5:15 PM - 6:45 PM
[MGI30-P03] Japan regional reanalysis data and reprocessing R/A precipitaton data as a software infrastructure in data-driven society
Keywords:regional reanalysis, Radar/Raingauge Analyzed Precipitation, societal applications, collaborative efforts of industry, academia, and government
Under JST's COI-NEXT program, a project called ClimCORE is underway at the University of Tokyo's Research Center for Advanced Science and Technology (RCAST), with Professor Hisashi Nakamura as project leader. Based on collaborated research with the Japan Meteorological Agency (JMA), the project is now preparing 5 km mesh MSM-based meteorological reanalysis (RRJ-ClimCORE) and 1km mesh Radar/Raingauge Analyzed Precipitation (hereafter R/A) for the period after 2001.
For RRJ-ClimCORE, data assimilation based on the JMA's operational MSM system (5 km horizontal), which was extended from 76 vertical layers to 96 layers in March 2022, will be output at 3-hour intervals. The output is bridged using 1-hour interval forecast values. Reprocessed R/A is calculated by improved algorithm to the JMA's operational one, with a particular focus on improving the quality of analysis over the ocean.
In this project, we are not only engaged in data creation but also in the collaborative efforts of industry, academia, government, and municipality to explore the societal applications of this data. The data generated by ClimCORE is much more regionally tailored to the needs compared to traditional global reanalysis (30-40km mesh). However, to meet the demands of end-users, it is crucial to further create the more tailored data based on the reanalysis, such as downscaling and reanalysis of ocean and land surface hydrology.
In these developments, progress cannot be made without the involvement of experts in meteorological modeling, oceanography, hydrology, satellite observations, AI/ML, and other relevant fields. Therefore, we would like to provide these experts with a trial version of a product that is close to the final product to promote research and development of a product that is close to the end user's needs. Specifically, in the case of RRJ-ClimCORE data, there are various options for providing data with model vertical levels or standard pressure levels, latitudinal-longitudinal grid coordinates or model computation grids, choices of forecast variables, analyzed values, forecast values, initial values, diagnostic quantities for physical processes, and data formats. Given limited resources, we also recognize the need to narrow down these options based on the utilization research status of this trial product.
Our attempt to create regional meteorological reanalysis and reprocessed R/A precipitation data sets is probably the very first challenge in the world. We expect that world-leading research will be conducted by combining these two sets of data. For example, as to disastrous rainfall unique to Monsoon Asia such as quasi-stationary bands of organized convective clouds (Senjo-kosuitai in Japanese), understanding the mechanisms, long-term trends associated with climate change, improvement of forecast accuracy through verification of many case studies, social applications using AI/ML, and disaster risk reduction practices using past case studies will be expected. We also believe that there are significant advantages to utilizing both meteorological reanalysis and reprocessed R/A data in the reanalysis of land surface hydrology and offshore ocean.
The trial version of data can be provided from July 2018 to December 2022 for RRJ-ClimCORE and from January 2017 to December 2020 for reprocessed R/A. When providing trial data, an application for data use stating the purpose of use, etc. is required due to the collaborative research with the Japan Meteorological Agency (JMA). Please send me an e-mail expressing your informal preferences and intentions about how you would like to use the data.
This work was supported by JST Grant Number JPMJPF2013.
For RRJ-ClimCORE, data assimilation based on the JMA's operational MSM system (5 km horizontal), which was extended from 76 vertical layers to 96 layers in March 2022, will be output at 3-hour intervals. The output is bridged using 1-hour interval forecast values. Reprocessed R/A is calculated by improved algorithm to the JMA's operational one, with a particular focus on improving the quality of analysis over the ocean.
In this project, we are not only engaged in data creation but also in the collaborative efforts of industry, academia, government, and municipality to explore the societal applications of this data. The data generated by ClimCORE is much more regionally tailored to the needs compared to traditional global reanalysis (30-40km mesh). However, to meet the demands of end-users, it is crucial to further create the more tailored data based on the reanalysis, such as downscaling and reanalysis of ocean and land surface hydrology.
In these developments, progress cannot be made without the involvement of experts in meteorological modeling, oceanography, hydrology, satellite observations, AI/ML, and other relevant fields. Therefore, we would like to provide these experts with a trial version of a product that is close to the final product to promote research and development of a product that is close to the end user's needs. Specifically, in the case of RRJ-ClimCORE data, there are various options for providing data with model vertical levels or standard pressure levels, latitudinal-longitudinal grid coordinates or model computation grids, choices of forecast variables, analyzed values, forecast values, initial values, diagnostic quantities for physical processes, and data formats. Given limited resources, we also recognize the need to narrow down these options based on the utilization research status of this trial product.
Our attempt to create regional meteorological reanalysis and reprocessed R/A precipitation data sets is probably the very first challenge in the world. We expect that world-leading research will be conducted by combining these two sets of data. For example, as to disastrous rainfall unique to Monsoon Asia such as quasi-stationary bands of organized convective clouds (Senjo-kosuitai in Japanese), understanding the mechanisms, long-term trends associated with climate change, improvement of forecast accuracy through verification of many case studies, social applications using AI/ML, and disaster risk reduction practices using past case studies will be expected. We also believe that there are significant advantages to utilizing both meteorological reanalysis and reprocessed R/A data in the reanalysis of land surface hydrology and offshore ocean.
The trial version of data can be provided from July 2018 to December 2022 for RRJ-ClimCORE and from January 2017 to December 2020 for reprocessed R/A. When providing trial data, an application for data use stating the purpose of use, etc. is required due to the collaborative research with the Japan Meteorological Agency (JMA). Please send me an e-mail expressing your informal preferences and intentions about how you would like to use the data.
This work was supported by JST Grant Number JPMJPF2013.