Techniques for Measuring Lava Dome Growth During the 2004 2006 Eruption of Mount St. Helens, Washington

Since the current eruption at Mount St. Helens began in October 2004, we have used vertical and oblique photogrammetry and pre-eruption lidar data to calculate volumes and extrusion rates of the growing lava dome and to document topographic changes in the surrounding crater. At approximately three-week intervals, vertical aerial photographs are acquired along a single flight line centered over the crater at a nominal 1:12000 scale. The photographs are scanned at 12-micron resolution, entered into a soft-copy stereoplotter system for orientation and aerotriangulation, and used to create 2-m-resolution digital elevation models (DEMs) of the dome and adjacent glacier. Pre-eruption GPS sites outside the area of active deformation and photogrammetric data passed to subsequent stereo model sets are used as ground control. Resulting positional accuracy is on the order of centimeters to decimeters. To quantify volumetric changes associated with dome growth, collapse, and glacier deformation, each DEM is compared to pre-eruption reference DEMs for 1986 (derived from topographic contours) and 2003 (from lidar), as well as to each preceding DEM. Volume and extrusion rate as functions of time are critical parameters used to constrain models of the magmatic system, relate measured emission rates of magmatic gases to the magma supply rate, interpret earthquake processes and changing patterns of seismicity, and constrain calculations of the loading effect of the growing dome on measurements of ground deformation. Five digital cameras installed on the crater rim and floor monitor the eruption, document short-lived events such as rockfalls and small explosions, and produce time-lapse animations of the changing topography. Recently, we have begun to measure deformation using images acquired every 10 minutes by these cameras to supplement the less-frequent DEMs. We employ software that uses epipolar equations to calculate coordinates of mobile points that are visible from two or more cameras. The changing coordinates are used to estimate linear extrusion rates and, assuming a constant vent area, to calculate volumetric extrusion rates. Lacking conventional ground control on steep crater walls and in the area of the growing dome, we establish control for the photographs by identifying stationary features that are well-defined in the photographs and in a nearly contemporaneous DEM; coordinates from the latter are used as control for the former. Initial results suggest that mobile features can be tracked in this way with a positional accuracy of ~1 m and that short-term extrusion rates derived from oblique photographs are a valuable supplement to longer-term rates from DEMs.