AI MONITORING OF GEOLOGICAL ACTIVITY IN ENVIRONMENT AND CLIMATE CHANGE

Journal of science. Lyon №72/2025 3 EARTH SCIENCES AI MONITORING OF GEOLOGICAL ACTIVITY IN ENVIRONMENT AND CLIMATE CHANGE Bayramov A. Kerimova R. Ibragimova L. Tagiyeva M. Republican Seismic Survey Center of the Azerbaijan National Academy of Sciences, Baku, Azerbaijan 25, N. Rafibeyli str., AZ1001 https://doi.org/10.5281/zenodo.17788322 Abstract Monitoring geophysical processes is important for studying and preventing the climate change phenomenon. With the transition to big data processing methods, automated measurements and algorithmic analysis, this industry in Azerbaijan has entered a new stage - the introduction of Artificial Intelligence systems. This is especially noticeable in projects related to the study of Caspian Sea resources, where AI plays a key role in modeling underground structures, analyzing seismic signals and predicting potentially promising areas. Thus, the evolution of technology has not only increased the accuracy of scientific forecasts, but also promotes environmentally responsible use of marine resources. At the present stage, there is an active use of machine learning and artificial intelligence technologies. For example, neural network algorithms are used to recognize geological anomalies, predict promising drilling sites, and assess risks. This is especially relevant when working with data from the Caspian shelf zone, which is characterized by complex geology. Keywords: monitoring, climate change, geophysical processes, Artificial Intelligence. Introduction. In the context of the transition to a green economy and digitalization, the oil and gas industry is faced with the need to rethink approaches to geological monitoring, especially in offshore regions where key resource extraction and transportation facilities are concentrated [10,4]. Digital monitoring of geological activity is not only a way to improve safety, but also an important element of sustainable natural risk management. Marine geophysics has undergone significant changes in recent decades due to the rapid development of digital technologies. With the transition to big data processing methods, automated measurements and algorithmic analysis, this industry in Azerbaijan has entered a new stage - the use of Artificial Intelligence (AI) systems [2,9]. This is especially noticeable in projects related to the study of Caspian Sea resources, where AI plays a key role in modeling underground structures, analyzing seismic signals and predicting potentially promising areas [1,5]. Thus, the evolution of technology has not only increased the accuracy of scientific forecasts, but also promotes environmentally responsible use of marine resources. At the present stage, there is an active use of machine learning and AI technologies. For example, neural network algorithms are used to recognize geological anomalies, predict promising drilling sites, and assess risks. This is especially relevant when working with data from the Caspian shelf zone, which is characterized by complex geology. Most of the oil fields in the Republic of Azerbaijan cover hundreds of square kilometers and some of them have hard-torecover reserves. Due to these factors and a number of other problems related to ensuring technological safety, there are difficulties in developing oil fields. As a result, oil companies need, more than ever, a high-reliability, digital infrastructure that delivers increased productivity and safety and controls the many oil and gas devices used throughout the field. AI offers a simple, structured approach to developing complex decision-making systems that allows the user to pose and solve problems of varying degrees of complexity. The application of AI in the oil and gas industry is rapidly developing and is gradually being introduced into various fields, such as: intelligent drilling, intelligent pipeline, intelligent refinery, etc. AI monitoring of geological activity. One of the most common ways of implementing AI is neural networks, which are a mathematical model, as well as its software and hardware. Machine Learning is used to identify geophysical anomalies, such as in magnetic or gravity data, that may indicate oil and gas deposits. This is especially effective when studying offshore areas with limited direct accessibility. Using AI in building geological models based on multicomponent data (seismic, acoustics, geochemistry) allows for more accurate prediction of the bottom structure, its stability and geological activity. Combining AI with historical and real-world data allows us to model drilling scenarios and select the best sites with minimal environmental risks. Digital monitoring of geological activity is a set of methods for