Extended Abstract (572K)3.5
Reconstrction of the oceanic precipitation from 1948 to the present
Mingyue Chen, RS Information Systems, Inc., Camp Springs, MD; and P. Xie, J. E. Janowiak, P. A. Arkin, and T. M. Smith
This paper presents an analysis of monthly precipitation on a 2.5 deg. latitude/ longitude grid over the global oceanic areas from 1948 to the present. The analysis is derived from an EOF reconstruction method which was applied successfully by Smith et al. (1996) in reconstruction of SST fields. This method involves interpolating sparsely distributed historical gauge observations onto spatial distributions of EOF patterns of satellite estimates for later years.
A series of simulation tests and cross validation have been conducted to design the best strategy for the reconstruction and to get insight into how well the reconstructed fields may represent large scale variabilities in precipitation. The results showed that, 1) Using the first 8 EOF modes, the reconstruction over the entire global domain (60S-75N) is able to retrieve precipitation variations associated with ENSO and major large-scale circulation patterns with reasonable accuracy; 2) Performing regional reconstruction on the residual components of the global reconstruction yields improvements to some extent for the regions with reasonable gauge data. A set of procedures are then developed and applied to modify the historical gauge obaservations so that they are consistent with the satellite estimates used to define the EOF functions of the target precipitation fields.
Based on these results, a test product of the global monthly oceanic precipitation anomaly is produced for an extended period from 1948 to the present and applied to verify the numerical model produced precipitation fields for the period before 1980 when satellite estimates over ocean are not available. Details of the reconstruction and its potential applications will be reported at the meeting.
Session 3, Observed Seasonal/Interannual Variability: I
Monday, 10 February 2003, 1:30 PM-5:30 PM
Previous paper Next paper