****************************************************************************************************************

第 245回 大気海洋物理学・気候力学セミナー のおしらせ

日 時: 7月 25日(木) 午前 09:30
場 所: 環境科学院 D棟2階 D201号室

発表者: 中山佳洋 (Alfred Wegener Institute, Germany)
\\Yoshihiro Nakayama
題名:On the difficulties of modeling the Amundsen Sea embayment

発表者: Pankaj Kumar Sinha (Faculty of Environmental Earth Science)
題名:Studies on Three Dimensional Variational Data Assimilation of Wind Observations in Simulating Meteorological Systems over the Indian Region

****************************************************************************************************************


On the difficulties of modeling the Amundsen Sea embayment \\(中山佳洋\\Yoshihiro Nakayama)発表要旨 :

In this presentation, I introduce the PhD program in Alfred Wegener Institute, Germany and importance of studying abroad. Then, I briefly present my on going work related to modeling Amundsen Sea, which is summarized below. The ice shelves and glaciers of the West Antarctic Ice Sheet (WAIS) are rapidly thinning, especially in the Amundsen Sea. The melting of Pine Island Ice Shelf (PIIS) increased since 1994. Strong basal melting of this small ice shelf is caused by the relatively warm (?~1°C) Circumpolar Deep Water (CDW) that advances to the PIIS cavity through submarine glacial troughs located on the Amundsen Sea continental shelf. The basal melting may have a large impact on ice sheet dynamics, sea-level rise, and changes in water mass properties of the surrounding ocean. Thus, we simulate the melting of Amundsen Sea ice shelves using the global Finite-Element Sea-ice Ocean Model (FESOM) to investigate the impact of melting on changes in shelf water mass properties. In FESOM, the Amundsen Sea is represented much colder (?-0.8°C at PIIS front) than in reality (?~1?°C). While this was previously considered to be caused by intensive sea-ice formation due to a cold bias in the atmospheric forcing data, a suite of sensitivity studies indicates that even with different forcing and without the influence of sea-ice formation the model still shows an Amundsen Seas much colder than in reality. In this presentation, we discuss the reasons for the cold-biased Amundsen Sea in FESOM by means of several numerical experiments using different forcing, resolution, and model parameters.

Studies on Three Dimensional Variational Data Assimilation of Wind Observations in Simulating Meteorological Systems over the Indian Region \\(Pankaj Kumar Sinha)発表要旨 :

Mesoscale simulation of monsoon depressions (MDs) and tropical cyclones (TCs) are challenging tasks since the errors in the model-predicted parameters are due to the errors in the initial conditions and deficiencies in the model physics. Data assimilation plays a vital role in obtaining the accurate initial state of the atmosphere in the atmospheric models for better model performance. This study investigates the impact of assimilation of ocean surface wind data from various satellites – Quick Scatterometer (QuikSCAT) and Oceansat-2 – and Doppler Weather Radar (DWR)’s reflectivity and radial velocity into the Advanced Research Weather Research and Forecast (ARW-WRF)-V.2.2 and V3.2 models using the three dimensional variational (3DVAR) data assimilation technique for the simulation of MDs and TCs over India. Results of simulations performed with and without data assimilation are compared with the Tropical Rainfall Measurement Mission (TRMM) observations, Joint Typhoon Warning Center (JTWC) observations, and GFS analysis (GFS- ANL) fields for the validation of the forecast. The results of the assimilation of QuikSCAT wind data show a clear impact in the simulations of mean sea level pressure (mslp) fields, 950 hPa wind vectors, thermal structure of the meteorological systems, relative vorticity profile, and the intensity and spatial distribution of the precipitation. The investigation of impact of 3DVAR assimilation of DWR data has indicated noticeable improvements in the simulation of above parameters. Track simulation results obtained due to assimilation of QuikSCAT and DWR data for the cyclone cases show good agreement with the best track of the JTWC while time series of minimum slp and maximum wind speed indicate the observed trend of intensification/weakening of the cyclone. Assimilating Oceansat-2 data for cyclone Bandu reveals improved simulation of an active cyclone, however, results of 3DVAR simulation of cyclones Phyan and Ward do not compare favorably with observations. The month long assimilation of the QuikSCAT wind on the short-range prediction shows improvements in the simulation of precipitation due to data assimilation while the results for the forecast impact (FI) on 850 hPa U- and V- components, FI on 850 hPa and 500 hPa temperature and relative humidity, reveal large positive improvements.

-----
連絡先

松村 義正 (Yoshimasa MATSUMURA)
mail-to: ymatsu@lowtem.hokudai.ac.jp