EXTRACTION OF DIGITAL ELEVATION MODELS USING by dandanhuanghuang

VIEWS: 3 PAGES: 1

									EXTRACTION OF DIGITAL ELEVATION MODELS USING PHOTOCLINOMETRIC METHODS
Jacquelyn A. Cordova and José M. Hurtado, Jr.
Department of Geological Sciences, University of Texas at El Paso, El Paso, TX 799868,
jacordova4@miners.utep.edu

Aside from retrieving rock samples and radiometrically dating them, the most widely used method for
dating the surface of planetary surfaces is through crater counting. This method is based on the simple
observation that older surfaces are more heavily cratered than younger surfaces. It assumes a relationship
between the rate of impacts over time and the size and distribution of the resulting impact craters. Because
this method relies on the population statistics of a large number of craters, it can only be applied to areas
large enough (and sufficiently old enough) for a statistically adequate number of craters to have
accumulated. Given image data of some minimum spatial resolution, this therefore limits the size of
geologic features that can be dated by crater counting. In addition, crater counting assumes that the impact
flux is known, while, in fact, it is not tightly constrained. Furthermore, it is desirable to be able to date
individual landscape features, such as single impact craters. To address all of these issues, a possible
alternative to crater counting is morphologic dating, which has been used on Earth to date fault scarps and
cinder cones. This method focuses on the changes that occur to the cross-sectional shape of a crater as it
erosionally degrades over time. To do this, detailed, high-resolution topographic data from impact craters
are necessary. We use a method of extracting this topographic data from images called photoclinometry.
Photoclinometry is a process in which a two-dimensional, high-resolution image is converted into a Digital
Elevation Model (DEM) by computationally modeling the geometric interaction between illumination and
an irregular surface that yields the image of that surface. We have successfully created a MATLAB
implementation of an existing “shape-from-shading” algorithm that rapidly that extracts DEMs from
imagery of the Moon. We are currently using the best data available, images from the Lunar
Reconnaissance Orbiter Camera (LROC), which have a resolution of 0.5 meters per pixel. We are
currently modeling craters from Balmer Basin, an impact basin on the nearside of the Moon, and from the
Highlands Feldspathic Terrane. Each model, with respect to their location on the Moon will be shown using
overlaid images, and KML files viewable in Google Moon. The purpose of this research is to compare the
morphologies of small impact craters, excavated in target materials of different types (e.g. mare basalt vs.
highlands). In addition, the crater-degradation for impacts of a wide range of sizes in both basaltic and
feldspathic targets will be investigated. At the Fall 2010 AGU Meeting, another group presented a study,
which uses profiles extracted from LROC NAC stereo images to study the relationships between crater
morphologies, ages and target materials. Using LOLA topography with ~1m vertical resolutions, they
calculated depth/diameter ratios for selected craters. We will compare their findings with our own to
ascertain the accuracy and efficiency of each method.

								
To top