DATA ACCEPTANCE PROCEDURES AND LEVELS OF SERVICE AT THE NATIONAL SNOW AND ICE DATA CENTER R. E. Duerr, R. L. Weaver National Snow and Ice Data Center Cooperative Institute for Research in Environmental Science University of Colorado at Boulder 1. INTRODUCTION All data are not created equal. Individual data sets vary from each other in a multitude of ways - from easily measured ways such as size, format, and complexity; to ways that require a more nuanced understanding, such as in the breadth and depth of a data set's potential user base, its "designated community" to use the terminology of the Reference Model for an Open Archival Information System [1]. Given limitations in the resources available, it is not surprising then that repositories, such as those of the National Snow and Ice Data Center, need to make choices about the level of support or services provided for each data set acquired. Such choices prioritize center activities. While such decisions have been an implicit part of NSIDC activities for many years, an effort was recently taken to explicitly define the Levels of Service supported at NSIDC. Baseline Levels of Service have been defined for all existing NSIDC data sets. In addition, Levels of Service considerations are a major component of the NSIDC Distributed Active Archive Center's (DAAC's) new data accessioning processes. In this paper we describe the Levels of Service currently supported at NSIDC and factors that affect the effort required to obtain a given level of service. We also discuss the process users should use if they wish to request that the NSIDC DAAC archive their data. 2. BACKGROUND The basis for most research papers published in the Earth sciences today is data - be it data acquired by an investigator as part of their research perhaps as a result of a field campaign, data obtained from one of the many remote sensing systems or observing networks operated by a governmental agency or other organization, or data obtained through access to one of the many models currently in use. However, the days where it was possible to include all the data needed in order to replicate the results of that research in the peer-reviewed literature are long gone, if indeed they ever existed. Yet, results replication is a core tenet of the scientific process - at the heart of the reason for public trust in and consequently funding for scientific research. As a result of increasing recognition of this disconnect, the global scientific community is beginning to call for submission and archival of