Preparing, storing, and distributing multi-dimensional scientific data

Data of all sizes, generated by simulation and observation (i.e., instruments and satellites) activities, should be collected, stored, and organized, along with associated tools and research results, so that they are easily discoverable and accessible. Most observational data capture conditions at an exact point in time and are thus not reproducible, therefore it is imperative that initial data be captured and stored correctly the first time. In this paper, we will discuss how NASA's Oak Ridge National Laboratory Distributed Active Archive Center (ORNL DAAC) is preparing, storing, and distributing large volumes of multi-dimensional scientific data using Daily Surface Weather Data and a corresponding Climatological Summaries Dataset (Daymet) as an example.