Fukui, S., T. Iwasaki, K. Saito, H. Seko, and M. Kunii, 2018: A feasibility study on the high-resolution regional reanalysis over Japan assimilating only conventional observations as an alternative to the dynamical downscaling. J. Meteor. Soc. Japan, 96, 565-585.
https://doi.org/10.2151/jmsj.2018-056 Graphical Abstract with highlights
- The feasibility of regional reanalysis assimilating only conventional observations was investigated through a one-month regional reanalysis experiment. The system was designed to assimilate surface pressure observations and radiosonde upper-air observations into the Japan Meteorological Agency's nonhydrostatic model (NHM) that was nested in the Japanese 55-year reanalysis, by using the local ensemble transform Kalman filter (LETKF).
- The regional reanalysis overcame the problems of dynamical downscaling approaches with a serial long-term time integration and with frequent reinitializations. It reproduced actual synoptic-scale systems (Fig.1) and precipitation intensity (Fig.2a) and patterns (Fig.2b) better than the downscalings.
- The higher-resolution regional reanalysis (dx = 5 km) reproduced frequency of heavy precipitation and anomalous local fields affected by the topography, such as near-surface circulations and solar radiation associated with low-level clouds, better than the coarser reanalyses.
- The NHM-LETKF was optimized for long-term reanalysis by sensitivity experiments in terms of first guess, lateral boundary perturbations and ensemble size.