35?: “WWII”!IWIHHIHWIIHHHHIIHHHHIHIWHIUI {“55“ a 1113 1:1 1111111111111111111111111111111111 11(111111111111111111 31293 018 1 This is to certify that the 1 thesis entitled ANIMATED MAPS OF TROPICAL STORMS presented by Catherine Nanci Van Vliet May has been accepted towards fulfillment of the requirements for M.A. degree in W Major pr essor Date May 1, 1996 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution ___ k- .____._ ._ ____ ,, __._ _.J LIBRARY Michigan State University PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE ms wlmmu ANIMATED MAPS OF TROPICAL STORMS By Catherine Nanci Van Vliet Mey A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTERS OF ARTS Department of Geography 1 995 ABSTRACT ANIMATED MAPS OF TROPICAL STORMS By Catherine Nanci Van Vliet Mey Geographic objects moving through space and time pose a cartographic problem that is difficult to solve without the aid of animation. This thesis uses several relatively inex- pensive software packages and cartographic techniques to animate the phenomena. Data on tropical storm movement in the North Atlantic Basin over a 50 year period are used for these illustrations. Results indicate that none of the software solutions tested here are ide- ally suited for these kinds of animations; however, showing movement over space through time is possible and the resulting maps and animations are intriguing. Copyright by Catherine Nanci Van Vliet Mey 1995 ACKNOWLEDGMENTS I would like to thank my committee: Richard Groop and Randy Schaetzl from the Department of Geography and Carrie Heeter from the Department of Telecommunica— tions. Richard Groop, as my advisor, for his time and effort, and for helping me develop a thesis out of all my crazy ideas. Randy Schaetzl for making the time to serve on my com- mittee and for all his help finding those little mistakes. Carrie Heeter, for taking the time out of her incredibly hectic schedule to help me with all my scripting nightmares and for giving me helpful feedback on my animation; but most especially for her enthusiastic sup- port when I really needed it. I would also like to thank Judy Olson for for all of her help and support in my devel- opment as a cartographer/geographer and also for her encouragement in my pursuit of multimedia cartography. A special thanks to Mike Lipsey for his technical support and his friendship. In addition I would like to thank the Department of Geography for its support of my thesis work, and all the people who were willing to take the time to be subjects for my interviews, for all their helpful comments. And finally, I would like to thank my husband Craig for his caring and support, through long hours and crabby moods. For his helpful suggestions and his encouragement. iv TABLE OF CONTENTS List of Tables ..................................................................................................................... vii List of Figures .................................................................................................................. viii 1. Chapter One: Hurricanes and Animation .............................................................. 1 Through Time and Space: Movement in Geography .................................................. 1 Tropical Storms .......................................................................................................... 2 Animation: A Method of Display ............................................................................... 4 Problem Statement ..................................................................................................... 6 11. Chapter Two: Animating Tropical Storms ............................................................. 7 Tropical Storm Data ................................................................................................... 7 Software Packages ..................................................................................................... 9 Transform ............................................................................................................. 9 HyperCard .......................................................................................................... 12 HyperCard Animation 1: Hurricane Debbie, 1969 ....................................... 13 HyperCard Animation 2: Tropical Storm Anna, 1969 .................................. 15 HyperCard Animation 3: 1992 Hurricane Season ........................................ 15 Director .............................................................................................................. 18 111. Chapter Three: Evaluating the Animations and Conclusions ............................. 20 Animation Technique ......................................................................................... 20 Data Handling ..................................................................................................... 21 Graphic Quality .................................................................................................. 22 The Learning Curve and Ease of Use ................................................................. 22 Summary ............................................................................................................. 23 Analysis of Interviews .............................................................................................. 24 HyperCard Evaluation ........................................................................................ 25 Transform Evaluation ......................................................................................... 26 Director Evaluation ............................................................................................ 27 Overall ................................................................................................................ 28 Technical and Design Considerations ....................................................................... 28 Conclusions .............................................................................................................. 30 Appendix A: Data Documentation ................................................................................. 32 Appendix B: HyperCard Instructions ........................................................................... 35 Appendix C: Transform Instructions ............................................................................ 40 Appendix D: Director Instructions ................................................................................ 42 Appendix E: Interview Materials ................................................................................... 48 Bibliography .................................................................................................................... 5 1 vi LIST OF TABLES Table l Saffir-Simpson scale of hurricane intensity ................................................................ 2 2 Summary of author’s evaluation ............................................................................... 23 3 Summary of subject evaluations ............................................................................... 28 vii Figure hWN \OOO\JO\Ut LIST OF FIGURES Three general paths of hurricane movement in the North Atlantic .......................... 3 Different types of static maps .................................................................................. 4 Base map of the North Atlantic Region frequented by tropical storms .................... 8 A one half degree grid superimposed over a map of Florida and the Eastern Gulf Coast ...................................................................................... 9 One frame of the Transform animation ................................................................ 10 Three frames from the Hurricane Debbie animation ...................... ' ...................... 1 4 Three frames from the Tropical Storm Anna animation ........................................ 16 Three frames from the 1992 Hurricane Season animation ................................... 17 A screen capture from the Director animation ...................................................... 19 viii Static views of the Earth are useful for descriptive purposes, but true understanding of movement and change is necessary for analysis. " Hardwick and Holtgrieve, 1990 ix CHAPTER ONE HURRICANES AND ANIMATION Through Time and Space: Movement in Geography Movement is an essential part of geography. Bunge (1966, p. 112) states that “in explaining how an object acquires its location it is difficult to avoid the notion of move- ment. Even such ‘static’ features as mountains and seacoasts are explained in terms of movements taking place over long periods.” Bunge also notes (p. 112) that “it is the geo- grapher Ullman who has struck upon the essential unity of spatial movement theories with his implicit question, ‘What makes objects move over the earth’s surface?’ It can be argued that Ullman’s question encompasses all geographic theory...” Geographers study the earth’s systems. These include the interactions within, among and between people, plants, animals, and the physical earth. “Elements of a system are recognizable entities with measurable attributes” (Abler, Marcus and Olson 1992, p. 234). Within a system connections “are the pathways by which energy, mass, people, goods, or information pass from one element to another. Taken together, the elements and their con— nections exist in a state that may change through time and that produces a mappable geo- graphic pattern” (Abler, Marcus and Olson 1992, pp. 2345). Through time, systems and patterns change. “Geographers begin with static descrip- tions, but their explanations inevitably incorporate dynamic processes characterized by change” (Abler, Marcus and Olson 1992, p. 234). Consequently, geographers are faced with the difficulty of representing spatial and temporal change within these systems. Movement, as described by Hardwick and Holtgrieve (1990, p. 310), is “transporta- tion or communication over space and time.” They describe four major forms of move- ment; the movement of people (migration), information (communication), ideas (diffu- sion), and goods (trade or transportation). These same concepts are discussed by Lowe and Moryadas (1975). However, these studies are limited to human—environment interac- 2 tion as the basis of movement; they do not address movement of or within physical sys- tems and thus ignore the concepts of energy and mass. Tropical Storms Tropical cyclones are an example