Login / Signup

Fire foci dynamics and their relationship with socioenvironmental factors and meteorological systems in the state of Alagoas, Northeast Brazil.

José Francisco de Oliveira JúniorWashington Luiz Félix Correia FilhoLaurízio Emanuel Ribeiro AlvesGustavo Bastos LyraGivanildo de GoisCarlos Antonio da Silva JuniorPaulo José Dos SantosBruno Serafini Sobral
Published in: Environmental monitoring and assessment (2020)
The objective is to evaluate the fire foci dynamics via environmental satellites and their relationship with socioenvironmental factors and meteorological systems in the state of Alagoas, Brazil. Data considered the period between 2000 and 2017 and was obtained from CPTEC/INPE. Annual and monthly analyzes were performed based on descriptive, exploratory (boxplot) and multivariate statistics analyzes (cluster analysis (CA), principal component analysis (PCA)) and Poisson regression models (based on 2000 and 2010 census data). CA based on the Ward method identified five fire foci homogeneous groups (G1 to G5), while Coruripe did not classify within any group (NA); therefore, the CA technique was consistent (CCC = 0.772). Group G1 is found in all regions of Alagoas, while G2, G5, and NA groups are found in Baixo São Francisco, Litoral, and Zona da Mata regions. Most fire foci were observed in the Litoral region. Seasonally, the largest records were from October to December months for all groups, influenced by the sugarcane harvesting period. The G4 group and Coruripe accounted for 60,767 foci (32.1%). The highest number of fire foci occurred in 2012 and 2015 (between 8000 and 9000 foci), caused by the action of the El Niño-Southern Oscillation. The Poisson regression showed that the dynamics of fire foci are directly associated with the Gini index and Human Development Index (models 1 and 3). Based on the PCA, the three components captured 78.8% of the total variance explained, and they were strongly influenced by the variables: population, GDP, and demographic density. The municipality of Maceió has the largest contribution from the fire foci, with values higher than 40%, and in PC1 and PC2 are related to urban densification and population growth.
Keyphrases
  • air pollution
  • electronic health record
  • machine learning
  • big data
  • cross sectional
  • artificial intelligence
  • deep learning
  • human health
  • energy transfer