Development of a high-resolution spatial inventory of greenhouse gas emissions for Poland from stationary and mobile sources

Greenhouse gas (GHG) inventories at national or provincial levels include the total emissions as well as the emissions for many categories of human activity, but there is a need for spatially explicit GHG emission inventories. Hence, the aim of this research was to outline a methodology for producing a high-resolution spatially explicit emission inventory, demonstrated for Poland. GHG emission sources were classified into point, line, and area types and then combined to calculate the total emissions. We created vector maps of all sources for all categories of economic activity covered by the IPCC guidelines, using official information about companies, the administrative maps, Corine Land Cover, and other available data. We created the algorithms for the disaggregation of these data to the level of elementary objects such as emission sources. The algorithms used depend on the categories of economic activity under investigation. We calculated the emissions of carbon, nitrogen sulfure and other GHG compounds (e.g., CO 2 , CH 4 , N 2 O, SO 2 , NMVOC) as well as total emissions in the CO 2 -equivalent. Gridded data were only created in the final stage to present the summarized emissions of very diverse sources from all categories. In our approach, information on the administrative assignment of corresponding emission sources is retained, which makes it possible to aggregate the final results to different administrative levels including municipalities, which is not possible using a traditional gridded emission approach. We demonstrate that any grid size can be chosen to match the aim of the spatial inventory, but not less than 100 m in this example, which corresponds to the coarsest resolution of the input datasets. We then considered the uncertainties in the statistical data, the calorific values, and the emission factors, with symmetric and asymmetric (lognormal) distributions. Using the Monte Carlo method, uncertainties, expressed using 95% confidence intervals, were estimated for high point-type emission sources, the provinces, and the subsectors. Such an approach is flexible, provided the data are available, and can be applied to other countries.