Comparison of Globally Complete Versions of GPCP and CMAP Monthly Precipitation Analyses
- Adler, Robert
- Physical Description:
- 1 electronic document
- Additional Creators:
- Curtis, Scott and Huffman, George
- hdl.handle.net , Connect to this object online.
- Restrictions on Access:
- Unclassified, Unlimited, Publicly available.
Free-to-read Unrestricted online access
- In this study two global observational precipitation products, namely the Global Precipitation Climatology Project's (GPCP) community data set and CPC's Merged Analysis of Precipitation (CMAP), are compared on global to regional scales in the context of the different satellite and gauge data inputs and merger techniques. The average annual global precipitation rates, calculated from data common in regions/times to both GPCP and CMAP, are similar for the two. However, CMAP is larger than GPCP in the tropics because: (1) CMAP values in the tropics are adjusted month-by month to atoll gauge data in the West Pacific, which are greater than any satellite observations used; and (2) CMAP is produced from a linear combination of data inputs, which tends to give higher values than the microwave emission estimates alone to which the inputs are adjusted in the GPCP merger over the ocean. The CMAP month-to-month adjustment to the atolls also appears to introduce temporal variations throughout the tropics which are not detected by satellite-only products. On the other hand, GPCP is larger than CMAP in the high-latitude oceans, where CMAP includes the scattering based microwave estimates which are consistently smaller than the emission estimates used in both techniques. Also, in the polar regions GPCP transitions from the emission microwave estimates to the larger TOVS-based estimates. Finally, in high-latitude land areas GPCP can be significantly larger than CMAP because GPCP attempts to correct the gauge estimates for errors due to wind loss effects.
- Other Subject(s):
- NASA Technical Reports Server (NTRS) Collection.
- Document ID: 20000057325.
- No Copyright.
View MARC record | catkey: 15970366