The North Atlantic Storage Trust: Maximizing Space, Preserving Collections

There are well-over 50 library book storage facilities across the United States and Canada, holding millions of monographs and older serials that are rarely read but often stored under environmental and security conditions superior to those in our circulating collections. Even as librarians add to these stored collections, our libraries grow more crowded; and we look for a low-cost alternative to adding additional full-service storage space on our central campuses. These storage facilities vary from planned "Harvard Model" facilities, built specifically to house collections in the most cost-effective manner, to repurposed buildings that range from old supermarkets to telephone substations. Some contain materials from several libraries and are managed consortially. Of these consortial storage collections, some "de-dupe" their collections for added efficiency, but many do not. Until now, no effort has been given to rationalize our geographically distributed storage facilities as a national storage system, whereby economies could be achieved and a structured plan for preservation could be created at very little investment.

In 2001 at an ARL/OCLC buildings workshop in Las Vegas, Don Kelsey, a building specialist from the University of Minnesota, posited that storage facilities are rarely weeded. The cost of weeding is just too great; and, once a book is there, it is there to stay. This observation led to some hallway suggestions about using this disinclination to weed our storage collections to our advantage. If our storage collections are not weeded, then we could build a "trust relationship," whereby libraries with storage collections would promise to retain these in perpetuity and loan materials to others, thus allowing other libraries to make withdrawal decisions based on this agreement. Since 2001, this idea has evolved into the current effort to create the North American Storage Trust (NAST).

Creating NAST could prove to be easier and more cost effective than expected. If many of the libraries that have storage facilities would report their holdings to a central accessible database of stored items and if they would also post their policies regarding retention and loan, then other libraries could check this database and make their withdrawal decisions based on the trust that those items would be accessible when needed. Once libraries knew that copies of little-used items in their collections were already held in perpetuity by a NAST member, they could establish their own comfort level about the number and geographic distribution of existing copies in storage before they weeded their collections. No commonly agreed upon national standard for numbers of copies would need to be set to drive individual library decisions, so the administration of this system could be very light weight.

Certain assumptions underlie the North American Storage Trust: that many copies of little-used books already reside in our storage collections, that those libraries considering weeding gravitate to the same titles assuming low readership and low interest, that the cost of weeding our storage collections is greater than the cost of retention, and that our full-service library facilities will not be able to grow in size to hold our collections. Also, as our legacy print serial collections age and are replaced by electronic archives, significant savings can be realized across the national grid of libraries by creating a national paper archive. To test these assumptions, nine ARL members of the Association of Southeastern Research Libraries (ASERL) agreed to report their storage holdings to OCLC to be analyzed by the Office of Research, using the previous WorldCat system. Each library simply ran a report of their storage holdings using their local library systems (LMS) and FTPed it to OCLC.

The results of the study were surprising in that there was a lower level of duplication among the nine ASERL libraries than expected. Indeed, only 15 titles were held by all nine libraries. However, when the holdings of the nine libraries' storage collections were matched against Vanderbilt's circulating collection, it was estimated that Vanderbilt could theoretically weed over 88,000 items, using the comfort level of two copies held in these nine southeastern storage collections. If one extrapolates this number based on a potential 50...