Open Access is an initiative that aims to make scientific research freely available to all. To date our community has made over 100 million downloads. It’s based on principles of collaboration, unobstructed discovery, and, most importantly, scientific progression. As PhD students, we found it difficult to access the research we needed, so we decided to create a new Open Access publisher that levels the playing field for scientists across the world. How? By making research easy to access, and puts the academic needs of the researchers before the business interests of publishers.
We are a community of more than 103,000 authors and editors from 3,291 institutions spanning 160 countries, including Nobel Prize winners and some of the world’s most-cited researchers. Publishing on IntechOpen allows authors to earn citations and find new collaborators, meaning more people see your work not only from your own field of study, but from other related fields too.
Earthquakes pose significant risks and challenges to human survival and societal development. Effectively assessing the imminent risk of strong earthquakes is crucial for societal and regional resilience. While the Sichuan and Yunnan regions of China are known for frequent earthquake activity, the North China region, despite historically fewer earthquakes, includes key areas such as Beijing, the capital of China, necessitating effective earthquake risk prevention. The Nowcasting method, successfully applied in the United States, Japan, and several big cities, offers a promising approach to earthquake risk assessment. This paper applies the Nowcasting method to the North China region, aiming to enhance the assessment of strong earthquake risks in this region, such as the Dezhou 5.5 earthquake and Dalian 4.6 earthquake, and investigate the effect on its performance from the aftershock events using the declustering method. In the end, we give a credible and scientific forward forecasting result after the last target earthquake in this region. Through comprehensive analysis, this study demonstrates the method’s effectiveness and emphasizes its potential for improving earthquake preparedness in regions with significant urban infrastructure but relatively lower seismic activity.
Institute of Earthquake Forecasting, China Earthquake Administration, Beijing, China
GNS Science, Avalon, Lower Hutt, New Zealand
Yongxian Zhang
Institute of Earthquake Forecasting, China Earthquake Administration, Beijing, China
*Address all correspondence to: 085012104@163.com
1. Introduction
Earthquakes are one of the most destructive natural phenomena, with the potential to cause significant loss of life and property. It is of the utmost importance to assess earthquake risks in order to enhance disaster preparedness and mitigate potential impacts. The Sichuan and Yunnan regions in China are well-known for their seismic activity, whereas the North China region has experienced fewer earthquakes historically but includes critical areas such as the capital Beijing. The 1976 Tangshan MW 7.6 earthquake serves as a stark reminder of the potential devastation in this region [1, 2]. North China is a region characterized by complex geological structures, which have significant implications for its seismic activity. The North China Craton, one of the oldest and most stable parts of the Earth’s crust, forms the core of the region. This craton has experienced multiple phases of tectonic activity, including rifting, subsidence, and uplift, which have collectively shaped its current structural configuration [3, 4]. The region is bounded by several active tectonic zones, including the Yanshan Fold and Thrust Belt to the north, the Taihang Mountains to the west, and the Bohai Bay Basin to the east. The North China Plain, lying between the Yellow River and the Bohai Sea, is a major structural feature formed by the subsidence of the craton’s eastern margin. Additionally, the TanLu fault, one of the most significant fault systems in East Asia, traverses the region and plays a crucial role in its seismicity [3, 5]. Historically, North China has experienced fewer earthquakes than regions like Sichuan and Yunnan. Nevertheless, the region is not immune to seismic hazards. The 1976 Tangshan earthquake was one of the deadliest earthquakes in the 20th century, causing extensive loss of life and property [2]. The aftershocks continue to occur up to the present time. Therefore, large earthquakes like the Tangshan earthquake highlighted the significant seismic risk in the region despite the relatively lower frequency of earthquakes.
