Improved dynamical downscaling methods with a general circulation model (GCM) bias corrections are developed and assessed over East Asia. A set of regional climate simulations is performed with the Global/Regional Integrated Model system (GRIMs) embedded in the Climate Forecast System (CFS) seasonal prediction data for 2005 winter. Four bias correction methods are considered: 1) ensemble average, 2) anomaly nesting, 3) anomaly nesting with standard deviation, and 4) ensemble anomaly nesting. The analysis reveals that the simulation with ensemble anomaly nesting method improves the downscaled climate in both seasonal mean and extreme events relative to the simulations with original CFS data without bias correction.

1. INTRODUCTION

2. DATA AND METHODOLOGY

2.1 Global/Regional Integrated Model System (GRIMs)

2.2 Bias correction methods

2.3 Experimental Design

3. RESEARCH RESULTS

3.1 Impact of bias correction methods on simulated climate

3.1.1 Seasonal mean

3.1.2 Intra‐seasonal variation

3.2 Evaluation of regional forecast with GCM bias correction

4. SUMMARY AND DISCUSSION

REFERENCE

APPENDIX