Dados do Trabalho

TÍTULO

MODELING THE GEOGRAPHIC DISTRIBUTION OF MYOCASTOR COYPUS (MAMMALIA, RODENTIA) IN BRAZIL: ESTABLISHING PRIORITY AREAS FOR MONITORING AND AN ALERT ABOUT THE RISK OF INVASION.

Resumo

Myocastor coypus is a large semi-aquatic rodent ranked between the 100 most invasive species in the world. This species can alter natural habitats by feeding on aquatic vegetation, destroying nests and preying on eggs of several aquatic birds,  besides feed on a variety of crops and weaken riverbanks through its burrowing activity. In Brazil, few ecological studies have been proposed to understand the relationship of this species with the environment. Due to the little information about this species occurrence in Brazil, it is difficult to estimate their relationship with environmental and develop management and conservation actions. Environmental niche models are a common goal in ecology, and a practical approach to understand potential species distribution, based on the relationship between occurrences and environmental information. Thus, the present study aimed to understand the geographic distribution of M. coypus and indicate areas of greater risk of establishment based on bioclimatic predictors and a surveillance map.Native species occurrence data were obtained from Global Biodiversity Facility Database. Species occurrences in Brazil were obtained searching published data available in scientific journals and Field observation. Climate data were obtained from the WorldClim database. In addition, because the presence of M. coypus in human associated landscapes, we choose to input as additional environmental information the human footprint environmental layer, that measure the Human Influence Index in the landscapes. We used three different classes of algorithm  (i.e. confidence interval, distance measure and machine learning) to model M. coypus. For each model, we use 75% of the occurrence data (Nnative = 78, NBrazil = 16) to calibrate the models, and 25% of the occurrence data (Nnative = 26, NBrazil = 6) to validate the models. The results demonstrated that M. coypus suitability and risk assessment areas are restricted to the southeastern and southern regions of Brazil. Variables presenting the highest relative importance values were ‘Human Footprint’, ‘Mean Temperature of the Coldest Quarter’, and ‘Minimum Temperature of the Coldest Month’ . Human Footprint was positively associated with M. coypus suitability (r = 0.6), while Mean Temperature of the Coldest Quarter (r = -0.48), Min Temperature of the Coldest Month (r = -0.35) and Annual Precipitation (r = -0.31) were negatively associated with the presence of this species. The surveillance map (i.e., the combination of binary maps) indicating that the prevalence observed for the south and Southern regions were classified at lower and moderate grid cell values. High risk classification cells are concentrated in the states of Santa Catarina and Rio Grande do Sul. Further, small high-risk patches were distributed along the state of Paraná. Due to the environmental impacts caused by this species, the monitoring in environments where it has been introduced is required. The model used herein presented efficient applicability and fit for Brazil. Preventive actions and the management of M. coypus in predicted regions prior to its establishment are recommended.

Palavras-chave

Biological distribution; ecosystem engineering; invasion; risk assessment; nutria.

Financiamento

This work was supported in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - [Proc. 1689817].