The North China region seems seismically active in recent years due to the occurrence of many aftershock events after large earthquakes, with several fault lines capable of generating moderate to strong earthquakes. For example, the Dezhou MS5.5 earthquake occurred on August 6, 2023, and the Dalian MS4.6 earthquake occurred on August 23, 2023. The region’s seismicity is influenced by the interaction between many main fault and sub-fault systems, which creates complex stress fields and fault movements [6]. The seismic activity in North China is typified by shallow earthquakes, which have the potential to cause severe ground shaking and extensive damage to infrastructure. However, there were few precursors in the fields of geochemistry, geomagnetism, geology, and others prior to the Dezhou MS5.5 earthquake and Dalian MS4.6 earthquake, which raises questions about the prospective evaluation of this capital region once again. Figure 1 illustrates the spatial distribution of earthquakes that occurred between January 01, 1970, and January 06, 2024, with particular emphasis on the two earthquakes that occurred in the previous year.
Figure 1.
Spatial distribution of earthquake epicenters occurred from January 01, 1970, to January 06, 2024. The red star indicates the two target earthquakes of the Dezhou MS5.5 earthquake occurred on August 6, 2023, and the Dalian MS4.6 earthquake occurred on August 23, 2023. The three circles surrounding two events present the region with a radius 200 km, 300 km, and 400 km, respectively. The green lines are the active faults in this region.
The term ‘Nowcasting’ was originally coined in the field of meteorology and refers to the prediction of events in the near future, typically within a few hours to a few days. In seismology, the ‘Nowcasting’ was used to estimate the current state of seismic hazard by calculating the likelihood of future earthquakes based on the recent history of earthquake sequence [7]. In the past years, this method has been widely employed in the United States, Japan, New Zealand, Greece, and Indonesia for earthquake risk assessment, providing timely and actionable information on the assessment of strong earthquakes [8, 9, 10, 11, 12, 13, 14]. The integration of machine learning (ML) and stochastic simulated techniques with Nowcasting has also opened new avenues for enhancing prediction accuracy [15, 16]. Meanwhile, this method has been preliminary applied in regions like Sichuan and Yunnan of China [17, 18] and used to assess the potential before significant events [19]. However, its use in North China, an area with fewer but potentially more impactful earthquake events, remains constrained. Simultaneously, the awareness and concern about the area’s new approach to forecasting significant events has increased in light of the Dezhou MS5.5 earthquake and the Dalian MS4.6 earthquake, especially in the capital region.
Therefore, the purpose of this study is to apply the Nowcasting approach to North China, emphasizing its usefulness in evaluating the potential prior to these two occurrences and improving its applicability to preparedness in this crucial region. As continuing research in the study region of China Seismic Experimental Site (CSES), we also investigate the impact of the clustering feature of the seismicity on the outcome of the Nowcasting approach. This chapter highlights the significance of improved forecasting models in earthquake preparedness and mitigation strategies, in addition to adding to our understanding of the seismic risks in North China. Furthermore, it is anticipated that the global implementation of state-of-the-art forecasting techniques will be facilitated in regions with relatively low seismic activity but hold significant social importance.
In order to conduct this study, we collected an earthquake catalog of uniform fast reports from the China Earthquake Networks Center (CENC)1. The CENC is responsible for the real-time monitoring and reporting of activity, including earthquake and non-earthquake events across China, providing a comprehensive and reliable database of earthquake research. The earthquake catalog includes detailed information on the time, location, magnitude, and depth of earthquakes, which are basic and crucial for our analysis. The data collected covers a significant period from January 01, 1970, to January 06, 2024, allowing for a thorough examination of seismic patterns and trends in this area. This extensive dataset enables us to apply advanced statistical methods, such as the Nowcasting method, to assess the potential for future large earthquakes. In this catalog, the magnitude of each earthquake was measured using various scales, which is crucial for understanding the energy release and potential damage. The appropriate magnitude measurements were selected, namely ML for small to medium-sized earthquakes with a magnitude below 4.5 and MS for larger earthquakes above 4.5. Furthermore, prior to the application of statistical forecasting models, the completeness magnitude level (Mc), which can be recognized as the lowest magnitude at which earthquakes in the catalog are reliably recorded, was considered. Understanding the Mc value is crucial for ensuring the dataset is complete and for accurately analyzing earthquake frequency and trends. Figure 2 depicts the fundamental analysis of the catalog utilized in this study. From the magnitude-time and magnitude-number plots, it is evident that there is a lack of data following strong earthquakes. For instance, during a short time after the 1976 Tangshan 7.6 earthquake, the aftershocks below 4.0 are not fully recorded. But from the perspective of magnitude 2.0 ~ 4.0, choosing magnitude 3.0 as the threshold can ensure the completeness level of the whole catalog. However, in this particular location, an earthquake with a magnitude of 4.0 is critical and can cause widespread alarm. As a result, as compared to the CSES parameters setting, the target magnitude level is significantly lower. To guarantee that enough small events participated in the computation and were assigned the required magnitude, we chose 3.0 as the magnitude threshold in the following Nowcasting process.
