The explosion of information is hitting the sciences, but who's saving the data?

As science becomes more data oriented, where are libraries -- and librarians -- to help preserve, share, organize and manage this information? Well, nowhere, found a few researchers in a 2012 white paper [pdf] published by the Association of College and Research Libraries. 

Only a few academic libraries in North America offer some sort of research data services, the researchers found. Depending on the type of research going on in the University, this seems to be a missed opportunity for libraries. For one, someone needs to oversee this data for reasons of preservation and knowledge management. But higher education is often decentralized, allowing departments and colleges to create their own rules as they go. It's not out of the question that a lot of information could get lost on forgotten servers in back hallways. 

From the white paper, written by Carol Tenopir, Ben Birch and Suzie Allard: 

The convergence of data-intensive science, technological advances, and library information expertise provides academic libraries with the opportunity to create a new profile on campus as a partner in knowledge creation, helping it expand beyond the traditional roles of libraries. This new environment allows libraries to take a more active and visible role in the knowledge creation process by placing librarians at all stages in the research planning process and by providing expertise to develop data management plans, identify appropriate data description, and create preservation strategies. 

 The bottom line: National Science Foundation and National Institute of Health requires grantees create data management plans for their work. 

For good reason. A late 2011 study [pdf] by the University of Houston Libraries found that corners can be cut regarding data management, where researchers allowed students to be responsible for data organization or data was not properly backed up.  

Another plus: The University of Houston Librarians found that carrying out data outreach -- through the Grants Office -- opens doors to areas not previously served by the Library. 

Academic Examples 
Purdue University created a model for information gathering called the Data Curation Profiles, where Librarians can ask researchers how they'd like to curate this information, the research focus, the narrative of the data and how they'd like to share it. 

It covers the basic questions for a typical research set, and (hopefully) gets the researcher to open up on what he/she would like to have done with the information. 

MIT Library has a very helpful website on data management and publishing, Unlike Purdue, this is aimed at researchers, giving them a variety of prompts to help describe to others what type of data it is, where it is stored, how to organize it and how they'd like it shared.