Environmental data such as carbon monoxide (CO), precipitation, air temperature, and traffic have recently drawn the attention of researchers. Several time series models such as Seasonal Autoregressive Integrated Moving Average (SARIMA) and Vector Autoregressive (VAR) models have been applied to forecast CO. A VAR model can study extrinsic and intrinsic variables together but it does not incorporate seasonality. An Airline model is a special case of SARIMA which uses only an intrinsic variable and incorporates seasonality. The purpose of this thesis is to propose a time series model which incorporates seasonality and uses an extrinsic variable to forecast an intrinsic variable. The model is called Airline Error Correction Model (AECM). This thesis uses AECM to forecast CO using traffic, precipitation and air temperature as extrinsic variables. The forecasts using different models of AECM are compared to forecasts using VAR and Airline models. The results of the study show that AECM does a better job on forecasting than VAR and Airline models.