MODELLING AIR POLLUTION AS AN INDICATOR OF THE IMPACT OF THE REGION'S ECOLOGICAL-ECONOMIC SYSTEM ON SOCIAL WELFARE

Air pollution poses a critical threat to public welfare due to its impact on public health and labour productivity. This paper refines and empirically validates a methodology for short-term forecasting of PM2.5 and PM10 concentrations in the air by analysing transport and environmental data using the ensemble method of extremely randomised trees. The authors previously tested the basic methodology on data from New York City. The scientific novelty lies in the extension of the methodology through the introduction of a three-component structure of time lags, the development of economic activity variables based on the interaction between transport demand and the daily rhythms of urban activity, and the confirmation of the generalisability of the approach using data from two geographically distant cities. The results form the basis for the development of cost-effective environmental monitoring systems and evidence-based decision-making in the fields of environmental policy, urban mobility management and the assessment of environmental and economic benefits in the context of public welfare.