Prediction of carbon monoxide (CO) atmospheric pollution concentrations using meterological variables

Abstract

This study proposes a two-stage procedure to better predict carbon monoxide (CO) concentrations in the Bay of Algeciras (Spain). In the first stage, a multiple regression model was employed to predict CO concentrations in different monitoring stations using historical data. In the second stage, a new regression scheme was used to forecast the CO concentrations using historical data together with weather forecasts. The experiment shows that two-stage models outperform the single models in the CO concentrations forecasting. The application of the proposed technique may become a supporting tool for the prediction of the values of CO concentrations in complex scenarios such as Bay of Algeciras (Spain).