Eastern Asian Regional Reanalysis for Surface Meteorological Variables in 1979–2018

Under climate change, researchers in earth science meets an obstacle that lacks a long term surface meteorological variables (e.g. near surface temperature) with high spatiotemporal resolution and high accuracy. High spatiotemporal resolution gridded surface meteorological products are obtained from two methods: one is extrapolated by high dense of meteorological observation network and the other is downscaling simulation by regional numerical weather prediction model. In Eastern Asia region, the gridded products of APHRO, CN05.1，ERA5，and GLDAS, which are made by two methods mentioned above, have low spatiotemporal resolution. Thus, these datasets could not well describe local atmospheric process and boundary layer phenomenon with appropriate spatiotemporal scale in complex terrain of Eastern Asia. The main purpose in this study is to make a long term and high spatiotemporal resolution surface meteorological variables in Easter Asia region (EARR). This study uses WRF (Weather Research Forecast Model) as data assimilation platform. The global atmospheric reanalysis product of ERA-Interim, which is assimilated with plenty of quality-controlled observation, is used as atmospheric forcing and best guess field of atmospheric state. And data from land surface data (e.g. GLDAS and CLDAS), remote sensing snow depth product, sea surface temperature, and density surface observation network are blended as best initial guess for land surface in WRF. Besides, this study considers much physical processes (lake model, topographic wind, etc.), and Four Dimension Data Assimilation (FDDA) with atmospheric nudging of ERA-Interim. In simulation, WRF only runs 36 hours for a day simulation. The data of first 12 hours, which is after 0h best analysis field is considered as spin-up initialization, is remove and data in subsequent 24 hours are kept for forecast field. In this way, this study obtains a grid of 5km with 1 hour interval surface meteorological data in 1979-2018. Compared with ERA5 reanalysis data，which the new generation global atmospheric reanalysis data, EARR is close to ERA5’s performance in China, but EARR can produce more spatiotemporal details (winter’s temperature inversion in mountains, and narrow tube of wind field, etc.).