Figure 2.
Different plots of the catalog used in the following analysis. The plots from top to bottom are magnitude-time, magnitude-events number, and magnitude-event number with a magnitude range of 2.0–4.0. The color in the bottom plot indicates the events with the same magnitude level.
Earthquakes often exhibit distinct spatiotemporal clustering characteristics, meaning that earthquakes tend to occur in groups both in space and time rather than as isolated events [20]. These clusters can be attributed to the underlying physical processes governing earthquake generation, such as stress accumulation and release along fault lines. The study of spatiotemporal clustering is crucial for understanding earthquake hazards [21, 22]. By analyzing the clustering patterns, seismologists were able to identify regions with heightened seismic risk and estimate the probability of future earthquakes. This information is of vital importance for developing effective mitigation measures, such as designing earthquake-resistant infrastructure and implementing Operational Earthquake Forecating (OEF) systems [23, 24]. In the field of statistical seismology, the process of earthquake declustering is a crucial process that aims to separate independent mainshocks from dependent aftershocks [25]. Declustering methods help in obtaining a clear pattern of the temporal and spatial distribution of seismic events, which is essential for the analysis of earthquake occurrences and for the development of forecasting models. The Gardner-Knopoff (G-K) method represents a widely used traditional approach to earthquake declustering [26]. Developed in the 1970s, this technique involves identifying aftershocks and foreshocks associated with a mainshock and removing them from the earthquake catalog. The G-K method employs a predefined temporal and spatial window around each earthquake event to determine whether subsequent earthquakes are aftershocks or independent events.
Although the North China region has historically experienced fewer seismic events compared to other parts of China such as Sichuan and Yunnan region, the occurrence of large earthquakes, such as the 1976 Tangshan MW7.6 earthquake, has resulted in a notable clustering of earthquakes in this region. To investigate the influence of these aftershocks on the analysis of the Nowcasting method, we applied the declustering technique to the North China earthquake catalog. We utilized the traditional G-K declustering method to filter out aftershocks and mainshocks in the earthquake catalog. The spatial distribution of declustering earthquake events and the aftershock events can be seen in Figure 3. The results of the declustering method helped us recognize that the aftershock events in our study region are primarily located in specific zones. These zones include the aftershock region of the Tangshan earthquake, the western part of the Liaoning region, and the western part of North China, where many clusters have existed since earlier time. The Tangshan earthquake zone, known for its historical seismic activity, remains a focal point for aftershock events. Similarly, the western parts of the Liaoning region and North China exhibit notable clusters of seismic activity, suggesting a persistent aftershock sequence that aligns with earlier seismic patterns. This spatial clustering provides valuable insights into the ongoing seismic activity and potential areas of heightened earthquake risk within these regions. As a consequence of the aforementioned recognition, the declustering process resulted in a refined catalog of independent earthquake events, which is essential for accurate subsequent analysis and modeling.
Figure 3.
Spatial distribution of the mainshocks and aftershocks distinguished using the G-K declustering method. The red and green dots indicate the mainshock events and aftershock events, respectively.
Nowcasting is a statistical technique employed to estimate the current rate of earthquake occurrence based on historical earthquake events. This method leverages the concept that small earthquakes can be transferred into the ‘natural time’ sequence through the analysis on the intervals divided by large earthquakes. Through the statistical analysis of the ‘natural time’ sequence, then the likelihood of larger earthquakes in the near future can be obtained within a given region [7, 11]. The equation here mainly used is the Gutenberg-Richter relation, which describes the relation between the frequency and the magnitude [27]. Here we give a basic explanation of the Nowcasting process:
Identify large and small events: The ‘natural time’ sequence is constructed from the earthquake catalog by dividing the time into intervals based on the occurrence of significant earthquakes. The number of small earthquakes in each interval is counted to create the natural time series. We can identify large and small earthquakes by defining the target magnitude Mλ for large earthquakes and magnitude threshold Mδ for small events. Then the large events can be used to divide the catalog into intervals.
