Geological and geophysical modeling on virtual globes using KML, COLLADA, and Javascript

Abstract

Keyhole Markup Language (KML) has become established as the universal standard for scientific research using virtual globes such as Google Earth and NASA World Wind. We use KML to render geological maps and link associated COLLADA models to represent data such as structural orientations, cross-sections, and geophysical moment tensor solutions. Many other geologists and geophysicists have draped digital maps over a virtual globe's terrain, but some have paid little attention to the problems of image file size and orientation conservation. To maintain application responsiveness and enhance interactivity, we have created data pyramids matching Google Earth's own image pyramids for terrain rendering. This is critical in areas of the world where the terrain is poorly rendered, such as our field areas in the west of Ireland. Following four weeks in the field, we constructed data pyramid structures by hand and coded KML to control data fade-in and blending. This resulted in geological maps that show content appropriate to the current level of detail as defined by the 〈Lod〉 tag in KML. Recent and current implementations of standard KML (versions 1.0 through 2.2) have limitations when it comes to mapping and modeling of subsurface geological and geophysical features. Initially, we overcame these limitations by rendering data in the style of seismic first motion "beach balls" hovering over their respective hypo-centers, and we developed KML code that permits COLLADA models to emerge from the subsurface using the Google Earth slider control. Our latest work includes subsurface views that are made possible by using the Google Earth application program interface (API) with multiple Javascript controls. The methods we have developed are applicable to a wide range of geological and geophysical mapping and modeling problems and have potential uses in both geoscience research and education.