Computers & Geosciences, Volume 24, Number 5, 1998

Holly Dodson
Jeff Johnson
Paula Levin
Randall Souviney
University of California, San Diego and San Diego State University

John C. Butler
Department of Geosciences
University of Houston
Houston, TX 77204

I have watched the Visualizing Earth Project evolve for the past year or so. This set of resources is beginning to tap the potential of the Internet and Multimedia in general as an important component in education.

Visualizing Earth Project

The Visualizing Earth Project is conducting research on the use of images and advanced visualization technologies to help students learn about Earth Science. Funded by the National Science Foundation : Applications of Advanced Technology Program, the team of educators, scientists, and visualization technology specialists is researching cognitive and perceptual issues in visualization, as well as technology development and pedagogical strategies in Earth Science. To that end, one activity of the Visualizing Earth Project is documenting the cognitive processes students use as they attempt to make sense of images of Earth taken from space. The project aims to inform the development of curricula and network-based tools that use visualizations for science teaching and learning. At the University of California, San Diego and San Diego State University, the Visualizing Earth Project activities include the development of several on-line tools to display and manipulate space-based imagery and other georeferenced (e.g., seismic) data to help students explore geological and geographic relationships. Another group of researchers at TERC is focusing on atmospheric and weather studies. A third set of researchers at Pennsylvania State Universityhave been analyzing the cognitive processes involved in visualization of graphic imagery, both printed and on-screen, in schools on both coasts.

At UC San Diego, project staff and classroom teachers are working together to design and implement model image-intensive geology lessons for elementary, middle and high school students. Observing these lessons has helped us to better understand the ways images of various types can be used by students to learn to interpret geographic displays, understand differences in fault types, recognize unique geology created by fault action, and understand why volcanoes and earthquakes are associated with fault movement. Currently one of our collaborating teachers is attempting to design activities appropriate for elementary level students, using this CD-ROM, VisEarth lessons and other materials available on the Web.

Teachers participating with the project have indicated strong support for access to network-based visualization tools in their classrooms. Project staff and middle school teachers have collaboratively designed Web-based lessons that have been field-tested in grades 3-10. Preliminary performance results indicate student improvement in both measures of improved visualization skills and science knowledge attainment. After completing these activities, students seem to be able to generalize their visualization skills to the identification and interpretation of unfamiliar visual images. Our observations in the classrooms and interviews with implementing teachers and students also indicate that students find these lessons highly motivating and effective in enhancing interest in learning science.

UCSD Visualizing Earth Web Sit

The UCSD Visualizing Earth Web site has four main elements:

  1. About VisEarth -- Learn about the goals of the Visualizing Earth project and the people working on the project.

  2. VisEarth Gallery -- Visit galleries of images, including an overview of current research activities.

  3. Fun Stuff -- Try out Web-based activities such as exploring the Aral and Salton Seas, making a 3-D representation of a space shuttle image, learning a little bit about geology, and Earth Detective (under construction), an investigation of the San Andrea Fault.

  4. Friends of VisEarth -- See what¼s happening in the Visualizing Earth classrooms. Lesson plans and activities are posted here as well as an on-line lesson plan submittal form.

Network-Based Assessment Tools

One component of the Visualizing Earth Project has been the development of a series of Web sites which allow students in classrooms to complete image interpretation tasks. Project personnel designed these on-line activities in order to collect data about students understanding of orientation, scale, and surface features of selected images. Space-based Earth images which are presented individually or in pairs. Student responses are automatically collected on-line in a database for later analysis.

Here are some examples of the tasks which assess student visualization skills. Image Task #1 begins with several close-up Shuttle images of the San Diego area. Students are asked to identify the places they recognize and also the general geographical features. They are also asked to identify specific image features such islands, space, lakes, bays, and rivers. They are then shown a series of images of the same region at different scales and orientations and asked to match features in all the images. Students also perform mental rotations of images so that they match the orientation of their paired image. After working with images of San Diego, students perform similar tasks with images of the Middle East. In order to elicit students¼ thinking in image interpretation, we point students to an unnamed feature of a Middle East image (in this case, the Suez Canal), and ask them to explain whether they think the feature is man-made or natural, and their reasons.

What's Next?

There are three new technologies currently under development by project staff at UC San Diego. o We have designed a html "preprocessor" called makeIT which allows teachers to easily create their own Image Tasks like those described above.

Using makeIT, teachers can assemble Image Tasks from text files containing questions and references to pictures. Thus, teachers can develop an image task for their own use and distribute it on the Web for collaboration with other teachers. Users do not need to know the Javascript and HTML to use the application. We will distribute makeIT to teachers who are interested if they will share their Image Tasks with others on our Web site. For more information on makeIT or if your want to obtain the program or the source code, check the VisEarth home page. Using an algorithm developed at JPL, we have also developed a prototype tool with allows students to view a 3-D anaglyph of a selected region using red and blue glasses.

Anaglyph are stereo-pairs of images that give a 3-D effect when viewed with red and blue glasses. A demonstration of a simple 3-D representation of numerous regions in California is included on our home page. Using an image and its corresponding USGS elevation data, the Stereo program maps all elevations onto a flat plane, mapping higher elevations to the side of their original locations. "Left" and "right" images are generated and are overlaid. Then, with a pair of red and blue glasses like those used in 3-D movies, a three dimensional image appears on the computer screen. The left eye (red) sees only the blue (left) image, while the right eye (blue) sees only the red (right) image. The program allows viewers to change elevation exaggeration and inclination to adjust the resulting 3D image.

Finally, in order to provide classrooms with some GIS capabilities, we are developing a functional Web-based interface to ERMapper, a powerful geophysical analysis suite. We are working with the publisher of ERMapper to design ERMapper On-line. Although ER Mapper On-line is in the early stages of development, you can get a better idea of what is possible by exploring the ERMapper Web site. We anticipate this effort will allow us to bring some of the power of georeferencing and the mosaicing of geographic data sets users with modest computer power and network bandwidth. The user will be able to choose an image from a given set and use the ERMapper running on a remote workstation to, for example, select an appropriate map to georeference with the image with a map. The program decides approximately which region of the Earth the image represents and then the user selects Ground Control Points (GCP) by linking points on the image to points on the map. ERMapper will estimate the error for each GCP the user chooses, and when the user has chosen enough GCP's, the program will attempt to georeference the image and the map. We think it will be possible to create mosaics of several neighboring regions and include other geographically referenced data such as digital elevation models (DEM) and seismic data.

Since September 7, 1998

Return to ANON Columns