The role of ocean‐atmosphere interaction in Typhoon Sinlaku (2008) using a regional coupled data assimilation system. Issue 5 (4th May 2017)

• Record Type: Journal Article
• Authors: Wada, Akiyoshi; Kunii, Masaru
• Is Part Of: Journal of geophysical research. Volume 122:Issue 5(2017)
• Page Start: 3675
• Page End: 3695
• Publication Date: 2017-05-04
• Subjects: air‐sea coupled data assimilation -- tropical cyclone -- sea surface temperature; Oceanography -- Periodicals; 551.4605
• DOI: 10.1002/2017JC012750
• Languages: English
• ISSNs: 2169-9275

Abstract

Abstract: For improving analyses of tropical cyclone (TC) and sea surface temperature (SST) and thereby TC simulations, a regional mesoscale strongly coupled atmosphere‐ocean data assimilation system was developed with the local ensemble transform Kalman filter (LETKF) implemented with the Japan Meteorological Agency's nonhydrostatic model (NHM) coupled with a multilayer ocean model and the third‐generation ocean wave model. The NHM‐LETKF coupled data assimilation system was applied to Typhoon Sinlaku (2008) along with the original NHM‐LETKF system to investigate the sensitivity of Sinlaku to SST assimilation with the Level 2 Pre‐processed (L2P) standard product of satellite SST. SST calculated in the coupled‐assimilation experiment with the coupled data assimilation system and the satellite SST (CPL) showed a better correlation with Optimally Interpolated SST than SST used in the control experiment with the original NHM‐LETKF (CNTL) and SST calculated in the succession experiment with the coupled system without satellite SST (SUCC). The time series in the CPL experiment well captured the variation in the SST observed at the Kuroshio Extension Observation buoy site. In addition, TC‐induced sea surface cooling was analyzed more realistically in the CPL experiment than that in the CNTL and SUCC experiments. However, the central pressure analyzed in each three experiments was overestimated compared with the Regional Specialized Meteorological Center Tokyo best‐track central pressure, mainly due to the coarse horizontal resolution of 15 km. The 96 h TC simulations indicated that the CPL experiment provided more favorable initial and boundary conditions than the CNTL experiment to simulate TC tracks more accurately. Plain Language Summary: Recent advances in computational technology have made it easier to express direct quantitative representation of extreme mesoscale atmospheric phenomena. Approach by data assimilation is one of promising ways to improve initial conditions by making most use of available observations. However, there is few research on coupled atmosphere‐ocean coupled data assimilation system such that that observational information is shared between the atmosphere and ocean models, and its application to extreme mesoscale atmospheric phenomena. This study shows the development of strongly coupled atmosphere‐ocean data assimilation system and its advantage of tropical cyclone and sea surface temperature analyses as well as their simulations for the first time. The coupled assimilation system well reproduced variations in in situ SST on a weather‐forecasting time scale and provided reasonable TC analysis. Key Points: Strongly coupled atmosphere‐ocean data assimilation system was developed The coupled assimilation system well reproduced variations in in situ SST Coupled assimilation contributes to the improvement of TC simulation