‘Natural time’ sequence: For each interval [ti, ti + 1], count the number of smaller earthquakes with Mλ ≥ M ≥ Mδ as Ni. For each interval, this count can be expressed as a sequence of [Ni].
Calculate the Cumulative Distribution Function (CDF): The Nowcasting method for earthquake potential assessment utilizes the CDF of ‘natural time’ sequence [Ni]. By employing the scientific mathematical approach described by Bevington and Robinson [28], the probability density function (PDF) and CDF for small events within each larger cycle are determined.
Estimation of Earthquake Potential Score (EPS): The current CDF is computed based on the frequency of small events, N(t), where t is the time elapsed since the recent large event. This current CDF is then used to define the earthquake potential score (EPS) at time t, providing a quantitative measure of the likelihood of a significant seismic event occurring.
The concept of ‘natural time’ has been used in many field, such as induced earthquake [29], geoelectric research [30], entropy change before a strong earthquake [31], and the order of parameters as a precursor change [32]. In comparison to the real-time sequence, the ‘natural time’ sequence offers a unique perspective by reordering events based on their occurrence intervals rather than their absolute times [33]. This approach may reveal hidden patterns and correlations that are not apparent in conventional time series analyses [34, 35]. By converting the sequence of events into a CDF based on their ‘natural time’ intervals, researchers can gain insights into the underlying dynamics of seismic activity. This approach is typically beneficial for identifying time period with elevated risk of earthquakes and for gathering important data on the temporal clustering of earthquakes.
Figure 4 presents the EPS values derived from the Nowcasting method prior to the Dezhou 5.5 earthquake. Figure 4(a) demonstrates that before the Dezhou 5.5 earthquake, 846 small events with magnitudes between 3.0 and 5.5 occurred within a 400 km radius since the last 5.6 earthquake on November 1, 1999. The temporal distribution analysis of these historical earthquakes yielded an EPS value of 98% via the Nowcasting method. This high EPS value indicates a significant likelihood of a forthcoming major event based on the frequency and distribution of smaller earthquakes in the region. Conversely, when the entire North China region was considered in the Nowcasting analysis, the EPS value peaked, as shown in Figure 4(b). This suggests a broader area of seismic activity that increases the number of sample in the computation and will give a more precise forecast to the study region.
Figure 4.
Nowcasting plots before the occurrence of the Dezhou 5.5 earthquake based on the non-declustering and declustering catalogs. (a) EPS of Dezhou region using non-declustering catalog. (b) EPS of local region using non-declustering catalog. (c) EPS of Dezhou region using declustering catalog. (d) EPS of local region using declustering catalog. This the case of radius 400 km.
The non-declustered results reveal numerous intervals demarcated by earthquakes exceeding the target magnitude. However, the application of a declustering catalog resulted in a reduction in the interval count, yet the EPS value remained high prior to the Dezhou 5.5 earthquake considering the 400 km circle region and the whole North China region, which is depicted in Figure 4(c) and (d). Similar to the case of using a non-declustering catalog, the high EPS values before the target event indicate a strong likelihood of occurrence based on the preceding seismic activity. The elevated EPS values preceding these target events suggest that the Nowcasting method remains effective in forecasting the potential for significant earthquakes, provided that only mainshock events were considered. Although the declustering method reduces the number of intervals divided by large events, the evolution and fluctuation of the local stress level could still be expressed using the mainshock events. With regard to the target magnitude level of 5.5 and above in the North China region, the declustering method exerts only a slight influence on the Nowcasting analysis when the ‘natural time’ sequence was used as the input.
