Evaluation of a statistical forecast model for size-fractionated urban particle number concentrations using data from five European cities

摘要: In this study we evaluated a forecast model for urban aerosol number concentrations against measurements made at five European cities: Helsinki, Stockholm, Copenhagen, Leipzig, and Athens. This requires learning data set with particle concentrations, traffic densities local meteorology. Additionally, in the forecasting process it same parameters from past week forecasted values of weather traffic. The performance was tested based on R2, index agreement (IA), mean square error (MSE), bias. We three modelling approaches: one fixed parameterisation two optimisations either Deviance or Akaike Information Criterion. Based hourly one-day forecasts background sites IA ranged 0.65 to 0.80 accumulation mode particles 0.68 0.87 ultrafine particles. best Helsinki Stockholm worst Leipzig Copenhagen. main reason is more pronounced diurnal variation Stockholm. Another that Copenhagen were not as complete other cities. approaches yielded similar performances, hence conclude simplest parametrisation be preferred.