Search

The evolution laws of LURR (Loading-Unloading Response Ratio) before strong earthquakes, especially the peak point of LURR, are described in this paper. The results of four methods (experimental, numerical simulation, seismic data analysis and with damage mechanics analysis) lead to a consistent conclusion-the evolution laws of LURR before strong earthquakes are that, at the early stage of the seismic cycle, LURR will fluctuate around 1 and in the late stage, it rises swiftly and to its peak point. At some time after this peak point, a catastrophic event or events occur. These do not occur at the peak point, but lag behind. The lag time which is denoted by T depends on the magnitude M of the upcoming earthquake among other factors. In order to consider the influence of geophysical parameters in a specific region such as $$ \dot{\gamma }, $$ E and J, where $$ \dot{\gamma } $$ is the shear strain rate of tectonic loading in situ, E is the sum of radiated energy of all earthquake occurring in a specific region measured during a long time duration (110 years in this paper) divided by the area of the region and the time duration, and J is a parameter denoting the LURR anomaly area weighted with Y (the value of LURR) and represents the expanse and degree of the seismogenic zone. The dimensional analysis method has been used to reveal the relation between M, T and other parameters in situ for more reliable earthquake prediction. [ABSTRACT FROM AUTHOR]

The MS8.0 earthquake that occurred in Wenchuan, Sichuan in 2008 provides an important case for the study of operational earthquake forecasting and short-term aftershock forecasting of major disaster-inducing earthquakes in China. This paper focuses on the comparative study of the applicability of the epidemic-type aftershock sequence (ETAS) model, the Reasenberg–Jones (R–J) model and the Omi–R–J model, which are widely adopted internationally for short-term aftershock forecasting and seismic hazard mitigation strategy research. We compare the stability of model parameters and aftershock occurrence rate forecasting, and evaluate the effectiveness of forecasting using the N-test and T-test with multiple time windows. The results show that the sequence parameters of the ETAS model, the R–J model and the Omi–R–J model tend to stabilize after 15.50, 15.50 and 6.00 days following the earthquake respectively, and the attenuation of the Wenchuan MS8.0 earthquake is rather normal. Compared to the ETAS model and the R–J model, the Omi–R–J model obtain steadier model parameters in a shorter time with significantly smaller parameters pORJ, cORJ, bORJ and standard deviations. Among the three models, the overall aftershock occurrence rate forecasted by the R–J model is the highest, followed by the Omi–R–J model, while that of the ETAS model is the lowest. N-test results show overall forecasting effectiveness of 93.8, 80.7 and 97.7% for the ETAS, R–J and Omi–R–J models, respectively, with the ETAS and Omi–R–J models superior to the R–J model, and the Omi–R–J model slightly better than the ETAS model. The overall "information gain per earthquake" calculation results show that the ETAS model is superior to the Omi–R–J and R–J models, while the Omi–R–J model is better than the R–J model; thus the combined use of the ETAS and Omi–R–J models by focusing on their respective strengths might ensure optimal performance. These "maneuverable" forecasting approaches to short-term aftershock model forecasting will play a vital role in efficient post-disaster relief, emergency management decision-making and post-disaster reconstruction. [ABSTRACT FROM AUTHOR]

Large earthquakes have semi-periodic behavior as a result of critically self-organized processes of stress accumulation and release in seismogenic regions. Hence, large earthquakes in a given region constitute semi-periodic sequences with recurrence times varying slightly from periodicity. In previous papers, it has been shown that it is possible to identify these sequences through Fourier analysis of the occurrence time series of large earthquakes from a given region, by realizing that not all earthquakes in the region need belong to the same sequence, since there can be more than one process of stress accumulation and release in the region. Sequence identification can be used to forecast earthquake occurrence with well determined confidence bounds. This paper presents improvements on the above mentioned sequence identification and forecasting method: the influence of earthquake size on the spectral analysis, and its importance in semi-periodic events identification are considered, which means that earthquake occurrence times are treated as a labeled point process; a revised estimation of non-randomness probability is used; a better estimation of appropriate upper limit uncertainties to use in forecasts is introduced; and the use of Bayesian analysis to evaluate the posterior forecast performance is applied. This improved method was successfully tested on synthetic data and subsequently applied to real data from some specific regions. As an example of application, we show the analysis of data from the northeastern Japan Arc region, in which one semi-periodic sequence of four earthquakes with M ≥ 8.0, having high non-randomness probability was identified. We compare the results of this analysis with those of the unlabeled point process analysis. [ABSTRACT FROM AUTHOR]

Hydrological and geochemical changes have been studied extensively in China for the purpose of earthquake prediction since 1966. This paper gives a brief overview of the findings, which have been reported mainly in Chinese so far. Parameters studied include water level; water temperature; and radon, hydrogen and mercury concentrations. Significant changes have been observed before some large or even intermediate-sized earthquakes, among many cases that did not show any such changes. Background noise caused by other environmental variables and possible mechanisms for the pre-earthquake anomalies are discussed. Due to the complexity of earthquake prediction, much more monitoring efforts as well as sophisticated analytic and synthetic efforts are needed in the future. [ABSTRACT FROM AUTHOR]

- An algorithm recently developed by RUNDLEet al. (2002) to find regions of anomalous seismic activity associated with large earthquakes identified the location of anM w = 5.6 earthquake near Calexico, Mexico. In this paper we analyze the regional seismicity before this event, and a nearbyM w = 5.7 event, using time-to-failure algorithms developed by BOWMANet al. (1998) and BOWMAN and KING (2001a,b). The former finds the radius of a circular region surrounding the epicenter that optimizes the time-to-failure acceleration of seismic release. The latter optimizes acceleration based on the expected stress accumulation pattern for a dislocation source. Both methods found a period of accelerating seismicity in an optimal region, the size of which agrees with previously proposed scaling relations. This positive result suggests that the Rundle algorithm may provide a useful technique to identify regions of accelerating seismicity, which can then be analyzed using signal optimization time-to-failure techniques. [ABSTRACT FROM AUTHOR]

