A new approach to dendrochronological data management

The procedures undertaken during the construction of chronologies often result in large quantities of data, typically in many associated files. The management of these data files, especially maintaining the relationships between processed and raw data, becomes increasingly difficult as a collection grows. To maintain a high level of accountability it is necessary to ensure that any derived or processed data can be replicated by independent researchers. Due primarily to practical limitations, some authors publish only subsets of the processed and/or raw data that constitute a chronology, making the process of independent scientific scrutiny difficult and on occasions impossible. There are associated problems when master chronologies are utilised for the purpose of dating specimens. Maintaining the relationships between chronologies and specimens is extremely important for accountability purposes, especially when errors are detected or revisions required. In these circumstances it is highly desirable to locate all specimens that have been dated with the original chronology to ascertain the implications of changes. This paper describes a new relational database design that addresses these problems and which has been implemented in the Corina dendrochronology application and web service. The new Corina system enables the data and analyses for all chronology building and cross-dating processes to be stored and documented, enabling scrutiny at every stage.