STRATIGRAPHIC THICKNESS UNCERTAINTY AND ERRORS IN BALANCED CROSS SECTIONS

A surprising outcome of our initial work on errors in balanced cross sections was the result that stratigraphic thickness uncertainties commonly account for at least 50% of the total shortening error. At that time, lacking any formal basis for assigning a value to stratigraphic uncertainty, we used an arbitrary 10% of the total stratigraphic thickness. We have now developed software that, in combination with freely available of scans of virtually all geologic maps published by the USGS and GSC, allows one to make hundreds of map thickness measurements in just a few hours. The program can calculate the strike and dip of bedding from a three point problem, propagating the errors from positional uncertainty to orientation uncertainty, or the user can specify strike and dip. With this software, we can collect enough data to determine true statistical uncertainty on map thickness of units within thrust plates.

The western North American Cordillera contains some of the most elegant and iconic balanced sections, as well as some of the finest geologic maps, ever published. We use sections from Price and Fermor, and Royse, to explore the effect of uncertainty in map thickness on area balances of these sections. 1 sigma stratigraphic thickness uncertainty, measured in homoclinal dip packages of prominent Paleozoic units within single thrust plates (and single quadrangles) ranges from 20-30% in Costigan plate in Canada and can be as high as 37% in the Darby and Absaroka plates of the Idaho Wyoming thrust belt. As the Canadian stratigraphic wedge does not have a uniform taper our basic area balance underestimates shortening; when we apply a new method to account for this discrepancy, the total shortening of ~94 km is close to that obtained by Price and Fermor (102 km for the same Paleozoic package). The uncertainty on shortening is greater than 40 km and nearly all of it due to the 1 sigma stratigraphic uncertainties. The Idaho-Wyoming section of Royse (with maps by the USGS) yields proportionately large errors in shortening magnitude. This study emphasizes the inherent uncertainty in the data that we use in our structural studies.