Real-time integration of multi-parametric observations is expected to accelerate the process toward improved, and operationally more effective, systems for time-Dependent Assessment of Seismic Hazard (t-DASH) and earthquake short-term (from days to weeks) forecast. However, a very preliminary step in this direction is the identification of those parameters (chemical, physical, biological, etc.) whose anomalous variations can be, to some extent, associated with the complex process of preparation for major earthquakes. In this paper one of these parameters (the Earth's emitted radiation in the Thermal InfraRed spectral region) is considered for its possible correlation with M ≥ 4 earthquakes occurred in Greece in between 2004 and 2013. The Robust Satellite Technique (RST) data analysis approach and Robust Estimator of TIR Anomalies (RETIRA) index were used to preliminarily define, and then to identify, significant sequences of TIR anomalies (SSTAs) in 10 years (2004-2013) of daily TIR images acquired by the Spinning Enhanced Visible and Infrared Imager on board the Meteosat Second Generation satellite. Taking into account the physical models proposed for justifying the existence of a correlation among TIR anomalies and earthquake occurrences, specific validation rules (in line with the ones used by the Collaboratory for the Study of Earthquake Predictability-CSEP-Project) have been defined to drive a retrospective correlation analysis process. The analysis shows that more than 93 % of all identified SSTAs occur in the prefixed space-time window around ( M ≥ 4) earthquake's time and location of occurrence with a false positive rate smaller than 7 %. Molchan error diagram analysis shows that such a correlation is far to be achievable by chance notwithstanding the huge amount of missed events due to frequent space/time data gaps produced by the presence of clouds over the scene. Achieved results, and particularly the very low rate of false positives registered on a so long testing period, seems already sufficient (at least) to qualify TIR anomalies (identified by RST approach and RETIRA index) among the parameters to be considered in the framework of a multi-parametric approach to t-DASH. [ABSTRACT FROM AUTHOR]

This paper aimed at assessing gravity variations as precursors for earthquake prediction in the Tibet (Xizang)-Qinghai-Xinjiang-Sichuan Region, western China. We here take a statistical approach to evaluate efficiency and possibility of earthquake prediction. We used the most recent spatiotemporal gravity field variation datasets of 2002-2008 for the region that were provided by the Crustal Movement Observation Network of China (CMONC). The datasets were space sparse and time discrete. In 2007-2010, 13 earthquakes (> M 6.0) occurred in the region. The observed gravity variations have a statistical correlation with the occurrence of these earthquakes through the Molchan error diagram tests that lead to alarms over a good fraction of space-time. The results show that the prediction efficiency of amplitude of analytic signal of gravity variations is better than seismicity rate model and THD and absolute value of gravity variation, implying that gravity variations before earthquake may include precursory information of future large earthquakes. [ABSTRACT FROM AUTHOR]

Among the schemes for earthquake forecasting, the search for semi-periodicity during large earthquakes in a given seismogenic region plays an important role. When considering earthquake forecasts based on semi-periodic sequence identification, the Bayesian formalism is a useful tool for: (1) assessing how well a given earthquake satisfies a previously made forecast; (2) re-evaluating the semi-periodic sequence probability; and (3) testing other prior estimations of the sequence probability. A comparison of Bayesian estimates with updated estimates of semi-periodic sequences that incorporate new data not used in the original estimates shows extremely good agreement, indicating that: (1) the probability that a semi-periodic sequence is not due to chance is an appropriate estimate for the prior sequence probability estimate; and (2) the Bayesian formalism does a very good job of estimating corrected semi-periodicity probabilities, using slightly less data than that used for updated estimates. The Bayesian approach is exemplified explicitly by its application to the Parkfield semi-periodic forecast, and results are given for its application to other forecasts in Japan and Venezuela. [ABSTRACT FROM AUTHOR]

Site effect is usually associated with local geological conditions, which increase or decrease the level of shaking compared with standard attenuation relations. We made an attempt to see in the macroseismic data of Italy some other effects, namely, hot/cold spots in the terminology of O lsen (in Bull. Seismol. Soc. Am. 90, 6B, 577-594, ), which are related to local fault geometry rather than to soil conditions. We give a list of towns and villages liable to amplify (+) or to reduce (-) the level of shaking in comparison with the nearby settlements. Relief and soil conditions cannot always account for the anomalous sites. Further, there are sites where both (+) and (-) effects are observed depending on the earthquake. The opposite effects can be generated by events from the same seismotectonic zone and along the same direction to the site. Anomalous sites may group themselves into clusters of different scales. All isolated anomalous patterns presented in this paper can be used in hazard analysis, in particular, for the modeling and testing of seismic effects. [ABSTRACT FROM AUTHOR]

The first part of the paper defines the terms and classifications common in earthquake prediction research and applications. This is followed by short reviews of major earthquake prediction programs initiated since World War II in several countries, for example the former USSR, China, Japan, the United States, and several European countries. It outlines the underlying expectations, concepts, and hypotheses, introduces the technologies and methodologies applied and some of the results obtained, which include both partial successes and failures. Emphasis is laid on discussing the scientific reasons why earthquake prediction research is so difficult and demanding and why the prospects are still so vague, at least as far as short-term and imminent predictions are concerned. However, classical probabilistic seismic hazard assessments, widely applied during the last few decades, have also clearly revealed their limitations. In their simple form, they are time-independent earthquake rupture forecasts based on the assumption of stable long-term recurrence of earthquakes in the seismotectonic areas under consideration. Therefore, during the last decade, earthquake prediction research and pilot applications have focused mainly on the development and rigorous testing of long and medium-term rupture forecast models in which event probabilities are conditioned by the occurrence of previous earthquakes, and on their integration into neo-deterministic approaches for improved time-variable seismic hazard assessment. The latter uses stress-renewal models that are calibrated for variations in the earthquake cycle as assessed on the basis of historical, paleoseismic, and other data, often complemented by multi-scale seismicity models, the use of pattern-recognition algorithms, and site-dependent strong-motion scenario modeling. International partnerships and a global infrastructure for comparative testing have recently been developed, for example the Collaboratory for the Study of Earthquake Predictability (CSEP) with test regions in California, Italy, Japan, New Zealand, and the Western Pacific. Algorithms and data bases are operated in a permanently learning and upgrading mode. Future perspectives and research requirements and the feasibility and possible problems encountered with the implementation of earthquake predictions in practice are briefly summarized. [ABSTRACT FROM AUTHOR]

