The Japan Meteorological Agency (JMA) launched a next-generation geostationary meteorological satellite, Himawari-8, on 7 October 2014 and began its operation on 7 July 2015. Advanced Himawari Imager (AHI) on board Himawari-8 has 16 observational bands (including three visible and three near-infrared wavelengths) that enable retrieval of full-disk map of aerosol optical properties (AOPs) including aerosol optical thickness (AOT) and Ångström exponent with unprecedented spatial and temporal resolution. Here we show an application of AOT derived from Himawari-8 to an aerosol assimilation/forecasting system for a heavy smoke event caused by a Siberian forest fire in May 2016. The assimilation of the Himawari-8 AOT compensates underestimates of the smoke in model simulation without data assimilation and provides a better forecast performance. These promising results suggest that the Himawari-8 AOPs would contribute not only monitoring but also improvement in forecasts of air quality.