Fractional Model of the Deformation Process
Abstract
:1. Introduction
2. Mathematical Model of the Deformation Process
3. Statistical Model of the Foreshock Mode
3.1. Algorithm for Constructing Sequences of Foreshocks
- (1)
- The time interval between the mainshock and the preceding event does not exceed the time scale : ;
- (2)
- The distance between the mainshock and the preceding event does not exceed the spatial scale : , where r is the radius-vector before the event;
- (3)
- The energy class of the preceding event is less than the class K of the mainshock.
3.2. Method for Constructing an Empirical Cumulative Foreshock Waiting Time Distribution Function for the Mainshock with a Given Energy
3.3. Processing the Data of the Catalog
4. Results and Discussion
5. Conclusions
Supplementary Materials
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Mogi, K. Active periods in the world’s shieft seismic belts. Tectonophysics 1974, 22, 265–282. [Google Scholar] [CrossRef]
- Kagan, Y.; Knopoff, L. Earthquake risk prediction as a stochastic process. Phys. Earth Planet. Inter. 1977, 14, 97–108. [Google Scholar] [CrossRef]
- Bak, P.; Christensen, K.; Danon, L.; Scanlon, T. Unified scaling law for earthquakes. Phys. Rev. Lett. 2002, 88, 178501-1–178501-4. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Keilis-Borok, V.I.; Soloviev, A.A. Nonlinear Dynamics of the Lithosphere and Earthquake Prediction; Springer: Berlin/Heidelberg, Germany, 2003; p. 337. [Google Scholar]
- Zaliapin, I.; Gabrielov, A.; Keilis-Borok, V.; Wong, H. Clustering Analysis of Seismicity and Aftershock Identification. Phys. Rev. Lett. 2008, 101, 018501. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Pisarenko, V.F.; Rodkin, M.V. Declustering of Seismicity Flow: Statistical Analysis. Izv. Phys. Solid Earth 2019, 55, 733–745. [Google Scholar] [CrossRef]
- Zaliapin, I.; Ben-Zion, Y. Earthquake declustering using the nearest-neighbor approach in space-time-magnitude domain. J. Geophys. Res. Solid Earth 2020, 125, 1–33. [Google Scholar] [CrossRef]
- Manna, S.S. Two-state model of self-organized criticality. J. Phys. A Math. Gen. 1991, 24, L363–L369. [Google Scholar] [CrossRef]
- Shebalin, P.N. Increased correlation range of seismicity before large events manifested by earthquake chains. Tectonophysics 2006, 424, 335–349. [Google Scholar] [CrossRef]
- Shebalin, P.; Narteau, C. Depth Dependent Stress Revealed by Aftershocks. Nat. Commun. 2017, 8, 1317–1318. [Google Scholar] [CrossRef] [Green Version]
- Shebalin, P.N.; Narteau, C.; Baranov, S.V. Earthquake Productivity Law. Geophys. J. Int. 2020, 222, 1264–1269. [Google Scholar] [CrossRef]
- Shevtsov, B.M.; Sagitova, R.N. A diffusion approach to the statistical analysis of Kamchatka seismicity. J. Volcanol. Seismol. 2012, 6, 116–125. [Google Scholar] [CrossRef]
- Shevtsov, B.M.; Sagitova, R.N. Statistical analysis of seismic processes on the basis of the diffusion approach. Dokl. Earth Sci. 2009, 426, 642–644. [Google Scholar] [CrossRef]
- Baiesi, M.; Paczuski, M. Complex networks of earthquakes and aftershocks. Nonlinear Process. Geophys. 2005, 12, 1–11. [Google Scholar] [CrossRef]
- Davy, P.; Sornette, A.; Sornette, D. Some consequences of a proposed fractal nature of continental faulting. Nature 1990, 348, 56–58. [Google Scholar] [CrossRef]
- Kagan, Y.Y.; Knopoff, L. Spatial distribution of earthquakes: The two-point correlation function. Geophys. J. R. Astr. Soc. 1980, 62, 303–320. [Google Scholar] [CrossRef] [Green Version]
- Saichev, A.I.; Zaslavsky, G.M. Fractional kinetic equations: Solutions and applications. Chaos 1997, 7, 753–764. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Carbone, V.; Sorriso-Valvo, L.; Harabaglia, P.; Guerra, I. Unified scaling law for waiting times between seismic events. Europhys. Lett. 2005, 6, 1036–1042. [Google Scholar] [CrossRef]
- Kagan, Y.Y. Observational evidence for earthquakes as nonlinear dynamic process. Phys. D 1994, 77, 160–192. [Google Scholar] [CrossRef]
- Metzler, R.; Klafter, J. The random walk’s guide to anomalous diffusion: A fractional dynamics approach. Phys. Rep. 2000, 339, 1–77. [Google Scholar] [CrossRef]
- Turcotte, D. Fractals and Chaos in Geology and Geophysics, 2nd ed.; Cambridge University Press: London, UK, 1997; p. 416. [Google Scholar]
- Shevtsov, B.; Sheremetyeva, O. Fractional models of seismoacoustic and electromagnetic activity. E3S Web Conf. Sol. 2017, 20, 02013. [Google Scholar] [CrossRef] [Green Version]
- The Geophysical Service of the Russian Academy of Sciences. Available online: http://www.gsras.ru/new/eng/catalog/ (accessed on 27 February 2022).
