Interactive 3D visualization of highly connected ecological networks on the WWW

Abstract

Food webs describe who eats whom among species within a habitat. These networks depict interlinked food chains and have long been a central ecological paradigm for studying biocomplexity. As food-web data encompasses more species, the network structure of the feeding links among those species results in a highly connected network that is difficult to visualize in 2D. Such simple representations often obscure important details critical for understanding the structure and function of ecosystems. Graph layout is thus critically important for enhanced understanding of complex food webs. In this paper, we describe 3D visualizations of food webs that use intuitive node placement, minimal edge-crossing and link length, hierarchical node aggregation, and analysis tool integration. We also describe our deployment of "Webs on the Web" (WoW) visualization and analysis tools on the WWW. These tools will facilitate food-web research, collaboration, and education. Given the variety of 3D visualization technologies on the WWW, our WoW visualization pipeline supports diverse formats and adapts to users' preferences by employing XML and a flexible architecture. Currently, WoW uses XML to semantically markup food-web data (FoodWebML), extracts visual information into X3D format, and then uses the information to create a direct display on VRML/X3D browsers or feed a Shockwave 3D visualization.