Abstract
Model error is an important source of numerical weather prediction (NWP) errors. Among model errors, the systematic diurnal bias plays an important role in high-resolution numerical weather prediction models. The main purpose of this study is to explore the characteristics of the systematic diurnal bias of a high-resolution NWP model in southern China and reduce the diurnal bias to improve the forecast results, hence providing a better background field for data assimilation. Based on the China Meteorological Administration Meso-scale (CMA-MESO) high-resolution NWP model, a 15-day sequential numerical weather prediction experiment was performed in southern China, and the forecast results were analyzed. A sequential bias correction scheme (SBCS) based on analysis increments was designed to reduce the systematic diurnal bias of the CMA-MESO model, and 15-day sequential comparative experiments were carried out. The analysis results showed that the CMA-MESO model has apparent systematic diurnal biases, and the characteristics differ among variables. A large diurnal bias was mainly found in the lower model layers, and it was concentrated in areas with a complex underlying surface for the horizontal distribution, such as the Qinghai-Tibet Plateau and South China Coast. The results based on the 15-day sequential experiment showed that the sequential bias correction scheme partly reduced the systematic diurnal biases of the CMA-MESO model. The mean biases of meridional wind, zonal wind, potential temperature, and water vapor mixing ratio were reduced by 45%, 35%, 20%, and 10%, respectively, and the root mean square errors (RMSEs) were reduced by approximately 5%. This study revealed the characteristics of the systematic diurnal bias of CMA-MESO model in southern China, which may be caused by the diurnal variation in the thermal and dynamic exchange on underlying surfaces. The effectiveness of the sequential bias correction scheme was also verified, and the results had good prospects for providing more reference information for high-resolution numerical prediction models and data assimilation.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.