IPWG Precipitation Validation / Intercomparison Study

Data Archive

Project overview

Several satellite precipitation algorithms are run operationally and semi-operationally from national centers and universities to produce rainfall estimates for time periods ranging from half-hourly to monthly. Many of these rainfall products are obtainable by the public via the web or FTP, and are being used for a variety of meteorological, climate, hydrological, agricultural, and other applications.

In order to use these rainfall estimates appropriately it is important to have an idea of their accuracy and expected error characteristics. This is done by validating the satellite precip­itation estimates against "ground truth" from rain gauge and/or radar observations. To get good estimates of absolute accuracy satellite products should be verified against very high quality radar and gauge data (from the TRMM verification sites, for example). To get estimates of regional and spatial accuracy it is necessary to use a much larger quantity of data, for example, from national rain gauge networks.

A thorough verification and intercomparison of satellite-based precipitation products should quantify their accuracy in a wide range of weather and climate regimes, give users information on the expected errors in the estimates, help algorithm developers understand the strengths and weaknesses of the satellite rainfall algorithms, including which aspects are in greatest need of improvement, monitor the performance of existing algorithms, and assist with evaluating algorithm upgrades.

In 2003 the International Precipitation Working Group began a project to validate and intercompare operational and semi-operational satellite rainfall estimates. In addition to providing users with product accuracy information, this project provides algorithm developers with an opportunity to test their algorithms and compare their output to other rainfall products. The focus so far has been on validating daily rainfall estimates against operational daily rainfall analyses (Australia and US) and summed radar rainfall estimates (US and Europe), with results updated on a daily basis. Rain forecasts from a few numerical weather prediction models (a potential alternative source of rainfall estimates for users) are also verified for comparison. The validation results can be viewed on the web at

http://www.bom.gov.au/bmrc/SatRainVal/sat_val_aus.html (Australia)
http://cics.umd.edu/~johnj/us_web.html (US)
http://kermit.bham.ac.uk/~kidd/ipwg_eu/ipwg_eu.html (Europe)

Scientists and precipitation data users are encouraged to participate in this ongoing satellite precipitation validation/intercomparison research. Opportunities for involvement include:

(a)  validation / intercomparison of satellite precipitation estimates in additional regions of the globe, using high quality and/or national rainfall reference data,
(b)  evaluation of satellite precipitation estimates as input to weather, climate, hydrological, and agricultural models and applications,
(c)  provision of satellite precipitation estimates for validation / intercomparison with other rainfall estimates,
(d)  provision of surface reference data for validation purposes.

Data archive

To facilitate the participation of the larger community, an archive of daily satellite precipitation estimates and surface reference data has been established at the Cooperative Institute of Climate Studies (CICS) at the University of Maryland. Included are IR, passive microwave, and combined microwave+IR satellite rainfall estimates from a large number of algorithms (see Table 1). Most of these satellite estimates have global or near-global coverage. Daily gauge analyses and radar analyses are provided for the continental U.S.; reference data for other regions may be added at a later date.

The archive starts in January 2004, with 24h accumulations valid at 00, 06, and 12 UTC. It is updated on a daily basis around midnight, USA Eastern Time.

The data are freely available via anonymous FTP:

ftp server:	cics.umd.edu
user name:	anonymous
password:	<your e-mail address>
directory:	/pub/DATA/Validation/

There are three subdirectories for 00Z, 06Z, and 12Z daily data.
Due to space limitations on the server, only data from January 2006 onward are included in the directory listed above. Data from 2004 and 2005 are available from the same FTP server in directories /pub/data1/old-data/IPWG2004 and /pub/data1/old-data/IPWG2005, respectively.

Click here to download a PDF document giving a brief description of each dataset including information on the nature of the product, contact information, rainfall units, grid dimensions, starting point and spatial resolution, value for missing data, and GrADS control files. This document is also available on the FTP site given above.

Detailed documentation for some of the satellite algorithms and datasets can be found at the IPWG web site, http://www.isac.cnr.it/~ipwg/algorithms/algorithms-invent.html.

Contact information

     IPWG precipitation validation / intercomparison project:

     Data archive:

Table 1. Datasets in the archive as of December 2004.

Dataset	Longitude range	Latitude range	Daily array size and starting point	Record length (byte)	00Z	06Z	12Z	Starting date (yyyymmdd)
Satellite precipitation estimates:
3B40RT	180oW –180 oE	90oN – 90oS	1440x720 (89.875oN, 179.875oW)	4	x			20040101
3B41RT		60oN – 60oS	1440x480 (59.875oN, 179.875oW)
3B42RT
3b40rt	0o – 360o	90oS – 90oN	Before 7 July 2004 1440x721 (90oS, 0o,) Since 7 July 2004 1440x720 (89.875oS, 0.125oE)	4			x	20031217
3b41rt		60oS – 60oN	Before 7 July 2004 1440x481 (60oS, 0o) Since 7 July 2004 1440x480 (59.875oS, 0.125oE)
3b42rt
NRLGEO	Before 7 July 2004     0o – 360o Since 7 July 2004 125oW – 65 oW	Before 7 July 2004 60oS – 60oN Since 7 July 2004  20oN – 50oN	Before 7 July 2004 1440x481 (60oS, 0o) Since 7 July 2004 241x121 (20.125oN, 124.875oW)	4			x	20031217
NRLBLD								20040226
GEO*	0o – 360o	60oN – 60oS	1440x480 (59.875oS, 179.875oW)	2	x		x	20040201
PMW*
AMSU	0o – 360o	60oS – 60oN	1440x481 (60oS, 0o)	4	x	x	x	00Z: 20031217 06Z: 20040524 12Z: 20031217
CMORPH	0o – 360o	60oS – 60oN	1440x481 (60oS ,0o)	4			x	20031217
IRRAIN	0 – 360o	60oS – 60oN	1440x481 (60oS, 0o)	4			x	20031217
MWCOMB	0 – 360o	60oS – 60oN	1440x481 (60oS, 0o)	4			x	20031217
CPC+COMB	0o – 360o	60oS – 60oN	1440x480 (59.875oS, 0.125oE)	4	x			20031205
GMSRA	125oW – 65 oW	20oN – 50oN	241x121 (20oN, 125oW)	4			x	20031217
GMSRAD	125oW – 65 oW	20oN – 50oN	241x121 (20oN, 125oW)	4			x	20040122
GPI	0o – 360o	60oS – 60oN	1440x481 (60oS, 0o)	4			x	20031217
HYDROE	125oW – 65 oW	20oN – 50oN	241x121 (20oN, 125oW)	4			x	20031217
EW	0o – 360o	60oS – 60oN	1441x481 (60oS, 180oW)	4	x		x	20040313
PERSIANN	125oW – 65 oW	20oN – 50oN	241x121 (20oN, 125oW)	4			x	20031216
Numerical weather prediction models:
NOGAPS	0 – 360	60oS – 60oN	Before 7 July 2004 1440x481 (60oS, 0o) Since 7 July 2004 1440x481 (59.875oS, 0.875oE)	4			x	20040201
NOGAPS*	0o – 360o	60oN – 60oS	1440x480 (59.875oS, 179.875oW)	2	x		x	20040201
GFS	0o – 360o	60oS – 60oN	1440x481 (60oS, 0o)	4			x	20031217
Surface reference data:
GAUGE	125oW – 65 oW	20oN – 50oN	241x121 (20oN, 125oW)	4			x	20031217
RADAR	125oW – 65 oW	20oN – 50oN	241x121 (20oN, 125oW)	4			x	20031217