The ability to successfully predict the future behavior of a system is a strong indication that the system is well understood. Certainly many details of the earthquake system remain obscure, but several hypotheses related to earthquake occurrence and seismic hazard have been proffered, and predicting earthquake behavior is a worthy goal and demanded by society. Along these lines, one of the primary objectives of the Regional Earthquake Likelihood Models (RELM) working group was to formalize earthquake occurrence hypotheses in the form of prospective earthquake rate forecasts in California. RELM members, working in small research groups, developed more than a dozen 5-year forecasts; they also outlined a performance evaluation method and provided a conceptual description of a Testing Center in which to perform predictability experiments. Subsequently, researchers working within the Collaboratory for the Study of Earthquake Predictability (CSEP) have begun implementing Testing Centers in different locations worldwide, and the RELM predictability experiment—a truly prospective earthquake prediction effort—is underway within the U.S. branch of CSEP. The experiment, designed to compare time-invariant 5-year earthquake rate forecasts, is now approximately halfway to its completion. In this paper, we describe the models under evaluation and present, for the first time, preliminary results of this unique experiment. While these results are preliminary—the forecasts were meant for an application of 5 years—we find interesting results: most of the models are consistent with the observation and one model forecasts the distribution of earthquakes best. We discuss the observed sample of target earthquakes in the context of historical seismicity within the testing region, highlight potential pitfalls of the current tests, and suggest plans for future revisions to experiments such as this one. [ABSTRACT FROM AUTHOR]

The objective of this paper is to quantify the use of past seismicity to forecast the locations of future large earthquakes and introduce optimization methods for the model parameters. To achieve this the binary forecast approach is used where the surface of the Earth is divided into l° × l° cells. The cumulative Benioff strain of m ≥ m c earthquakes that occurred during the training period, Δ Ttr, is used to retrospectively forecast the locations of large target earthquakes with magnitudes ≥ m T during the forecast period, Δ Tfor. The success of a forecast is measured in terms of hit rates (fraction of earthquakes forecast) and false alarm rates (fraction of alarms that do not forecast earthquakes). This binary forecast approach is quantified using a receiver operating characteristic diagram and an error diagram. An optimal forecast can be obtained by taking the maximum value of Pierce’s skill score. [ABSTRACT FROM AUTHOR]

The concept of state vector stems from statistical physics, where it is usually used to describe activity patterns of a physical field in its manner of coarsegrain. In this paper, we propose an approach by which the state vector was applied to describe quantitatively the damage evolution of the brittle heterogeneous systems, and some interesting results are presented, i.e., prior to the macro-fracture of rock specimens and occurrence of a strong earthquake, evolutions of the four relevant scalars time series derived from the state vectors changed anomalously. As retrospective studies, some prominent large earthquakes occurred in the Chinese Mainland (e.g., the M 7.4 Haicheng earthquake on February 4, 1975, and the M 7.8 Tangshan earthquake on July 28, 1976, etc) were investigated. Results show considerable promise that the time-dependent state vectors could serve as a kind of precursor to predict earthquakes. [ABSTRACT FROM AUTHOR]

Five large earthquakes of M ≥ 7.0 (based on the magnitude scale of the China Earthquake Networks Center) occurred in and near the Tibetan Plateau during 2008-2014, including the Wenchuan M8.0 earthquake on May 12, 2008 (BJT). In this paper, the Tibetan Plateau was chosen to be the study region, and calculating parameters of pattern informatics (PI) method with grid of 1° × 1° and forecasting time interval of 8 years were employed for the retrospective study according to the previous studies for M7 earthquake forecasting. The sliding step of forecasting interval was 1 year, and the hotspot diagrams of each forecasting interval since 2008 were obtained year by year. The relationships among the hotspots and the M ≥ 7.0 earthquakes that occurred during the forecast intervals were studied. The predictability of PI method was tested by verification of receiver-operating characteristic curve (ROC) and R score. The results show that the successive obvious hotspots occurred during the sliding forecasting intervals before four of the five earthquakes, while hotspots only occurred in one forecasted interval without successive evolution process before one of the five earthquakes, which indicates that four of the five large earthquakes could be forecasted well by PI method. Test results of the predictability of PI method by ROC and R score show that positive prospect of PI method could be expected for long-term earthquake forecast. [ABSTRACT FROM AUTHOR]

We investigate the intermediate-term earthquake forecasting of the central China North–South Seismic Belt for the last quarter-century using pattern informatics (PI) of seismicity. While the number of earthquakes above the cut-off magnitude is typically used in the original PI analysis, we use the mean magnitude of earthquakes as an alternative index of the 'pattern'. This alternative index can be understood in physics considering that the mean magnitude of earthquakes, related to the b value in the sense of the maximum-likelihood estimation, is associated with the stress level. While the original PI considers the number of earthquakes with different magnitudes above the cut-off magnitude, the PI of micro-seismicity considering earthquake magnitude, noted as PIm, assigns more weight to the larger events. Compared with the original PI, the PIm, with a different distribution of 'hotspots', does not demonstrate better forecast performance, implying that the simple number counting approach of PI effectively grasps the key physics of earthquake dynamics. Meanwhile, the combination of the original PI and the PIm, noted as PIJ, significantly reduces the number of hotspots and thus can reduce the false alarms. [ABSTRACT FROM AUTHOR]

An introduction is presented in which the editor discusses various reports within the issue on topics including research in statistical seismology methods and applications, and methods, techniques and outputs related to statistical analysis of earthquake occurrence and probability forecasting.

