The EFFIS forest fire atmospheric emission model: Application to a major fire event in Portugal

2014 
Abstract Forest fires are a major contributor of gaseous and particulate compounds to the atmosphere, impairing air quality and affecting human health. A new forest fire emissions module was developed and integrated into the European Forest Fire Information System (EFFIS), which systematically compiles, since 2000, series of burnt area statistics mapped from satellite imagery. This new forest fire emission model was built on classical methodologies of fuel-map based emission estimation that were improved, especially on burning efficiency, fuel consumption estimation and emission factors. It makes the best use of EFFIS near-real time and detailed information on forest fires, mainly concerning products with a high temporal resolution, which is needed to simulate smoke dispersion and chemical transformation in the atmosphere. A case study of a forest fire event in the north of Portugal on October 14, 2011, with a total of 4400 ha of burnt area, was selected to test this forest fire emission model. The fine scale information used in this study included: (1) 3-h resolution meteorological fields; (2) daily evolution of the cumulative fire perimeter from the EFFIS rapid damage assessment system; and (3) a fine spatial resolution fuel map, forest type map and topography. The 3-h evolution of pollutant emissions was calculated for gas and particulate species based on the evolution of the burnt area increase and fuel consumption. The estimated forest fire emissions represent more than 90% of the total annual (anthropogenic and natural) emissions over the study region. The impact of these forest fire emissions was analyzed in terms of air quality, using observational data from the nearest air quality monitoring station. High peaks of NO 2 and SO 2 were registered simultaneously during the period 06–09 a.m. and a later peak of PM from 07 a.m. to 15 p.m.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    26
    References
    2
    Citations
    NaN
    KQI
    []