Interactive Visual Analytics of Carbon Cycle Science

Tracking the flow of carbon (C) through the Earth’s terrestrial biosphere remains a major challenge to understanding how ecosystems respond to environmental change . To build an understanding of this flow, researchers have recently developed models that include thousands of variables whose intricate inter-dependencies can make them difficult to interpret. To open this scientific black box, we partnered with C cycle scientists at the NASA Jet Propulsion Lab who developed the Carbon Data-Model (CARDAMOM) framework. This paper presents a design study of CLOVE, a C science visual analytics application that encodes model dimensions in an intuitive way by using visual metaphors that correspond to the natural world. We describe how CLOVE encapsulates C states and fluxes in complex models, and explains how C cycle scientists can use it to diagnose their model outputs.