An improved pattern informatics (PI) method, inspired by the soup-of-groups model, is developed in this study. This improved method reveals its potential in reducing the noise that, when the classical PI is applied, can yield misleading conclusions about the existence of "hotspots" in the vicinity of the epicenter of a strong earthquake. The application of this new method to Italian seismicity from 1985 to 2018 properly reveals "hotspots" around the two strongest earthquakes that occurred in 2009 (L'Aquila earthquake) and 2016 (Norcia earthquake), suggesting its utility in studies devoted to earthquake prediction. [ABSTRACT FROM AUTHOR]

It is necessary to have reliable data on the distributions of earthquake hypocenters in the study area to solve many geological and geophysical problems, in particular, the forecast of strong earthquakes, seismic zoning and earthquake-proof construction. In turn, the completeness of the data on the distribution of earthquake hypocenters and the accuracy of their determination largely depends on the effectiveness of the observing system. The efficiency of the modern Ukrainian network of seismological observations that consists of 28 stations is estimated in this article. By the effectiveness of the seismological observation network we mean the ability of the network to record events with minimal magnitudes at certain epicentral distances and allowing us to determine its main parameters (time in the focus and coordinates of hypocenters) with minimal errors depending on the position of the earthquake hypocenter. Calculation of the minimum magnitude of earthquakes for the Ukrainian seismological network shows if each seismic station increase of susceptibility to 50,000 units then the seismological the network confidently registers earthquakes with a magnitude of 3 or more throughout the Ukraine. The errors in the determination of the coordinates of epicenters of earthquakes in latitude (δφ) (km) and in longitude (δλ) (km) do not exceed 1.0 km in the center of the network and 5.0 km on the periphery of the network. The values of errors in determining the depths of the earthquake (H) (km) centers in the center of the network does not exceed 5.0 km but on the whole territory of Ukraine does not exceed 20 km. The optimal configuration of the network of Ukrainian seismic stations based on the C-criterion of optimality, introduced earlier to the theory design of optimal experiments by one of the authors of the article, is proposed. [ABSTRACT FROM AUTHOR]

The Gutenberg–Richter (G–R) relationship can be derived as the Gibbs distribution. For a given earthquake set (all earthquakes in a given region, time period, magnitude range, tectonic settings) the Gibbs probability density function for magnitudes, with a given b value in its exponent, is the most uniform distribution under the constraints of the magnitude range and mean value. Therefore, it represents our limited knowledge about the system output: the only pieces of information are the mean value and the magnitude range. Honest earthquake forecasts can be based on such a distribution, since it represents all and only available information about the seismic system. The b value can change among different earthquake sets (in time, space, magnitude ranges, or tectonic settings), since it is related to earthquake rupture dynamics, or seismic source characteristics. The relationship between the b value and the exponent β in the rupture area vs. maximum slip scaling, A ∝ D β , results from viewing earthquake recurrence time in connection with the slip budget. This makes a link between earthquake statistics (the G–R law) and physics (fault characteristics). Specifically, the relationship enables us to explain different ranges of b values at megathrust faults, in dependence on interplate asperity and coupling distributions, as well as on amounts of sediments and fluids in subduction channels. The approach differs from common interpretations of the G–R law in that the b value becomes a field variable, not a constant. It is always the Gibbs distribution for a given magnitude range that we use due to our ignorance about the system outcome, and it is the b value that variates, depending on our knowledge about the system physics. This is important for seismic forecasts, which are mostly based on the G–R relationship. First, because the physical processes leading to the largest earthquakes can be revealed by observing the b value variations. Second, because earthquake generation process can be thought of as sampling with constraints, where the b value and the magnitude range are the constraints. [ABSTRACT FROM AUTHOR]

Rapid finite fault source determination is critical for reliable and robust tsunami early warnings. Near-field Global Navigation Satellite System (GNSS) observations have shown value to constrain the source inversion, but real-time GNSS stations are sparse along most of the active faults. Here we propose an automatic earthquake finite source inversion (AutoQuake Inversion) algorithm jointly using near-field (epicentral distance < 1000 km) GNSS data and mid-range (epicentral distance from 30° to 45°) teleseismic P displacement waveforms. Neither the near-field GNSS nor the mid-range teleseismic data clip or saturate during large earthquakes, while the fast-traveling P-waves are still essential to constrain the source in regions where very few or no GNSS stations are available. Real-time determination of the fault geometry remains to be the main challenge for rapid finite source inversion. We adopt a strategy to use the pre-defined geometry Slab2 for earthquakes within it or to forecast a focal mechanism based on near-by historical events for earthquakes without Slab2 prior. The algorithm has been implemented successfully in the prototype of JPL's GPS-Aided Tsunami Early-Detection system and tested for many real events recently. This article provides the framework of the algorithm, documents the retrospective and real-time results, and discusses remaining challenges for future improvements. [ABSTRACT FROM AUTHOR]

