Analytical Propagation of Emission Uncertainties into CAMSPolicy Products

We present a methodology and first results for analytical propagation of emissions uncertaintiesthrough the EMEP MSC-W chemical transport model (CTM) and an application of these uncertaintyestimates to policy products provided by the Copernicus Atmosphere Monitoring Service(CAMS) for European cities. CTMs are widely employed in atmospheric modeling to simulatethe transport and transformation of pollutants, but uncertainties in emission estimates cansignificantly impact the accuracy of air quality predictions. Our study systematically analyzesthe propagation of uncertainties arising from emissions. The emissions’ uncertainties are consistentwith the CAMS regional emissions product and are calculated using detailed, countryspecificuncertainty estimates in activity data and generic emission factor uncertainties. Theuncertainties are calculated per source sector and country. The Local Fractions/Sensibilities [1]methodology available in the EMEP MSC-W model is a tool that allows computation of sourcereceptorrelationships more efficiently. In conjunction with analytical methods for uncertaintypropagation, we deliver air quality predictions with uncertainty estimates at a fraction of thecomputational cost and with increased traceability compared to modern surrogate modelingtechniques. In our study we focus on PM2.5 and PM10, and first results will be presented for theimpact of emission uncertainties on forecasted PM concentrations in European cities, as well asuncertainties in contributions from different source sectors and countries. By integrating emissionuncertainty propagation, our study aims to provide decision-makers with a more accurateassessment of the reliability of CAMS policy products under various atmospheric conditionsand in the future provide these estimates as part of their operational delivery.     References [1] P. Wind, B. Rolstad Denby, and M. Gauss, “Local fractions – a method for the calculationof local source contributions to air pollution, illustrated by examples using the emepmsc-w model (rv4 33),” Geoscientific Model Development, vol. 13, no. 3, pp. 1623–1634,2020. [Online]. Available: https://gmd.copernicus.org/articles/13/1623/2020/