Evaluation of the UKMO's Unified Model (UM) utilizing the Coupled Large-scale Aerosol Simulator for Studies in Climate (CLASSIC) scheme to forecast global aerosols for real-time applications.

The global aerosol environment is extremely complex, making it difficult to provide a real-time operational forecast that includes the state of the aerosols in the atmosphere. The UKMO's Unified Model (UM) has the capability to include the Coupled Large-scale Aerosol Simulator for Studies in Climate (CLASSIC) aerosol scheme to conduct climatological studies, but this study shows that it can also be applied to real-time numerical weather prediction forecasts. The CLASSIC scheme models up to eight aerosol types including sulfates, biomass burning, fossil fuel black carbon, fossil fuel organic carbon, sea salt, mineral dust, nitrates, and secondary biogenic aerosols. CLASSIC is not a full diagnostic chemistry model. Instead, the scheme uses a mix of parametrizations and simplified models to capture the majority of the important chemical impacts without incurring the numerical cost of fully modeled chemistry.

The UM running the CLASSIC scheme was evaluated on its performance in calculating the input parameters for the 550 nm Aerosol Optical Depth (AOD). A basic data assimilation structure with fire data from the Global Fire Assimilation System (GFAS) and initializing the model from the preceding 6 hour forecast was employed. The results were compared to a global 10-year Flow-following finite-volume Icosahedral Model with Chemistry module (FIM-Chem) climatology for each day of the study period that is currently in operations. Truth data for this analysis was the NASA Moderate-resolution Imaging Spectroradiometer (MODIS) satellite-based instrument, the Aerosol Robotic Network (AERONET) ground-based network, and the NASA Modern-Era Retrospective Analysis for Research and Applications, Version 2 (MERRA-2) reanalysis. The model performance was tested in unique aerosol locations to highlight its performance with respect to dust, wildfire smoke, controlled seasonal burning, and sea salts. The analysis will show results of the study comparing the dynamical model to climatology.