Co-variability of smoke and fire in the Amazon basin. (May 2015)

• Record Type: Journal Article
• Authors: Mishra, Amit Kumar; Lehahn, Yoav; Rudich, Yinon; Koren, Ilan
• Is Part Of: Atmospheric environment. Volume 109(2015)
• Page Start: 97
• Page End: 104
• Publication Date: 2015-05
• Subjects: Smoke -- Fire -- AOD -- MODIS -- AERONET; Air -- Pollution -- Periodicals; Air -- Pollution -- Meteorological aspects -- Periodicals; 551.51
• DOI: 10.1016/j.atmosenv.2015.03.007
• Languages: English
• ISSNs: 1352-2310

Abstract

Abstract: The Amazon basin is a hot spot of anthropogenically-driven biomass burning, accounting for approximately 15% of total global fire emissions. It is essential to accurately measure these fires for robust regional and global modeling of key environmental processes. Here we have explored the link between spatio-temporal variability patterns in the Amazon basin's fires and the resulting smoke loading using 11 years (2002–2012) of data from the Moderate Resolution Imaging Spectroradiometer (MODIS) and the Aerosol Robotic Network (AERONET) observations. Focusing on the peak burning season (July–October), our analysis shows strong inter-annual correlation between aerosol optical depth (AOD) and two MODIS fire products: fire radiative power (FRP) and fire pixel counts (FC). Among these two fire products, the FC better indicates the amount of smoke in the basin, as represented in remotely sensed AOD data. This fire product is significantly correlated both with regional AOD retrievals from MODIS and with point AOD measurements from the AERONET stations, pointing to spatial homogenization of the smoke over the basin on a seasonal time scale. However, MODIS AODs are found better than AERONET AODs observation for linking between smoke and fire. Furthermore, MODIS AOD measurements are strongly correlated with number of fires ∼1 0 –2 0 to the east, most likely due to westward advection of smoke by the wind. These results can be rationalized by the regional topography and the wind regimes. Our analysis can improve data assimilation of satellite and ground-based observations into regional and global model studies, thus improving the assessment of the environmental and climatic impacts of frequency and distribution variability of the Amazon basin's fires. We also provide the optimal spatial and temporal scales for ground-based observations, which could be used for such applications. Highlights: Joint statistical analysis of fire and smoke is conducted in the Amazon basin. Analysis shows strong inter-annual correlation between smoke (AOD) and fire (FC). Spatial homogenization of smoke are found over the basin on a seasonal time scale. MODIS-AODs have a stronger correlation with fire properties than AERONET-AODs. Provided optimal spatial-temporal scales for AOD data for different applications.