of a physical phenomenon that incorporates move- ment of energy and mass through space and time. In the North Atlantic basin and the eastern Pacific Ocean, strong tropical cyclones are referred to as hurricanes. These same storms in Australia and India are called cyclones, in the western north Pacific, typhoons. Hurricanes form from tropical storms, usually originating between 5 and 20 degrees of latitude. They are violent storms with wind speeds in excess of 74 miles per hour. On rare occasion, wind speeds may even reach speeds in excess of 180 miles per hour. Hurricanes are rated on a intensity scale called the Saffir-Simpson scale of intensity. The scale uses ranges of central pressure, wind speed and storm surge to designate storms into five categories of hurricane, from category 1, a minimal hurricane, to category 5, a cata- strophic hurricane (Table 1). Table 1 Saffir-Simpson scale of hurricane intensity Category Central Pressure Winds Storm Surge Damage (mb) (mi/hr) (ft) 1 > 980 74-95 4-5 Minimal 2 965-979 96- l 10 6-8 Moderate 3 945-964 1 1 1-130 9-12 Extensive 4 920-944 131-155 13-18 Extreme 5 < 920 >155 >18 Catastrophic During the peak of the Atlantic Basin hurricane season (late August to early October), hurricanes often form as tropical depressions off the coast of Africa and travel westward toward the Caribbean islands. Hurricanes can also form in the Caribbean Sea and travel 3 north into the Gulf of Mexico. Storms that originate in the eastern Atlantic tend to either travel westward through the Caribbean Sea into the Gulf of Mexico or swing north as they approach the Caribbean Islands and move northward along the coast, finally swinging back towards Europe (Figure 1). Not all hurricanes follow these paths; hurricanes are well-known for their erratic path behavior. Figure 1 Three general paths of hurricane movement in the North Atlantic In addition to variations in storm path, there are also variations in the number of storms that form each year, the strength of these storms, and their temporal patterns. An average of between nine and ten tropical storms form each year in the North Atlantic basin and on average about six of these storms develop into full-fledged hurricanes. From the year 1944 until the year 1993, the maximum number of tropical storms, including hurri- canes, that formed in any one year was eighteen and the minimum was four. For this same time period, the maximum number of tropical storms that reached hurricane status was twelve and the minimum was two. Although the database that will be used for this project extended back to the year 1886, it is recommended that data before the year 1944 not be 4 relied on since it was not until this year that reconnaissance aircraft were first used to help track storms (Appendix A). Tropical Storms are now tracked by satellite. Measurements of wind speed and pressure are either measured directly, interpolated or estimated from satellite imagery. Because hurricanes have such spatial and temporal variation, it would be both inter- esting and useful to have a visualization tool that allowed for the portrayal of their many characteristics simultaneously to achieve a better understanding of them. As noted by MacEachren (Abler, Marcus and Olson 1992, p. 101) “Recent interest in scientific visual- ization has emphasized development of new and more powerful visualization tools such as computer animation and three dimensional modeling.” Computer animation as a visu- alization tool is here evaluated as a technique for better understanding phenomena that move through space and time. Animation: A Method of Display Becauseof limitations imposed by traditional cartographic techniques, the convention in cartography has been the production of static maps (Campbell and Egbert 1990). “Maps ...have long been used to depict change. With maps, there are two traditional strategies: single static maps that depict limited aspects of change in or across space and multiple sta- tic maps that can be visually compared” (Abler, Marcus and Olson 1992, p. 120). Single static maps depicting change can be divided into two groups (Figure 2). The first group Migration Map Change Map Series of Static Maps Figure 2 Different types of static maps 5 includes maps that represent movement between places, such as maps of migration or trade. The second group includes maps that display change over time in a fixed place, such as a map of population growth and decline (Abler, Marcus and Olson 1992). A third way that is often used to depict change (Figure 2), uses a series of static maps. This method is often used when showing phenomena through time since each frame can be a specific point in time. This method most closely resembles the animated map, however the map reader is left to interpolate what happens between each image. Thrower in his article “Animated Cartography in the United States” discusses the limitations of static maps for showing change: It is widely recognized that a single illustration, however, good in itself, has definite limitations for indicating change through a considerable period of time. For showing progressive development in the physical and cultural landscape it has been customary to use maps or diagrams in series. But, valuable as these maps in series may be for showing changing phenomena, they too have limitations. This is so because, while in reality changes proceed more or less continuously, the maps in a series show only instantaneous cross sections of an area, or snapshots. ...However, by the use of ani- mated cartography we can bridge the gap and give an impression of continuous change (Thrower 1961, pp. 20-1). It is this impression of continuous change, of movement, that makes animation such a powerful tool in the visualization of geographic phenomena. We need animation “to cope with the amount and complexity of information. Our curiosity inspires us to see what sto- ries these huge banks of data have to tell” (Dorling 1992, p. 225). In order to satisfy that curiosity we must be able to see, or visualize, those phenomena and cartographic animation allows us to do that. One appealing aspect of animation in cartography is “the ability to show change through time in a way that is more powerful than the creation of a series of static maps” (Moellering 1980, p. 67). While “it may be true that impressions of animated images sel- dom persist if they stand alone, ...with timely reinforcement an animated map is one of the best ways to communicate ideas about spatial relationships and their changes through time” (Gersmehl 1990, p. 3). Problem Statement This thesis will use three animation software packages that employ different visual— ization and cartographic techniques to produce animations of tropical cyclone data from the North Atlantic basin. The animation packages are Macromedia Director, Spyglass Transform, and Apple HyperCard, all operating on a Macintosh computer. Each package approaches animation in a unique way, and each has different strengths and weaknesses. However, all three have several attributes in common. They are all relatively inexpensive, capable of producing animation, not extraordinarily difficult to operate, commonly used, and they all run on a desktop computer. Macromedia Director is primarily an animation package, an authoring tool for multi- media productions. It is the most expensive, as well as the most complex, of the three soft- ware packages. Spyglass Transform, on the other hand, is used more often as a data visu- alization tool. It incorporates animation only as an small component of a larger visualiza- tion package. HyperCard, which is produced by Apple Computer, is the least expensive of the three packages. It is a tool that can be used to construct small independent programs or stacks that link information together. In the past, HyperCard has been a popular tool for creating simple animations and multimedia productions. Each of the packages was used to create animations of tropical storm and hurricane data. The animations were created with the intention of using the strengths and capabili- ties of each program to help in the visualization of tropical storm and hurricane data. This thesis will examine the advantages and disadvantages of these three software packages in illustrating, through animation, a geographical phenomena that moves through space and time. Discussion of the results will include both a subjective evaluation by the author, as well as objective comments of independent observers. CHAPTER TWO AN IMATING TROPICAL STORMS Tropical Storm Data Hurricane data were acquired via the intemet from Colorado State University. However, as stated in the documentation that accompanies the file, the data are “main- tained by the forecasters and researchers at the National Hurricane Center in [Coral Beach], Florida.” and “permission to make free access of this data and documentation [is] provided by Dr. Robert Sheets - director of NHC and by Mr. Neal Lott of the National Climatic Data Center in Ashville, NC.” A copy of this documentation, which includes descriptions of the data format is included in Appendix A. The file itself includes a large amount of information. For every known tropical storm in the North Atlantic Basin, from 1886 to 1993, the following information are recorded: - Year, month, and day the storm was first recorded 0 Duration of the storm in days 0 Storm number for that year (including subtropical storms) 0 Storm number since the beginning of record (1886) 0 Name of the storm (if available) 0 Maximum intensity of the storm (hurricane, tropical, subtropical) 0 Whether or not the storm made US landfall as a tropical storm or hurricane - Highest category (Saffir-Simpson scale) of the storm, at landfall, if the storm made landfall - State or region of US landfall and the Saffir-Simpson category at that landfall The following information is included at six hour intervals (002, 062, 122, 182): 0 Month and day 0 Storm stage - Latitude and longitude of the storm 0 Maximum sustained surface wind speed in knots - Central surface pressure in mb (if available) A short program was written in Basic to sort through and organize the data into a for- mat which could be read into Microsoft Excel spreadsheet and into a Microsoft FoxPro relational database where more processing took place. 