Pequenos mamíferos não voadores do Parque Nacional Cavernas do Peruaçu - MG: caracterização e modelagem ecológica

Abstract

With limited resources allocated to the protection of biological diversity, understanding the species present in an area is crucial for developing effective conservation strategies. In the diagonal of open formations in South America, predominantly composed of the Cerrado and Caatinga biomes, despite harboring rich biodiversity and endemism, there is a significant neglect in scientific research, especially compared to tropical forests. These ecosystems are in a race against time due to rapid biodiversity loss, undervalued in terms of conservation. The Cavernas do Peruaçu National Park (CPNP), located in northern Minas Gerais, stands out as a key region for conservation due to its high biodiversity and a significant number of endemic species, including mammals. However, the CNPC mammalian fauna lacks comprehensive studies, especially concerning non-flying small mammals. This study aims to fill this gap by exploring species composition in different environments and investigating factors influencing their occurrence and detection. Encompassing the main phytophysiognomies of the CPNP, such as typical Cerrado, gallery forest, and karstic dry forest, this study used rarefaction curves to assess sampling efficiency and explore the estimated species diversity in the community. Similarity in species composition between sampled environments was evaluated using the Jaccard similarity index. Additionally, occupancy and detection models were constructed, incorporating phytophysionomy and trap type variables, to examine the effect of these parameters on species occupancy and detection probability. The results reveal a total of 16 species in the CPNP, including two species (Rhipidomys macrurus and Wiedomys pyrrhorhinos) not previously recorded in the Vale do Peruaçu region. Rarefaction analysis indicates that both the gallery forest and dry forest reached the asymptote of their curves, while the typical Cerrado did not stabilize completely. Regarding species similarity, there is moderate differentiation between cave environments and typical Cerrado areas compared to other environments. In the constructed models, trap type and head-body length variables had the highest cumulative weights. However, none received robust support. We observed a mean increase in captures of small mammals as the Normalized Difference Vegetation Index (NDVI) increased, suggesting a positive, albeit not significant, relationship. Despite the advances provided by this sampling, coupled with captures from previous studies, we acknowledge the limitation in covering the potentially present species in the CPNP. Model uncertainty underscores the importance of considering imperfect species detection and the need for adjustments in sampling strategies, such as the inclusion of pitfall traps. The concentration of sampling in the central area of the CPNP suggests opportunities in less represented locations, such as the northern region bordering the Xakriabá Indigenous Land. Although NDVI does not reveal a strong relationship with capture rates, it highlights the importance of incorporating seasonality in future studies to better understand the role of vegetation integrity in capture rates. The results of this study enrich the understanding of the richness of non-flying small mammal species in the CPNP, emphasizing the importance of unique environments in the region and the presence of various endemic species from the Cerrado and Caatinga biomes. Furthermore, we reinforce that studies aiming to include seasonal variation, along with the possibility of adjusting sampling strategies through the inclusion of complementary methods, emerge as a promising approach to enhance future collections na fully explore the diversity potential in the Vale do Peruaçu region.