The casualties and financial losses caused by large earthquakes have led to an awareness of prediction importance of such earthquakes. The earthquake prediction is divided into four categories: long term, intermediate term, short term, and immediate. The M8 algorithm is one of the intermediate-term middle-range prediction algorithms primarily used to predict earthquake of magnitude 8 or more and is applied later for smaller magnitudes. The Iranian Plateau is less exposed to earthquake with magnitude 8 or more and it is observed that the seismicity rate in this region is generally low. Thus; the original M8 is not suitable for applying in this region. The objective of this study is to modify the M8 algorithm for Prediction of Major M7.0+ Earthquakes in the Iranian Plateau. The major earthquake of magnitude 7 or more in the Iranian plateau from 1975 to 2018 is considered as the target earthquakes. The hit rate times 1 minus alarm rate is defined as objective function and the particles swarm optimization meta-heuristic algorithm is used to maximize it. The optimum M8 could predict 14 out of 17 large earthquakes in the Iranian plateau while occupying 31.7% of the spatio-temporal space as the alarm. The results show that by employing an optimization algorithm, we can modify the M8 algorithm for efficient prediction of the target magnitudes less than 8 in the regions with low seismicity rate. [ABSTRACT FROM AUTHOR]

In this study, we documented a case involving large-scale macroscopic anomalies in the Xichang area, southwestern Sichuan Province, China, from May to June of 2002, after which no major earthquake occurred. During our field survey in 2002, we found that the timing of the high-frequency occurrence of groundwater anomalies was in good agreement with those of animal anomalies. Spatially, the groundwater and animal anomalies were distributed along the Anninghe-Zemuhe fault zone. Furthermore, the groundwater level was elevated in the northwest part of the Zemuhe fault and depressed in the southeast part of the Zemuhe fault zone, with a border somewhere between Puge and Ningnan Counties. Combined with microscopic groundwater, geodetic and seismic activity data, we infer that the anomalies in the Xichang area were the result of increasing tectonic activity in the Sichuan-Yunnan block. In addition, groundwater data may be used as a good indicator of tectonic activity. This case tells us that there is no direct relationship between an earthquake and these anomalies. In most cases, the vast majority of the anomalies, including microscopic and macroscopic anomalies, are caused by tectonic activity. That is, these anomalies could occur under the effects of tectonic activity, but they do not necessarily relate to the occurrence of earthquakes. [ABSTRACT FROM AUTHOR]

Recurrent groundwater radon anomalous declines were observed from well measurements in the Antung hot spring area (eastern Taiwan) prior to five of six earthquakes that occurred between 2003 and 2011 ( M range 5.0-6.8). The relationship between the detectability of radon anomalies and the first motions of P-waves was investigated. Based on the first motions of P-waves recorded near the investigated well, a precursory decrease in groundwater radon can be detected only when the first motion is compression. No precursory change in groundwater radon concentration was observed for the downward first motion of P-waves. [ABSTRACT FROM AUTHOR]

In the present study, a number of atmospheric and some ionospheric anomalies are analyzed, which were recorded prior to the Mw 8.3 Illapel earthquake of September 16, 2015. This very large earthquake occurred in Central Chile, close to the coast, as the result of thrust faulting on the interface between the Nazca Plate and South American continent. Using remotely sensed data extracted from NASA/Giovanni, NOAA/NCEP, and NOAA/NGDC, atmospheric and ionospheric anomalies were observed that co-registered 35-40 and 25-30 days prior to the main shock, respectively. With reference to long-term time series over the epicentral area, significant atmospheric anomalies were recorded for cloud cover, geopotential height, precipitation rates, surface air pressure, omega, stream function, and wind vectors-all in the time window of August 5-10, 2015, 35-40 days prior to the main shock. Anomalous TEC maps were recorded for the same time period. Satellite images indicate the formation of an unusual cyclone, presumably triggered by air turbulences and abnormal atmospheric conditions over the epicentral area, including strong vertical winds. Data from the Jicamarca radio observatory in Peru, more than 2000 km to the North, reveal anomalous ionospheric variations on August 15-20, 2015 with respect to international reference ionosphere thickness parameters and the altitude of the F layer. The observed anomalies are consistent with processes that occur at the ground-to-air interface due to the stress activation of peroxy defects in the hypocentral volume. The flow of positive hole charge carriers to the Earth surface expected to have led to massive air ionization, generating at first primarily positive airborne ions, then negative air ions plus ozone. Understanding the sequence of processes inside the Earth's crust and at the ground-to-air interface provides information not previously available about the causal and temporal linkages between the various pre-earthquake phenomena and the future seismic event. [ABSTRACT FROM AUTHOR]

Foreshock activity is considered as one of the most promising precursory changes for the main shock prediction in the short term. Averaging over several foreshock sequences has shown that foreshocks are characterized by distinct 3D patterns: their epicenters move towards the main shock epicenter, event count accelerates, and b-value drops. However, these space-time-size patterns were verified so far only in a very few individual cases mainly due to inadequate seismicity catalogue data. We have investigated 3D foreshock patterns before the M 8.8 Maule in 27 February 2010, M 8.1 Iquique in 1 April 2014, and M 8.4 Illapel in 16 September 2015 great earthquakes in the Chile subduction zone. To avoid biased results, no a priori spatiotemporal definitions of foreshocks were inserted. The procedure was based on pattern recognition from statistically significant seismicity changes in the three domains. The pattern recognition in one domain was independent of the pattern recognition in another domain. We found and verified with two independent catalogue data sets (CSN, IPOC) that within a critical area of ca. 65 km from the main shock epicenter, the 2014 event was preceded by distinct foreshock 3D patterns. A nearly weak foreshock stage (20 January-14 March 2014) was followed by a main-strong stage (15 March-1 April 2014) highly significant in all domains, although foreshock activity slightly decreased in about the last 5 days. Seismic moment release also accelerated in the last stage due to the occurrence of a cluster of very strong foreshock events. Foreshock activity very likely occurred in the hanging-wall fault domain on the South American Plate overriding Nazca Plate. The 2014 foreshock activity was quite similar to the one preceding the 6 Apr. 2009 L' Aquila (Italy) M 6.3 earthquake associated with normal faulting. Using the 2014 earthquake as a reference event, we observed that similar foreshock 3D patterns preceded the 2010 and 2015 earthquakes within critical distances of about 170 and 50 km, respectively. However, the foreshock activities were only weak in both the cases likely because of poor catalogue completeness. [ABSTRACT FROM AUTHOR]