7 8 There were three types of data lines in the original file (see Appendix A for a full description). Line types “A” and “C” contained summary information about each storm and line type “B” contained the detailed locational and strength information at six hour intervals for each day of the storm. The relational database made it possible to connect these two types of information with a common variable and to select records from either file. For example, one could ask for the name and year of any storm whose maximum winds were greater than 135 knots, a class five hurricane. The program would then return a list of storms and their years which met the criteria. The base map that was used in each of the animations, for location reference, was a cylindrical equidistant projection (Figure 3). The reason this map projection was chosen is that the latitude and longitude lines form perfect squares which makes it easy to plot data using an x,y coordinate system. Although there are some distortions in higher lati- tudes (east/west distances are more exaggerated nearer the pole), for the purposes of this ”a é in rm project, the base map provided a useful reference for the data. i it i” VL)’ \ lhva is.» *7 i? Figure 3 Base map of the North Atlantic Region frequented by tropical storms. Software Packages Transform Transform is a program created by Spyglass, Inc. that works as a tool for “analyzing matrix and image data” (Spyglass 1993). Data can be entered either directly in matrix form or in column form which Transform then modifies in order create a matrix. With this matrix of data the user can then quickly and easily generate images. Transform works by first creating a matrix of numbers and then assigning a color to each cell based on the data value. “Data for Spyglass Transform usually consists of two independent variables and one dependent variable arranged as a 2-D array of numbers.” “Color raster imaging, along with a spreadsheet-like display of the numbers, surface plots, vector plots, contour plots, and interactive line graphing, presents a powerful list of tools...” available to the user (Spyglass, 1993, p. 3). In the case of the images created for this project, columns of numbers (latitude, lon— gitude and wind speed) were imported and modified to create a matrix of longitude by lat- itude in 0.5 degree increments with wind speed, in knots, as the value within each cell. Figure 4 shows a 0.5 by 0.5 degree grid superimposed over Florida and the Gulf Coast to Louisiana. Figure 4 A one half degree grid superimposed over a map Florida and the Eastern Gulf Coast 10 Next the data set was smoothed using the “smooth” command which creates a new data set. Smoothing is accomplished by averaging every point with its eight surrounding neighbors for a user specified number of iterations (Spyglass 1993). Once the new smoothed data set has been created, an image can be generated using either the “Generate Image” command or the “Interpolate Image” command. When the image is displayed with one pixel representing each value in the matrix, the generated image looks smooth and one cannot see individual cells. However, if the “zoom” is set to greater then one, each value is represented by a group or block of pixels. This “chunky” appearance can be avoided by interpolating the image, which smoothes the transition between blocks of pixels representing different data values. Once an image has been generated, there are a variety of tools within Transform to help fashion the image. The size of the image can be set by either selecting the size of the image in pixels or by selecting the “zoom” factor. When the “zoom” is equal to one, then one pixel represents each data cell in the matrix. When the zoom for X is equal to the zoom for Y then the blocks of pixels representing each data block form squares. An “Axes Jan Feb I “,, Oct Net—Dec Figure 5 One frame of the Transform animation ll tool” allows the creation and display of text along the axes and a title for the image. There is a long list of different color tables that allows changing the colors assigned to each value. It is also possible to import a custom color table. There is also a “fiddle” tool that allows setting of colors to white, or to compress and shift the color range and rotate the colors. Once the image has been created, an overlay can be added to enhance, annotate, reference or add new pieces of information to the existing image. In the case of the images for this project a base map was overlaid for reference purposes. Once a series of images have been created, Spyglass View can be used to create ani- mations from these images. Each image becomes a frame within the movie and by dis- playing these frames in rapid succession the illusion of movement is created. A sequence of procedures was followed to create a series of images that could be “played” to create an animation (Appendix C). In each of the images created, latitude and longitude were used as the X and Y coordinates and wind speed as the Z value. The data were aggregated by compressing a fifty year time span into one year. For example, for each image in the first animation I selected all the storm positions, and their correspond- ing wind speeds, that fell within the month of September for every year from 1944 to 1993. Where a given pair of x,y coordinates were associated with more than one wind speed value (i.e. multiple storms passed over the same coordinates), the wind speeds were added together. The values in the resulting matrix represent an index of storm severity and frequency at different locations. To test this procedure, data were aggregated by month over the fifty year period for the first animation. A new image was created for each of the twelve months of the year using only those storms which reached hurricane strength. This same process was followed again in creating the second animation, but this time using weekly data and all storm types. These initial animations were not acceptable because of “choppiness” in the display and because of color allocation problems. Adjustments in data sequence and 12 minimum and maximum color assignments and interpolation were performed and a more acceptable set of images were created for use in the animation. In this re-creation of the animation I employed a moving average to smooth the fre- quency data. The moving average used a four week time span in the following way: weeks 23, 24, 25, & 26 were averaged for frame F 1; weeks 24, 25, 26, & 27 were averaged for frame F2; weeks 25, 26, 27, & 28 were averaged for frame F3 and so forth. The result was a 52 frame year each containing about one month worth of data. This end product was pro- duced as “Animation lzTransform.” HyperCard HyperCard is a card-based system that works rather like a stack of index cards. Stacks, cards, buttons, and fields can be created to call other stacks, cards, buttons and fields, or even other programs. For this reason, HyperCard is very good at setting up simple hyper- links, but for animation it is somewhat crude. Although HyperCard does have a “foreground/background” set-up feature, for pur- poses of animation, the program works similar to a flip-book -- movement is shown by “flipping” through a series of cards in rapid succession. This works well if the images are already constructed and can be imported in from another program. However, because HyperCard is monochromatic, the images are black and white, regardless of the colors in the original images. The program is not capable of moving objects within HyperCard based on external data; objects can only be moved on the screen by “flipping” already- created images. The foreground / background option allows one to place objects that will be the same from frame to frame in the background so that it does not have to be placed on each card. This also saves program speed and memory space. See Appendix B for detailed instructions for creating an animated HyperCard stack. 13 HyperCard Animation 1: Hurricane Debbie, 1969 After searching the database for years when there were many strong storms, two years, 1969 and 1992, were extracted. From these years seven storms, five from 1969 (from a total of 18), and two from 1992 (from a total of 7), were used. I extracted the latitude, lon- gitude and wind speed for these storms and transferred them into “Transform.” Transform was used because it has a feature that allows equal spacing on the X and Y axes and because the base map was formatted for this program. However, any program with plot- ting capabilities could have been used. After the data had been converted, I plotted the map and adjusted the colors so that values appeared as gray tones on a white background. Hunicane Debbie was selected for the first animation because it was a fairly simple and common storm track (coming off the coast of Africa, westward through the tropics, swinging north along the west side of the Atlantic, then east as it reached the northern lat- itudes). The image of the storm track was processed in Freehand by placing the image in a background layer and drawing a line that represented the storm track with a node at each of the storm positions that had been plotted in Transform. I also drew a line around the border of the map, so that I could more easily and accurately position the storm track over the map in HyperCard. In HyperCard, I set up a stack with a few buttons, a title, and a map in the background. I made thirteen copies of the card and starting with the last one, I pasted on the storm track of Hurricane Debbie. I then pasted this information to the appropriate card in the stack, working backwards with the first card showing no track at all. I then scripted the cards in order with a one-second delay between each card. The resulting animation gives the impression that the storm track grew with each passing day. Although this animation was crude and took a lengthy time to construct, it did show the storm track as it moved across the Atlantic (Figure 6). See Appendix B for more detailed instructions on the creation of these HyperCard animations. I969: Debbie I Play | I969! Debbie flugwt I9 I 9693 Debbie 9.9.1! I! I Start Over I @3 Figure 6 Three frames from Hurricane Debbie animation 15 HyperCard Animation 2: Tropical Storm Anna, 1969 This HyperCard animation was created much like the one for Hurricane Debbie. The data points were plotted in Transform and then imported into Aldus Freehand. However, instead of drawing a line that connected the points, a hurricane symbol was created and positioned at each of the plotted locations. The symbol was then “colored” in patterns of gray to represent the wind speed at each location. Tropical Storm Anna last for a duration 12 days, with location information every