资源环境科技发展态势分析平台

In this study, we have developed a global in situ daily precipitation dataset based on quasi-real-time sub-daily observations of precipitation totals for the 0000-0000 UTC (Co-ordinated Universal Time) day everywhere in the world. The sub-daily precipitation data from meteorological stations are obtained via the World Meteorological Organization's (WMO) Global Telecommunication System (GTS) and China Meteorological Administration Net (CMANet) archived by the National Meteorological Information Centre (NMIC) in China and the Integrated Surface Database (ISD) released by the National Centers for Environmental Information (NCEI) in the USA. We have combined these three sources into a global dataset, referred to as NMIC. Accumulated precipitation totals (depending on the country and the WMO region) are transmitted at a variety of times on the GTS. Of these, about 4,500 stations report daily for the 0000-0000 UTC day. Here, we significantly add to this, by developing two-way accumulation algorithms to decompose other reported sub-daily totals to shorter intervals, and then re-cumulate them where possible to the 0000-0000 UTC day. Using these algorithms, we increase by 51.1% of the number of stations during 2009-2016 to around 6,800 day(-1). Additionally, date boundary adjustment (sliding between 1 and 6 hours either side of 0000 UTC) raises the data volume to between 7,800 and 8,700 day(-1). We compare our NMIC product with the First Guess Daily (FGD) product from the Global Precipitation Climatology Centre (GPCC) and GHCN-Daily from NCEI (National Centers for Environmental Information). Root mean square differences between our NMIC and GPCC FGD products over the 2009-2016 period are around 3.4-3.7 mm center dot day(-1) and the average consistency percentage is about 75.1-76.8%. Greater differences between NMIC and GHCN-daily are found which are probably due to the non-uniform date boundary in GHCN-Daily.