Area, length and thickness conservation: Dogma or reality?

The basic assumption of quantitative structural geology is the preservation of material during deformation. However the hypothesis of volume conservation alone does not help to predict past or future geometries and so this assumption is usually translated into bed length in 2D (or area in 3D) and thickness conservation. When subsurface data are missing, geologists may extrapolate surface data to depth using the kink-band approach. These extrapolations, preserving both thicknesses and dips, lead to geometries which are restorable but often erroneous, due to both disharmonic deformation and internal deformation of layers. First, the Bolivian Sub-Andean Zone case is presented to highlight the evolution of the concepts on which balancing is based, and the important role played by a decoupling level in enhancing disharmony. Second, analogue models are analyzed to test the validity of the balancing techniques. Chamberlin's excess area approach is shown to be on average valid. However, neither the length nor the thicknesses are preserved. We propose that in real cases, the length preservation hypothesis during shortening could also be a wrong assumption. If the data are good enough to image the decollement level, the Chamberlin excess area method could be used to compute the bed length changes. ► The dip-domain extrapolation method often leads to the wrong subsurface geometry. ► In the sand box model, volume is preserved but not length or thickness. ► Data on decollement depth may allow computation of length decrease during compression.