GSA Annual Meeting in Denver, Colorado, USA - 2016

Paper No. 176-17
Presentation Time: 9:00 AM-6:30 PM


EDWARDS, Benjamin R.1, POLLOCK, Meagen2, KOCHTITZKY, William1 and ENGEN, Carl-Lars3, (1)Department of Earth Sciences, Dickinson College, 28 N. College Street, Carlisle, PA 17013, (2)Department of Geology, The College of Wooster, 944 College Mall, Wooster, OH 44691, (3)Department of Geology, Beloit College, 700 College Street, Box 777, Beloit, WI 53511,

Documentation of the internal structures of volcanoes are critical for understanding how edifices are built over time, especially for glaciovolcanoes, which have rarely formed historically and are inaccessible during eruptions. We have been unraveling the internal structure of a complex glaciovolcanic ridge (tindar) in southwestern Iceland for the past 5 years in order to better understand the sequence of events that built the ridge. Undirhlidar ridge is ~5 km long, and has been dissected by two different aggregate mines along its axis. The northern mine (Undirhlidar quarry) is inactive and has walls up to 40 m in height that fully expose several critical stratigraphic relationships including multiple sequences of separate pillow lava flows, cross-cutting dikes that locally feed overlying pillow flows, and ridge parallel, continuous massive jointed basaltic units that may be the remnants of internal lava supply networks. The second quarry, ~3 km to the southwest (Vatnsskard quarry) is presently active and continually has new exposures. This quarry only penetrates halfway through the width of the ridge but has ~500 m of exposure along strike. It also has remnants of what appears to be the internal magma distributary system, and many components clearly show evidence that they were (and some still are) open lava tubes. While both quarries contain excellent exposures, many of the structures are difficult to safely access or are inaccessible due to mining activity. In order to overcome access issues, we have used Structure-from-Motion techniques to make 3-D maps of the quarry walls. A series of overlapping pictures were taken from points constrained with D-GPS using a Trimble GeoXH data logger and external antennae. The image locations with corrected positions were imported into Photoscan software to create a point cloud representative for each quarry and to derive a Digital Elevation Model with a reported vertical resolution of less than 1 m. Field testing of a preliminary, low resolution DEM shows that measurements of dyke widths on the DEM have errors of ~5% relative to measurements on the ground. Measurements made from the field-generated DEM will provide significantly better constraints on deposit thicknesses and volume estimates compared to traditional methods of estimating unit thicknesses on vertical faces.