six hours, providing 46 different points that needed to be positioned, “colored,” copied and pasted into HyperCard, each on a separate card. As with the first animation, a “play” button was scripted to display the data in order with a one second delay between cards. This gave the impression of the storm moving across the ocean, changing from light to dark as the storm intensity changed (Figure 7). HyperCard Animation 3: 1992 Hurricane Season The last of the HyperCard animations shows an entire hurricane season as it pro- gresses from the middle of August through the end of October, with six storms. Again the images were created much like the first two: data points were plotted and imported into Freehand, lines were drawn connecting the points of the storms together into one day seg- ments, and segments were “colored” in gray patterns to represent the tail of the storm, which got lighter with each passing day. The current day segment was black. These line segments were cut and pasted into HyperCard so that the storms happened at the correct temporal position. The first storm completed its path in the beginning of the season, followed by a period of no storm activity, followed by an explosion of storm activity as four storms moved about in different parts of the ocean, then another break in the storm activity, and finally, one last storm of the season. Although this took a large amount of copying and pasting to complete the animation, the result showed how the storms moved in relation to each other throughout the storm season (Figure 8). ”69: none m. as hour o ) When-nod I it, mm murmur — wr I969: ‘1an Mg :1 hour l8 -' ,_., W'ndspood I m (IL, (MPH) 3on 403 60*: box I969: none 1ng :9 hour 18 v' w Wildspeed I {'1 (MPH) 30's 10's 50-: 60's ,5. Figure 7 Three frames from Tropical Storm Anna animation I999: Storm Season flugul :I <5 I992: Storm Season W as I992: Storm Swen W n Figure 8 Three frames from 1992 Storm Season animation 18 Director Macromedia Director is an authoring tool for multimedia productions (MacroMedia 1994). Director uses the “Stage and Play” metaphor as described by Gersmehl (1990, p. 6). The first step is to design the “stage.” I enlarged the size of the map so that it filled the stage area and placed it on the top portion of the screen. The legend was later placed on the bottom portion of the stage. When creating animations in a “scripting” environment like the one in Director, it is easiest to progress slowly through the process, scripting one element at a time, adjusting it until it works, and then moving on to the next step. This allows debugging the animation as it is created. If all the scripts were written at once and not checked, the program may not work and finding the problem in the animation could be difficult. The first step was to import data from outside text files into Director and use these data to move a circle around on the stage. Once this was accomplished I could then create sep- arate files for longitude and latitude, with numbers that could be read by Director, and direct the symbol where to move on top of the map. However, since Director has its own coordinate system, I needed to transform the numbers to correspond to the screen coordi- nate system. Once I had the symbol moving about the stage as I wanted, I needed to add a third dimension, the wind speed. This Z value was assigned a color and size; this allowed the symbol to move across the stage changing size and color proportional to wind speed and strength. I also wanted the user to be able to see, in number form, what the latitude, longitude and wind speed were and also to see the category, date, and time. These variables were written to text boxes on the screen and integrated with the moving symbols. C File Storm Speed Trall '~\..‘ f\,,;r't‘;..,.u‘. J ‘- .3-_ ' r}:‘j- .‘v' E LONGITUDE: ~46 Degrees ,1 lATITUDE: 58 Degrees i WINDSPEED: 46 MPH STRENGTH: Tropical Storm “.0.“me .r-- LA,59I..)~..~ r Lay» . -.-r. ~ Figure 9 A screen capture from the Director animation As additional storms were added, custom menus were created that gave the user a choice of storms, a choice of animation speed, and a choice of storm trail or no storm trail behind it (Figure 9). See Appendix D for detailed instructions on data preparation and operation of Director. CHAPTER THREE EVALUATING THE ANIMATIONS AND CONCLUSIONS The intent of this project was to compare three different software packages based on their ability to animate tropical storms, geographic phenomena that move through space and time. In the first part of this chapter, I discuss the analysis of these animations from the point of view of cost, ease of use, learning curve, animation technique, ability to han- dle data input, and their graphic quality and animation quality. This section will be followed by a summary of responses given by other observers of the animations during interviews. Animation Technique HyperCard and Transform are both card-based animations, what Gersmehl (1990) refers to as “flipbook” style animation. Gersmehl (p. 6) effectively sums up my experi- ence of using this type of animation, stating: “...the flipbook technique is labor-intensive; each scene must be drawn (or at least assembled) in its entirety, and a convincing illusion of smooth motion usually requires about a dozen scenes per second. Image cloning and electronic modification can lessen this burden considerably, but with today’s equipment the list of options is large, and the flipbook metaphor is seldom the easiest way to add motion to a map.” Macromedia Director on the other hand uses what Gersmehl terms a “stage and play” form of animation. In this type of animation the ‘stage’ usually has a backdrop that remains stationary throughout the animated sequence and the actors that move about, sometimes changing form and color, in front of this backdrop. In Director, the stage is the screen, and cast members are placed on the screen either in the score or as directed by a series of scripts. Images of cast members that appear on the screen are referred to as sprites. Sprites in Director, not to be confused with Gersmehl’s sprites, can move about the screen, changing color, size, and any number of other attributes. This type of 20 21 animation allows for much more complex animations that can be easily adjusted by changing either the cast member or script for different effects. Thus from the point of view of animation ability, the technique utilized by Director is far more advanced than that utilized by the other two programs. Data Handling Although Spyglass Transform technically creates a card-based form of animation, the images for these animations are easily designed by manipulating a matrix of numbers and applying colors based on the numbers in each cell. Although each frame of the animation must be designed separately, once a desired format is established, creating each frame goes quickly. However, as with all card-based animations, making changes, regardless of how small, is a time consuming process; each image must be opened, altered and re-saved. The Transform animation designed for this project contained fifty-two separate images. Although HyperCard can keep track of a large amount of data, it does not seem to allow for importing data from an external source; nor does it allow objects to be moved around based on data within the program. These limitations force the animation designer to create separate still images and manually move them around on each card to create the illusion of movement. Director does allow for the import and export of data and allows objects to be moved and changed based on this data. For my Director animation, this meant that once one storm had been scripted, new storms could be added just by creating new text files for the program to read. And since the size, color and position of objects are designated within the script, changes can easily be made by adjusting a few line of the script. Thus, from a data handling point of view, Transform’s abilities are more advanced then the abilities of the other two. However Director, with the use of its scripting language Lingo, does employ some data handling abilities. 22 Graphic Quality Of the three programs that I used, HyperCard’s graphics were the most unrefined. Since Apple has not yet integrated color into this program, all the stacks created with HyperCard are in black and white. Even shades of gray are simulated with a coarse pat- tern of black dots. The other two programs both fully integrate color, which allows for more dramatic visual effects. The animation in Transform is definitely less complicated than that of Director, however the ability to easily generate new images and quickly change between more than a dozen built-in color tables, makes Transform more flexible in a graphic sense. The Learning Curve and Ease of Use Transform was the easiest of the three software packages to learn. The manuals are well-written, and the program is quite intuitive. Although the programs uses matrix data, column data are easily transformed and configured with user-defined specifications. The user defines these specifications by answering a series of questions with answers in the form of check boxes, pull-down menus and “fill-in-the-blanks.” In a matter of less then fifteen minutes one can create a striking first image. HyperCard was more difficult to learn. With a brief pass through the tutorials, one could make simple hyperlinked stacks, using predefined clip art, buttons and a little imag- ination. The manuals were fairly informative and easy to use, and HyperCard’s scripting language, HyperTalk, was easy enough to learn if one already knew the basics of how pro- gramming languages work. There is also a separate ‘Script Language Guide’ that deals specifically with the more complex concepts of HyperTalk. Since the scripting I needed to use to play my animations was very basic, it was reasonably easy to learn the necessary HyperTalk commands. The most difficult program to master was Macromedia Director. With its large num- ber of options and capabilities, Director is the most complex of the three programs. 