Multivariate cyclical data visualization using radial visual rhythms: A case study in phenology analysis

Abstract

Phenology is a traditional science that investigates the periodic phenomena of plants and animals and their relations to environmental conditions. Typically plant phenological studies are based on observations made by phenology experts in the field over time and the correlation with climate data collected by weather sensors. Although within the visualization community several approaches have been proposed for visualizing data that vary over time, many of them have a specific purpose and cannot be applied to phenology studies. Besides that, phenology experts increasingly need tools for managing appropriately long-term time series with many variables of different data types, as well as to identify cyclical temporal patterns. In this work, we propose a novel approach to visualize phenological data by combining radial visual structures along with visual rhythms. Radial visual structures are used to provide contextual insights regarding cyclical phenomena, while the visual rhythm encoding is used to summarize long-term time series into compact representations. We developed, evaluate, and validate our proposal with phenology experts using plant phenology direct observational data both at individuals and species levels.