Interpretation of Magnetic Anomalies over 2D Fault and Sheet-Type Mineralized Structures Using Very Fast Simulated Annealing Global Optimization: An Understanding of Uncertainty and Geological Implications

Abstract

Identification of intraterrane dislocation zones and associated mineralized bodies is of immense importance in exploration geophysics. Understanding such structures from geophysical anomalies is challenging and cumbersome. In the present study, we present a fast and competent algorithm for interpreting magnetic anomalies from such dislocation and mineralized zones. Such dislocation and mineralized zones are well explained from 2D fault and sheet-type structures. The different parameters from 2D fault and sheet-type structures such as the intensity of magnetization (k), depth to the top (z1), depth to the bottom (z2), origin location (x0), and dip angle (θ) of the fault and sheet from magnetic anomalies are interpreted. The interpretation suggests that there is uncertainty in defining the model parameters z1 and z2 for the 2D inclined fault; k, z1, and z2 for the 2D vertical fault and finite sheet-type structure; and k and z for the infinite sheet-type structure. Here, it shows a wide range of solutions depicting an equivalent model with smaller misfits. However, the final interpreted mean model is close to the actual model with the least uncertainty. Histograms and crossplots for 2D fault and sheet-type structures also reveal the same. The present algorithm is demonstrated with four theoretical models, including the effect of noises. Furthermore, the investigation of magnetic data was also applied from three field examples from intraterrane dislocation zones (Australia), deep-seated dislocation zones (India) as a 2D fault plane, and mineralized zones (Canada) as sheet-type structures. The final estimated model parameters are in good agreement with the earlier methods applied for these field examples with a priori information wherever available in the literature. However, the present method can simultaneously interpret all model parameters without a priori information.

Fault Structure Reconstruction From Magnetic Anomalies Using an Improved Backtracking Search Optimization Algorithm

“…The length of the survey profile is 50 km, which was digitized at 1 km intervals from Radhakrishna Murthy et al (2001). Table 10 shows a brief comparison for the evaluation of this anomaly considering the results of the imBSA application here and other solutions from previous studies using different metaheuristics, including WOA (Gobashy et al, 2020) and SA (Biswas & Rao, 2021). In the comparison table presented here, the first thing to notice is that the z1 depth estimated by imBSA is relatively lower than WOA and SA, but the z2 depth is higher.…”

Inversion of Geomagnetic Anomalies Caused by Ore Masses Using Hunger Games Search Algorithm

“…However, because of the well‐known ill‐posedness and non‐uniqueness nature of the geomagnetic data inversion problem, explanation of anomaly sources, that is, model parameter estimations, necessitate some special strategies and efficient approaches (Ekinci et al., 2019). Over the recent years, instead of derivative‐based local optimizers, derivative‐free nature‐inspired global optimizers and metaheuristics such as Particle Swarm Optimization (PSO) (Essa, Abo‐Ezz, et al., 2022; Essa & Elhussein, 2020; Fernández‐Martínez et al., 2010; Pallero et al., 2015; Roy et al., 2022; Santos, 2010), Very Fast Simulated Annealing (VFSA) (Biswas, 2016; Biswas & Acharya, 2016; Biswas & Rao, 2021), Ant Colony Optimization (Liu et al., 2014, 2015; Srivastava et al., 2014); Gray Wolf Optimizer (Agarwal et al., 2018; Chandra et al., 2017), Genetic‐Price Algorithm (Di Maio et al., 2020), Cuckoo Search Algorithm (Turan‐Karaoğlan & Göktürkler, 2021), Differential Search Algorithm (Alkan & Balkaya, 2018; A. Balkaya & Kaftan, 2021; Özyalın & Sındırgı, 2023), Bat Algorithm (Essa & Diab, 2022; Gobashy et al., 2021), Differential Evolution Algorithm (Ç. Balkaya, 2013; Du et al., 2021; Ekinci, Balkaya, & Göktürkler, 2020; Ekinci et al., 2023; Göktürkler et al., 2016; Hosseinzadeh et al., 2023; Roy et al., 2021a; Sungkono, 2020); Backtracking Search Algorithm (Ekinci, Balkaya, & Göktürkler, 2021), Manta‐Ray Foraging Optimization and Social Spider Optimization (Ben et al., 2022a, 2022b, 2022c), Barnacles Mating Optimization (BMO) (Ai et al., 2022) have gained increasing attention in geophysical inversion applications. Unlike local search algorithms, these stochastic optimizers do not need a well‐designed starting point in the model space to reach the global minimum (Sen & Stoffa, 2013; Tarantola, 2005).…”

“…However, because of the well-known ill-posedness and non-uniqueness nature of the geomagnetic data inversion problem, explanation of anomaly sources, that is, model parameter estimations, necessitate some special strategies and efficient approaches (Ekinci et al, 2019). Over the recent years, instead of derivative-based local optimizers, derivative-free nature-inspired global optimizers and metaheuristics such as Particle Swarm Optimization (PSO) (Essa, Abo-Ezz, et al, 2022;Essa & Elhussein, 2020;Fernández-Martínez et al, 2010;Pallero et al, 2015;Roy et al, 2022;Santos, 2010), Very Fast Simulated Annealing (VFSA) (Biswas, 2016;Biswas & Acharya, 2016;Biswas & Rao, 2021), Ant Colony Optimization (Liu et al, 2014(Liu et al, , 2015Srivastava et al, 2014); Gray Wolf Optimizer (Agarwal et al, 2018;Chandra et al, 2017), Genetic-Price Algorithm (Di Maio et al, 2020), Cuckoo Search Algorithm (Turan-Karaoğlan & Göktürkler, 2021), Differential Search Algorithm (Alkan & Balkaya, 2018;A. Balkaya & Kaftan, 2021;Özyalın & Sındırgı, 2023), Bat Algorithm (Essa & Diab, 2022;Gobashy et al, 2021), Differential Evolution Algorithm (Ç.…”

Magnetic data interpretation for 2D dikes by the metaheuristic bat algorithm: sustainable development cases

“…However, most of these methods have several defects such as individual subjectivity, use of only a few data points along with the measurement profile, hypersensitivity to noise, and influence of adjacent effect (which might degrade the accuracy of the results). Moreover, they require initial model parameters depending upon the geological data, which the final solutions often get trapped in local minima than global minima 45 and depended on a priori knowledge, which is not always accessible 18 .…”