Vol.57 No.6, December 31, 2024 Article TypeResearch PaperReviewSpecial Research Paper on “Applications of Data Science and Artificial Intelligence in Economic and Environmental Geology”

• Special Research Paper on “Applications of Data Science and Artificial Intelligence in Economic and Environmental Geology” 2024-12-31

  0 105 15

  Full Waveform Inversion Using the Hypergradient Descent Method

  Jun Hyeon Jo, Wansoo Ha

  Econ. Environ. Geol. 2024; 57(6): 665-680

  Abstract

  Full Text Cite Share

  Abstract : Optimizing step length or learning rate is crucial for efficient gradient-based inversions, including seismic full waveform inversions and deep learning. Hypergradient descent methods, initially proposed for deep learning, update hyperparameters using gradient descent techniques. We applied the hypergradient descent method to update the step length in full waveform inversion. While this approach still requires selecting an appropriate learning rate for hypergradient descent, it eliminates the need to manually tune and schedule the step length in full waveform inversion. We implemented the hypergradient descent method with the Adam optimizer to invert seismic data and compared the results to those obtained using a line search method. Numerical examples demonstrated that the hypergradient descent method accelerated full waveform inversion and produced results comparable to those from the conventional line search method.

• Special Research Paper on “Applications of Data Science and Artificial Intelligence in Economic and Environmental Geology” 2024-12-31

  0 89 25

  Application of Machine-Learning for Detecting Gas Indicator Distribution from Seismic Data

  Jongpil Won, Hyunggu Jun

  Econ. Environ. Geol. 2024; 57(6): 681-699

  Abstract

  Full Text Cite Share

  Abstract : The bright spot and seismic chimney, which are prominent gas indicators observed in seismic data, exhibit distinct geophysical features such as high amplitude, phase reversal, low continuity, and frequency attenuation. As a result, seismic attribute analysis has been widely applied to derive gas distributions within subsurface media. However, seismic attribute analysis has limitation that it is difficult to clearly distinguish gas indicators from other strata with similar geophysical properties. Therefore, this study proposes a machine-learning method to predict the distribution of bright spot and seismic chimney within seismic data. To effectively predict gas indicators in complex seismic data, the study constructed training data by simultaneously using noise-reduced seismic data and various seismic attribute analysis results. The proposed method was applied to 3D seismic survey data acquired from the F3-block in the North Sea, Netherlands to verify the effectiveness of the proposed method. The gas indicator distribution predicted by the trained model demonstrated higher accuracy and consistency compared to traditional multi-seismic attribute analysis results. Additionally, sensitivity analysis and a forward selection method were applied to optimize the selection of input data, confirming that the prediction accuracy was improved when input with low sensitivity was removed.

• Special Research Paper on “Applications of Data Science and Artificial Intelligence in Economic and Environmental Geology” 2024-12-31

  0 90 24

  Application of a Deep Learning Phase Picker to Improve the Performance of Deep Borehole Seismic Data Analysis

  Yoontaek Hong, Dong-Hoon Sheen

  Econ. Environ. Geol. 2024; 57(6): 701-708

  Abstract

  Full Text Cite Share

  Abstract : Deep learning has demonstrated significant advancements in various fields that previously required human intervention. In particular, artificial intelligence (AI) has shown superior efficiency in areas where it is difficult to define clear rules or theories, outperforming traditional methods. However, these technologies have the drawback of requiring huge amounts of training data. The field of seismology is particularly well-suited for the development and application of deep neural networks due to the vast amount of digital seismic data accumulated alongside the analysis of skilled professionals. In this study, a deep learning method was applied to seismograms recorded at deep borehole seismometers installed for microseismic monitoring, and potential future applications of AI technology were explored. The deep borehole seismometers are installed at depths of 500 m and 1000 m, recording ground motions at a sampling rate of 1,000 sps. Since a deep neural network-based seismic phase picker has been trained on seismic data sampled at 100 sps, the deep borehole seismic data were resampled for phase picking. While 94 % of P-waves and 93 % of S-waves picked by visual inspection could be detected, the technique also detected many more phases that were not identified in the process of visual inspection, and approximately 131 % more earthquakes were detected compared to manual analysis. This indicates that while the existing deep learning-based seismic phase picker performs very well, there is still room for improvement. Furthermore, if deep neural networks trained using deep borehole seismic data to eliminate the preprocessing steps required for resampling, it is expected that earthquakes could be analyzed more rapidly and precisely.

