Wednesday, 15 January 2020
Hall B (Boston Convention and Exhibition Center)
Gerald L. Potter, NASA GSFC, Greenbelt, MD; and L. Carriere, J. Hertz, G. J. Huffman, T. P. Maxwell, J. Peters, and Y. Shen
In order to simplify access to multiple reanalyses, the NASA/NCCS Collaborative REAnalysis Technical Environment (CREATE) has repackaged selected fields from the world’s major reanalysis efforts and published 6-hour, daily and monthly data in the Earth System Grid Federation (ESGF) and on the NCCS THREDDS server. One of the benefits of this service is the addition of a Multi Reanalysis Ensemble (MRE3) mean and spread consisting of the major reanalyses including MERRA, MERRA2, CFSR, ERA-Interim, JRA-55 and most recently, ERA5. For some fields, the MRE3 is more accurate than any single reanalysis. The spread of each variable provided in the MRE3 gives some indication of the uncertainty associated with those variables.
Because precipitation is such a critical variable, it is vital that we understand the limitations of using the reanalysis precipitation products. This presentation demonstrates the use of the CREATE service to extract appropriate data using Earth System Grid Federation (ESGF) or THREDDS, and then analyze using the Earth Data Analysis Service (EDAS), and plot using standard plotting tools with examples from a collection of Jupyter notebooks.
Using tools and resources from the CREATE service, we will present some preliminary analyses of daily precipitation for the MRE3 and the individual reanalyses compared with daily Global Precipitation Climatology Project (GPCP), Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks (PERSIANN-CDR) and Integrated Multi-satellitE Retrievals for GPM (IMERG) data sets. We explore how best to compare the reanalysis precipitation rate with the latest observational data on a variety of time scales.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner