MarshVis: Visualizing Real-Time and Historical Ecological Data from a Wireless Sensor Network

QianSheng Li, Gershon Dublon, Brian Mayton, Joseph A. Paradiso


We have been developing a dense sensor network to document ecological processes resulting from a large-scale wetland restoration taking place at Tidmarsh Farms, a decommissioned 577-acre cranberry farm in southern Massachusetts. Every 30 seconds, thousands of data points are streamed to a server, capturing a rich picture of the environment in flux as the restoration proceeds. It has been imperative for us to find ways to represent this information graphically for a variety of users and audiences, ranging from research collaborators studying wetland ecosystems to the visiting public. To achieve this aim, we built MarshVis, a system that visualizes data from the sensor network, highlighting spatiotemporal and inter-sensor relationships while also exposing the system operation. We implemented a number of web-based apps and developed strategies for real-time and historical exploration, as well as dynamic mapping. Our work is motivated by the need for interactive graphical tools that shed light on the delicate, interdependent ecological processes that make a natural environment sustainable. How can we expand the boundaries of public perceptions of natural phenomena at every scale?

Related Content