Under the auspices of the Committee on Science and Public Policy (COSPUP) of the National Academy of Sciences, J.R. Cole and S. Cole have recently completed phase 2 of a study of peer review in the National Science Foundation (NSF). A summary of their results appeared in a Science (November 14, 1981) article by S. Cole, J.R. Cole, and G. Simon. We have no wish to take issue with the results and conclusions of this study, but we would like to describe to our AGU colleagues the NSF peer review process used by the Division of Earth Sciences (EAR) and how we view it.
EAR is one of the few NSF divisions that currently employs a review panel in addition to the ad hoc mail reviewers. All proposals are sent to four or more mail reviewers selected by the program director. These proposals are also sent to all panel members in a given discipline. In the past there were three program panels of five members each, serving geology, geochemistry-petrology, and geophysics. Under the new EAR organization (Eos, December 22, 1981) of eight programs, each of the sixteen panel members will serve on two program panels. These review panels meet in Washington with the program director three times per year to evaluate the proposals with the aid of the mail reviews.