23 However, once one grasps the overall design of the process, one can create some interest- ing animations. Although the manuals were fairly well laid out and basics about the pro- gram were clearly described, I found that descriptions of many of the Lingo (Directors scripting language) commands were confusing and that, occasionally, even the sample scripts did not work. Summary In summary, each of the programs has its own strengths and weaknesses that are a combination of the topics discussed above. In Table 2 I have summarized my evaluations for each of the programs. Table 2 Summary of author’s evaluation Director 'Ii'ansform HyperCard Graphic Quality very good excellent poor Ease of Use difficult easy moderate Data Handling fair excellent poor Animation Technique excellent crude crude Cost * $800 $399 $99 Overall Evaluation recommend recommend not recommend * Approximate Retail Price: November, 1995 Director has its strength in its multitude of options and its powerful features which allow the user to design many different types of animations that are visually appealing. However these strengths come at a price, both monetary and in learning time. Transform takes very little time to learn. It is able to handle large data sets and to quickly transform them into an image with stunning colors. Its one major limitation is that the animation is card-based; to change a small part of the display, one has to change it on each individual frame, a time-consuming process for the perfectionist. 24 HyperCard’s main attraction is its lower price. Although it is fairly easy to create hyperlinks between different cards or stacks, HyperCard’s limitation of black-and-white and its inability to allow objects to be moved around based on a coordinate system, make it difficult to create animations that effectively illustrate the desired concepts. Analysis of Interviews After all the animations had been finished, I conducted a number of informal inter- views with subjects who viewed each of the animations to get a more objective view of their quality and effectiveness. I interviewed nine people: three fellow cartographers, three climatologists, and three geography students who had not done extensive work in either cartography or climatology. Each of the three groups of three contained at least one male and female; there were a total of five males and four females. Each subject was told that they would be viewing three sets of animated maps, the sub- ject of which was tropical cyclones in the North Atlantic Basin; and that after viewing each set, they would be asked a series of questions. The animations were shown in a dif- ferent order to each of the subjects within each group. Care was taken to make sure that the order of the animations was not repeated within an individual group. Before each animation was shown, the subjects were read a short description of that animation. They were then allowed to view the animation and ask questions if they wished. After that, the following series of questions were asked: 0 Do you think this animation would be useful in teaching? scientific visualization? 0 If the animation were to be used in teaching, what age group do you think it should be used with? 0 What aspects of this animation did you like the best? 0 What aspects do you think could be improved? After all the animations were viewed, the subjects were asked to comment on which of the animations they preferred and why. 25 Responses between the groups of climatologists, cartographers, and students differed somewhat and these variations are noted in the following discussion. HyperCard Evaluation HyperCard seemed to be the least-liked of the three sets of animations. Both the stu- dents and the cartographers thought that these animations would be useful in teaching. The climatologists were more hesitant about this point. They thought that the animation of the 1992 storm season possibly could be used for educational purposes, but were unsure about the others. Most of the subjects seem to like the 1992 animation best, of the three HyperCard animations. When asked whether the animations would be useful for scientific visualization, both the students and the cartographers seemed uncertain how to respond, however the clima- tologists agreed on “no.” It is likely that this was a difficult question for the subjects to answer, because the term itself was not well understood. When considering what age group the HyperCard animations might be most appro- priate for, the responses were divided. Several thought they would be good for younger children because of the simplicity of the animations. Others thought that they would not be good for younger children because they did not have color, and children would lose interest. All of the subjects agreed that they could be used for students from junior high school through introductory college. When asked what they liked best about this set of animations the majority of the sub- jects said they like the storm season animation because of the way it showed the storms in real time. Several subjects said they liked the Anna animation. Most liked it because it showed the wind speed, while one preferred it because of the hurricane symbol. A unanimous conclusion was that HyperCard would benefit from the use of color. Other suggestions included using different wind speed designations, a clearer legend, and making the symbols in Anna better. Additional suggestions involved adding wind speed to 26 the storm season animation, adding some kind of symbol that would designate one day increments to all three animations, and leaving faint one color trails behind the symbols in both the Anna animation and the storm season animation. It was noted that in several of the cases, the trails in the storm season animation were confused with wind speed; how- ever this was not the case for all the subjects. A few thought that the fading trails was a “nice” effect. Transform Evaluation When asked if the Transform animation would make a good tool for either teaching or scientific visualization, the answer was a resounding “yes.” All subjects thought it would be best used in junior high or high school at the earliest because of the complexity of the subject and its symbolization. They also thought that it would need a great deal of explanation. When asked what they liked best about this animation, the color scheme was men- tioned most often. Although one person thought that the colors seemed dim, most seemed to think that the blue-purple-red-yellow color scheme was quite effective. Other praise included pattern recognition and detail, and animation smoothness as a result of the moving average approach. Several of the subjects did have some suggestions for improvements. One subject thought that the animation should be done in a year by year fashion instead of aggregat- ing all fifty years. Others thought that making it larger and adding country boundaries, to help designate land from water would help. And one subject was concerned about having the frequency and wind speed combined onto one map. It was felt that it would be better to have two separate animations, one that showed frequency and one that showed intensity. 27 Director Evaluation The subjects seemed to have more opinions about the Director animation than either of the other two sets of animations. Everyone thought that this animation could definitely be used in teaching and visualization tasks. Most of the subjects seemed absorbed in watching it, some playing the storm animation over and over. Most thought it would be good for scientific illustration but probably not for research because the storm patterns shown were “expected;” that is, little new information about hurricanes was added by the animations for research purposes. It was noted, by more than one subject, that the animation gave one a feel for where hurricanes originate, when and where they gain their strength, and the effect that landfall has on them. Again, subjects were mixed on what they thought was a good age group for this animation. Several said any age, others thought perhaps middle school and up. Everyone agreed that they could definitely be used for college age students. A majority agreed that color, visual impact, and overall design were some of the best aspects of this animation. Other positive comments included: the use of color to designate hurricane strength, the trails feature, the size of the animation, the amountof data pre- sented, and the use of circle size to represent wind speeds. Not all subjects agreed on the graduated circle issue. Some thought that the circle size was misleading, that it gave one the impression that the storm itself was changing size. Other improvement issues that were mentioned included: adding more storms, a smoother progression of symbol size, smaller symbols, adding names to the storm trails, and changing the colors in the graduated circle legend to match the colors in the strength legend. Other suggestions included overlaying satellite imagery, changing the symbols to match the storm’s actual size, adding landmarks such as cities, and having the trails become transparent or turn gray so that it is easier to see the current storm. 28 Overall Table 3 shows a summary of the evaluations of the subjects that were interviewed. Table 3 Summary of subject evaluations HyperCard Transform Director Visual Appeal poor excellent excellent Usefulness in Teaching good good excellent Age Group divided High school+ Middle school+ Usefulness in poor excellent good Scientific Visualization Usefulness in fair excellent excellent Scientific Illustration Overall Preference not as well liked well liked well liked Overall, the subjects preferred either the Director or the Transform animation over the HyperCard set of animations. One exception was one of the climatology subjects, who preferred the HyperCard animation that showed the entire 1992 storm season. Director was chosen mainly for its colors, its symbology, and its overall design. Transform was pre- ferred for its color scheme and its ability to show when and where tropical storms occur. Technical and Design Considerations There are a variety of things that one must consider when making animated maps as opposed to printed static maps. These considerations range from technical decisions con- cerning software and hardware to design decisions which affect the “look” of the anima- tions. One of the first things to be considered is what computer platform to use. This choice should be made based on what is available, and the audience that will use the animations. If a Macintosh computer is chosen then screen size is an important choice. Most often a 29 13 inch screen size with a pixel resolution of 640 by 480 is chosen. This is a fairly stan- dard screen size but it does exclude people with smaller monitors. Because the screen resolution on most Macintosh monitors is 72 dpi, information that can be shown on