Multi-hierarchical approach of potential species distribution modeling and selection of environmental filters of invasive species in the State of São Paulo

Author(s)

Augusto Hashimoto de Mendonça

• Document type: Doctoral Thesis
• Press: São Carlos.
• Institution: Universidade de São Paulo (USP). Escola de Engenharia de São Carlos (EESC/SBD)
• Defense date: 2015-07-13

Examining board members

Giselda Durigan; Alessandra Tomaselli Fidelis; Natalia Macedo Ivanauskas; Antônio Carlos Galvão de Melo; Marinez Ferreira de Siqueira

• Advisor: Giselda Durigan

Abstract

In recent decades, as a result of globalization and international trade, the movement of species away from their native regions has grown in frequency and extent, increasing the risk of biological invasions, which can significantly impact the economy, the functioning of ecosystems and cause biodiversity loss. The eradication of invasive species it is not an easy task, usually with little chance of success, so that prevent the invasion is still the best management alternative. Predictive species distribution modeling techniques have generally been applied to predict the potential distribution of invasive species. In Brazil, studies on biological invasions are recent and insufficient to understand the current invasive status. To enhance understanding about invasive plants and its distribution in the state of São Paulo, in this study we apply predictive modeling techniques to 10 species considered invasive. For each of the selected species we seek to characterize the invasion pattern and identify the active environmental factors that limit or leverage its distribution through multihierarchical ecological niche models. For this purpose, we collect information for these species occurring worldwide and recorded geographical coordinates characteristics of populations and environments invaded by these species in all vegetation types and in all regions of the state of São Paulo. Based on these data, we characterize the invasion pattern of each species and the invasiveness of each vegetation type studied. We applied the multihierarchical ecological niche modeling methodology through MaxEnt algorithm in macro scale for the entire globe and meso scale for the state of São Paulo. Despite the peculiarities of species and vegetation types, our results highlight the influence of ecosystem conservation status and position in the landscape on the severity of the invasion and on the invasiveness of vegetation types on a local scale. Even the most resistant types of vegetation, such as the restinga and the tropical rain forest, can become susceptible to invasion because of disturbances and seedlings pressure. In conserved fragments, however, exotic species were rarely able to settle down and become a real threat to conservation. The more open forms of the Cerrado vegetation types appeared as the most susceptible to invasion by most of the studied species. In macro scale, the ecological niche models identified the potential areas of invasion and revealed the physiological limits of temperature and precipitation for each species, while in meso scale, niche models refined these projections and revealed new patterns associated with the distribution of species on the scale of São Paulo. This study provides important contributions in terms of information about the characteristics and potential areas of invasion for managers and decision makers in prevention and control of invasions and identifies limitations and environmental factors that contribute to a better understanding of biological invasions in Brazil. In general, the multihierarchical approach proved to be a powerful tool to explore patterns of distribution at scales compatible with conservation objectives, prevention and control of alien species. (AU)

• FAPESP's process: 11/05930-0 - Multi-hierarchical modeling of species potential distribution and selection of environmental filters of invasive species in São Paulo
• Grantee: Augusto Hashimoto de Mendonça
• Support Opportunities: Scholarships in Brazil - Doctorate