Research on mesoscale data assimilation and ensemble forecast

Subprogram 1　Development of advanced mesoscale data assimilation methods
Subprogram 2　Development of utilization techniques of observation data
Subprogram 3　Development of mesoscale ensemble prediction methods


Period F. Y. 2009 – F. Y. 2013
Chief Scientist Kazuo Saito (Forecast Research Department)　

Departments in charge
Forecast Research Department, Typhoon Research Department, Meteorological Satellite and Observation System Research Department

Purposes
　 Advanced data assimilation methods which utilize remote sensing observation data are developed to improve the short range quantitative precipitation forecasts in the local municipal scale. Mesoscale ensemble prediction methods are developed to add information of reliability and probability to the short range weather forecast.

Objectives		 　
  1. Development of advanced mesoscale data assimilation methods
    • Development of NHM-4DVAR.
    • Development of a variational data assimilation method using ensemble predictions.
    • Development of the Local Ensemble Transformed Kalman Filter (LETKF) techniques.
    • Improvement of initial fields for typhoon predictions and data assimilation studies in the tropics.

  2. Development of utilization techniques of observation data
    • Development of data assimilation techniques for microwave radiometer data.
    • Development of data assimilation techniques for GPS data.
    • Data assimilation studies using NHM-4DVAR.
    • Data assimilation studies using the Local Ensemble Transformed Kalman Filter (LETKF).

  3. Development of mesoscale ensemble prediction methods
    • Development of initial / lateral boundary perturbation methods.
    • Development of physical process perturbation methods.
    • Evaluation of reliability and probability information in predictions.
    • Studies of ensemble prediction in the tropics.