screen is much less than can be shown in a printed map that has a reso- lution of 600 dpi (common laser printer output). It is best to keep detail to a minimum; if a symbol is not necessary, then remove it. This is also important since the viewer should concentrate on what is happening in the animation. If there is extra detail on the map, this becomes confusing and the viewer may be distracted from the primary message. Color is also an important consideration. It allows more variety and is more interest- ing for the viewer than monochrome displays. Pale pastel colors work better for back- ground, while bright bold colors work better for symbols. This scheme makes the symbols stand out and not “disappear” into the background. If there are too many colors (or too many bright colors) or if it is not clear what is symbol and what is background, then the viewer tends to be confused. Symbols that move across the screen may be situated on different background colors; symbol colors should be chosen so that they will be readily visible on any background. Another factor is the speed of the animation. Selecting an appropriate speed can be a problem because different viewers sometimes prefer different speeds. Occasionally a vari- ety of effects can be seen by speeding up or slowing down the animation. My experience indicates that it is best if some changeability or user interactivity can be built into the animation, so that the user can adjust the speed for their comfort or to see different pat- terns. I also found that many of the users became more interested in the animation if they could interact with it. For example, being able to change things such as the speed, the objects displayed, the color scheme, trails behind moving symbols, stopping, starting, for- ward or backward play, all can increase user interest and learning. 30 Legends are also important. Just like static maps, animated maps need legends explaining symbols; however, unlike static maps, legends showing the passage of time are usually a crucial but problematic part of the animation. There have been many versions of a time “clock” used in different animations. These include the timeline, the clock, the cal- endar, and the counter. There is little research on the effectiveness of these different time- legend devices and care should be taken that they are informative but not distracting to the users. The same principle applies to other parts of the legend. The information must be eas- ily understandable yet not distract from the rest of the animation. In the different anima- tions that I made, I tried having the legend information in different places. I found that having the information all together on the lower part of the screen worked out the best. I also found that information on the bottom right or the top left side of the screen was more likely to be seen then information in the opposite comers. Another aspect that must be thought about is the use of labels. Since the objects on the screen are often moving around, it is difficult to give them labels. I found the use of labels to be distracting in most cases. If labels are needed, an option is to give the user the choice of turning them “on” or “off.” Otherwise, it is probably best to make them as simple and as unobtrusive as possible. Finally, I would emphasize that simplicity in animations is a virtue; complexity is distracting. Conclusions In this thesis I have examined the advantages and disadvantages of using three differ- ent software packages in illustrating, through animation, a geographical phenomenon that moves through space and time. I have given a subjective evaluation of the packages based on my experiences and I have summarized comments given by independent observers. Based on the above, it is my opinion that one should not pursue the use of HyperCard as 31 a cartographic animation tool. Because of its crude construction technique, creating ani- mations in HyperCard is so time-consuming that the products are not worthwhile. Without color support, there is very little visual appeal in the animations. However, both Director and Transform should be used as tools for creating cartographic animations. Both of these packages created cartographic animations that were intriguing, visually stimulating, and useful in both scientific visualization and illustration for teaching or for research. It would also be useful to explore the use of Director and Transform to animate other movement phenomena, such as migration and trade, or a phenomenon like rainfall which does not move through space but does change over time. In general, current desktop animation techniques are primitive, time-consuming, and somewhat crude, however they should not be discarded as a method of producing carto- graphic animations. Since the use of desktop computers to create cartographic animations is still a relatively new technique, there are still many questions both about how the ani- mations should be created and also about what the animations should include. As the tech- nique evolves, and as the software evolves, using desktop animation techniques to create cartographic animations will only become easier. APPENDICES APPENDIX A DATA DOCUMENTATION ATLANTIC BASIN BEST TRACK DOCUMENTATION ..... BY CHRIS LANDSEA (2/16/95) This documentation is based upon "A tropical cyclone data tape for the North Atlantic basin, 1886-1983: Contents, limitations, and uses", _NOAA Technical Memorandum NWS NHC 22_, 1984 written by B.R. Jarvinen, C.J. Neumann, and M.A.S. Davis. The North Atlantic 'best track' (so named as it is the 'best' track and intensity estimates of tropical cyclones as determined in a post-analysis of all available data) is maintained by the forecasters and researchers at the National Hurricane Center in Miami, Florida. Permission to make free access of this data and docu- mentation were provided by Dr. Robert Sheets - director of NHC and by Mr. Neal Lott of the National Climatic Data Center in Asheville, NC. Questions regarding the format and/or errors in the documentation/data should be directed to Chris Landsea (landseanowndry.atmos.colostate.edu). Currently, the database extends from 1886 to 1993. The database filename is 'tra86t093.atl. Updates for the coming year will be provided as soon as the analysis is final at NBC. Eh» I hope this data provides some insight into the workings of the Atlantic hurricane for you as it, to some degree, has for me ........ CWL Note that the 'tra86t093.at1' file is UNIX compressed to save space on our computer so that the filename is 'tra86t093.atl.Z'. To read the file after you've ftp'ed it, you need to UNCOMPRESS the file. Otherwise, you can't access the data. $$$$$$$SS$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ P.S. Though the data extends back to 1886, one is cautioned in deriving seasonal statistics back that many years. The advent of aircraft recon- naissance in 1944 is suggested (both by myself and C.J. Neumann) to be the start of reliable seasonal statistics on the frequency and duration of storms. However, as reported in Landsea (1993) ["A climatology of intense (or major) Atlantic hurricanes", _Mon. Wea. Rev._, 121, 1703— 1713], there is a bias in the best track data in that strong hurricanes were over-reported in windstrength for the years 1944 to 1969 (and pos- sibly before 1944 as well). This bias amounted to a five knot overesti- mation at the 100kt threshold for major hurricane status. The amount of overestimation is even worse for yet stronger storms. $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ 32 33 There are three basic types of datalines in the Best Track. TYPE A: 92620 08/16/1992 M=l3 2 SNBR= 899 ANDREW XING=1 SSS=4 Card# MM/DD/Year Days S# Total#... Name ........ US Hit.Hi US category TYPE B: 92580 04/2282450610 30 100382490615 45 100282520620 45 100252550624 45 1003* Card# MM/DD&LatLongWindPress&LatLongWindPress&LatLongWindPress&LatLongWindPress TYPE C: 92760 HRCFL4BFL3 LA3 Card# TpHit.Hit.Hit. ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo TYPE A: Card# = Sequential card number starting at 00010 in 1886 MM/DD/Year = Month, Day, and Year of storm Days = Number of days in which positions are available (note that this also means number of lines to follow of type B and then one line of type C) S# = Storm number for that particular year (including subtropical storms) Total# = Storm number since the beginning of the record (since 1886) Name = Storms only given official names since 1950 US Hit =’1' = Made landfall over the United States as tropical storm or hurricane, '0' = did not make U.S. landfall Hi US category = '9' = Used before 1899 to indicate U.S. landfall as a hurricane of unspecified Saffir-Simpson category '0' = Used to indicate U.S. landfall as tropical storm, but this has not been utilized in recent years '1' to '5' = Highest category on the Saffir-Simpson scale that the storm made landfall along the U.S. '1' is a minimal hurricane, '5' is a catastrophic hurricane ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo TYPE 3: Card# = As above. MM/DD = Month and Day of Storm & = 'S' (Subtropical stage), '*' (tropical cyclone stage), 'E' (extratropical stage), 'W' (wave stage — rarely used) LatLong = Position of storm: 24.5N, 61.0w 34 Wind = Maximum sustained (1 minute) surface (10m) windspeed in knots (in general, these are to the nearest 5 knots). Press = Central surface pressure of storm in mb (if available). Since 1979, central pressures are given everytime even if a satellite estima- tion is needed. Positions and intensities are at 002, 06Z, 12Z, 18Z TYPE C: Card# = As above. Tp = Maximum intensity of storm ('HR' = hurricane, 'TS' = tropical storm, 'SS‘ = subtropical storm) Hit = U.S. landfallings as hurricane ('LA' = Louisiana, etc.) and Saffir- Simpson category at landfall ('1' = minimal hurricane...'5' = super hur- ricane). (Note that Florida and Texas are split into smaller regions: 'AFL' = Northwest Florida, 'BFL' = Southwest Florida, ’CFL' = Southeast Florida, 'DFL' = Northeast Florida, 'ATX' = South Texas, 'BTX' = Central Texas, 'CTX' = North Texas.) Here are two sample storms from the original data: 80310 08/25/1979 M=15 4 SNBR= 777 DAVID XING=1 SSS=2 80320 08/25* 0 0 0 0* 0 0 0 0*117 361 25 1008*117 382 25 1007* 80330 08/26*1l7 403 30 1006*116 422 35 1005*116 440 40 1003*116 455 45 998* 80340 08/27*117 470 55 990*118 485 65 980*118 500 80 966*119 515 95 954* 80350 08/28*122 529 115 947*125 544 125 941*128 557 130 938*132 569 125 941* 80360 08/29*137 580 120 944*142 592 120 942*148 603 