- Fedotov, S.A. Regularities of the distribution of strong earthquakes in Kamchatka, the Kurile Islands, and northeastern Japan. Tr. Inst. Phys. Earth. Acad. Sci. USSR 1965, 36, 66–93. [Google Scholar]
- Laskin, N. Fractional Poisson processes. Commun. Nonlinear Sci. Numer. Simul. 2003, 8, 201–213. [Google Scholar] [CrossRef]
- Uchaikin, V.V. Self-similar anomalous diffusion and stable laws. Phys. Uspekhi 2003, 46, 821–849. [Google Scholar] [CrossRef]
- Dobrovolsky, I.R.; Zubkov, S.I.; Myachkin, V.I. Estimation of the size of earthquake preparation zones. Pure Appl. Geophys. 1979, 117, 1025–1044. [Google Scholar] [CrossRef]
- Davis, J.C. Statistics and Data Analysis in Geology, 2nd ed.; John Wiley and Sons. Inc.: New York, NY, USA, 1986; p. 267. [Google Scholar]
- Broccardo, M.; Mignan, A.; Wiemer, S.; Stojadinovic, B.; Giardini, D. Hierarchical Bayesian Modeling of Fluid-Induced Seismicity. Geophys. Res. Lett. 2017, 44, 11–357. [Google Scholar] [CrossRef] [Green Version]
- Khajehdehi, O.; Eaton, D.W.; Davidsen, J. Spatiotemporal Clustering of Seismicity in the Kiskatinaw Seismic Monitoring and Mitigation Area. Front. Earth Sci. 2022, 10, 894549. [Google Scholar] [CrossRef]
- Vorobieva, I.A.; Gvishiani, A.D.; Dzeboev, B.A.; Dzeranov, B.V.; Barykina, Y.V.; Antipova, A.O. Nearest Neighbor Method for Discriminating Aftershocks and Duplicates When Merging Earthquake Catalogs. Front. Earth Sci. 2022, 10, 820277. [Google Scholar] [CrossRef]
Mainshock Class, K | Sample Size of Mainshocks | Sample Size of All Foreshocks | Foreshock Class, | Sample Size of Foreshocks |
---|---|---|---|---|
12.0 | 116 | 963 | 8.5 | 88 |
8.9 | 71 | |||
12.3 | 95 | 1245 | 8.5 | 107 |
9.1 | 75 | |||
12.7 | 63 | 2390 | 8.6 | 174 |
9.0 | 140 | |||
12.9 | 62 | 3875 | 8.5 | 322 |
8.8 | 262 | |||
9.0 | 236 | |||
9.8 | 137 |
K | Approximating Function | Approximation Error | Stream Density , | ||||
---|---|---|---|---|---|---|---|
12.0 | 8.5 | 0.113 | 12.56 | 0.055 | |||
0.031 | 7.33 | 0.074 | 0.81 | 0.85 | |||
12.0 | 8.9 | 0.183 | 17.27 | 0.046 | |||
0.011 | 7.14 | 0.061 | 1 | 0.61 | |||
12.3 | 8.5 | 0.064 | 11.1 | 0.042 | |||
0.006 | 5.03 | 0.05 | 0.96 | 0.8 | |||
12.3 | 9.1 | 0.033 | 9.97 | 0.039 | |||
0.003 | 3.55 | 0.044 | 0.97 | 0.84 | |||
12.7 | 8.6 | 0.057 | 7.27 | 0.0196 | |||
0.033 | 6.84 | 0.021 | 1 | 0.94 | |||
12.7 | 9.0 | 0.069 | 9.35 | 0.02 | |||
0.01 | 6.58 | 0.023 | 0.95 | 0.85 | |||
12.9 | 8.5 | 0.023 | 3.9 | 0.017 | |||
0.004 | 3.6 | 0.018 | 0.99 | 0.93 | |||
12.9 | 8.8 | 0.013 | 2.94 | 0.018 | |||
0.008 | 3.91 | 0.019 | 1 | 0.95 | |||
12.9 | 9.0 | 0.022 | 6.54 | 0.017 | |||
0.022 | 6.54 | 0.017 | 1 | 1 | |||
12.9 | 9.8 | 0.054 | 9.23 | 0.017 | |||
0.006 | 2.72 | 0.019 | 0.99 | 0.87 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Sheremetyeva, O.; Shevtsov, B. Fractional Model of the Deformation Process. Fractal Fract. 2022, 6, 372. https://doi.org/10.3390/fractalfract6070372
Sheremetyeva O, Shevtsov B. Fractional Model of the Deformation Process. Fractal and Fractional. 2022; 6(7):372. https://doi.org/10.3390/fractalfract6070372
Chicago/Turabian StyleSheremetyeva, Olga, and Boris Shevtsov. 2022. "Fractional Model of the Deformation Process" Fractal and Fractional 6, no. 7: 372. https://doi.org/10.3390/fractalfract6070372
APA StyleSheremetyeva, O., & Shevtsov, B. (2022). Fractional Model of the Deformation Process. Fractal and Fractional, 6(7), 372. https://doi.org/10.3390/fractalfract6070372