Large devastating events in systems such as earthquakes, typhoons, market crashes, electricity grid blackouts, floods, droughts, wars and conflicts, and landslides can be unexpected and devastating. Events in many of these systems display frequency-size statistics that are power laws. Previously, we presented a new method for calculating probabilities for large events in systems such as these. This method counts the number of small events since the last large event and then converts this count into a probability by using a Weibull probability law. We applied this method to the calculation of large earthquake probabilities in California-Nevada, USA. In that study, we considered a fixed geographic region and assumed that all earthquakes within that region, large magnitudes as well as small, were perfectly correlated. In the present article, we extend this model to systems in which the events have a finite correlation length. We modify our previous results by employing the correlation function for near mean field systems having long-range interactions, an example of which is earthquakes and elastic interactions. We then construct an application of the method and show examples of computed earthquake probabilities. [ABSTRACT FROM AUTHOR]

Instrumentally recorded seismicity from 1970/01/01 to 2014/01/01 of the central China north-south seismic belt (21.0°-41.5°N, 97.5°-107.5°E) was analyzed, emphasizing the applicability of the predictive algorithms based on the assumptions of meta-stable equilibrium. The seismicity in this region was shown to exhibit ergodicity from 1980 to the present, with sub-region dependence, and interrupted by the 2008 Wenchuan earthquake. pattern informatics algorithm, a statistical physics-based predictive model for five-year time scale, is put to forward forecast test for the period 2014/01/01 to 2019/01/01. [ABSTRACT FROM AUTHOR]

Properties of seismoelectric waves in relation to natural earthquakes have been investigated. The electromagnetic disturbances were analyzed to test the hypothesis that pulse-like electric variations are directly related to microcracks as source. Because variation is very difficult to detect, there have been few quantitative field investigations. We used selected events with clear S and P phases from the data catalog obtained before the Tohoku earthquake in 2011. The electric strength of the fast P wave ( P), S wave ( S), and electromagnetic wave (EM) associated with formation of cracks of tensile mode were estimated. The co-seismic electric signal accompanied by the S wave has the largest strength, well above the noise level, and the EM wave has the lowest strength. Analytical estimation of the ratio of the strengths of the P and EM phases to that of the S phase by use of Pride's equations gave results partially in agreement with observation (the order was Ap > A > A). The strength of the observed electromagnetic mode is approximately two orders of magnitude larger than that estimated from the theory. We suggest this greater strength can be attributed to the converted modes at layer contracts or to the effect of the boundary between free atmosphere and crust. Overall agreement between observations and theoretical estimates suggests that electromagnetic anomalies, crustal deformation, and groundwater changes can be investigated on the basis of the unified equations for the coupled electromagnetics, acoustics, and hydrodynamics of porous media. [ABSTRACT FROM AUTHOR]

We developed a synth esis approach to augment current techniques for seismic hazard evaluation by combining four previously unrelated subjects: the pattern informatics (PI), load/unload response ratio (LURR), state vector (SV), and accelerating moment release (AMR) methods. Since the PI is proposed in the premise that the change in the seismicity rate is a proxy for the change in the tectonic stress, this method is used to quantify localized changes surrounding the epicenters of large earthquakes to objectively quantify the anomalous areas (hot spots) of the upcoming events. On the short-to-intermediate-term estimation, we apply the LURR, SV, and AMR methods to examine the hazard regions derived from the PI hot spots. A predictive study of the 2014 earthquake tendency in Chinese mainland, using the seismic data from 1970-01-01 to 2014-10-01, shows that, during Jan 01 to Oct 31, 2014, most of the M > 5.0 earthquakes, especially the Feb 12 M7.3 Yutian, May 30 M6.1 Yingjiang, Aug. 03 M6.5 Ludian, and Oct 07 M6.6 earthquakes, occurred in the seismic hazard regions predicted. Comparing the predictions produced by the PI and combination approaches, it is clear that, by using the combination approach, we can screen out the false-alarm regions from the PI estimation, without reducing the hit rate, and therefore effectively augment the predictive power of current techniques. This provided evidence that the multi-method combination approach may be a useful tool to detect precursory information of future large earthquakes. [ABSTRACT FROM AUTHOR]

In a previous publication, the seismicity of Japan from 1 January 1984 to 11 March 2011 (the time of the $$M9$$ Tohoku earthquake occurrence) has been analyzed in a time domain called natural time $$\chi.$$ The order parameter of seismicity in this time domain is the variance of $$\chi$$ weighted for normalized energy of each earthquake. It was found that the fluctuations of the order parameter of seismicity exhibit 15 distinct minima-deeper than a certain threshold-1 to around 3 months before the occurrence of large earthquakes that occurred in Japan during 1984-2011. Six (out of 15) of these minima were followed by all the shallow earthquakes of magnitude 7.6 or larger during the whole period studied. Here, we show that the probability to achieve the latter result by chance is of the order of $$10^{-5}$$ . This conclusion is strengthened by employing also the receiver operating characteristics technique. [ABSTRACT FROM AUTHOR]

Earthquake Data Enhanced Cyber-Infrastructure for Disaster Evaluation and Response (E-DECIDER) is a NASA-funded project developing new capabilities for decision making utilizing remote sensing data and modeling software to provide decision support for earthquake disaster management and response. E-DECIDER incorporates the earthquake forecasting methodology and geophysical modeling tools developed through NASA's QuakeSim project. Remote sensing and geodetic data, in conjunction with modeling and forecasting tools allows us to provide both long-term planning information for disaster management decision makers as well as short-term information following earthquake events (i.e. identifying areas where the greatest deformation and damage has occurred and emergency services may need to be focused). This in turn is delivered through standards-compliant web services for desktop and hand-held devices. [ABSTRACT FROM AUTHOR]

