Developments in 3D Photorealistic mapping, modeling and analysis of sedimentary features at scales from centimeter to kilometer: Application to Pennsylvanian Breathitt Group, Appalachian Mountains, Pikeville, Kentucky

By: Miao Wang

Research Supervisor: Carlos Aiken


Three Dimensional digital mapping of outcrops provide a robust and flexible dataset that can be saved and archived to be re-interpreted multiple times subsequent to data collection. Three dimensional photorealistic models of geology are being uniquely created at UTD and specific digital analysis tools are also being developed mainly for structural geology. The central issues in this study are the building of sedimentary feature extraction tools from such 3D photorealistic outcrop models, in this case a site in the western Kentucky Appalachians. The mapping and analysis techniques on an ArcGIS platform will be extended to extract individual architecture elements from sedimentary outcrops. The benefits of this workflow are significant, resulting in more accurate interpretations and the ability to conduct and test hypotheses. A test area is the Pennsylvanian deltaic outcrop of Breathitt Group in Pikeville, Kentucky that is under analysis and mapping by the Univ. of South Carolina. This Appalachian road-cut shows a wide variation of facies and a multitude of heterogeneities represented within channel complex systems, which is a challenge to map and analyze quantitatively in 2D. At least 30 channels and associated facies of different scales have been identified. The crescent (arc) shaped outcrop is about 1.5 km in length and 20 m in thickness which can lend itself to 3D digital analysis at small and large scales. The outcrop’s small thickness relative to its length makes it difficult to manually correlate the architecture and facies through the entire outcrop and it is a dilemma as to how to display the features at scales from centimeters to kilometers.  The outcrop has been mapped and analyzed in 2D. Hypotheses have been formulated which cannot be effectively tested without 3D quantitative information. Much of the outcrop has now been preliminarily captured by a UTD/USC team and the ultimate goal is to create a high quality and resolution 3D model and then use standalone ‘graphic tools’ on an ArcGIS platform to identify the various architectural elements used in the interpretation of the sedimentary depositional settings. Centimeters to kilometer scale features which will be defined, digitized, extracted and analyzed as 3D models consisting of the photorealistic terrain surfaces bounded with geometries consisting of surfaces fit to the data. Then the various elements can then be placed in a lateral, vertical, and time context.


The aim of this project therefore is to provide efficient methods and software to improve the accuracy and resolution of the current photorealistic outcrop model by automatically draping higher quality photos (105mm lens) in particular interest areas using image matching, and removing noise from vegetation from models by means of reflectance spectra analysis, to develop new efficient "graphic tools" that include automatic identification of different geologic fabrics and user defined manual identification required for the interpretation of sedimentary depositional settings, and build architectural elements that can be systematically extracted from the section in a reverse order of their accumulation to be chronologically reassembled.

Study Area

The study section located on US 23 – 4.5 km south of Pikeville, Kentucky (Fig.2) has been recommended (C. Kendall, personal communication) as an ideal site for such study. The Latitude and Longitude from west to east are N37°26'36.48", W82°31'3.01" and N37°26'49.76", W82°30'17.1". The 1.2 miles long and 15.25 meters thick road-cut section exposes sequences of fluvio-deltaic clastic sediments - the Appalachian Basin Breathitt Group.


Regional setting

The sediment of this road-cut is in Breathitt Group, middle age of Pennsylvanian, along the Appalachian Mountain of eastern Kentucky, including about 30 channels and associated facies of different scales. The arc shaped outcrop is 1.5km long and 20m thick. It provides suitable constraints for laser scanned capture and subsequent incorporation into 3-D reconstruction of the geology and sequence stratigraphic framework of the outcrop (Chris Kendall and Sumanta Chatterjee, University of South Carolina).



There is workflow and software for the development of photorealistic 3D outcrop models and tools for analysis [Xu 2000; Xu et al., 1999, 2000; Alfrahan et al., 2008; Oldow 2008; White 2008; Aiken et al., 2009]. These methods and tools work well for modeling and analyzing structural geology.


Related Materials

Video of 3D photorealistic model of Pikeville (Partial)


Issue with vegetation (Master Project in GIS, 2009)


Images of the model: