Representing Flow Patterns by Using Streamlines with Glyphs David H.F. Pilar and Colin Ware Abstract—Most professional wind visualizations show wind speed and direction using a glyph called a wind barb in a grid pattern. Research into flow visualization has suggested that streamlines better represent flow patterns but these methods lack a key property—unlike the wind barb, they do not accurately convey the wind speed. With the goal of improving the perception of wind patterns, and at least equaling the quantitative quality of wind barbs, we designed two variations on the wind barb and designed a new quantitative glyph. All of our new designs space glyph elements along equally spaced streamlines. To evaluate these designs, we used a North American mesoscale forecast model. We tested the ability of subjects to determine direction and speed using two different densities each of three new designs as well as the classic wind barb. A second experiment evaluated how effectively each of the designs represented wind patterns. The results showed that the new design is superior to the classic, but they also showed that the classic barb can be redesigned and substantially improved. We suggest that flow patterns with integrated glyphs may have widespread application in flow visualization. Index Terms—Streamline placement, wind barb, glyph, flow visualization, weather maps, multivariate visualization Ç 1 INTRODUCTION M ODERN weather visualizations normally indicate wind speed and direction using a glyph called a wind barb [13]. The basic wind barb design has a shaft oriented in the wind direction and a set of bars and/or pennants to encode speed in 5 knot intervals (Fig. 1). Each half bar encodes 5 knots, and each full bar encodes 10 knots. The triangular pennant encodes 50 knots. The direction of the wind is from the tail (the part with the bars and pennants codes) to the tip. In weather displays (e.g., Fig. 2), wind barbs are either drawn on a regular grid or at the locations of wind measurement stations with the measurement location given by the position of the barb tip. The 5 knot code of the wind barb means that they can be read to an accuracy of þ=  2:5 knots. The speed code is easy to read and to learn, and wind barbs are a standard feature of meteorological maps. However, a previous study of modeled hurricane data by Martin et al. [5] showed that wind barbs lead to substantial errors in area wind speed estimation and also produce systematic biases in the perception of wind orientation. In the following discussion, we argue that wind barbs have several design features that interfere with the representation of wind direction and regional wind patterns. The problem may be especially acute in portrayals of computational weather models that generate dense continuous vector fields. In the present paper, we report on an effort to develop a method for representing wind data that combines the virtues of the wind barb in the representation of quantity with the virtues of flow visualization methods that better show overall wind patterns for representing model output. Our goal is the effective portrayal of the dense wind patterns that are generated by operational wind forecast models. Our solution, we believe, can have widespread application in flow visualization. We begin with an analysis of the strengths and weaknesses of wind barbs as well as alternative methods that have been developed to represent flow patterns. We organize our analysis around a breakdown of the components of a two-dimensional (2D) vector. Vectors are often defined in terms of two components: direction and magnitude. For the purposes of analyzing the effectiveness of flow visualization, it is useful to break the concept down even further, separating direction into two parts, orientation and direction sign [11]. The orientation is simply the angle as expressed by a line segment, and the direction sign differentiates the two ends of that line segment. A stream- line trace, for example, shows orientation at every point along its length, but no direction sign. Arrowheads are one method for encoding direction sign. 1.1 Orientation With a wind barb, the shaft orientation indicates wind orientation at the tip location. There are a number of perceptual problems with this. First, untrained observers may judge wind orientation to occur in the middle of the barb, or at some other point, perhaps even the tail where the visual weight is greatest. Second, the bars and pennants create their own orientation pattern at approximately 45 degree angles to the flow. Having the measurement location at the tip may be responsible for the systematic orientation errors recorded by Martin et al. [5]. Bars also introduce sharp changes in contour direction that break continuity and make it difficult to identify patterns such as wind fronts or cyclones. Third, wind patterns curve continuously but because of its long straight shaft, only a very small part of the wind barb contour is actually aligned with the wind direction, and this point is far from the visual center of the glyph. IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS, VOL. 19, NO. 8, AUGUST 2013 1331 . The authors are with the Center for Coastal and Ocean Mapping, University of New Hampshire, Durham, NH 03824. E-mail: {cware, dpilar}@ccom.unh.edu. Manuscript received 6 Jan. 2012; revised 26 July 2012; accepted 19 Dec. 2012; published online 23 Jan. 2013. Recommended for acceptance by R. Machiraju. For information on obtaining reprints of this article, please send e-mail to: tvcg@computer.org, and reference IEEECS Log Number TVCG-2012-01-0006. Digital Object Identifier no. 10.1109/TVCG.2013.10. 1077-2626/13/$31.00 ß 2013 IEEE Published by the IEEE Computer Society