Figure 5 depicts the Nowcasting analysis results pertaining to the Dalian 4.6 earthquake. Both Figure 5(a) and (c) show outcomes similar to those observed for the Dezhou earthquake in Figure 4(a) and (c), indicating the effectiveness of the Nowcasting methodology preceding the occurrence of the Dalian event. However, when expanding the analysis to encompass the entire North China region, whether using the non-declustering or declustering catalog, the EPS values remained notably low. This was primarily influenced by the scarcity of subsequent small earthquake events following the Dezhou 5.5 earthquake. Specifically, only 6 and 1 small events were recorded prior to the target events across the entire region, respectively. This paucity of precursor events contributed significantly to the lower EPS values derived from the historical natural time sequence analysis. Table 1 lists all the EPS results before two target events with different circular regions, using both the non-declustering and declustering catalogs.
Figure 5.
Nowcasting plots before the occurrence of the Dalian 4.6 earthquake based on the non-declustering and declustering catalog. The other caption is the same as in Figure 4.
Target Events
Mλ
Radius
EPS result
non-declustering
declustering
2024-2108-06 Dezhou MS5.5
5.5
200 km
90%
86%
300 km
95%
100%
400 km
98%
100%
2024-2108-23 Dalian MS4.6
4.6
200 km
100%
99%
300 km
100%
99%
400 km
98%
94%
Table 1.
EPS before two target events with different circular regions radius R, using non-declustering and declustering catalogs.
5.2 Forward Nowcasting to North China
Based on the above analysis, the Nowcasting method shows its special performance in the assessment of the target earthquakes. Once the local database of the North China region is constructed, then we can calculate the forward computation to the next target events. Figure 6 presents the forward Nowcasting results for the entire North China region since the occurrence of the last significant event using the non-declustered catalog. For instance, Figure 6(a) indicates that since the Dezhou 5.5 earthquake, it has occurred 89 small seismic events recorded up to the present time, yielding an EPS value of 55% based on the historical ‘natural time’ sequence. At the same time, for a target magnitude of 4.6, the EPS is markedly higher at 97%, offering valuable insight into the potential for the next target earthquake. From these results, it can be inferred that for target earthquakes exceeding 5.5, the accumulation of small events following the Dezhou 5.5 earthquake was insufficient, indicating that more time is required to gather a sufficient number of small events to reach a critical state. This suggests that the earthquake activity in the region has not yet reached the necessary conditions for another large earthquake of similar or greater magnitude in the immediate future. Conversely, for target events with magnitudes above 4.6, the EPS value is sufficiently high, indicating that the system is nearing a critical threshold state at this magnitude level. This high EPS value suggests that the accumulation of small seismic events has progressed enough to indicate a higher probability of an impending earthquake around this magnitude.
Figure 6.
Forward Nowcasting to the North China region since the time of the last target event using non-declustering catalogs. (a) Potential for earthquake above 5.5 since the occurrence of the last target event (Dezhou 5.5 earthquake). (b) Potential for earthquake above 4.6 since the occurrence of last target event (Dalian 4.6 earthquake).
The implication of these findings is significant for earthquake preparedness and risk mitigation strategies in North China. It underscores the importance of continuous monitoring and the use of advanced forecasting models like Nowcasting to assess earthquake potential accurately. By understanding the accumulation patterns of small seismic events and their impact on the EPS value, people can better anticipate and prepare for potential seismic events, thereby enhancing the overall resilience of the region to earthquake hazards.
This study applies the Nowcasting approach to assess earthquake activity in North China, demonstrating its effectiveness in forecasting significant earthquakes such as the Dezhou 5.5 and Dalian 4.6 events. Prior to the Dezhou 5.5 earthquake, Nowcasting using non-declustering catalogs achieved a high EPS of 98%, indicating a strong likelihood of major seismic events following 846 small seismic events within a 400 km radius since the last 5.6 earthquake in 1999. The method maintained high EPS values even with declustering catalogs, emphasizing its reliability in forecasting earthquakes based on historical seismic patterns. Similarly, the analysis of the Dalian 4.6 earthquake confirmed the effectiveness of Nowcasting methodology in forecasting earthquake events leading up to significant occurrences. However, EPS values before the Dalian 4.6 earthquake computed for the entire North China region, whether using non-declustering or declustering catalogs, were lower due to the scarcity of small events following the Dezhou 5.5 earthquake. The forward potential assessment indicates that the current EPS value to the target earthquake above 5.5 and 4.6 is 55% and 97%, respectively, which provides a valuable and informative reference result for understanding the risk of the next significant events in this region.
