Global Precipitation Climatology Project (GPCP)

1. Global Precipitation Climatology Project(GPCP) Arnold Gruber Director of the GPCP WCRP Workshop on Detrermination of Solid Precipitation in Cold Climate Regions 9-14 June 2002, Fairbanks, Alaska

2. Global Precipitation Climatology Project • Organized in 1986 • Component of the Global Energy and Water Cycle Experiment (GEWEX) of WCRP • Objectives: • Improve understanding of seasonal to inter-annual and longer term variability of the global hydrological cycle • Determine the atmospheric heating needed for climate prediction models • Provide an observational data set for model validation and initialization and other hydrological applications

3. Global Tropics NOAA - National Weather Service Global Precipitation Climatology Project IR Component Data Processing Centres GMS Meteosat GOES NOAA JAPAN EUROPE UNITED STATES MW Component CAL/VAL Component Polar Satellite Precipitation Data Center Geostationary Satellite Precipitation Data Centre Surface Reference Data Centre emission scattering (EVAC - U.OK) (ocean) (land+ocean) Algorithm Intercompararison Program NASA-GSFC NOAA-NESDIS New Observations GPC Merge Development Centre Merged Global Analysis NASA - GSFC Station Observations (CLIMAT, SYNOP National Collections) Gauge - Only Analysis Global Precipitation Climatology Centre DWD - GERMANY

4. Remote Sensing Estimates used in GPCP • Infra –red • GOES • RR linearly related to fractional pixels Tcld<235K • Most effective for deep convective clouds, used only in 40N,S zone • High spatial and temporal resolution • false signatures, insensitive to warm top rain • TOVS • Regression between cloud parameters and rain gauges • Used in high latitudes where MW techniques are poor • Microwave (SSM/I) • Closely related to hydrometeors • Emission from cloud drops ( 29 GHz). Most effective over water surfaces ( Tsfc <<Tcld) • Scattering by ice particles over land over land ( 89, Tcld< Ta) • only ice clouds over land, low resolution, no estimate over snow and ice

5. IMPORTANT POINT Algorithms are designed for liquid precipitation Gauges Used to produce a gridded analysis, incorporates water equivalent of solid precipitation Final GPCP Precipitation Field satellite estimates adjusted to large scale gauge analysis ( water equivalent of solid precipitation incorporated in this stage) Satellite data merged with gauge analysis using inverse error variance weighting

6. Global Precipitation Climatology Project • Current Products • Monthly mean 2.5°x2.5° latitude/longitude • Merged satellite and gauge, error estimates • Satellite components: microwave and infrared estimates, error estimates • Gauge analysis, error estimates • Intermediate analysis products, e.g., merged satellite estimates • Daily 1 x 1 degree, Pentad July 1987 and continuing -Version 1 1985 1990 1995 2000 1979 & Continuing- Version 2 , Pentad 1997 Daily 1x 1, deg

7. Global Precipitation Climatology Project Annual Mean Precipitation mm/day

9. 1 x 1 degree, daily precipitation January 1, 1998 mm/day

10. Summary • Needs for solid precipitation • Atmosphere: Latent heat of fusion is important diabatic heat source • Surface: albedo affects land atmosphere energy exchange; important for surface hydrologic applications. e.g., floods, water resources, etc. • GPCP • No direct measure of solid precipitation rate – only water equivalent when adjusted to gauges • Can lead to bias where there are no gauges • Need validation and feedback on GPCP precipitation estimates over land areas • Techniques being developed to identify solid precipitation – need observed rates for calibration/validation

11. Global Precipitation Climatology Data • Monthly Mean 2.5 x 2.5 degree – 1979 and continuing • Pentad ( 5 day) 2.5 x 2.5 degree – 1979 and continuing • Daily, 1 x 1 degree - 1997 and continuing • Available On Line from World Data Center A at The National Climatic Data Center: • http://lwf.ncdc.noaa.gov/oa/wmo/wdcamet-ncdc.html

12. Potential of detecting falling snow over land using AMSU-B • Preliminary study (ongoing) using AMSU-B (89, 150, 183+1, 3, 7 GHz) • Great Plains U.S (flat, homogeneous) • Cases where no snow existed, active snowfall and remaining snowcover after precipitation event • Ancillary data: • NEXRAD composites • Synoptic weather reports/first order stations • Hourly precipitation amounts • QC of AMSU & surface reports to insure that proper surface and weather types have been classified

13. Preliminary Findings • Use of AMSU-B 150 and 176 GHz appears to be best set of channels • Single channel inadequate • More channels may not add much more information • Application to case studies seems promising, but • Need to consider false alarm rate • Need to determine global applicability • Need to determine sensitivity to snowrate, cloud physics, etc.

15. Algorithm Enter Rain Rate Algorithm Snow on Ground? NO YES Snow Index = 6.4 + 0.213*TB150-0.043*TB176 No Falling Snow Snow Index > 0.60? NO YES TB176 > TB180? NO YES Snow is Falling