This is the README file for version ("v9906") of the CAMS_OPI data.
The data are available through the ncep ftp site
(ftp.ncep.noaa.gov/pub/precip/data-req/cams_opi_v9906).

The following changes have been made to the original method:

	1)  A new version of the OPI (OLR Precipitation Index) which provides
	precipitation estimates where raingauge data are not available is
	used.  
		
	A problem had been discovered with the original OPI version  in the
	tropical East Pacific during strong ENSO episodes.  This problem
	affected primarily the  precipitation accumulations (not the
	anomalies).  We noticed that the depiction of the ITCZ in the far
	eastern Pacific is shifted northward relative to mean OLR during the
	December 1997 through February 1998 period although THE ANOMALIES
	ARE IN GOOD AGREEMENT between the CAMS_OPI and OLR in this region

	The problem with the accumulated totals in the original OPI version
	stems from the fact that the satellite algorithm ("OPI") computes
	anomalous precipitation (from anomalies of OLR) and adds the anomaly to
	climatology to form the monthly estimated precipitation accumulation. 
	It appears the problem is associated with the  large changes in
	precipitation distribution in the tropical far-eastern  Pacific that
	are associated with strong ENSO events (such as 1982/83, 1997/98).


	2)  More quality-control was used on the raingauge data.

	3)  Raingauge analysis procedure was modified to limit the influence of
	raingauge data to < 500 km. 	



The "CAMS_OPI" data are 2.5 degree lat/lon spatial means of monthly
mean precipitation derived from station gauges and satellite estimates.
Each data 'record' is a 144 x 72 array, oriented from SOUTH -> North
and Eastward from 1.25E.  


The array orientation is:


     0     2.5E   5.0E        2.5W     0
     |      |      |           |       |
     -----------------------------------  South Pole
     |      |      |           |       | 
     |(1,1) | (2,1)|   ....    |(144,1)|.............88.75S
     |      |      |           |       |
     -----------------------------------  87.5S
     |      |      |           |       |
     |(1,2) | (2,2)|   ....    |(144,2)|.............86.25S
     |      |      |           |       |
     -----------------------------------  85.0S
			.
			.
			.
     -----------------------------------  87.5N
     |      |      |           |        |
     |(1,72)|(2,72)|   ....    |(144,72)|............88.75N
     |      |      |           |        |
     -----------------------------------   Nort Pole
         .      .                  .
         .      .                  .
         .      .                  .
       1.25E  3.75E               1.25W
     
     Note that the the coordinates for each grid correspond to the center of
     each box since the data values are spatial averages.


A separate data file for each month exists.  The data arrays are in a FORTRAN
77 "direct", binary records ("IEEE") and are grouped as follows:

	Record 1: Precipitation analysis based on raingauge data only (CAMS)
	       2: number of CAMS gauges 
               3: Precipitation analysis based on OPI estimates only  
               4: The blended analysis CAMS_OPI monthly total precip. analysis
               5: For Internal Use only
               6: CAMS_OPI monthly anomaly (1979-1995 base period) 
               7:    "        "       "    expressed as % of gamma distribution
	       
The units are "mm/day" for the means and anomalies.


REFERENCE
			
Janowiak, J. E. and P. Xie, 1999:  CAMS_OPI: A Global Satellite-Rain Gauge
Merged Product for Real-Time Precipitation
Monitoring Applications.  J. Climate, vol. 12, 3335-3342.
			

			John Janowiak  (jjanowiak@ncep.noaa.gov)
			Climate Prediction Center/NCEP/NWS/NOAA