In the traditional analysis of the China annual consultation meeting with regard to the prospective outlook for the subsequent one or three years, a multitude of techniques were typically used [36, 37, 38]. Compared with other approaches applied in the work, the Nowcasting method presents an indirect way to assess the stress state around the local region since the last target events and could provide us with the likelihood of the next target events in the near future. As is known, statistical forecasting models like probability models were helpful in contributing crucial insights for earthquake preparedness and mitigation strategies [39]. In addition to the current lack of significantly effective analytical methods, techniques such as the Nowcasting method, which can give probabilistic analyses and thus aid in decision-making, are particularly important. On the other hand, the reliability of forecasting models such as the Nowcasting method is intricately linked to the size and quality of the earthquake event sample. The availability of larger and more comprehensive datasets of small events is essential for improving accuracy and reducing uncertainties in forecasting future earthquakes. Moving forward, expanding earthquake sequence to include more extensive records of precursor events will provide a more robust foundation for predictive analytics. Furthermore, the refinement of models through the integration of advanced statistical techniques and machine learning algorithms can enhance the analysis of complex data, thereby improving forecasting capabilities across a range of magnitudes and geographical regions. For the benefit of the general public, the development of real-time monitoring systems with the capacity for continuous data integration and model validation is of the utmost importance to enhance early capabilities and optimize disaster preparedness strategies. These advancements are critical for strengthening societal resilience to earthquakes in North China and beyond, ultimately mitigating risks and minimizing the impact of significant earthquakes.
The earthquake sequence data was provided by the China Seismic Networks Center (CENC). This work was supported by the National Natural Science Foundation of China (42004038), Earthquake Tracking Orientation Tasks of CEA (2024020104), the Special Fund of IEFCEA (CEAIEF2022030206), and the China Scholarship Council (CSC) exchange program (202204190019). We also acknowledge the valuable feedback from peer reviewers and the editor, which significantly improved the quality of this chapter.
References
1.Sun QZ, Wu SG. Development of the Earthquake Monitoring and Prediction in China during 1966∼2006. Beijing: Seismological Press; 2007
2.Chen Y, Liu M, Wang H. Aftershocks and background seismicity in Tangshan and the rest of North China. Journal of Geophysical Research: Solid Earth. 2021;126:e2020JB021395. DOI: 10.1029/2020JB021395
3.Xia B, Thybo H, Artemieva IM. Seismic crustal structure of the North China craton and surrounding area: Synthesis and analysis. Journal of Geophysical Research: Solid Earth. 2017;122:5181-5207. DOI: 10.1002/2016JB013848
4.Zhai MG. Comparative study of geology in North China and Korean peninsula: Research advances and key issues. Acta Petrologica Sinica. 2016;32:2915-2932. DOI: 10.0000/892be8cb 68d94c55bd8072353a9c7251
5.Xu Y, Zeyen H, Hao T, Santosh M, Li Z, Huang S, et al. Lithospheric structure of the North China craton: Integrated gravity, geoid and topography data. Gondwana Research. 2016;34:315-323. DOI: 10.1016/j.gr.2015.03.010
6.Li H, Tian Y, Zhao D, Yan D. Anatomy of large earthquakes in North China. Journal of Asian Earth Sciences. 2022;237:105342. DOI: 10.1016/j.jseaes.2022.105342