• Review 2024-12-31

  0 87 26

  Geological Characteristics, Ore-forming Processes, and Mineral Resources of Rare Earth Element Deposits in Mongolia

  Seungyeol Lee, Sunjin Lee

  Econ. Environ. Geol. 2024; 57(6): 709-720

  Abstract

  Full Text Cite Share

  Abstract : This study investigates the geological characteristics, ore-forming processes, and mineral resources of rare earth element (REE) deposits in Mongolia. REE deposits in Mongolia are primarily associated with two distinct geological settings: Mesozoic carbonatites and Paleozoic peralkaline granitic rocks. Carbonatite-related deposits are found within alkaline silicate-carbonatite complexes in the Gobi-Tien Shan rift zone of southern Mongolia. These complexes consist of nepheline syenite and their volcanic equivalents. In these carbonatites, REE are mainly associated with bastnäsite, synchysite, parisite, and apatite. REE mineralization linked to peralkaline granitoid rocks is mainly located in the Mongolian Altai region in the northwest. Here, REE is found in minerals like elpidite, monazite, xenotime, and various fluorocarbonates. These minerals typically form during the final stages of magma differentiation. Mongolia hosts several promising rare earth deposits, with three showing significant economic potential: the Devonian Khalzan Buregtei peralkaline granites, and the Mesozoic Mushgai Khudag and Khotgor carbonatites. Many other promising locations have yet to be fully investigated. Further geological investigations and exploration efforts could substantially expand Mongolia's known rare earth element resources.

• Review 2024-12-31

  0 69 29

  Manganese Mineralizations of West Africa: Geology, Genesis and Distribution

  Hamidou GARBA SALEY, Moussa KONATÉ, Olugbenga Akindeji OKUNLOLA

  Econ. Environ. Geol. 2024; 57(6): 721-734

  Abstract

  Full Text Cite Share

  Abstract : In this review paper, we present the compilation of research work carried out on eight (8) manganese mineralizations, which represent the main manganese mineralizations of the West African Craton (WAC) and its surrounding areas. These Mn mineralizations were selected based on bibliographic availability and the importance of their grade and extend. A compilation of data from outcrop description and mineralogical analyses allowed to highlight the emplacement of mineralizations, their origin and evolution within the WAC. These analyses made possible to make comparisons between the different manganese mineralizations. Depending on the periods of emplacement, at least two contexts emerge: (i) the manganese mineralizations which are hosted in the Mesozoic dolomites, like Imini and Bouarfa (Morocco) manganese mineralizations, (ii) and the Paleoproterozoic manganese mineralizations which are located in volcano-sedimentary rocks of greenstone belts: this is the case of the manganese mineralizations of Téra (Niger), Nsuta (Ghana), Mokta (Côte d'Ivoire), Tambão-Beliata (Burkina Faso) and Ansongo (Mali). The main manganese oxides noted at Imini are pyrolusite and the hollandite group (cryptomelane, hollandite and coronadite) while pyrolusite-manganite-hausmannite are preponderant at Bouarfa manganese mineralization. Mineralogically, the Mokta and the Téra mineralizations are close and show little or no rhodochrosite in the primary manganese minerals which are composed of manganese-bearing silicates (spessartine and braunite), whereas rhodochrosite is dominant in the other Paleoproterozoic mineralizations (Nsuta, Tambão-Beliata and Ansongo).

• Research Paper 2024-12-31

  0 78 18

  Petrogenesis and Geochemical Evolution of Rocks and Pegmatites in Kwarra Area, Northcentral Nigeria: Implications for Rare Metal Mineralization

  Adamu, Lukman Musa, Sunday, Adedeji Ebenezer, Ohiemi, Adukwu Fabian, Ayuba, Rufai, Ugbena, Kelvins Godfrey, Baba, Yahaya, Abraham, Templeman, Ogunkolu, Bolade Ayodeji, Ebeh, Austine

