A NEW METHOD FOR PICKING SHORELINE HEIGHT VARIATIONS FROM ELEVATION PROFILES, APPLIED TO PLEISTOCENE LAKE BONNEVILLE

Shoreline elevation datasets that are currently available for Lake Bonneville were obtained prior to the dawn of space-based geodesy and LiDAR technologies. The locations for these older data are sparse and highly uncertain and the method used for picking shoreline elevations relied on a landform found only in a few places, making correlations of shorelines uncertain (Currey, 1982). We propose an algorithm analogous to that used for automated picking of seismic phases, based on cross-correlation analysis of elevation profiles, that more accurately determines differences in shoreline elevations over large scales and that enables more accurate correlation of shorelines with those of the same age elsewhere.

We first convert elevation profiles to signals consisting of slope or curvature vs. height. Our algorithm then uses a cross-correlation metric to identify the shift in height required to maximize the similarity of arbitrary pairs of signals. The algorithm has been applied to both synthetic data and actual Bonneville shoreline profiles obtained from GPS surveys and available LiDAR. Our planned efforts include applying this algorithm to new survey data with the goal of better constraining uplift associated with unloading of Lake Bonneville. The resulting improved data-set of shoreline height changes through time will be modeled using the isostatic response of a viscoelastic medium to removal of the lake water load. The modeling will be used in turn to evaluate rheological flow laws describing depth- and temperature-dependence of viscosity.

Currey, D.R., Lake Bonneville: Selected features of relevance to neotectonic analysis, U.S. Geological Survey Open File Report, 82-1070, 31 pp., 1982.