Keywords

: Air pollution; Discrete time Markov model; Long-term forecasting; Modeling; Scenario analysis; Nitrogen dioxide (NO2)

Start Date

28-6-2018 9:00 AM

End Date

28-6-2018 10:20 AM

Abstract

Abstract: Air pollution management and control are key factors in maintaining sustainable societies. Air quality forecasting is a key factor in these tasks. While short-term forecasting, few days into the future, is a well-established research domain, no method exists for long-term forecasting (e.g., what will be the pollution levels distribution in the next year), which is essential for environmental planning and management as well as epidemiological and exposure studies. This research defines long-term air pollution forecasting and presents a Discrete-Time-Markov-based model for forecasting ambient nitrogen oxides patterns. Specifically, the model provides an estimate for the chance for tomorrow’s pollution level, given today’s level. Thus, it provides an understanding of pollution’s temporal behaviour. The model was applied on two distinctive regions in Israel and Australia, giving a solid base for evaluation. The model did manage to forecast accurately the future transition probabilities.

Stream and Session

F3: Modelling and Decision Making Under Uncertainty

Download

Share

COinS
 
Jun 28th, 9:00 AM Jun 28th, 10:20 AM

Long-Term Forecasting of Nitrogen Dioxide Ambient Levels in Metropolitan Areas Using the Discrete-Time Markov Model

Abstract: Air pollution management and control are key factors in maintaining sustainable societies. Air quality forecasting is a key factor in these tasks. While short-term forecasting, few days into the future, is a well-established research domain, no method exists for long-term forecasting (e.g., what will be the pollution levels distribution in the next year), which is essential for environmental planning and management as well as epidemiological and exposure studies. This research defines long-term air pollution forecasting and presents a Discrete-Time-Markov-based model for forecasting ambient nitrogen oxides patterns. Specifically, the model provides an estimate for the chance for tomorrow’s pollution level, given today’s level. Thus, it provides an understanding of pollution’s temporal behaviour. The model was applied on two distinctive regions in Israel and Australia, giving a solid base for evaluation. The model did manage to forecast accurately the future transition probabilities.