APIFLAME v1.0: high-resolution fire emission model and application to the Euro-Mediterranean region

• Main Authors: S. Turquety, L. Menut, B. Bessagnet, A. Anav, N. Viovy, F. Maignan, M. Wooster
• Format: Article
• Language: English
• Published: Copernicus Publications 2014-04-01
• Series: Geoscientific Model Development
• Online Access: http://www.geosci-model-dev.net/7/587/2014/gmd-7-587-2014.pdf

This paper describes a new model for the calculation of daily, high-resolution (up to 1 km) fire emissions, developed in the framework of the APIFLAME (Analysis and Prediction of the Impact of Fires on Air quality ModEling) project. The methodology relies on the classical approach, multiplying the burned area by the fuel load consumed and the emission factors specific to the vegetation burned. Emissions can be calculated on any user-specified domain, horizontal grid, and list of trace gases and aerosols, providing input information on the burned area (location, extent), and emission factors of the targeted species are available. The applicability to high spatial resolutions and the flexibility to different input data (including vegetation classifications) and domains are the main strength of the proposed algorithm. The modification of the default values and databases proposed does not require any change in the core of the model. <br><br> The code may be used for the calculation of global or regional inventories. However, it has been developed and tested more specifically for Europe and the Mediterranean area. A regional analysis of fire activity and the resulting emissions in this region is provided. The burning season extends from June to October in most regions, with generally small but frequent fires in eastern Europe, western Russia, Ukraine and Turkey, and large events in the Mediterranean area. The resulting emissions represent a significant fraction of the total yearly emissions (on average amounting to ~ 30% of anthropogenic emissions for PM_2.5, ~ 20% for CO). The uncertainty regarding the daily carbon emissions is estimated at ~ 100% based on an ensemble analysis. Considering the large uncertainties regarding emission factors, the potential error on the emissions for the various pollutants is even larger. Comparisons with other widely used emission inventories show good correlations but discrepancies of a factor of 2–4 in the amplitude of the emissions, our results being generally on the higher end.