Prior to an earthquake, natural seismicity is correlated across multiple spatial and temporal scales. Many studies have indicated that an earthquake is hard to accurately predict by a single time-dependent precursory method. In this study, we attempt to combine four earthquake prediction methods, i.e. the Pattern Informatics (PI), Load/Unload Response Ratio (LURR), State Vector (SV), and Accelerating Moment Release (AMR) to estimate future earthquake potential. The PI technique is founded on the premise that the change in the seismicity rate is a proxy for the change in the underlying stress. We first use the PI method to quantify localized changes surrounding the epicenters of large earthquakes to objectively quantify the anomalous areas (hot spots) of the upcoming events. Next, we delineate the seismic hazard regions by integrating with regional active fault zones and small earthquake activities. Then, we further evaluate the earthquake potential in the seismic hazard regions using the LURR, SV and AMR methods. Retrospective tests of this new approach on the large earthquakes ( M > 6.5) which have occurred in western China over the last 3 years show that the LURR and SV time series usually climb to an anomalously high peak months to years prior to occurrence of a large earthquake. And, the asymptote time, t, 'predicted' by the AMR method correspond to the time of the actual events. The results may suggest that the multi-methods combined approach can be a useful tool to provide stronger constraints on forecasts of the time and location of future large events. [ABSTRACT FROM AUTHOR]

The purpose of using loss functions for earthquake prediction and disaster preparedness is to minimize expected costs when destructive earthquakes occur. This study focuses on developing a greater understanding of the interrelation between earthquake prediction characteristics and economic parameters. Equations for estimating economic losses and optimizing both earthquake prediction and disaster preparedness are presented in different dimensional forms to improve the understanding of parametric relationships between prediction and preparedness. The equations expand upon previously presented loss functions by explicitly considering loss as a function of both time and space and the cost parameters are clearly described to allow for practical application. Derivations reveal the close relationship between the loss function in terms of earthquake prediction characteristics and the benefit-cost analysis commonly used for disaster preparedness. An optimal preparedness scheme is presented based on a concept of unpreventable damage in extreme events and is shown to be a function of the level of damage prevented by taking action in response to an earthquake prediction. The formulations show that alarm durations are optimal relative to the type and time to implement different actions and the alarm area is optimal relative to the potential earthquake size and related geohazards. The presentation shows that earthquake prediction need not be constrained at a point in space to be useful for disaster preparedness and that mitigation activity is more economically feasible the smaller the area of prediction is with respect to the potential earthquake source size. Examples are used to show how loss functions can be utilized to determine if an algorithm may be useful to implement into practice and how earthquake prediction strategies can be implemented in coordination with other risk reduction strategies to make cost effective mitigations. Optimized earthquake prediction algorithms will greatly aid disaster managers and decision makers in their preparations once a prediction is made. The loss functions help to develop a greater understanding between earthquake prediction research and disaster preparedness implementation, allowing for future improvements in earthquake disaster prevention. [ABSTRACT FROM AUTHOR]

The Sannio seismogenic area turns out to be responsible for the highest peak ground accelerations (PGA) and seismic response spectra (SRS) at Napoli. The 1688 earthquake is considered representative of the area, and realistic synthetic seismograms have been computed for this scenario earthquake at the historical center and the eastern sector of Napoli. The use of a hybrid technique based on mode summation and finite-difference methods is fully justified by the available detailed knowledge about the geological and geophysical properties of the Neapolitan subsoil. This modeling makes it possible to recognize that amplifications of ~2 for PGA and >3 for SRS are to be expected because of the pyroclastic soil cover. Based on the information contained in the available catalogs, different magnitudes have been considered. Taking into account the correlation, valid for the Italian territory, between synthetic PGA and observed intensities, it turns out that the most probable magnitude ( M) of the 1688 earthquake is 6.7, while M = 7.3 should be assigned to a conservative scenario earthquake. Comparison of the computed response spectra for the 1688 scenario earthquake with the Italian seismic building code shows that the code is adequate with respect to the expected effects at the historical center of Napoli, but that it underestimates the possible ground motion at the eastern sector, in particular at the newly developed area built after the 1980 earthquake. [ABSTRACT FROM AUTHOR]

In this study we present a characteristic kinematic rupture model for an intraslab earthquake of the type of the 2001 Geiyo earthquake. The procedure for developing the characteristic rupture model follows the recipe developed by I rikura and M iyake (Proceedings of the 8th US National Conference on Earthquake Engineering, San Francisco, ) with a few modifications. We used the functional form of N akamura and M iyatake [Zisin (J Seism Soc Jpn), 53, 1-9, ] to model the spatial and temporal variability of the slip rate function. In our procedure the earthquake source is specified by the kinematic description of the fault model that incorporates spatial heterogeneity in slip and rise time and constant rupture velocity. In the proposed characteristic model the asperity locations and rupture initiation point were the only parameters that were constrained by the available earthquake rupture models of the target earthquake. The quality of the characteristic rupture model was assessed by comparing recorded and synthetic ground motion time histories from the 2001 Geiyo earthquake calculated with a standard broadband (0.1-10 Hz) ground motion simulation technique. Our analyses show that the characteristic rupture model performs well in reproducing the recorded ground motion, in spite of its simple representation of the fault geometry and kinematic rupture complexity. The analyses of ground motion sensitivity to relative location of asperities, and their stress drop contrast suggest that the overall ground motion goodness of fit remains the same at short periods, but decreases slightly at periods longer than 1 s. We suspect that, due to the depth of the intraslab source (40-50 km), the higher-frequency ground motions are relatively insensitive to the details of the slip distribution and slip contrast. On the other hand, even for intraslab events, the lower-frequency motions are influenced by rupture directivity effects and slip distribution. [ABSTRACT FROM AUTHOR]

