Date of Award

Spring 2012

Project Type


Program or Major

Computer Science

Degree Name

Master of Science

First Advisor

Colin Ware


In weather visualizations, it is common to see vector data represented by glyphs placed on grids. The glyphs either do not encode magnitude in readable steps, or have designs that interfere with the data. The grids form strong but irrelevant patterns. Directional, quantitative glyphs bent along streamlines are more effective for visualizing flow patterns.

With the goal of improving the perception of flow patterns in weather forecasts, we designed and evaluated two variations on a glyph commonly used to encode wind speed and direction in weather visualizations. We tested the ability of subjects to determine wind direction and speed: the results show the new designs are superior to the traditional. In a second study we designed and evaluated new methods for representing modeled wave data using similar streamline-based designs. We asked subjects to rate the marine weather visualizations: the results revealed a preference for some of the new designs.