A rose analysis method relating air chemistry to fire distribution in tropical Africa

• Published in: Atmospheric research Vol. 50; no. 2; pp. 81 - 104
• Main Authors: Brivio, P.A, Grégoire, J.M, Cros, B, Galy-Lacaux, C, Lacaux, J.P
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01.02.1999; Elsevier
• Subjects: Africa; Air; Chemical composition and interactions. Ionic interactions and processes; Earth, ocean, space; Exact sciences and technology; External geophysics; Fire; Meteorology; Africa; Africa; Fire; Air; Gas emission; Analysis method; Spatial distribution; Atmospheric chemistry; Satellite observation; Radiometry; Canopy fire
• ISSN: 0169-8095; 1873-2895
• DOI: 10.1016/S0169-8095(98)00096-9

Emissions from vegetation fires in the tropics contribute significantly to atmospheric chemistry at continental and global scales. Africa is the continent where the largest number of vegetation fires annually occurs. The use of Global Area Coverage (GAC) data, from the Advanced Very High Resolution Radiometer (AVHRR) sensor onboard the NOAA satellites, allows a good description at continental level of the spatial distribution of intense active fire patterns and a unique temporal perspective over multi-annual periods of time. In this study, 5 years of active fire maps, from November 1984 to October 1989, derived from GAC data for the African continent were considered. In order to enhance the relationships between spatially distributed phenomena, such as regional patterns of active fires, and point data, such as experimental sites of atmospheric chemistry measurements, the rose diagram technique as derived from directional statistics was exploited. This methodological approach allowed the spatial characteristics of vegetation fires, which is considered the main source of atmospheric pollutants in Africa, to be described using measures of frequency and distance for each reference point. Comparison between rose diagrams of fires and nitrate (NO 3 −) content in precipitation as measured at Enyele (Congo), at the heart of the equatorial forest during both dry and wet seasons, confirmed the usefulness of satellite observations and the spatial analysis technique proposed.