both species .Our result showed that C1 at larger dispersal abilities was most important core area for both species. For this regards, considering this core area is essential as the most important landscape conservation area for promoting network connectivity.
Our resistant kernel analysis showed that between 32 – 41 % of identified core habitats for grey wolf are covered by Protected Area networks depending on dispersal ability. For the golden jackal we found a lower contribution of the PAs as core habitats (15-21%), which was explained by species’ association with human dominated landscapes. Our results for golden jackal connectivity are aligned with the outcomes of Shahnaseri et al (2019). For this canid, the highest overlap between core habitats and protected sites was observed for C1 and C2 incorporating a considerable number of villages. In contrast, the highest overlap between grey wolf core habitats and Protected Areas was observed for core numbers C1, C2, C3 and C4 (Figure 4), with 40 % of core habitats intersecting with the protected area network (Figure 4). However, the coverage of Protected Areas is not sufficient, particularly for those cores in the northern and western parts of the study area, due to the small size and wide separation of protected areas in that part of the study landscape.
The largest protected area in the southern part of the study area supports large numbers of natural prey species including wild goat, wild sheep and Persian gazelle and our model predicted this area is an important habitat core area and connectivity node for both species of canid in central Iran. These core habitats have also been documented to have a high potential for supporting other carnivore species such as red fox (Vulpes vulpes ) and Persian leopard (Panthera pardus saxicolor ). However, of Protected Areas’ coverage is not sufficient to safeguard core habitats in the northern and western portions of the study area. Therefore, we believe that the proportion of protected area networks should be increased and located along with strategic locations where the new PAs protect both important unprotected core areas and lie along the most important connectivity corridors through the system.
Similar results and recommendations were produced by Moqanaki and Cushman (2016) and Khosravi et al (2018), who found that Protected Area status is the most important predictor of the occurrence and dispersal of Asiatic cheetah and sympatric carnivores (Asiatic cheetah, Persian leopard, caracal, wild cat, sand cat and grey wolf), respectively. Therefore, protected area network coverage should be accompanied by protected connected land to increase functional landscape connectivity for carnivores.
Most protected areas networks in developing countries, such as Iran, are fragmented by roads, and road collisions present a serious threat for carnivores. In this research, vulnerable parts of the connectivity network were found in the southern part of the study area (C1 and C2) where roads intersected important movement corridors (Cushman, McRae, et al., 2013). The vulnerability of these locations is related to the high potential for grey wolf and golden jackal vehicle collisions. Our findings are similar to Moqanaki and Cushman (2016) (Moqanaki & Cushman, 2017) and Khosravi et al (2018) (Khosravi et al., 2018). They showed that primary and secondary roads cross the predicted corridor paths between the core patches. One of our study’s most novel aspects is the validation of our predicted connectivity maps with independent data on road mortality and crossing locations of both species. Relatively few studies have independently validated connectivity predictions with movement (Cushman et al., 2014), density (Puyravaud, Cushman, Davidar, & Madappa, 2017) mortality, or genetic data (Mateo-Sánchez et al., 2015; Zeller et al., 2018). Our spatial randomization approach provided strong support for our predicted connectivity value. Predicted connectivity is highly related to the actual patterns of observed road mortality and crossing in the study area for both species, giving important independent validation of our predictions. This significantly strengthens their utility for decision-making.