b'AEGC 2023Short abstractsmost of geomechanical and geotechnical assessments, stressconsidered as main input parameters in any geomechanical and information from outside of the stress measurement domaingeotechnical analysis. Direct measurement of these parameters (i.e., stress-data gap) is required. There are different methods tois not frequent due to low recovery of core in fractured intervals predict and model the stress states and other geomechanicaland high cost of operation and subsequent analysis. Hence, parameters in these data gap regions. Herein, we present aother methods are being used to estimate these parameters state-of-the-art 3D geomechanical modelling approach tofrom commonly acquired data such as geophysical logs.develop a 3D geomechanical-numerical model for the Northern Bowen Basin, which is considered as an energy-rich basin thatIn recent years, machine learning approaches have been shown plays a significant role in Australias economy. to be an invaluable tool for rock property prediction. This paper examines a fuzzy logic approach to make relationships This paper presents the most comprehensive quality ranked, basin- between geophysical logs (i.e., input data) and laboratory scale stress map across the world according to the World Stressrock mechanical tests (output data). In order to do that all the Map database, where we have analysed over 240 km of boreholedata (2400 data points) were classified into testing (600) and image logs (inferred from 1000 boreholes) to describe maximumtraining data (1800) in which the training data is used for the horizontal stress (S Hmax ) of the Northern Bowen Basin with 1272construction of the model, and testing data is used to evaluate data records. In addition, we examine the horizontal stressthe models predictions. The results of the best-fit models magnitudes of the study area with 466 pointwise data recordsshow that the rock mechanical properties can be predicted in (inferred from hydraulic fracturing, overcoring and leak-off tests). acceptable ranges (i.e., correlation coefcients 0.9) when we Compilation and analysis of a massive amount of geologicaluse DT, NPHI, and RHOB as input data. Hence, the final model data (over 2000 borehole data and 600 2D seismic lines)can be used as a predictive model of rock mechanical properties enabled us to develop a comprehensive 3D static geologicalin the absence of core measurements in this reservoir.model for the northern Bowen Basin, which makes a framework for construction of a detailed 3D geomechanical model forGeology-geophysics integrated inversion in Mount Isa the region. The 3D geomechanical-numerical model which isSouth, Queensland.calibrated with point-wise stress data provides a continuous description of the geomechanical parameters for the NorthernMahtab Rashidifard 1 Jeremie Giraud2, Mark Jessell1, Mark Bowen Basin that can be used by different industries toLindsay3 and Vitaliy Ogarko1complement their geotechnical and geomechanical analysis.1 University of Western Australia 2 RING Team, GeoRessources, Universit de LorraineDrilling sensor systems for trajectory control toward a3 CSIROdistant mineral target while drilling .Aruni Rajanayake 1,2 Brett Harris1,2, Andrej Bona1, HoangTaking advantage of all available information including Nguyen1,2, Michael Carson1,2 and Fiona Best3,2 geological and geophysical datasets is important to understand the subsurface. Integration of geological and geophysical 1 Curtin Universitydatasets is commonly limited to the interpretation of area-2 MinEx CRCspecific case studies and has seldom been addressed by 3 South32 automated techniques. Considering, the geometry of rock Steering towards a mineral target while drilling would appear tounits as the unknown for both problems, one advantage of the be highly desirable. However, one significant challenge is thatautomation of integration approaches is the parametrisation conventional drill rig requires rod changes every 3 or 6 m makingof the subsurface which can be identical for both geological continuous communication along the drill string and bottommodelling and geophysical inversion. In this study, we introduce hole assembly (BHA) problematic. New coil tubing drilling cana new approach that allows us to automate geophysical accommodate direct real-time high-speed communicationscooperative inversion workflow with an implicit geological from the BHA. We explore the possibilities for real-time trajectorymodelling engine. We consider the contacts between units as control based on sensor combinations set in the BHA of a coiledthe quantity linking geology and geophysics and generate a tubing rig, to delineate distant mineral targets. Parameters suchmodel from the geometry of the rock unit boundaries.as mineralisation style and potential seismic/EM sensor systemWe use a generalised level-set inversion algorithm integrating are simulated to assess viability and benefits for trajectory control.the implicit geological modelling engine called LoopStructural, For example, modelling the seismic response with the signal fromwhich is used to define a regularisation constraint term for continuous percussion drilling in the subsurface demonstrates ageophysical inversion. The geological plausibility of the inverted significant potential advantage in delineating mineral targets. Thismodel is continually evaluated and ensured during inversion along with the integration of existing surface geophysics data setsbased on correction factors from LoopStructural. Constraints presents a potential new tool for mineral discovery and mining. from other geophysical datasets or existing knowledge are also encapsulated in weighting matrices through constraining Prediction of geomechanical properties fromterms. The introduced methodology has been tested in the geophysical logs using machine learning approach. Boulia region (Southern Mount Isa, Queensland), using several 2D seismic transects and high-resolution gravity datasets. This Rasoul Ranjbarkarami 1, Parisa Tavoosi2 and Mojtaba Rajabi1 study focuses on recovering the geometry of major structures of the basement units on a regional scale. The utilisation of seismic 1 School of Earth and Environmental Sciences, University ofprofiles and gravity datasets with geological modelling allows Queensland, Saint Lucia, Queensland, Australiaus to generate a 3D model of the area containing distinct rock 2 School of Geology, University of Tehran, Tehran, Iran units and boundary geometries that automatically produce Rock mechanical parameters including Youngs modulus,geologically realistic gravity and seismic responses. The legacy Poissons ratio, shear modulus, and bulk modulus aredatasets and existing results from the previous modelling have 133 PREVIEW FEBRUARY 2023'