125 938*153 616 125 933* 80370 08/30*156 628 130 929*160 642 140 925*163 652 145 924*166 662 150 924* 80380 08/3l*168 673 145 927*170 683 145 928*172 691 145 927*179 697 150 926* 80390 09/01*188 704 130 953*193 720 100 978*197 737 65 1002*206 746 60 1002* 80400 09/02*213 752 65 997*219 755 70 990*230 763 70 984*239 774 75 979* 80410 09/03*246 782 80 976*253 791 80 974*263 796 85 973*272 802 85 972* 80420 09/04*280 805 85 971*291 808 85 970*302 809 85 970*315 812 80 970* 80430 09/05*325 811 65 972*335 809 55 976*349 806 45 980*362 801 40 984* 80440 09/06*376 795 40 987*392 785 40 989*415 763 40 9913433 737 40 992* 80450 09/07E450 700 45 991E465 660 50 988E475 615 50 987E500 570 55 986* 80460 09/08E525 525 60 985* O 0 0 0* 0 0 0 0* 0 0 0 0* 80461 HRCFLZDFLZ GA2 SC2 83220 11/03/1981 M: 5 11 SNBR= 804 KATRINA XING=0 83230 11/03*169 812 25 1005*172 813 25 1005*175 814 25 1004*178 814 30 1002* 83240 11/04*l81 814 30 1001*183 814 35 1000*186 813 40 998*189 812 50 996* 83250 11/05*192 811 60 993*196 808 65 988*200 805 75 980*204 801 70 988* 83260 11/06*209 795 65 996*216 783 55 999*224 770 50 1001*232 755 45 1001* 83270 11/07*240 733 40 1001*250 706 40 1002*259 675 40 1002*268 645 40 1002* 83280 HR APPENDIX B HYPERCARD Instructions for Creating HyperCard Animations The HyperCard animations were created by placing a series of images on cards with- in each stack. Once the images had been placed, HyperText scripts were written to cycle the images in an animated fashion, and to run the “clock.” The following are the scripts used in each of the HyperCard animations. HyperText Scripts Hurricane Debbie - Script of play button: on mouseUp put 13 into theDay go to the next card show bg field month show bg field date set hilite of bg button play to true repeat 11 times add 1 to theDay put theDay into bg field date wait 1 seconds go to next card end repeat hide bg button play end mouseUp 0 Script of Start Over button: on mouseUp hide bg field month hide bg field date set hilite of bg button play to false show bg button play go to card 1 end mouseUp 35 36 Tropical Storm Anna 0 Script of Play button: on mouseUp put put put put put put 24 into theDay "July" into theMonth 0 into theHour theDay into bg field date theMonth into bg field month theHour into bg field hour go to the next card show bg field month show bg field date show bg field hour2 show bg field hour set hilite of bg button play to true repeat 12 times put 0 into theHour put theHour into bg field hour show bg field hour if theDay > 30 then put 1 into theDay put "August" into theMonth put theMonth into bg field month put theDay into bg field date wait 20 ticks go to next card else add 1 to theDay put theDay into bg field date show bg field date wait 20 ticks go to next card end if repeat 3 times add 6 to theHour put theHour into bg field hour put theDay into bg field date show bg field month show bg field date show bg field hour wait 20 ticks go to the next card end repeat end repeat hide hide hide hide hide bg field month bg field date bg field hour bg field hour2 bg button play end mouseUp 37 ° Script of Start Over button: on mouseUp hide bg field month hide bg field date set hilite of bg button play to false show bg button play go to card 1 end mouseUp 1992 Storm Season - Script of Play button: on mouseUp put 15 into theDay put "August" into theMonth go to the next card show bg field month show bg field date set hilite of bg button play to true repeat 19 times if theDay > 30 then put 1 into theDay put "September" into theMonth put theMonth into bg field month put theDay into bg field date wait 25 ticks go to next card else add 1 to theDay put theDay into bg field date wait 25 ticks go to next card end if end repeat repeat 13 times put theMonth into bg field month add 1 to theDay put theDay into bg field date show bg field month show bg field date wait 30 ticks end repeat go to the next card put theMonth into bg field month show bg field month show bg field date repeat 23 times if theDay > 29 then put 1 into theDay put "October" into theMonth put theMonth into bg field month 38 put theDay into bg field date wait 25 ticks go to next card else put theMonth into bg field month add 1 to theDay put theDay into bg field date wait 25 ticks go to next card end if end repeat repeat 12 times put theMonth into bg field month add 1 to theDay put theDay into bg field date show bg field month show bg field date wait 30 ticks end repeat go to the next card put theMonth into bg field month show bg field month show bg field date repeat 15 times if theDay > 30 then put 1 into theDay put "November" into theMonth put theMonth into bg field month put theDay into bg field date wait 25 ticks go to next card else put theMonth into bg field month add 1 to theDay put theDay into bg field date wait 25 ticks go to next card end if end repeat 'hide bg button play end mouseUp 0 Script of Start Over button: on mouseUp hide bg field month hide bg field date set hilite of bg button play to false show bg button play go to card 1 end mouseUp 39 For all three animations - Script of Quit button: on mouseUp answer "Quit HyperCard?" with "OK" or "Cancel" if it is "OK" then doMenu "Quit HyperCard" else answer "Glad you reconsidered" with "No problem" end if end mouseUp ° Stack Script: on open go to card 1 end open APPENDIX C TRANSFORM To import the data into Transform I went through the following steps: 0 Open the text (.txt) file 0 Select “text columns” from the list data types that can be imported [] ° For the delimiters select “commas only” [] 0 Configure using the information below: Longitude -110 10 0.5 Latitude 5 65 0.5 Wind speed 10 165 ° Specify Data Placement: sum values Background Value: -1 0 Fill Method: No fill click (OK> and the data matrix appears - select 0 a new matrix appears 0 select ° an image appears 0 set the min and max for best use of color and replot ' adjust the position of the image to 0,0 0 adjust the , zoom=2 : overlay outline map image from scrapbook ° deselect : from the tables menu select a color table (“Space”) ' Now save the image as a transform file and then under a different name as a pict file I used the Open command to open the text (.txt) file. Then I selected “text columns” from the list data types that can be imported. For the delimiters I selected “commas only.” To configure the file so that each image would be the same I used the information listed above. I then specified [Data Placement = sum values] so that Z values that fell on top of one another would be added together and [Background Value = -1] so that the background value would be different from all the data. By specifying [Fill Method = No fill] my data was placed in the matrix, and cells without data were not filled in. By clicking and the data matrix appears. Once the data matrix appeared, I created a new matrix of 40 41 smoothed values by selecting . This smoothed my data values with five iterations. When the new data matrix appeared, I selected . This created an interpolated image of my data. I then went through a series of adjustments to enhance the look of the image and to make sure that each image had the same parameters as the others. I set the minimum and maximum values for the color range, adjusted the position of the image to (0,0), adjusted the rectangle size to zoom=2, overlaid an outline map from the “scrapbook” onto the image, and deselected so that the numbers along the x and y axis disappeared. From the tables menu, I selected the color table “Space.” I then saved the image as a transform file and then under a different name as a pict file. I followed the above sequence for each image to be used in the final animation. I then re-opened each pict file and placed the timeline, created in Freehand, into the overlay. This allowed the map user to see what time of year different patterns appeared or disappeared. Once all the fames had been created, the animation was viewed in a companion program called “View.” This program allows the map user to control the speed of the animation, to start or stop the animation on any given frame, and to change the color table used. APPENDIX D DIRECTOR Instructions for Creating the Director Animation After the castmembers had been imported or created in Director, and after they had been place in appropriate positions on the stage, the following Lingo Scripts were used to create the animation. Lingo Scripts - First Frame (Score) on enterFrame puppetsprite 4, TRUE end on exitFrame installMenu 40 global t set t = 20 set the checkMark of menuItem 3 of menu "speed" to true set the text of Cast 29 set the enabled of menuItem 1 of menu ”Trail" to TRUE set the enabled of menuItem 2 of menu ”Trail" to FALSE set the textSize of field 29 = 18 set the textFont of Field 29 = "Gill Black 881" end exitFrame 0 Second & Fourth Frame (Score) on exitFrame go to the frame end on enterFrame set the text of cast 10 = EMPTY set the text of cast 11 = EMPTY set the text of cast 12 = EMPTY set the text of cast 13 = EMPTY set the text of cast 30 = EMPTY set the text of cast 31 = EMPTY set the text of cast 32 = EMPTY end 42 43 0 Menu (Cast Member) menu: @ About... z about menu: File Quit/Q = quit menu: Storm 1992, Andrew = set the text of cast 29 = “ANDREW" 1969, Camille = set the text of cast 29 "CAMILLE" 1988, Gilbert = set the text of cast 29 "GILBERT" 1985, Gloria = set the text of cast 29 = "GLORIA" 1989, Hugo = set the text of cast 29 = "HUGO" menu: Speed Very Fastz put vfast() into t Fast = put fast() into t Medium = put med() into t Slow z put slow() into t Very Slow = put vslow() into t menu: Trail Trail On z trailOn Trail Off = trailOff Clear Trails/C = clear 0 Apple menu scripts on about set the fileName of window "aboutme" to "aboutme" set the rect of window "aboutme" =rect(75, 50, 275, 150) set the windowType of window "aboutme" to 3 open window "aboutme" end about 0 Speed Menu Scripts on vFast set t=1 repeat with i=1 to 4 set the checkMark of menuItem i of menu "speed" to false end repeat set the checkMark of menuItem 1 of menu "speed" to true return t end vfast on Fast set t=10 repeat with i=1 to 4 set the checkMark of menuItem i of menu "speed" to false end repeat set the checkMark of menuItem 2 of menu "speed" to true return t end fast on Med set t=20 repeat with i=1 to 4 set the checkMark of menuItem i of menu "speed" to false end repeat set the checkMark of menuItem 3 of menu "speed" to true return t end Med on Slow set t=40 repeat with i=1 to 4 set the checkMark of menuItem i of menu "speed" to false end repeat set the checkMark of menuItem 4 