  Econ. Environ. Geol. 2024; 57(6): 735-768

  Abstract

  Full Text Cite Share

  Abstract : Although pegmatite rare metal deposits are significant sources of rare metals, their metallogenic mechanisms remain a topic of debate. Nigeria's Basement Complex and Younger Granite provinces have identified pegmatites, which comprise a variety of rock types. We classify pegmatites into two categories: quartz-muscovite pegmatite and quartz-feldspar pegmatite. The study area has three primary stress orientations: N-S, NE-SW, and NNE-SSW, with secondary trends of ENE-WSW and E-W. There are more high-ferromagnetic elements (HFS) in quartz-muscovite pegmatites than there are in the crust as a whole. These elements include Zr, Ga, Nb, and Ta. They have a high concentration of the lithophile element Rb but are deficient in Sr and Ba. Quartz-feldspar pegmatite doesn't have many high-field-strength (HFS) elements, but it has a lot of rubidium (Rb), though not as much as quartz-muscovite pegmatites. The pegmatites and albitized granite were subjected to sodic metasomatism, which changed the feldspars from K to Na and made them contain more uranium. The albitized rock types contain low total rare earth elements (REE), strontium (Sr), and barium (Ba), while the amount of rubidium (Rb) decreases with increasing albitization. It's not clear how the columbite-tantalite series or the Ta and Cs minerals formed, but the high Sn concentration in pegmatites is linked to the formation of cassiterite minerals. Rare metal pegmatite is formed when deeply buried S-type fertile granite is partially melted. On the other hand, barren quartz-feldspar pegmatite is formed from late-stage residual melts from a less evolved granitic parent.

• Research Paper 2024-12-31

  0 81 22

  Early Permian to Late Triassic Metamorphism in the Southeastern Gyeonggi Massif (Wonju-Yeongwol-Hoengseong area): Implication for the Permian-Triassic Tectonic Evolution of the Central Part of the Korean Peninsula

  Byung Choon Lee, Chang Whan Oh

  Econ. Environ. Geol. 2024; 57(6): 769-792

  Abstract

  Full Text Cite Share

  Abstract : In this study, we conducted metamorphic P-T estimation and zircon U-Pb age dating for the Chiaksan gneiss in the Wonju-Yeongwol-Hoengseong area in the southeastern Gyeonggi Massif to examine Permian-Triassic metamorphic events occurred in the region. The results of metamorphic P-T estimation using conventional geothermobarometer and pseudosection analysis show that the Chiaksan gneiss underwent intermediate-P/T (656-696°C/8.3-8.7 Kbar) and low-P/T (756-759°C/6.3-6.9 Kbar) metamorphism during prograde and peak metamorphic stages, respectively. After the peak metamorphism, the Chiaksan gneiss experienced retrograde metamorphism (567-612°C/2.2-2.8 Kbar). By combining the results of metamorphic P-T estimation with mineral paragenesis and texture, it was confirmed that successive metamorphic events, from prograde to retrograde metamorphism, occurred along a clockwise P-T path. Additionally, it was identified that partial melting of Chiaksan gneiss occurred due to the dehydration reaction of muscovite during prograde to peak metamorphism. From zircon U-Pb age analysis, metamorphic ages corresponding to the Paleoproterozoic era (1923 Ma) and Permian-Triassic (278 Ma, 229.9 Ma) are recognized in the Chiaksan gneiss, and the peak and retrograde metamorphism of the Chiaksan gneiss were confirmed to have occurred at 278 Ma and 229 Ma, respectively. Considering the metamorphic geothermal gradient identified in this study and the results of previous studies, the regional metamorphism related to the orogenic event occurred in the southeastern Gyeonggi Massif during the Permian-Triassic period. This metamorphic event occurred in an arc-related tectonic setting that was developed in the southeastern part of the Korean Peninsula at that time.

• Research Paper 2024-12-31

  0 67 14

  Slip-sense Inversion Recorded in the Fault Rocks of the Northern Yangsan Fault, Yeonghae Area, Yeongdeok, South Korea

  Hakyung Lee, Raehee Han, Youngbeom Cheon

  Econ. Environ. Geol. 2024; 57(6): 793-801

  Abstract

  Full Text Cite Share

  Abstract : The Yangsan Fault is known to have experienced kinematic changes due to tectonic stress variations since its formation in the Late Cretaceous. However, direct kinematic evidence based on the microstructure of fault rocks is rare. In this study, we analyzed the fault rocks of the Yangsan Fault using core samples collected from Yeonghae-eup, Yeongdeok-gun, in Southeast Korea, and aimed to obtain kinematic information through observations of rock slab-scale and micro-scale structures. In the inclined core samples with a length of 73.2 meters, fault rocks appear over a width of approximately 25 meters, consisting of foliated cataclasite derived from gneiss and foliated gouge derived from sedimentary rock. Rock slab-scale observations revealed dextral shear deformation in the cataclasite and sinistral shear deformation in the fault gouge. Micro-scale observations identified a principal slip zone approximately 1 mm wide, along which three units with different foliation orientations were found, indicating a dextral-sinistral-dextral sequence of slip sense changes. These findings provide geological evidence of slip-sense reversals in the northern segment of the Yangsan Fault, alongside similar reversals observed in the Bogyeongsa area, Pohang, approximately 30 km from the study area, and the identified sequence of slip senses is crucial for establishing the kinematic history of the Yangsan Fault.