7.Rundle JB, Turcotte DL, Donnellan A, Grant Ludwig L, Luginbuhl M, Gong G. Nowcasting earthquakes. Earth and Space Science. 2016;3:480-486. DOI: 10.1002/2016ea000185
8.Luginbuhl M, Rundle JB, Turcotte DL. Natural time and Nowcasting earthquakes: Are large global earthquakes temporally clustered? In: Williams CA, Peng Z, Zhang Y, Fukuyama E, Goebel T, Yoder MR, editors. Earthquakes and Multi-Hazards around the Pacific Rim. Vol. II. Cham: Springer International Publishing; 2019. pp. 137-146. DOI: 10.1007/978-3-319-92297-3_11
9.Pasari S, Neha. Nowcasting-based earthquake Hazard estimation at major cities in New Zealand. Pure and Applied Geophysics. 2022;179:1597-1612. DOI: 10.1007/s00024-022-03021-z
10.Pasari S, Simanjuntak AVH, Neha SY. Nowcasting earthquakes in Sulawesi Island, Indonesia. Geoscience Letters. 2021;8:27. DOI: 10.1186/s40562-021-00197-5
11.Rundle JB, Donnellan A, Fox G, Crutchfield JP, Granat R. Nowcasting earthquakes: Imaging the earthquake cycle in California with machine learning. Earth and Space Science. 2021;8:e2021EA001757. DOI: 10.1029/2021EA001757
12.Rundle JB, Luginbuhl M, Khapikova P, Turcotte DL, Donnellan A, McKim G. Nowcasting great global earthquake and tsunami sources. Pure and Applied Geophysics. 2020;177:359-368. DOI: 10.1007/s00024-018-2039-y
13.Pasari S. Nowcasting earthquakes in the bay of Bengal region. Pure and Applied Geophysics. 2019;176:1417-1432. DOI: 10.1007/s00024-018-2037-0
14.Pasari S, Mehta A. Nowcasting earthquakes in the northwest Himalaya and Surronuding regions. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. 2018;XLII-5:855-859. DOI: 10.5194/isprs-archives-XLII-5-855-2018
15.Rundle JB, Yazbeck J, Donnellan A, Fox G, Ludwig LG, Heflin M, et al. Optimizing earthquake Nowcasting with machine learning: The role of strain hardening in the earthquake cycle. Earth and Space Science. 2022;9:e2022EA002343. DOI: 10.1029/2022EA002343
16.Rundle JB, Baughman I, Zhang T. Nowcasting earthquakes with stochastic simulations: Information entropy of earthquake Catalogs. Earth and Space Science. 2024;11:e2023EA003367. DOI: 10.1029/2023EA003367
17.Zhang SF, Zhang YX. China seismic experimental site: Seismicity, ergodicity and Nowcasting earthquakes. In: Li Y-G, Zhang Y, Wu Z, editors. China Seismic Experimental Site: Theoretical Framework and Ongoing Practice. Singapore: Springer Nature Singapore; 2022. pp. 197-213. DOI: 10.1007/978-981-16-8607-8_10
18.Zhang SF, Zhang YX. The “natural time” method used for the potential assessment for strong earthquakes in China seismic experimental site, in Natural Hazards - New Insights, M Dr. Mohammad, Editor. 2023, IntechOpen: Rijeka. p. Ch. 6. DOI: 10.5772/intechopen.110023
19.Zhang SF, Wu ZL, Zhang YX. Is the September 5, 2022, Luding Ms6.8 earthquake an ‘unexpected’ event? Earthquake Science. 2023;36:76-80. DOI: 10.1016/j.eqs.2023.02.004
20.Zaliapin I, Ben-Zion Y. Perspectives on clustering and Declustering of earthquakes. Seismological Research Letters. 2021;93:386-401. DOI: 10.1785/0220210127
21.Gurjar N, Basu D. On the declustering methods of seismic catalogue — An application over Indian subcontinent. Journal of Seismology. 2022;26:1077-1103. DOI: 10.1007/s10950-022-10105-9
22.Eroglu Azak T, Kalafat D, Şeşetyan K, Demircioğlu MB. Effects of seismic declustering on seismic hazard assessment: A sensitivity study using the Turkish earthquake catalogue. Bulletin of Earthquake Engineering. 2017;16:3339-3366. DOI: 10.1007/s10518-017-0174-y
23.Marzocchi W, Jordan TH, Woo G. Operational earthquake forecasting and decision making. Annals of Geophysics. 2015;58:RW0434. DOI: 10.4401/ag-6756
