ClairCity is a four year European project (2016-2020) working directly with citizens and local authorities in six countries around Europe with the aim to improve air quality. Nine research institutes, six municipalities and one national statistical institute – Statistics Netherlands - work together on models, scenarios and policies to help cities decide on the best local options for a future with clean air and lower carbon emissions. This paper explains the ClairCity project in brief and highlights the enabling role of official statistics in this international, multi-disciplinary project. Six cities are partners in the project; Amsterdam in the Netherlands; Bristol in the UK; Ljubljana in Slovenia; Sosnowiec in Poland; the Aveiro region in Portugal and the Liguria region around Genoa in Italy. Each city faces different issues and causes of air pollution, but all of them are working to improve their air quality. From an official statistics point of view this creates an interesting challenge. Although local circumstances differ, a generic model based approach is applied which asks for good quality data on demography, citizens behaviour, traffic patterns, energy (both production and consumption) and other aspects that may influence local air quality now and in future. A key element of the ClairCity approach is to use existing data to drive the modelling activities. It uses data that is already collected by local, national and European institutes to develop new models of urban air pollution and carbon emissions and scenarios to reduce emissions in future. Although initially developed for the six participating pilot cities / regions, the outputs will be generic so that they can be reused in every European city with over 50,000 residents. This will make it easier for cities to identify changes that they can make to reduce emissions and make a positive change in peoples’ lives. A consequence of this approach is that the project has a very broad and extensive data demand. Moreover, the data needed is usually scattered around multiple data sources and usually not very standardised. That is where official statistics come in. Both the identification of data, the management of multiple - maybe incompatible - data sources and their metadata as well as the visualization of scenario effects are fields where official statistics can - and should - add their knowledge and experience.