collecting, processing and visualizing geophysical data using digital technologies [3,6,7,8]. It includes: • Seismological networks: seafloor seismographs, broadband sensors; • Fiber optic cables: distributed sensors for deformation recording; • Construction of digital models of the seafloor with superimposed seismic data; 4 Journal of science. Lyon №72/2025 • Visualization of subduction zones, epicenters and wave propagation; • Satellite systems: GNSS, InSAR for tracking crustal displacements; • Digital platforms: SeisComP, Antelope, ArcGIS for analysis and visualization; • Real-time data transmission via satellite channels; • Seismic wave modeling in ArcGIS; • Earthquake scenario animation. Digital monitoring of geological activity in the marine environment is not just a technological trend, but a necessary step towards increasing the resilience of coastal regions and the safety of marine infrastructure. The integration of seismological data, GIS models and visualization tools opens new horizons for scientific analysis and public understanding of geodynamic processes. The use of artificial intelligence, machine learning, data analytics, cloud computing and robotic automation has paved the way for the digital transformation of the oil and gas industry. Application of AI to analyze the impact of the Earth's magnetic field on seismicity in geology processes. Predicting the time, location and magnitude of an earthquake is a challenging job as an earthquake does not show specific patterns resulting in inaccurate predictions. Techniques based on Artificial Intelligence (AI) are well known for their capability to find hidden patterns in data. In the case of earthquake prediction, these models also produce a promising outcome. This work systematically explores the contributions made to date in earthquake prediction using AI-based techniques. These studies include a range of AI techniques including rule-based methods, shallow machine learning and deep learning algorithms. Covering all existing AI-based techniques in earthquake prediction, this article provides an account of the available methodologies and a comparative analysis of their performances. Using comparative analysis of performances, the paper aims to facilitate the selection of appropriate techniques for analyze the impact of the Earth's magnetic field on seismicity and earthquake prediction. The problems concerning the connection between the seismic situation in one or another region of the Earth and geomagnetic disturbances are among the most pressing in the field of interaction of geophysical fields and the physics of earthquake precursors. Mechanical effects at the source of an earthquake, variations in the electrical properties of the medium, excitation of powerful seismic signals and other factors influence the regimes of geophysical fields, including the geomagnetic field. Despite the large number of scientific publications, there is still no generally accepted consensus on solving this problem. Therefore, there is a need for additional research in this direction. At the Republican Center for Seismological Service at the National Academy of Sciences of Azerbaijan, with the support of the Science Development Fund, among others, research is being conducted on the influence of geomagnetic disturbances of the Earth on the intensity of earthquakes in the territory of the Azerbaijan Republic. For the period from January to December 2024, measurements were taken of the intensity of earthquakes with a magnitude ML in the range of 3.0 - 5.3 and the intensity of a geomagnetic storm of variations in the Earth's magnetic field with an impact index G in the range of 2 - 5. The analysis of the first research results shows that, as a rule, 5–8 days before magnetic storms with G = 3–4, earthquake foreshocks occur at depths from 50 km to 70 km, and after magnetic storms, aftershocks occur at shallower depths from 13 km to 20 km. The authors put forward two probable reasons for these phenomena: 1) geomagnetic storms, interacting with powerful flows of electric fields in the Earth’s crust, cause aftershocks and shifts of tectonic plates; 2) geomagnetic storms cause high temperature gradients in the Earth's interior at the edges of tectonic plates: 𝑔𝑟𝑎𝑑𝑇=∇𝑇=𝜕𝑇 𝜕𝑥𝑒𝑥+𝜕𝑇 𝜕𝑦𝑒𝑦+𝜕𝑇 𝜕𝑧𝑒𝑧 This factor may be associated with the Ettingshausen effect: the effect of the appearance of a temperature gradient in a conductor located in a magnetic field through which an electric current flow. At the same time, if the current flows along the X axis, the magnetic field is directed along the Y axis, then a temperature gradient will appear along the Z axis. The temperature gradient, in turn, creates tension forces between structures (tectonic plates). As a result, aftershock