High-resolution Dynamical Downscaling for Shenzhen Microclimate

Shenzhen is a major city located in the Pearl River Delta in southern China, with a 230-kilometer-long coastline and an area of about 2,000 square kilometers, including urban and rural areas. In order to improve the weather service to the city, Meteorology Bureau of Shenzhen Municipality (SZMB) has implemented a high-density observation network with advanced remote sensing instruments in the Shenzhen metropolitan and surrounding area. The observation system includes ultra-dense surface Automatic Weather Station (AWS), wind profiler, radiometer, met-tower, Doppler radar, the Global Positioning System (GPS), lightning, and other network platforms. These observational systems/networks provide unique ultra-high spatiotemporal resolution observations for leveraging urban-scale numerical weather prediction. NCAR (National Center for Atmospheric Research) has developed a WRF-based Climate-FDDA (Four Dimensional Data Assimilation) system assimilating all available observation, including SZMB high-resolution observational networks. The NCAR Climate-FDDA has been deployed at SZMB and configured for the Shenzhen area. The system is configured with four nested grid domains with grid sizes of 27 km, 9 km, 3 km and 1 km, respectively. The 1 km domain covers the Shenzhen municipality, Hong Kong, and neighboring areas. The system generates hourly microclimatological reanalysis for the past six years. NCEP CFSR (climate forecast system reanalysis) global reanalysis data are used as the boundary condition and to drive the dynamical downscaling reanalysis. In this paper, we will present the research on: a) adequately customizing and localizing the WRF model (physics) for the costal city region; b) evaluation of data impact on the climate reanalysis through downscaling; and c) initial statistical analysis of several microclimatology features, including city heat island, land/sea breeze, and complex-terrain flows as revealed by the microclimatology reanalysis dataset.