of menu "speed" to true return t end Slow on VSlow set t=60 repeat with i=1 to 4 set the checkMark of menuItem i of menu "speed" to false end repeat set the checkMark of menuItem 5 of menu "speed" to true return t end VSlow : Trail Menu Scripts On trailOn go to frame 5 set the enabled of menuItem 2 of menu "Trail" to TRUE set the enabled of menuItem 1 of menu "Trail" to FALSE end trailOn On trailOff puppetTransition 49, 0, 2, FALSE go to frame 2 set the enabled of menuItem 1 of menu "Trail" to TRUE set the enabled of menuItem 2 of menu "Trail" to FALSE end trailOff on clear puppetTransition 49, 0, 2, FALSE end clear 45 ° Reading Files on readFile fname put FileIO(mNew, "read", fname) into gVarObject set theVText = EMPTY put gVarObject(mreadline) into theline repeat while (theline <>EMPTY) if char length(theline) of theline=RETURN then put "" into char length(theline) of theline end if set theVText=theVText & theline put gVarObject(mReadLine)into theline end repeat gVarObject(mdispose) return theVText end - “Click Me” — Running the animation on mouseUp -- sets up global variables global gVarObject global dataList global n -- calls subroutines from movie scripts put the text of cast 29 into n if n=" " then alert "Before playing this animation you must first select the storm you wish to View from the STORM menu." else readFile (n & "x.txt") set dataXList=the result readFile (n & "y.txt") set dataYList=the result readFile (n & "z.txt") set dataWList=the result readFile (n & "d.txt") set dataDList=the result readFile (n & "h.txt”) set dataHList=the result 46 -— calculates the length of the file and sets a value for the variable r put length(dataXList) into noChars set r=1 set q=1 set p=1 —- sets up a repeat structure that repeats from 1 to the total number of Longitudes in the file reading a new set of numbers each time -- it sets size, color and position of the hurricane symbol —- it shows on the screen the windspeed (stage, Latitude and Longitude) Repeat with counter = 1 to (noChars/3) set c=EMPTY put chars(dataXList, r, r+2) into put chars(dataYList, r, r+2) into put chars(dataWList, r, r+2) into put chars(dataDList, q, q+1) into put chars(dataDList, q+2, q+3) into D put chars(dataDList, q+4, q+5) into E put chars(dataHList, p, p+l) into H 3 N & x Set 2: Z * 1.15 set the text of cast 12 = integer(Z) && "MPH" set the text of cast 10 = string(integer((value(X)/5.3)-110)) && "Degrees" set the text of cast 11 = string(integer((((value(Y)—30)/5.3)-65)*(-1))) && "Degrees" if value(M) = 01 then set the text of cast 30 = "January" else if value(M) = 02 then set the text of cast 30 = "February" else if value(M) = 03 then set the text of cast 30 = "March" else if value(M) = 04 then set the text of cast 30 = "April" else if value(M) = 05 then set the text of cast 30 = "May" else if value(M) = 06 then set the text of cast 30 = "June" else if value(M) = 07 then set the text of cast 30 = "July" else if value(M) = 08 then set the text of cast 30 = "August" else if value(M) = 09 then set the text of cast 30 = "September" 47 else if value(M) = 10 then set the text of cast 30 = "October" else if value(M) = 11 then set the text of cast 30 = "November" else if value(M) = 12 then set the text of cast 30 = "December" end if set the text of cast 31 =string(integer(D)&","&&"19"&E) set the text of cast 32 =string("HR:"&&H) if value(Z) < 74 then set c: 5 set the text of cast 13 = "Tropical Storm" else if value(Z) >= 74 and value(Z) < 96 then set C: 11 set the text of cast 13 = "Class 1 Hurricane" else if value(Z) >= 96 and value(Z) < 111 then set c: 17 set the text of cast 13 = "Class 2 Hurricane" else if value(Z) >= 111 and value(Z) < 131 then set C: 23 set the text of cast 13 = "Class 3 Hurricane" else if value(Z) >= 131 and value(Z) < 155 then set C: 29 set the text of cast 13 = "Class 4 Hurricane" else if value(Z) >= 155 then set C: 35 set the text of cast 13 = "Class 5 Hurricane" end if set the locH of sprite 5 to Value(X) set the locV of sprite 5 to Value(Y) set the forecolor of sprite 5 to c set the height of sprite 5 to value(((Z-10)/5)+10) set the width of sprite 5 to value(((Z-10)/5)+10) updateStage startTimer repeat while the timer < Value(t) nothing end repeat set r=r+3 set q=q+6 set p=p+2 end repeat end if end APPENDIX E INTERVIEW MATERIALS Copy of the Consent Form Consent Form 1. I have freely consented to take part in a study being conducted by Catherine Van Vliet Mey. I will be participating in a study on animated map design. 2. The study has been explained to me, and I understand the explanation that has been given and what my participation will involve. 3. I understand that I am free to discontinue my participation at anytime without penalty. I understand that the expected length of my participation is approximately 10 minutes. 4. I understand that the results of my participation in the study will be kept in strict confidence, as will those of all other individuals participating. In other words, all partici- pants will remain anonymous in the reporting of the results. 5. I understand that, at my request, I can receive additional explanation of the study after my participation is completed. Signed: Date: 48 49 Copy of the Interview Information & Questions Subject Name: Date: Climatologist-Cartographer—Student Animation: 1) Do you think this animation would be useful in: Teaching? Scientific Visualization? What age group would you use it with? 2) What do you think this animation is trying to show? 3) What did you like best about this animation? Why? 4) What do you think could be improved? Why? The same questions were asked for each set of animations and then the following ques- tions was asked: Which of the three sets of animations do you prefer? Why? The following information was given to the subjects: I will ask you a series of questions as you view three different sets of animated maps. The subject of each of these animations will be tropical cyclones in the North Atlantic Basin. Read to them the consent form. Transform: This animation shows 50 years worth of hurricane information aggregated into a one year time span. Please note the timeline along the bottom the animation. Each frame of the animation shows a four week block of time (ie it shows all the storms that occurred in that four week block from the year 1944 to 1993). For each frame that one year block moves forward one week. (so if this frame was showing week 1, 2, 3, & 4; the next frame would show week 2, 3, 4, & 5) The colors are an index of windspeed and frequency— the blues represent low wind- speed and low frequency, the yellows indicate high windspeed and high frequency. You can change the speed of the animation by changing the number at the bottom. 50 H yperCard: This set of animations contains three different animations. One shows hurricane Anna, one shows hurricane Debbie, and one shows all the storms in the 1992 hurricane season. Hurricane Anna: This animation shows hurricane anna, note the colors indicate wind- speed in knots, and there is a clock in the upper right hand comer (a knot = one nautical mile per hour or about 1.15 mph) Hurricane Debbie: This animation shows hurricane debbie, this animation shows the track of the storm as it move across the atlantic. Clock in upper right hand corner. 1992 Hurricane Season: This animation shows all the storms that occurred during the 1992 storm season. Note that there are two time periods when there are no storms occur- ring however the clock (in the upper right hand comer) continue to count off the days. Director: In this animation you view one storm at a time however you can choose which storm to view (from a group of five famous storms) by picking a storm from the storm menu. You can also change the speed of the animation and whether or not you want it to leave a trail of the storm. Note you can see statistics about the storm on the bottom of the screen. Click the button that says click me to start the animation. BIBLIOGRAPHY BIBLIOGRAPHY Abler, Ronald E, Melvin G. Marcus, and Judy M. Olson (eds.) 1992. Geography’s Inner Worlds: Pervasive Themes in Contemporary American Geography. New Brunswick, New Jersey: Rutgers University Press. Ahrens, C. Donald. 1994. Meteorology Today: An Introduction to Weather; Climate, and the Environment. (Fifth Edition) St. Paul: West Publishing Company. Apple HyperCard, ver 2.2. 1993. (Manuals and Computer Software for the Apple Macintosh) Cupertino, CA: Apple Computer, Inc. Bunge, William. 1966. Theoretical Geography. Lund Studies in Geography Ser. C. General and Mathematical Geography No. 1. Lund, Sweden: C.W.K. Gleerup. Campbell, Craig S. and Stephen L. Egbert. 1990. “Animated Cartography / Thirty Years of Scratching the Surface.” Cartographica 27, 2:24-46. Cross, John A. 1992. “Natural Hazards within the West Indies.” Journal of Geography 91, 52190-199. Dorling, Daniel. 1992. “Stretching Space and Splicing Time: From Cartographic Animation to Interactive Visualization.” Cartography and Geographic Information Systems 19, 42215-227, 267-270. Gersmehl, Philip J. 1990. “Choosing Tools: Nine Metaphors of Four-Dimensional Cartography.” Cartographic Perspectives 5:3-17. Hardwick, Susan Wiley, and Donald G. Holtgrieve. 1990. Patterns on our Planet: Concepts and Themes in Geography. New York: Macmillian Publishing Company. Lowe, John C., and S. Moryadas. 1975. The Geography of Movement. Boston: Houghton Mifflin Company. 51 52 Lutgens, Frederick K. and Edward J. Tarbuck. 1986. The Atmosphere: An Introduction to Meteorology. (Third Edition) Englewood Cliffs, New Jersey: Prentice-Hall, Inc. MacroMedia Director, ver 4.0.4. 1994. (Manuals and Computer Software for the Apple Macintosh) San Francisco, CA: Macromedia, Inc. Moellering, Harold. 1980. “The Real-Time Animation of Three-Dimensional Maps.” The American Cartographer 7, 1:67-75. Moran, Joseph M. and Michael D. Morgan. 1991. Meteorology: The Atmosphere and The Science of Weather: New York: MacMillan Publishing Company. Neumann, Charles J, et al. 1993. Tropical Cyclones of the North Atlantic Ocean, 1871- 1992. Historical Climatology Series 6-2. Ashville, North Carolina: National Climatic Data Center. Peterson, Michael P. 1995. Interactive and Animated Cartography. Englewood Cliffs, New Jersey: Prentice-Hall, Inc. Spyglass Transform, ver 3.0 and Spyglass View, ver. 3.0. 1993. (Manuals and Computer Software for the Apple Macintosh) Champaign, IL:Spyglass, Inc. Thrower, Norman J.W. 1961. “Animated Cartography in the United States.” International Yearbook of Cartography 1:20-28. "IlllllllllllllllllllllS