Is the Monte Carlo search method efficient for a paleostress analysis of natural heterogeneous fault-slip data? An example from the Kraishte area, SW Bulgaria

The efficiency of the Monte Carlo search method (MC) in separating successfully heterogeneous fault-slip data is evaluated using the separation and stress inversion TR method (TRM). Fault-slip data from the Kraishte area of SW Bulgaria have been used from which two extensional stress regimes with stress axes oriented as in Andersonian state of stresses have been published with the MC. The examination with the TRM that is based on constraints from the Slip Preference Analysis (SPA), indicates that the MC in the absence of any tectonostratigraphic criteria, might well constrain the orientation of the stress axes, but fails to correctly separate the fault-slip data into homogeneous groups, and to correctly define the stress ratio of the stress tensors. The problem is enlarged when a small number of fault-slip data is used for the determination of the stress tensor as occurs in sites of limited size. The resolved stress tensors vary significantly from site to site, because they strongly depend on the number of the recorded fault-slip data and their combinations and, more precisely, on the number of the non-TR compatible faults (i.e. faults without geometric, kinematic and dynamic compatibility according to SPA) that have been considered in their calculation.