The established practice of geomechanical testing includes the procedures of core sampling and marking, as well as core plug drilling out and testing. The key challenge is that a selection of core plugs is physically or economically limited. The classical core analysis offers a limited and discrete data set. Along with that, the tests carried out on a set of core plugs are dependent upon the core quality and the accuracy of selection of the core plugs sharing identical properties. These constraints impair accuracy and reliability of geomechanical models. This paper describes controlled core scratching (CST) as a method of geomechanical testing of full-size cores. Physically, the CST method is based on resistance measurements of the of full-size cores to scratching. Depending on the cutter penetration depth and the core sample scratching rate, the force characterizes unconfined compressive strength (UCS) and/or fracture resistance of the rock. The principal CST advantage is that it provides accurate and diverse continuous geomechanical rock properties. The application of the CST method enhances comprehensive core analysis accuracy, synthetic physical and mechanical parameters reliability and, therefore, geomechanical modeling precision. In practice, the CST method applied in conjunction with the classical geomechnical core analysis enhances the quality of drilling, completion, frac'ing, and field development support in general. This paper presents a comparative study of geomechanical models that incorporate synthetic curves for unconfined compressive strength (UCS), Brazilian tensile strength (BTS) and cohesion strength (CS), derived from the CST and well logging data, using a well that is drilled through halocarbonates and clastics as an example. The comparative study of the CST-based and log-based geomechanical models demonstrated that the CST-based model outputs robust correlations and detailed distinctive curves of physical and mechanical properties and clear stratification boundaries.
