EVALUATING UNCERTAINTY IN BALANCED CROSS-SECTIONS: A CRITICAL STEP FOR RELATING THRUST-BELTS TO PLATEAU UPLIFT

Continental plateaus such as the Altiplano of the Central Andes are thought to have formed primarily by horizontal shortening and vertical thickening of the crust. The primary tool for assessing horizontal shortening is the balanced cross section, which may be constructed just for the bordering foreland thrust belt or for the entire orogen. In the past, these sections were arduously constructed by hand, in a process so time consuming that individual investigators seldom drew more than one for any particular area. Individual sections are commonly presented as representing the “minimum shortening,” even though other geologists in the same area would produce sections with different “minimum” estimates. In the Central Andes, the differences in these “minimum” estimates are on the order of 200 km or more! What are we to make of these differences?

Balanced cross sections are essentially forecasts of subsurface geometry and should be treated as models which are “best fit” to the available data. As such, the uncertainty of the fit is of critical importance. For a limited class of fold-fault kinematic models, the confidence limits on the estimated model parameters (i.e., those which produce the shortening estimate) can be obtained by Monte Carlo simulation, as demonstrated by the trishear model of individual structures. For cross sections of entire regions or orogens, such rigorous treatment is not yet possible. Nonetheless, computer-aided section construction now makes it possible for an individual structural geologist to explore qualitatively the complete range of viable models which fit the available data and their attendant uncertainties. This should be regarded as a necessary step before when evaluating the contribution of horizontal shortening to plateau uplift. For whole orogen crustal-scale balanced sections, the uncertainties may be so great as to render the entire exercise meaningless. Where rates are important, as in the comparison of thrust shortening to GPS velocities, uncertainties in timing and duration must also be taken into account. In the few well documented cases in the Central Andes, uncertainty in duration of deformation can be as much as a factor or 4 between minimum and maximum estimates whereas uncertainty in shortening magnitude is seldom more than a factor of 2.