Graphical Abstract

Hirose, H., S. Shige, M. K. Yamamoto, and A. Higuchi, 2019: High temporal rainfall estimations from Himawari-8 multiband observations using the random-forest machine-learning method. J. Meteor. Soc. Japan, 97, https://doi.org/10.2151/jmsj.2019-040.
Early Online ReleaseGraphical Abstract with highlights

Plain Language Summary: The Himawari-8 Rainfall estimation Algorithm (HRA) was developed by using the Random Forest (RF) machine-learning method. Training data is obtained from an Infrared (IR) radiometer of a third-generation geostationary meteorological satellite (GEO), Himawari-8, and Ku-band precipitation radar on board the Global Precipitation Measurement (GPM) core observatory. HRA could estimate heavy rainfall from warm-type precipitating clouds in Kanto-Tohoku heavy rainfall events although the Global Satellite Mapping of Precipitation (GSMaP_NRT) could not estimate such heavy rainfall when microwave satellites were unavailable.

Highlights: