High Resolution Air Quality Information for Policy

Limited access to reliable scientific information on concentrations of air pollutants and population exposure is an impediment to robust assessments of risks related to air pollution, and for the planning and adoption of effective solutions. The AirQuip project has developed a new computational method, uEMEP ('urban EMEP'), which in combining comprehensive data on local emission sources with an advanced air quality model system (the EMEP model), provides maps of air pollution on high resolution (down to 50m resolution). Information on long-range transported air pollution is obtained daily from the Copernicus Atmosphere Monitoring Service (https://atmosphere.copernicus.eu). uEMEP provides the basis for an operational air quality forecast service for all of Norway (https://luftkvalitet.miljostatus.no), which was established in 2018 and formally launched in January 2019, and which many Norwegian municipalities have started using. Since autumn 2019 the forecasts are also shown on YR (https://www.yr.no/nb) which is run by MET Norway and NRK and which has a large number of users in the Norwegian population. In addition, uEMEP is used in a special service for advanced users (https://www.miljodirektoratet.no/luftkvalitet-fagbruker), analyzing air pollution over longer periods in the past and providing information on exceedances of limit values, contributions from different emission sources, and population exposure. This is valuable information to underpin policy decisions, area planning, etc. In 2019 AirQuip conducted a thorough evaluation of uEMEP against measurement data, for the first winter season with operational forecasts (2018/2019) and for a rerun over several years back in time (2016-2019). The model performs well in general, but there are some deviations, especially during periods of high particle concentrations due to road dust, and for NO2 during cold winter periods when emissions from road traffic are underestimated. In regard to population exposure to poor air quality, uEMEP gives more accurate results than common regional models on coarser resolution. During AirQuip, the Norwegian Environment Agency was tasked to update the knowledge base for limit values for particulate matter and to suggest any necessary modifications. MET has used uEMEP to calculate population exposure over all of Norway and performed a number of scenario calculations to assess how the new limit values could be met. The results were further used by the Norwegian Institute of Public Health to calculate possible health benefits. For dissemination and promotion of uEMEP two seminars were organized by AirQuip in 2018 and 2019 for all Norwegian municipalities to give an introduction about the forecasting service, learn about user experience, and to obtain improved input data from the municipalities (local emission sources, routines for road cleaning and salting, speed limits, reroutings, use of studded tires, etc.). Local knowledge is essential to enhancing the quality of uEMEP further. As a part of AirQuip, the Norwegian Computing Centre developed an algorithm in 2018 and 2019 to detect cars from aircraft and satellite images, supposed to be used to improve available data on traffic volumes and thus road emission data. However, on aerial photographs it remains challenging to distinguish vehicles parked along roads from moving vehicles, leading to an incorrect bias of road traffic towards residential areas and small roads where parking along the road is allowed. For the time being we thus continue using the Norwegian Road database to count how many cars there are on the road and generate exhaust. In collaboration with the Chinese partners of AirQuip, uEMEP is tailored for use in Chinese cities and provinces with poor air quality. In February 2019 Jinan University organized a two-day workshop where uEMEP was introduced to local institutions dealing with air quality (Guangzhou Environmental Monitoring Centre). In addition, calculations with new and improved emission data provided by Tsinghua University were performed (geographical location of point sources, data on road traffic, temporal variation in emissions in China, etc.). We have continued the regional air quality forecasts for all of China, originally developed by the EU project PANDA. EMEP is one of the models contributing to that ensemble (http://www.marcopolo-panda.eu/forecast). Up to now the EMEP model does not perform as well in China as in Norway, but progress has been made since 2018, especially in regard to land use and emissions data. We continue to make uEMEP more user-friendly so that it can be used also beyond Norway and China, e.g. in the Netherlands or parts of South America, where the tool has been tested successfully already.

AirQuip har utviklet beregningsverktøyet uEMEP som ligger til grunn for den nye operasjonelle og åpent tilgjengelige varslingstjenesten for luftkvalitet i hele Norge. Varsling av luftkvalitet øker bevisstheten om problemet blant befolkningen og hjelper beslutningstakere til å vurdere og iverksette tiltak på kort sikt. Brukerne av varslingstjenesten, særlig i risikogrupper, kan tilpasse sin adferd når høy forurensning varsles. Med sin høye romlige oppløsning gir uEMEP forbedret kunnskap om hvor stor del av befolkningen er utsatt for høy luftforurensning, og dermed gi nyttig inngangsdata til helsekonsekvensutredelser, arealplanlegging, osv. Scenarieberegninger hjelper beslutningstakere til å vurdere langsiktige tiltak mot luftforurensning og kan i tillegg brukes til å estimere kostnader og nytteeffekter, for eksempel av nye grenseverdier. Når det gjelder det etablerte samarbeidet med Kina så vil det fortsette utover AirQuip, og på sikt vil samme nytteverdier også gjelde der.

Air pollution has received considerable attention in recent years, especially in China as a result of rapid economic growth, but also in Norway, in particular during winter time. And yet, reliable scientific information on air pollution on local scales remains a limiting factor in accurate population exposure calculations, health impact assessment and air quality policy support. Regional scale data lack the necessary spatial resolution, while local scale data are limitted in coverage and do not capture long-range transported air pollution correctly. AIRQUIP will bridge the gap between regional and local scales by developing a novel downscaling method that relies on a state-of-the-art regional air quality modeling and the acquisition of proxy data on local scales, from high-resolution satellite images and other available sources. AIRQUIP will build on the experience from the European CAMS and PANDA projects, which have been providing operational regional air quality forecasts for Europe and China, respectively. Output from these projects will be downscaled to the local scale for all of Norway and for selected megacities in China frequently hit by severe air pollution. AIRQUIP will benefit from collaboration with partners in China who possess considerable expertise in acquisition of local data and in air pollution modelling. The downscaling method to be developed in AIRQUIP will be equally applicable in all countries where the necessary proxy data are available, notably in Europe. Dissemination of local scale air pollution data will occur through various channels (web pages, reports, workshops), both in Norway and in China. Particular attention will be given to experts on health impact, who require high resolution population exposure data, that will be provided by AIRQUIP. The project will also simulate exposure for future scenarios assuming different sets of emission reduction measures, in order to advise air quality policy makers on short and long time scales.