occur with a delay of several days. Note that this is only one of the factors that causes earthquakes. Fuzzy Sets application for joint analysis of solar and seismic events data. In recent years, research has been actively conducted on the influence of processes occurring on the Sun on seismic processes occurring in the depths of the Earth. Until now, scientists have not received an unambiguous answer to this problem. One of the effective mathematical methods for the joint analysis of these processes is Fuzzy Sets. Given the relevance of this problem, this work provides an analysis of Fuzzy Set methods and geoinformatics. Note that in this case, geoinformatics includes research on the creation of methods and algorithms that allow automating the simultaneous solution of problems in the field of Earth sciences and, in particular, solar physics based on initial observation data. The solution is understood as an adequate modeling of the logic of an expert who analyzes data and makes decisions. It is the observation systems and the recorded data on the processes occurring in the depths of the Earth and on the Sun that are the basis for fundamental joint research in the field of solar physics and seismology. This raises the problem of effective analysis of large arrays of observational data. The complexity of this problem lies in the joint processing of these volumetric data on processes occurring on the Sun and seismic processes occurring in a particular region of the Earth. For the effective use and processing of large arrays of observation data, and to obtain qualitatively new results on this basis, it is proposed to use adequate automated methods of complex analysis and data Journal of science. Lyon №72/2025 5 processing, in particular Fuzzy Logic methods. In many cases, information about the disturbances being sought is very limited and concerns only general ideas about their shape. The shape of the anomaly is a rather vague concept, and its correlation properties are unknown. Since the nature of the phenomena reflected in the recorded data is a priori unknown and changes over time, the methods must be highly adaptive. Any classical subset 𝐴  𝑈 can be considered as a Fuzzy Set on U with membership 𝜇𝐴(𝑥)={1,если 𝑥 𝜖 𝐴 0,если 𝑥  𝐴 function. Let supposed U is a finite or integer set. Then, 𝐴={𝐴1 𝑥1,𝐴2 𝑥2,…,𝐴𝑛 𝑥𝑛,…}. where, 𝐴𝑖=0, or 𝐴𝑖 is a Fuzzy Set of the first type and is defined by the formula 𝐴𝑖={𝜇𝐴𝑖(𝑦1) 𝑦1,…,𝜇𝐴𝑖(𝑦𝑘) 𝑦𝑘,…}, here, 𝑦𝑖∈[0,1] и 𝜇𝐴𝑖(𝑦𝑗) 𝜖 [0.1]. If the universal set U is the set of real numbers R, then the following formula is used 𝐴= ∫ (𝜇𝐴(𝑥) 𝑥)𝑑𝑥 ⬚ 𝑈 These basic formulas can be used to construct a model based on Fuzzy Sets for the joint analysis of processes occurring on the Sun and in the bowels of the Earth. References: 1. Bahaloo S., Mehrizadeh M., NajafiMarghmaleki A. Review of application of artificial intelligence techniques in petroleum Operations. Petroleum Research. Vol.8, 167-182, 2023/ 2. Bayramov A.A., Abdullayev F.N., Suleymanov S.S., Karimova R.D., Safarov H.N., Rzayev E.A. Multifactor model for seismological research Seismoprognosis observations in the territory of Azerbaijan. Vol.26, № 2, 2024. pp.28-33. https://doi.org/10.59849/2219-6641.2024.2.28 3. Beroza GC, Segou M, Mousavi SM. 2021. Machine learning and earthquake forecasting—next steps. Nat.Commun. 12:4761 4. Etirmishli G.Dj, Valiyev G.O., Kaz;mova S.E., Ismailova S.S., Kazimov I.E. Residual oil production technologies. Geology and Geophysics of the South of Russia. 2019. Vol. 9 № 1. pp. 84-96. DOI: 10.23671/VNC.2019.1.26790. 5. Honore T., Vorobev A.E., Vorobyev K.A. Digitalization of the oil industry: basic approaches and rationale for "intelligent" technologies. The Eurasian Scientific Journal, 2018, Vol. 10, No 2, pp.1-7. 6. Gitis, V.G.; Derendyaev, A.B. Machine Learning Methods for Seismic Hazards Forecast. Geosciences 2019, 9, 308. 7. Lähivaara, A. et al., Estimation of groundwater storage from seismic data using deep learning. Geophys. Prospect.67, 2115–2126 (2019). doi: 10.1111/1365-2478.12831 8. Pawley S, Schultz R, Playter T, Corlett H, Shipman T, et al. 2018. The geological susceptibility of induced earthquakes in the Duvernay play. Geophys. Res. Lett. 45:1786–93 9. Suleymanov S.S., Bayramov A.A., Abdullayev F.N. Study of seismic processes using Aİ technology Current directions of development of information and communication technologies and control tools. Proceedings of 15-th International Scientific and Technical Conference. Xarkov, April 24 – 25, 2025, Volume 1: sections 1, 5. P.47. doi: https://doi.org/10.32620/ICT.25.t1 10. Veliyev E.F., Shirinov S.V., Mammedbeyli T.E. Intelligent oil and gas field based on artificial intelligence technology. SOCAR Proceedings. No.4 (2022) 070-075.