Animating Geospatial Data (Spatial Tech Column)31 Dec, 2006 By: James L. Sipes
Add motion to GIS data to discern patterns that aren't there in standard images.
Animation is an illusion. You display a series of still images in rapid succession, and the illusion is that you have captured the dynamic nature of an object or scene. GIS professionals have long been interested in displaying dynamic changes that occur over time to geospatial information. One reason is to present information in a format that is exciting and easy to understand. Another is that animating geospatial information reveals patterns in the landscape that may be invisible when studying static images.
Types of Animation
We tend to use the term animation to refer to any graphic representation that creates the illusion of motion. Animations can be created in numerous ways. The old-fashioned approach to creating animation, called cel animation, involves drawing, inking, painting and photographing each individual image. Traditional cel animation typically requires 24 images for each second of finished animation, so the process can be tedious and time consuming.
In this article
Most current 3D animation programs use a technique called keyframing, which involves defining critical points in an animation and then letting the program interpolate the frames in between. This process reduces the time it takes to create 3D animation.
Stop-motion is a technique in which small changes are made to physical models and then photographed one frame at a time. Morphing involves blending images together to create the illusion of motion. Procedural-based animation is used to simulate environmental features: particle systems are used to model rain, water, fire and explosions, and volumetric sampling creates atmospheric effects.
The most frequently used types of animation in GIS are a fly-by or walkthrough, in which the viewer moves across or through a site, and a temporal animation that illustrates dynamic changes over time. Regardless of the technique, digital animation typically uses 30 frames per second.
Hurricanes are a way of life along the Gulf Coast, and at times the impacts can be devastating. That was certainly the case in 2005, when Hurricanes Katrina and Rita slammed into the coast, causing significant damage to cultural and natural resources. My colleagues and I wanted to create a comprehensive mapping system that would identify the most vulnerable areas along the coast to show planners what areas should be rebuilding priorities and which ones were simply too risky from a human and economic standpoint. We determined an area's vulnerability by analyzing data such as storm surge, flooding, demographic vulnerability, high wind risks, land loss and development patterns.
As part of the project, we animated all severe storms that have hit the Gulf Coast since 1851 to show the magnitude of the problem. The base GIS maps were created from the Historical North Atlantic Hurricane Tracks dataset published by the U.S. Geological Survey. This dataset contains all known Atlantic Basin major storms that made landfall in the United States between 1851 and 2002. These storms are classified on the Saffir–Simpson hurricane scale as Category 3, 4 or 5 at the time of landfall. The data was collected by the NOAA' (National Oceanic and Atmospheric Administration's) Tropical Prediction Center/National Hurricane Center.
Geospatial models often involve massive amounts of data, including high-resolution orthoimagery, terrain models, 3D objects and environmental features. Among the data for each hurricane or storm are name, latitude, longitude, wind speed, pressure, category of storm, basin and hour, day, month and year of each event. Any given year can have several hurricanes or severe storms; 2005 had 26, including Hurricanes Katrina and Rita.
As we all know, not all geospatial data is the same, and that is true for hurricane tracks as well. Aircraft reconnaissance of storms was first introduced in 1944, and this data was considerably more accurate that that of previous years. In the mid-1960s, continuous weather satellite surveillance was introduced, and accurate data was available in a more timely manner. These two developments mean that more recent storm records have a higher degree of accuracy than those prior to 1944.
Using 3ds Max 9
Virtually all major GIS vendors are adding 3D visualization tools that can be used to do some sort of animation. As mentioned earlier, most of the animation tools focus on fly-by or walkthrough sequences where the viewer moves across or through a site. Perhaps because of my background in 3D modeling and animation, I prefer to use high-end programs such as Autodesk's 3ds Max for animating geospatial data (figure 1). 3ds Max is a sophisticated 3D modeling, rendering and animation program used to create games, movies, animations and visualizations. 3ds Max is part of the Autodesk 3D product portfolio that includes Maya, MotionBuilder and VIZ. By combining these programs, you can tackle almost any 3D modeling and animation task.
Figure 1. 3ds Max 9 can depict terrain and the camera path. The Track View Curve Editor shows animation curves (above), and the Material Editor shows gradient material (right). Images courtesy of Autodesk.
The animation capabilities of 3ds Max 9, the latest version available, really can't be compared with those found in GIS programs. To start, 3ds Max 9 provides 64-bit executable files, and the ability to access this much memory makes it easier to handle large, complex scenes that you might expect when working with geospatial datasets.
3ds Max provides a broad range of tools for keyframe and procedural animation that can be used to animate all components of a scene. GIS users probably will find its Curve Editor and Dope Sheet to be a little confusing, but experienced animators will tell you these tools are the best ways to control and edit animation tracks. Key-based and parametric controllers help control the position, rotation and scale of objects, and Bezier and TCB controllers can be used to interpolate movements. Constrained animation techniques can link objects and control their movements. Animation components can be separated into layers—this simplifies the editing process. You can even blend new keys with existing motion. This function is particularly useful when working with multiple animation paths such as hurricane tracks.
One of the major differences between 3ds Max 9 and GIS programs is the ability to render realistic scenes. GIS programs typically use a standard flat lighting scheme with one light source. This approach works fine when you are producing 2D maps, but isn't effective when trying to create 3D scenes.
GIS users may also find that customizing 3ds Max software using third-party plug-ins will allow them to simplify the process of animating geospatial data. Autodesk provides a list of certified animation plug-ins that have been reviewed and approved for use with 3ds Max.
GIS users may find that 3D modeling and animation programs such as 3ds Max are an effective way to depict dynamic changes in geospatial data. The process is far from seamless—3ds Max still isn't geared to working with georeferenced data, and GIS users will find the topological approach to be significantly different that that of GIS programs. But I can't imagine not using the animation tools in 3ds Max for geospatial animation projects.
James L. Sipes is a senior associate with EDAW in Atlanta, Georgia, and the founding principal of Sand County Studios in Seattle, Washington. Reach him at firstname.lastname@example.org.