Observations and models of the global atmosphere are more available and capable than ever before. However, our understanding of climate system variations and changes depends crucially on our ability to integrate the observations and models into complete descriptions, or analyses, of the atmosphere. In this presentation I will describe the evolution of atmospheric analysis techniques and the development and implementation of reanalysis as a vital resource for climate science. Atmospheric reanalyses in the U.S. and elsewhere have proven to be an exceptionally valuable tool for studying regional and global climate variability, and the continuation of atmospheric analyses, both regional and global, as Climate Data Assimilation Systems is essential to operational monitoring of such phenomena as El Niņo. I will present the current status of atmospheric reanalysis activities and describe a proposed U.S. National Program for Earth System Analysis.