Combining Machine Learning and Satellite Observations to Predict Spatial and Temporal Variation of near Surface OH in North American Cities

The hydroxyl radical (OH) is the primary cleansingagent in the atmosphere. The abundance of OH in cities initiatesthe removal of local pollutants; therefore, it serves as the keyspecies describing the urban chemical environment. We propose amachine learning (ML) approach as an efficient alternative to OHsimulation using a computationally expensive chemical transportmodel. The ML model is trained on the parameters simulated fromthe WRF-Chem model, and it suggests that six predictiveparameters are capable of explaining 76% of the OH variability.The parameters are the tropospheric NO2column, the tropo-spheric HCHO column, J(O1D), H2O, temperature, and pressure. We then use observations of the tropospheric NO2column andHCHO column from OMI as input to the ML model to enable measurement-based prediction of daily near surface OH at 1:30 pmlocal time across 49 North American cities over the course of 10 years between 2005 and 2014. The result is validated by comparingthe OH predictions to measurements of isoprene, which has a source that is uncorrelated with OH and is removed rapidly andalmost exclusively by OH in the daytime. We demonstrate that the predicted OH is, as expected, anticorrelated with isoprene. Wealso show that this ML model is consistent with our understanding of OH chemistry given the solely data-driven nature