Association and causality analyses between climate and wildfires using network science

• Grant number: 19/00157-3
• Support type: Scholarships abroad - Research Internship - Post-doctor
• Effective date (Start): April 01, 2019
• Effective date (End): March 31, 2020
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Elbert Einstein Nehrer Macau
• Grantee: Leonardo Nascimento Ferreira
• Supervisor abroad: Jurgen Kurths
• Home Institution: Instituto Nacional de Pesquisas Espaciais (INPE). Ministério da Ciência, Tecnologia, Inovações e Comunicações (Brasil). São José dos Campos , SP, Brazil
• Local de pesquisa : Humboldt University, Germany
• Associated to the scholarship: 17/05831-9 - Analysis of climate indexes influence on wildfires using complex networks and data mining, BP.PD

Abstract

Wildfires are part of a complex system formed by climate, vegetation, and human factors. The fire changes the vegetation and soil, destroys buildings, affects the biodiversity, and influences human activities. Recent studies project that we will face more severe wildfires in the next years and that the traditional fire forecasting methods might be ineffective. Given this pessimist scenario and the high complexity involving fire dynamics, we need better models to study global and regional fire activity. In the last decade, complex networks have emerged as a powerful model for the analysis of complex systems. This model permits the quantification of the features and dynamics of the system in different scales. In the context of climate sciences, complex networks have been used to study many climatological phenomena, but not wildfires. Given the potential of this tool and the complexity of global fire dynamics, the goal of this project is to develop network-based methods to quantify the influence of the climate on fire activity. We will focus on temporal climate networks constructed using two approaches: statistical association measures (correlation networks) and causal inference (causal effect networks). In both cases, we will divide the underlying region into grid cells and use nodes to represent them. The major difference is the way how the links between nodes are established. First, we want to quantify the impacts of climatological phenomena, like El-Niño and extreme events, in the network topologies. Then we will search for temporal patterns in the network topologies using graph mining tools. We expect to better understand the relationship between climate and fire.