Ocorrência das espécies de carnívoros e artiodáctilos do Parque Nacional de Itatiaia e suas relações com o habitat

Abstract

The biodiversity is directly or indirectly affected by chances in systems ecological systems, which modify the composition, distribution and interaction between species. Developing methods to evaluate the consequences of the impact related to such changes, is necessary. Non-invasive sampling methods associated with molecular and landscape analyses allows us to identify and obtain information about wild animals that may suffer the consequences of new environmental. In the context, we investigated the selection and preferential use of the habitat by carnivorous and artiodactyls at a conservation unit of integral protection, as well as analyze the relationship between species living in this park, the Itatiaia National Park (PNI). Between June/2017 and April/2018 samples (feces) with compatible morphology of carnivorous and artiodactyls were collected and its geographic coordinate was recorded using a GPS device. These samples was evaluated through molecular identification, sequencing a fragment of the gene 12S and ATP6 for carnivorous and COI for artiodactyls. We identified 175 of 244 samples collected, the species found: C. brachyurus (38,5%), L. guttulus (21,3%), C. familiaris (5,73%), C. thous (1,22%) e S. scrofa (4,91%). The following variables were analyzed: land use and cover; digital elevation model; slope and Euclidean distances from water bodies and from the nearest buildings, to develop the landscape analysis. The relationship of landscape and topographic variables with less sampled species was performed descriptively while for species widely sampled (more than 50 samples) logistic regression was used. Considering that a sample can only belong to one species and using model selection Akaike Method, Backward method and biological sense for validation. Samples of Chrysocyon brachyurus and Leopardus guttulus have a wide distribution in the PNI, totalizing 59,7% of the identified samples, meanwhile Cerdocyon thous, Canis lupus familiaris and Sus scrofa were sampled at specific locations. The preferred habitat of C. brachyurus and L. guttulus depends on altitude and land use and land cover. Higher altitudes represent a bigger probability to find the species C. brachyurus, with higher frequency in regions of grassland formation. Although the occurrence of L. guttulus being related with lower altitudes, we found samples of this species at altitudes of 2544 meters, the highest found to date. L. guttullus shows a highter frequency in forest formations. There is a correlation between the two variables in question, however, this relationship is not strong enough to influence the estimates of the model parameters, justifying the use of both variables. The artiodactyls samples were found only in mosaic of agriculture and pasture, but traces were seen in others areas of PNI. Non-invasive sampling complemented with molecular tools are efficient for the detection of species with elusive behaviors such as C. brachyurus and L. guttulus. Landscape analysis evidenced that these animals have preferential habitats within the PNI.