Characterization of Ground Conditions at Seismic Stations in the North Caucasus Using Machine Learning Methods

Cover Page

Cite item

Full Text

Open Access Open Access
Restricted Access Access granted
Restricted Access Subscription or Fee Access

Abstract

To extend the capabilities of using records of local earthquakes (for constructing regional ground motion prediction equations, assessing seismic hazard, etc.), the classification of seismic stations in the North Caucasus by the ground conditions was performed. A technique has been developed that allows assessment of ground conditions by comparing spectra of weak earthquakes selected in narrow ranges of magnitudes and hypocentral distances, at different stations. The use of machine-learning methods showed the complexity of the problem, but at the same time, the application of logical operations and techniques allowed us to find the most effective approaches to solve it. As a result, 70 seismic stations of the North Caucasus were classified according to the ground conditions; the conditions were characterized by one dimensionless parameter based on the calculation of spectral characteristics. We are planning to refine the estimates in the future.

Full Text

Restricted Access

About the authors

T. S. Savadyan

Schmidt Institute of Physics of the Earth, Russian Academy of Sciences

Author for correspondence.
Email: olga@ifz.ru
Russian Federation, Moscow

O. V. Pavlenko

Schmidt Institute of Physics of the Earth, Russian Academy of Sciences

Email: olga@ifz.ru
Russian Federation, Moscow

References

  1. Акимов В. А. и др. Карты сейсмической опасности Северо-Западного и Центрального Кавказа в детальном масштабе // Вопросы инженерной сейсмологии. 2019. Т. 46. №. 4. С. 57–74.
  2. Векслер В. А. Машинное обучение на основе алгоритма “k-ближайших соседей”. Вызовы цифровой экономики: итоги и новые тренды. 2019. С. 110–115.
  3. Виноградова Е. П., Головин Е. Н. Метрики качества алгоритмов машинного обучения в задачах классификации. Научная сессия ГУАП. 2017. С. 202–206.
  4. Габсатарова И. П. и др. Северный Кавказ // Землетрясения Северной Евразии. 2018. №. 21 (2012). С. 79–94.
  5. Габсатарова И. П. и др. Северный Кавказ. Землетрясения России в 2020 году. Обнинск: ФИЦ ЕГС РАН. 2022. 204 с.
  6. Генрихов И. Е., Дюкова Е. В. Классификация на основе полных решающих деревьев //Журнал вычислительной математики и математической физики. 2012. Т. 52. №. 4. С. 750–761.
  7. Гусев А.А., Мельникова В.Н. Связи между магнитудами — среднемировые и для Камчатки // Вулканология и сейсмология. 1990. № 6. С. 55–63.
  8. Дьяконов И. Д., Новикова С. В. Решение задачи прогнозирования при помощи градиентного бустинга над решающими деревьями. Научный форум: технические и физико-математические науки. 2018. С. 9–12.
  9. Кузьмина С. В., Ефимов А. И. Актуальные методы машинного обучения в области классификации. Актуальные проблемы современной науки и производства. 2018. С. 34–38.
  10. Наумов В. Н., Жиряева Е. В., Падерно П. И. Анализ данных и машинное обучение. Методы и инструментальные средства. 2020.
  11. Павленко О. В. Сейсмические волны в грунтовых слоях: нелинейное поведение грунта при сильных землетрясениях последних лет. Науч. мир. 2009.
  12. Пруцкий Н. И. и др. Геология и минерагения Северного Кавказа-современное состояние (Геологический атлас Северного Кавказа м-ба 1: 1 000 000) // Региональная геология и металлогения. 2005. №. 25. С. 27–38.
  13. Рогожин Е.А. Сейсмическая опасность на Северном Кавказе // Экологический Вестник научных центров ЧЭС. 2012. № 1. С. 124–128.
  14. Boore D.M. Simulation of Ground Motion Using the Stochastic Method // Pure Appl. Geophys. 2003. V. 160. P. 635–676.
  15. Brink H., Richards J., Fetherolf M. Real-world machine learning. Simon and Schuster. 2016.
  16. Oppenheim A. V. Discrete-time signal processing. Pearson Education India. 1999.

Supplementary files

Supplementary Files
Action
1. JATS XML
2. Fig. 1. Location of seismic stations in the North Caucasus [Gabsatarova I. P. et al.].

Download (825KB)
3. Fig. 2. Map of earthquake epicenters in the North Caucasus from 2001 to 2020 (from [Gabsatarova et al., 2022]).

Download (944KB)
4. Fig. 3. An example of recording an earthquake cycle diagram on the eastern component (SHE) of the Sunzha station (SNJ in Fig. 1). On the vertical axis — the amplitude in relative units, on the horizontal axis — the time in seconds.

Download (241KB)
5. Fig. 4. An example of the spectrum of a bicycle diagram and its envelope. The vertical axis shows the amplitude in relative units, and the horizontal axis shows the frequency in hertz.

Download (319KB)
6. Fig. 5. The spectrum of the local earthquake cycle recorded by the Stavdurt station installed on rocky ground. The vertical axis shows the amplitude in relative units, and the horizontal axis shows the frequency in hertz.

Download (220KB)
7. 6. Examples of cycling spectra of a group of earthquakes with magnitudes 4.2–4.4 and hypocentral distances of 100-125 km. The vertical axis shows the amplitude in relative units, and the horizontal axis shows the frequency in hertz. The spectra for Stavdurt, Gofitskoye, Khunzakh, Krasnodar and Urkarakh stations are shown in blue, green, red, yellow and lilac colors, respectively.

Download (385KB)
8. Fig. 7. The spectrum of the Stavdurt station and its envelope. The red circles mark the calculation points of the spectral parameters.

Download (244KB)
9. 8. The scattering diagram of objects with the signs “Deviation” and “Amplitude at 5 Hz” for 5 classes.

Download (221KB)
10. 9. The scattering diagram of objects with the signs “Maximum frequency” and “Amplitude at 2 Hz” for 5 classes.

Download (263KB)
11. 10. The scattering diagram of objects with the signs “Amplitude at 2 Hz” and “Amplitude at 14 Hz” for 5 classes.

Download (223KB)
12. 11. The scattering diagram of objects with the signs “Amplitude at 5 Hz” and “Amplitude at 2 Hz” for 5 classes.

Download (334KB)
13. Fig. 12. The logic of the algorithm of decision trees. Here x, y are the signs of the objects.

Download (266KB)
14. Fig. 13. The logic of the gradient boosting algorithm. As the number of iterations increases, the number of decision trees increases.

Download (186KB)

Copyright (c) 2025 Russian Academy of Sciences