NEXRA: A weather research analysis system based on the Nonhydrostatic Icosahedral Atmospheric Model with a local ensemble transform Kalman filter

Abstract This article reviews the first version of the NICAM–LETKF JAXA Research Analysis (NEXRA), a weather research analysis system, which was developed from the fusion of the Nonhydrostatic Icosahedral Atmospheric Model (NICAM) and the local ensemble transform Kalman filter (LETKF) using various observational data, such as NCEP PREPBUFR, radiances from Advanced Microwave Sounding Unit (AMSU)-A, and precipitation obtained from Global Satellite Mapping of Precipitation (GSMaP). The development of NEXRA started in 2017 as a collaboration between the Japan Aerospace Exploration Agency (JAXA), the Atmosphere and Ocean Research Institute at the University of Tokyo (AORI), and the RIKEN Center for Computational Science (R-CCS). NEXRA routinely provides 2-D graphical weather information on the web site for general users and ensemble analysis data for research purposes. Routine operation of the first version of NEXRA started on the JAXA Supercomputer System Generation 2 (JSS2) in June 2018 and ended in February 2021 when JSS2 ceased operation. Routine operation of the second version of NEXRA (NEXRA_2.0) started in April 2021 on the JAXA Supercomputer System Generation 3 (JSS3). This article reviews the development and operation of the first version of NEXRA and describes the use of the ensemble analysis data produced by NEXRA as the atmospheric initial conditions for predictability research in the analysis of three events: the record-breaking rainfall in Japan that occurred in July 2018, Typhoon Krosa in August 2019, and the rapid intensification of Typhoon Haishen in August 2020. Finally, some details of the collaborations between NEXRA and other projects and of the development of the current version of NEXRA (NEXRA_2.0) are given, and a future vision for NEXRA is outlined.