Can your data come to my lab to play?

Scientific data are very diverse. A wide variety of data formats and metadata standards are used by different scientific communities. Users attempting to answer broader scientific questions such as factors inducing climate change require interdisciplinary data. Discovering the data they need, confirming the usability, and integrating them in their analysis is often time consuming and tedious work. This talk will cover some of the current issues in making data available to broader communities and offer some practical solutions being developed for Oak Ridge National Laboratory data center projects. This talk will include key concepts such as 1) metadata strategies to improve the distributed data discoverability by effectively using metadata standards such as ISO 19115, FGDC, EML and DIF 2) creating controlled keyword vocabularies using various keyword standards 3) enhancing keywords using various ontology repositories such as SWEET, Biocomplexity thesaurus and others 4) providing variable level mappings to help users find various representations of a single parameter and 5) effectively processing different data formats in a data analysis tool. This talk will also share some experiences from a recent effort in publishing observational data into a model data portal to facilitate model - observational intercomparison studies.