19th International CODATA Conference
Category: Interoperability

Data Sharing: the experience of the speciesLink network in Brazil

Vanderlei Perez Canhos (vcanhos@cria.org.br), Sidnei de Souza, Renato de Giovanni, Mauro Munoz, Alexandre Marino, Fabrício Pavarini, Dora Ann Lange Canhos (dora@cria.org.br)
Centro de Referência em Informação Ambiental, Brazil
http://www.cria.org.br/


Natural history collections have been sharing data among peers for centuries basically working on documenting, preserving, identifying and classifying specimens. Specimen and species data has become fundamental for a number of applications such as conservation management, recuperation of degraded areas, sustainable use, prevention of diseases, etc. It has therefore become important to a number of users of different fields, demanding the use of different tools to help compare, integrate, analyze, synthesize, and disseminate data.

Helping biological collections to make their data available without disrupting existing structures and without imposing uniform platforms has been a challenge for decades. Biological collections in Brazil, like in most parts of the world, have different levels of infrastructure and expertise. In order to help promote data sharing, information systems must be designed in such a way that data providers have complete control over their data and have full recognition and acknowledgement of data users.

The speciesLink network is a distributed information system integrating primary specimen data from biological collections at real time. The project makes use of the most current advances in communication protocols such as DiGIR - Distributed Generic Information Retrieval, and SOAP - Simple Object Access Protocol and database management systems. The information system is being developed using free and open source software, mirroring techniques, and Internet 2 connectivity. The project is developing tools to help in the cleaning, utilization and visualization of data from 36 associated collections. The tools that are being developed include openModeller, a species distribution modeling framework, spOutlier, an on-line tool to detect outliers in latitude, longitude and altitude, and geoLoc, a geo-referencing tool.