Experimental GPCP Pentad (5-Day) Precipitation Analysis 


1.	Overview 

	An analysis of global pentad (5-day) precipitation has been constructed 
for the Global Precipitation Climatology Project (GPCP).  It is created on a 
2.5 degree latitude/longitude grid and covers a period from 1979 to near current
time.  This pentad analysis is a companion to the version 2.2 product of the 
GPCP global monthly precipitation analysis.  Users are recommended to utilize 
the GPCP monthly analysis for applications of monthly and longer time scales 
and use this pentad data set only for studies where sub-monthly and intra-
seasonal phenomena are involved.   


2.	Methodology 

	The GPCP pentad precipitation analysis is defined by merging several 
kinds of observation-based individual data sets, i.e., the GTS gauge 
observations and the estimates inferred from satellite observations of IR, 
OLR, MSU and SSM/I.  

	A three-step algorithm is adpoted.  First, to reduce the random error 
inherent in the individual data sources, the satellite estimates are combined  
linearly through the Maximum Likelihood Estimation method, in which the 
weighting coefficients are inversely proportional to the individual error 
variance.  

	To remove the bias, the output of the first step is then blended with 
the gauge data through the method of Reynolds (1988), in which the first-step-
output and the gauge data are used to define the 'shape' and the magnitude of 
the precipitation fields, respectively.  Finally, the blended analysis is 
adjusted by the version 2.2 products of GPCP monthly precipitation analysis 
so that the accumulation of the pentad analysis matches that of the monthly 
analysis. 


3.	Products   

        Pentad GPCP Version 2.2 is defined by adjusting the pentad CMAP against 
the MONTHLY GPCP Version 2.2.

	The current release of the GPCP pentad precipitation analysis is an 
experimental product and covers a period from pentad #1 of 1979 to 
pentad #11 of 2015. 

	The data set is composed of yearly files named as:
	
	gpcp_pen_v2.2.sgi.yyyy
	
where yyyy is a 4-digital integer indicating the corresponding year (e.g. 
1979 for the data file of the year 1979).

	Each yearly file consists of 73 fields of pentad precipitation over 
the globe. Each global field is an array of 144 x 72 4-byte real numbers in 
big_endian byte order.

	Starting from (1.25E; 88.75S), the global array goes from west to east
and then from south to north in 2.5 degree latitude/longitude increment.

	The pentad precipitation analysis values are output in units of mm/day 
with -999.0 indicating missing. 
	
	Below is a sample Fortran 77 program to read the pentad data set for 
1979:
	
      ###################################################################
      

c     program	:	sample.f 
c     objective	:	sample program to read the pentad file for 1979
	
      dimension	rain (144,72) 
c
      open  (unit=10, 
     #       file='gpcp_pen_v2.2.sgi.1979', 
     #       access='direct',recl=144*72) 
     
      do 2001 kk = 1, 73
        read  (10,rec=kk)  rain 
        write (6,2951)  kk,rain (72,36) 
 2001 continue
 2951 format  (i8,f8.2)  
c
      stop 
      end
   
      #################################################################### 
      

4.	Contact Point 

      If you have any problems or questions concerning this pentad
precipitation data set, please contact the data set creator at:

	Dr. Pingping Xie 
	Climate Prediction Center
        5830 University Research Court
        College Park, MD 20740
        Tel:  (301)683-3388
        Fax:  (301)763-8125
        E-mail:  Pingping.Xie@noaa.gov