PERSIANN-CCS-CDR: A Global Precipitation Climate Data Record for Hydro-climate Studies

Abstract:

PERSIANN-CCS-CDR (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks-Cloud Classification System-Climate Data Record) is a newly developed high-resolution precipitation dataset for hydro-climate studies. This data covering from 60^oS-60^oN globally and from 1983 to near current time was developed by merging PERSIANN-CCS and Global Precipitation Climatology Project (GPCP) monthly precipitation observations. The spatial-temporal resolution of the product is 0.04^ox0.04^o lat-long and three-hourly. PERSIANN-CCS, the main algorithm, is used to extract spatial features of cloud top temperature to the surface rainfall field. Precipitation estimation is provided at every 3-hours for the data period from 1983 to February 2000 using NOAA National Centers for Environmental Information (NECI) GridSat-B1 longwave infrared (IR) image, while half-hour data is estimated after March 2000 and then accumulated to 3-hourly using NOAA Climate Prediction Center (CPC) Global Merged IR data. In order to provide a consistent product over the climate period, monthly estimation is adjusted based on the GPCP monthly estimation and then downscaled to the 3-hour time period.

The PERSIANN-CCS-CDR product is evaluated over various spatial-temporal scales using ground and other satellite observations. Our evaluation of PERSIANN-CCS-CDR using ground observations shows that the proposed dataset is capable of tracking extreme precipitation events, which is critical for hydro-climate analysis. In this presentation, the process of developing PERSIANN-CCS-CDR, the data sources used to generate the product, and product evaluation, as well as the plan for data distribution, will be presented.