top of page

Carbon Monoxide (CO)

We developed machine learning ensemble models (random forest, light gradient boosting, and neural network) to predict monthly CO concentrations in South Korea.

Model Information
We constructed three machine learning-based models, namely random forest, light gradient boosting, and neural network, to predict monthly CO averages using a 1 km × 1 km grid during 2002–2020. A total of 112 predictor variables collected from the GEE, SEDAC, regional socioeconomic database, and others were used as input variables, and monthly average CO concentrations were predicted as the outcome values. Please refer to the following figures and tables for more specific information.

Table 1. Test results by R2 and rooted mean squared error (RMSE) for each study year and area.

CO_Table1.jpg

Figure 1. Monitoring Map (Above) vs. Prediction Map (Below)

CO_Figure1.jpg

Figure 2. Trend of Predicted CO Concentrations (ppm)

CO_Figure2.jpg

Figure 3. Monthly Average CO Concentration Trend at Monitoring Station

CO_Figure3.jpg
bottom of page