Geovisualization is the visual representation of the surface and inside of the earth, an inherently three-dimensional and evolving entity, to understand a geospace and how variables in that space change with time. In this post and some future ones, I will explore earth visualization methods and philosophies, and invite you to help extend this community’s capabilities.
Two weeks ago, I attended the Where2.0 conference in San Jose, California, the fifth annual three-day gathering of location-based technologists. To know what got other attendees excited, search #where20 in Twitter and watch the conference videos. There was much talk about the latest platform or API with which to add location awareness to a web or mobile application, but the scientist in me was enthralled with novel ways to visualize data through imagination and related programming. Specifically, I was impressed by the presented work of Maps From Scratch by Stamen Design, Urban Mapping’s transit trees and animations (video) and Sense Networks’ CitySense 4D heatmaps (video).
Velodyne Lidar Demo at Where 2.0
While these are beautiful, innovative mapping techniques, they remind you not to depend on Google or another major map API to the point where your perception is trained into certain ways of viewing our world. In other words, the data and their relationships ought to dictate how we look at it, rather than a pre-canned visualization defining the data. Again, I fully support (open) standards, but let’s not standardize our way into a small dark box. Dynamic and relevant visualization as well as “transforming access, usability and collaboration,” to quote ESRI’s Jack Dangermond, is all too important in the location-based, four-dimensional and data-intensive field of geoscience.
Another geovisualization concern relates to the use of 3D and simulation tools at the end of a workflow, instead of through its entirety. Until recently, I worked for an oil major where many interpreters of 3D and time-lapse seismic data cubes insisted on working with 2D maps, cross-sections and well plans, using 3D visualization as just that: to rotate, pan and zoom through the end product of their analysis. This turns out to be an industry-wide phenomenon, brought up and bemoaned at conferences annually. Companies have made various attempts to enable analyzing the data, not just looking at it, in 3D through its lifecycle from seismic acquisition up until and including hydrocarbon production and well abandonment.
3D seismic interpretation of a deepwater petroleum system
Products such as GeoProbe, GigaViz, 3DDrillView, 3DMove and other internal tools are researched, generated, tested and deployed frequently, but their usage numbers are small compared with 2D counterparts. Most common user complaints are data load/refresh times, effort and time required to climb the learning curve associated with a 3D-interpretation application and “What’s the point if we can do this in 2D?” All this is to emphasize my earlier point of dynamic and relevant user tools along with access and usability. While great strides have been made in the areas of graphics, access of memory and terabytes of storage, software must constantly be updated to reflect these improvements and user interfaces made relevant to geoscientists to whom time is a lot of money. As Michael Heck says in “3D Visualization for Oil and Gas Evolves”:
Advances in both hardware and software are coming together to enable larger data sets, more automated analysis, and more effective presentation of the data on single workstations. Taking advantage of these advances will be challenging for software developers and will require some re-thinking of application architectures and user interfaces.
Also necessary is a strong corporate commitment to 3D data analysis and ensuring these tools are germane, do help provide new insights and keep the user efficient and productive, not using 3D for 3D’s sake.
At Where 2.0, Brandon Martin-Anderson of Urban Mapping said “A map must correspondingly warp and bend according to position and time.” As must all 3D visualizations, according to space, variables and time. Data is king, not the software application. As we go forward in our line of work, let us keep this in mind and our decision-making tools usable, accessible and suitable. If you are working on something cool and new in earth systems visualization, please share with us here at VizWorld.
Petrel is now much stronger in Geophysics than either GigaViz or GeoProbe, you should check it out if you get a chance.
Many students are coming out of university with a good knowledge of the the Windows based applications and some companies are starting to reflect that in there application choice. I completely agree with you data is king, you can not do 3D just for 3D’s sake that is why it is important that when we develop applications we have a full depth of functionality with the option to see (or interpret) the data either in 3D or 2D.
Stephen,
Thanks for the observation. Generally, I agree that the younger generation picks up on 3D tools and incorporates them into the workflow quicker, but there is a roadblock and that is mentoring: Young people are mentored in concepts & tools by those who are used to working with an older set and, to get up to speed and begin producing work, the younger workers mimic what their mentors do. I see this cycle breaking but it requires some adventurous young workers.
GigaViz, which I mentioned in the post, is the Schlumberger 3D volume interpretation tool. Petrel is good for geological modeling, but is lacking in geophysical robustness. Again, data is king – I’d much rather work in 2D and honor the data than leave the science behind to do 3D for 3D’s sake.
Good article. I find that some parts of the earth science workflow have been predominantly 3D for some time, a good example of this is 3D geological modeling which in it nature needs to be in 3D. This is a fairly new domain so no precedent was set to work in 2D. It also tends to be done by a new generation of geoscientists. Many older geoscientists learnt to work in 2D and feel more comfortable there, it is my experience that the younger generation pick up working in 3D more easily. In the past compute and graphics restrictions lead to the development of specialist 3D visualization tools to work also side the existing tools. Now fast compute and graphics make 3D visualization available at every desktop the time of specialist 3D applications is over, the trick is to make software that provide both 3D and 2D windows for viewing the same data. This type of visualization will ‘ease in’ the older generation while enabling the benefits of 3D visualization and enable work to be done in 3D not just visualization. Check out Schlumberger’s Petrel for a good example of this