Recent dramatic advances in computer networks and information technologies have created exciting new possibilities for managing scientific research data. Years of accumulated information, which until recently were accessible only to individual scientists in their laboratories, can now be shared in a very efficient way through distributed research networks. The amount of information contained in such assembled databases provides opportunities that were previously out of reach. Meanwhile, the arrival of new and user friendly spatial visualization tools (the so called "virtual globes", such as World Wind or Google Earth) brings revolutionary new ways to interact with geographic information. This project consists of designing a Web application taking advantage of these new technologies. The tool is designed to facilitate the sharing of scientific data within a research network, as well as offer both spatial and non-spatial methods to quickly and efficiently visualize them. Two major components form the system: a database built from the information submitted by the members, and a module featuring novel and sophisticated data visualization techniques. Users can explore the database and visualize their queries (a) in a spatio-temporal fashion with the "virtual globe", and/or (b) as graphics, using the dynamic features of the prefuse software package. Researchers can download the appropriate datasets for further analysis. By enabling scientists to share results among themselves, the system provides a platform for important meta-analyses and the analysis of broad-scale patterns. Here we present the most recent developments in the building of this platform for the Sustainable Forest Management Network, a pan-Canadian center of excellence. We focus on the description of the application as well as the challenges to scientific data sharing and on the new opportunities provided by a collective database.