Occupancy model with continuous-time detection and single-season occupancy model: methods comparison to evaluate invasive-carnivore species interactions

• Grant number: 24/13501-2
• Support Opportunities: Scholarships abroad - Research Internship - Master's degree
• Start date: February 01, 2025
• End date: July 31, 2025
• Field of knowledge: Biological Sciences - Ecology - Applied Ecology
• Principal Investigator: Rita de Cassia Bianchi
• Grantee: Letícia Rodrigues Moretto
• Supervisor: Roland Kays
• Host Institution: Faculdade de Ciências Agrárias e Veterinárias (FCAV). Universidade Estadual Paulista (UNESP). Campus de Jaboticabal. Jaboticabal , SP, Brazil
• Institution abroad: North Carolina State University (NC State), United States
• Associated to the scholarship: 23/12855-2 - Spatiotemporal interaction between invasive species (Canis familiaris and Sus scrofa) and wild canids (Cerdocyon thous and Chrysocyon brachyurus) in areas from Cerrado and Atlantic Forest., BP.MS

Abstract

One of the primary goals for ecologists in this century is to mitigate the biodiversity loss, and the principal step in this task is understanding their causes. Invasive species are one of the drivers of this biodiversity loss, increasingly exerting negative effects on the environment. In Brazil, two of the world's most harmful invasive species, domestic dogs (Canis familiaris) and wild boar (Sus scrofa scrofa) have already infiltrated our protected areas. These species can adversely impact native fauna through competition, predation, and disease transmission. However, accurately quantifying and evaluating these adverse effects poses significant challenges. Traditional methods in mammalian ecology often rely on data from camera traps and occupancy analyses, which may introduce biases due to the grouping of observation occasions. To suppress this limitation, a recent analysis was developed to adopt the concept of continuous time. This approach allows us a more precise model of animal's interactions and can offer a nuanced understanding of how invasive species impact native species. This study aims to compare the outcomes derived from traditional methods with those obtained through the new analysis, that applies the Markov-modulated Poison process (MMPP) to generate continuous-time detection process occupancy models. To compare this, I will use data collected from camera traps deployed across 15 sites in regions of the Cerrado and Atlantic Forest. The research will focus on the interactions between domestic dogs, wild boar, and other carnivores, such as the maned wolf (Chrysocyon brachyurus), crab-eating raccoon (Procyon cancrivorus), and ocelots (Leopardus pardalis). The findings are expected to not only shed light on the immediate effects of invasive species on the native fauna, but also contribute to the development of more robust methodologies to analyze the interactions between species in ecological studies. (AU)