• Research Paper 2024-12-31

  0 59 20

  Optimum Conditions of Thermal Treatment for Mercury Contaminated Agricultural Soil in Abandoned Mine Areas

  Il-Ha Koh, Min-Hyeon Lee, Yo Seb Kwon, Ju In Ko, Won Hyun Ji

  Econ. Environ. Geol. 2024; 57(6): 803-810

  Abstract

  Full Text Cite Share

  Abstract : In this study, we investigated the feasibility and optimum treatment conditions (temperature and retention time) of thermal desorption for agricultural soil contaminated by mercury in an abandoned mine area. With a retention time of 2 hours, mercury concentration decreased as heating temperature, 100 ℃ to 700 ℃, increased. Eventually, a temperature of 200 ℃ with a retention time of 3 hours was suggested as the optimum condition of thermal desorption to meet the Korean environmental standard and maintain soil functions for agricultural use. 54 % of mercury existed as elemental mercury in the studied soil, and this species was dramatically decreased after the treatment. Therefore, thermal desorption with a lower temperature range and longer retention time has a feasibility for remediation of mercury contaminated soil in abandoned mine areas.

• Research Paper 2024-12-31

  0 67 21

  Geochemical Characteristics of the Taehwa River Water: Impacts of Weathering and Human Activities

  Duk-Hee Jo, Yeongkyoo Kim

  Econ. Environ. Geol. 2024; 57(6): 811-821

  Abstract

  Full Text Cite Share

  Abstract : This study systematically analyzed the geochemical characteristics of the Taehwa River water to examine the natural factors and human activities influencing changes in its major chemical composition. The Taehwa River flows through the major industrial city of Ulsan and eventually into the sea, where it is impacted by both anthropogenic pollution and natural processes such as rock weathering. In the upstream region, the weathering of igneous and sedimentary rocks plays a dominant role, resulting in high ratios of ions such as Ca, Mg, and HCO₃. Additionally, the weathering of igneous rocks in the upper reaches contributes to elevated Si ratio. However, as the river flows downstream, the ratios of Cl and SO₄ ions increase greatly due to human activities, such as the discharge of industrial effluents, domestic wastewater, and agricultural runoff. In the downstream areas, the mixing of seawater causes a significant rise in Na and Cl ratios, indicating the influence of seawater on the river water. Regarding heavy metals, their concentrations generally increase from upstream to downstream. Zn and As, in particular, show relatively high levels due to nearby industrial activities and abandoned mines, while other heavy metals exhibit irregular variations depending on the location. This study identifies the spatial distribution and origins of major ions and heavy metals in the Taehwa River, providing fundamental data for understanding how rivers respond to natural weathering and human activities.

• Research Paper 2024-12-31

  0 55 11

  The Formation of Phosphate in Archaeological Site: Material Characteristics of Settlement Deposits from the Village Before the Construction of the Southern Wall of Wolseong, Gyeongju

  Sungyoon Jang, Hong Ju Jin, Serin Park, Myeong Seong Lee

  Econ. Environ. Geol. 2024; 57(6): 823-835

  Abstract

  Full Text Cite Share

  Abstract : A consolidated grayish-white layer was identified in the excavation site of land formation for settlements located beneath the southern wall of Wolseong, Gyeongju, the royal castle of the Silla Kingdom. Mineralogical and chemical analyses revealed high concentrations of nitrate, sulfate, and chloride within this layer. Although their mineral compositions were similar to that of the surrounding soil samples, significant differences were observed in the microstructure and chemical composition. The upper samples of the consolidated layer contained herbaceous plant remnants in soil minerals, while the lower samples were predominantly composed of the rice husk, which is primarily Si-based. This consolidated layer, interposed between them, exhibited significant enrichment of Fe, P, and Ca within their microstructure, suggesting the presence of amorphous phosphate minerals. Notably, fragments of blackened wood, from the oak family, were distributed throughout the consolidated layer, with the enrichment of Fe, P, and Ca inside the pores of their microstructure. It is thought that the decayed wooden fragments and rice husks have absorbed chemicals contributing to the formation of iron phosphate. Furthermore, the identification of vivianite on the upper wall structure means that this area experienced flooding, low-oxygen conditions, supply of P and Ca, and microbial activity. These factors facilitated the formation of band-shaped consolidated layers of amorphous phosphate minerals, including Ca-Fe phosphates on the rice husk layers.