24.Jordan TH, Marzocchi W, Michael AJ, Gerstenberger MC. Operational earthquake forecasting can enhance earthquake preparedness. Seismological Research Letters. 2014;85:955-959. DOI: 10.1785/0220140143
25.Zaliapin I, Gabrielov A, Keilis-Borok V, Wong H. Clustering analysis of seismicity and aftershock identification. Physical Review Letters. 2008;101:018501. DOI: 10.1103/PhysRevLett.101.018501
26.Gardner JK, Knopoff L. Is the sequence of earthquakes in Southern California, with aftershocks removed, Poissonian? Bulletin of the Seismological Society of America. 1974;64:1363-1367. DOI: 10.1785/bssa0640051363
27.Gutenberg B, Richter CF. Frequency of earthquakes in California. Bulletin of the Seismological Society of America. 1944;34:185-188. DOI: 10.1785/BSSA0340040185
28.Bevington PR, Robinson DK. Data Reduction and Error Analysis for the Physical Sciences. 3rd ed. Boston (Mass.): McGraw-Hill; 2003
29.Luginbuhl M, Rundle JB, Turcotte DL. Natural time and nowcasting induced seismicity at the Groningen gas field in the Netherlands. Geophysical Journal International. 2018;215:753-759. DOI: 10.1093/gji/ggy315
30.Sarlis NV, Skordas ES. Study in natural time of Geoelectric field and seismicity changes preceding the M(w)6.8 earthquake on 25 October 2018 in Greece. Entropy (Basel). 2018;20:882. DOI: 10.3390/e20110882
31.Sarlis N, Skordas E, Varotsos P. A remarkable change of the entropy of seismicity in natural time under time reversal before the super-giant M9 Tohoku earthquake on 11 march 2011. EPL (Europhysics Letters). 2018;124:29001. DOI: 10.1209/0295-5075/124/29001
32.Varotsos P, Sarlis N, Skordas E. Natural time analysis: Important changes of the order parameter of seismicity preceding the 2011 M9 Tohoku earthquake in Japan. EPL (Europhysics Letters). 2019;125:69001. DOI: 10.1209/0295-5075/125/69001
33.Rundle JB, Luginbuhl M, Giguere A, Turcotte DL. Natural time, nowcasting and the physics of earthquakes: Estimation of seismic risk to global megacities. In: Earthquakes and Multi-Hazards around the Pacific Rim. Vol. II. Pageoph Topical Volumes. Cham: Birkhäuser; 2019. pp. 123-136. DOI: 10.1007/s00024-017-1720-x
34.Rundle J, Stein S, Donnellan A, Turcotte DL, Klein W, Saylor C. The complex dynamics of earthquake fault systems: New approaches to forecasting and Nowcasting of earthquakes. Reports on Progress in Physics. 2021;84:076801. DOI: 10.1088/1361-6633/abf893
35.Fildes RA, Turcotte DL, Rundle JB. Natural time analysis and Nowcasting of quasi-periodic collapse events during the 2018 Kīlauea volcano eruptive sequence. Earth and Space Science. 2022;9:e2022EA002266. DOI: 10.1029/2022EA002266
36.Zhang YX, Wu ZL, Zhang XT, Li G. Annual earthquake potential consultation: A real forward prediction test in China. In: Earthquake and Disaster Risk: Decade Retrospective of the Wenchuan Earthquake. Singapore: Springer; 2019. pp. 117-134. DOI: 10.1007/978-981-13-8015-0_5
37.Wu FT. The annual earthquake prediciton conference in China (National Consultative Meeting on seismic tendency). Pure and Applied Geophysics. 1997;149:249-264. DOI: 10.1007/BF00945170
38.Huang F, Li M, Ma Y, Han Y, Tian L, Yan W, et al. Studies on earthquake precursors in China: A review for recent 50 years. Geodesy and Geodynamics. 2017;8:1-12. DOI: 10.1016/j.geog.2016.12.002
39.Marzocchi W, Lombardi AM. Real-time forecasting following a damaging earthquake. Geophysical Research Letters. 2009;36:L21302. DOI: 10.1029/2009gl040233
Notes
CEA database: http://10.5.160.18/uniteDayCatalog/index.action.
Written By
Shengfeng Zhang and Yongxian Zhang
Submitted: 15 July 2024Reviewed: 24 July 2024Published: 26 August 2024
Open access peer-reviewed chapter - ONLINE FIRST
