Study of the impact of wildfire emissions, through MOPITT total CO column, at different spatial scales

Abstract

The estimation of total carbon monoxide (CO) column has been identified as essential to improve our understanding of its role in the global climate system. The Earth Observing System (EOS) Science Steering Committee and the World Meteorological Organization (WMO) has suggested that a satellite-borne CO sensor, which would operate for extended periods, would be useful for that task. Measurements of Pollution in the Troposphere (MOPITT), on board the Terra spacecraft, is a correlation radiometer for estimating CO vertical profiles and total CO column in the lower atmosphere, through the thermal radiance received in the 4.7 μm spectral region. One of the main sources of CO in the atmosphere is the fires and global biomass-burning emissions that are produced when combustion is not complete, especially in the smouldering phase. This article presents a methodology based on a Fourier technique and spatial analysis in order to estimate the total CO column contribution of wildfires at three different spatial scales. First, in a seasonal study, a Mediterranean country (Spain) is selected, and the main regions affected by fire during four years in the summer season are analysed. Second, in order to estimate CO emissions at a local scale, a large fire (in Spain) and a cluster of fires (in North China) are selected. Third, for a global study at large scale and for comparing with CO and carbon dioxide (CO₂) data from Scanning Imaging Absorption Spectrometer for Atmospheric Chartography (SCIAMACHY), locations in North China, equatorial Africa, and Amazonia are selected. Results obtained show that MOPITT data are suitable to assess and to discriminate CO emissions at local spatial scales. Finally, a qualitative agreement between CO behaviour obtained by MOPITT and CO and CO₂ obtained by SCIAMACHY is found.

Acknowledgements

This work was carried out in the framework of the Project Cat-1 ID C1P_4414 funded by ESA. The authors thank the Atmospheric Science Data Center of Langley ASDC User Services, NASA, for providing the MOPITT Level-2 data set; the MODIS Adaptive Processing System (MODAPS), for providing the data of forest fire detections coming from MODIS; and Prof. Buchwitz, University of Bremen, Germany, for providing the SCIAMACHY CO and CO₂ data. This article was funded by the Junta de Castilla y León, project VA089A08, and the CGL2004-00173 project, from the Spanish Science Ministry and the European FEDER funds.