Spatial and Temporal Analysis of Fire Patterns and Their Determinants in Amazonia Using Open Earth Observation (EO) Data

Abstract:

Owing to a combination of land cover changes and changes in weather patterns, the fire dynamics in the Amazon basin have undergone considerable change over the past few decades. Extreme climatic events such as El-Nino too play an important role in exacerbating forest fires in the Amazon basin. Human modification also increases the vulnerability of the Amazonian forests to fire events. The main objective of my research is to undertake a detailed spatial and temporal analysis of fire events, their characteristics (and their natural and anthropogenic drivers) in the different countries that overlap with the Amazon rainforest (Brazil, Bolivia, Peru, Colombia, Guyana and Ecuador) using a combination of freely available global Earth Observation (EO) data sources and Artificial Intelligence (AI). The main aims of the research are to identify (1) If and how fire dynamics and their bioclimatic drivers (precipitation, maximum temperature, soil moisture and drought severity) varied between dry and wet seasons of the Amazon basin from 2003-16 (2) If variations in fire and bioclimatic driver dynamics were more pronounced in the El Nino years (3) Identify if and how climatic, topographic, forest structure and human modification variables drive fire dynamics the El-Nino and non-El-Nino years (from 2003-16). These research will identify the relative importance of both the anthropogenic and natural drivers of the different fire attributes between the El-Nino and non- El Nino years. Additionally,the variation in anthropogenic and natural drivers(and their interactions) across the different Amazon countries will be quantified. Examining the variation in fire dynamics and their natural and anthropogenic drivers across the different parts of the Amazonian basin will facilitate a cross-country examination of the variation in fire dynamics. These predictor data included bioclimatic and anthropogenic variables related to the magnitude of human impact. My preliminary research has discovered that for both El-Nino and non-El-Nino years, the magnitude of human modification was the important driver of fire frequency. Further, a positive association was discovered between fire frequency and human modification, indicating areas of the higher magnitude of human modification had a higher fire frequency.