Impact of four-dimensional data assimilation (FDDA) on urban climate analysis

This study employs the NCAR WRF-FDDA (four-dimensional data assimilation) technology to develop an urban-scale microclimatology database for Shenzhen area (~200x200km2, centered at the Shenzhen city). Shenzhen is a rapidly developing metropolitan at the southern coast of China. The NCAR WRF-based Climate FDDA system (C-FDDA) is an innovative dynamical downscaling regional climate analysis system that assimilates various regional observations. The Shenzhen area features uniquely high-density observations including ultrahigh-resolution surface AWS network, wind profilers, radiometers, GPS stations, Doppler weather radars and other weather observation platforms. The C-FDDA has been employed to effectively assimilate these data to produce a five-year multi-scale high-resolution microclimate reanalysis. The C-FDDA system contains four nested-grid domains with grid sizes of 27, 9, 3 and 1 km, with the 1-km grid domain covering Shenzhen metropolitan, Hong Kong, and the surrounding costal regions. This research focuses on the impact of assimilating high-resolution observation data in reproducing the climatological features of urban-scale circulations. Two experiments were conducted with a five-year run using CFSR (climate forecast system reanalysis) as boundary conditions: one with FDDA and the other without. A third experiment was also conducted to assess the data assimilation impact through data-withhold study. Data impact on the micro-climate analysis is investigated with observed station data in terms of traditional statistical metrics, such as bias, root mean square error (RMSE), and mean absolute error (MAE), for both the domain averaged and individual stations. The result indicates that C-FDDA is able to reproduce the microclimatology features such as urban-rural-coastal circulation, land/sea breezes, and local-hilly terrain thermal circulations.