Ground motions of the 1923 Kanto Earthquake inside the Kanto Basin are numerically simulated in a wide frequency range (0-10 Hz) based on new knowledge of the earthquake's source processes, the sedimentary structure of the basin, and techniques for generating broadband source models of great earthquakes. The Kanto Earthquake remains one of the most important exemplars for ground motion prediction in Japan due to its size, faulting geometry, and location beneath the densely populated Kanto sedimentary basin. We reconstruct a broadband source model of the 1923 Kanto Earthquake from inversion results by introducing small-scale heterogeneities. The corresponding ground motions are simulated using a hybrid technique comprising the following four calculations: (1) low-frequency ground motion of the engineering basement, modeled using a finite difference method; (2) high-frequency ground motion of the engineering basement, modeled using a stochastic Green's function method; (3) total ground motion of the engineering basement (i.e. 1 + 2); and (4) ground motion at the surface in response to the total basement ground motion. We employ a recently developed three-dimensional (3D) velocity structure model of the Kanto Basin that incorporates prospecting data, microtremor observations and measurements derived from strong ground motion records. Our calculations reveal peak ground velocities (PGV) exceeding 50 cm/s in the area above the fault plane: to the south, where the fault plane is shallowest, PGV reaches 150-200 cm/s at the engineering basement and 200-250 cm/s at the surface. Intensity 7, the maximum value in the Japan Meteorological Agency's intensity scale, is calculated to have occurred widely in Sagami Bay, which corresponds well with observed house-collapse rates due to the 1923 event. The modeling reveals a pronounced forward directivity effect for the area lying above the southern, shallow part of the fault plane. The high PGV and intensity seen above the southeastern corner of the fault plane and further east are largely due to this effect. Waveforms above the fault plane contain both short- and long-period components, but the short-period components are not observed further afield. Away from the fault, long-period waves (>2 s) dominate the ground motion, and in areas where the base of the third layer is relatively deep, the predominant period is >5 s. Levels of long-period ground motion in the southern part of the study area, around Sagami Bay and the southern parts of Boso Peninsula and Tokyo Bay, exceed that recorded at Tomakomai during the 2003 Tokachi-oki earthquake, when large oil storage tanks collapsed in response to sloshing generated by strong long-period motions. [ABSTRACT FROM AUTHOR]

systematic analysis is made of static Coulomb stress changes and earthquake occurrence in the area of the North Aegean Sea, Greece, in order to assess the prospect of using static stress changes to construct a regional earthquake likelihood model. The earthquake data set comprises all events of magnitude M ≥ 5.2 which have occurred since 1964. This is compared to the evolving stress field due to constant tectonic loading and perturbations due to coseismic slip associated with major earthquakes (M ≥ 6.4) over the same period. The stress was resolved for sixteen fault orientation classes, covering the observed focal mechanisms of all earthquakes in the region. Analysis using error diagrams shows that earthquake occurrence is better correlated with the constant tectonic loading component of the stress field than with the total stress field changes since 1964, and that little, if any, information on earthquake occurrence is lost if only the maximum of the tectonic loading over the fault orientation classes is considered. Moreover, the information on earthquake occurrence is actually increased by taking the maximum of the evolving stress field since 1964, and of its coseismic-slip component, over the fault orientation classes. The maximum, over fault orientation classes, of linear combinations of the tectonic loading and the evolving stress field is insignificantly better correlated with earthquake occurrence than the maximum of the tectonic loading by itself. A composite stress-change variable is constructed from ordering of the maximum tectonic loading component and the maximum coseismic-slip component, in order to optimize the correlation with earthquake occurrence. The results indicate that it would be difficult to construct a time-varying earthquake likelihood model from the evolving stress field that is more informative than a time-invariant model based on the constant tectonic loading. [ABSTRACT FROM AUTHOR]

Rigorous predictability experimentation requires a statistical characterization of the performance metric used to evaluate the participating models. We explore the properties of the area skill score measure and consider issues related to experimental discretization. For the case of continuous alarm functions and continuous observations, we present exact analytical solutions that describe the distribution of the area skill score for unskilled predictors, and we also describe how a Gaussian distribution with known mean and variance can be used to approximate the area skill score distribution. We quantify the deviation of the exact distribution from the Gaussian approximation by specifying the kurtosis excess as a function of the number of observed target earthquakes. For numerical earthquake predictability experiments that involve discretization of the study region and observations, we explore simulation procedures for estimating the area skill score distribution, and we present efficient algorithms for various experimental scenarios. When more than one target earthquake occurs within a given space/time/magnitude bin, the probabilities of predicting individual events are not independent, and this requires special consideration. Having presented the statistical properties of the area skill score, we describe and illustrate a preliminary procedure for comparing earthquake prediction strategies based on alarm functions. [ABSTRACT FROM AUTHOR]

Forecasts of future earthquake hazard in the San Francisco Bay region (SFBR) are dependent on the distribution used for the possible magnitude of future events. Based on the limited observed data, it is not possible to statistically distinguish between many distributions with very different tail behavior. These include the modified and truncated Gutenberg–Richter distributions, and a composite distribution assembled by the Working Group on California Earthquake Probabilities. There is consequent ambiguity in the estimated probability of very large, and hence damaging, events. A related question is whether the energy released in earthquakes is a small or large proportion of the stored energy in the crust, corresponding loosely to the ideas of self-organized criticality, and intermittent criticality, respectively. However, the SFBR has experienced three observed accelerating moment release (AMR) cycles, terminating in the 1868 Hayward, 1906 San Andreas and 1989 Loma Prieta events. A simple stochastic model based on elastic rebound has been shown to be capable of producing repeated AMR cycles in large synthetic catalogs. We propose that such catalogs can provide the basis of a test of a given magnitude distribution, via comparisons between the AMR properties of the real and synthetic data. Our results show that the truncated Gutenberg–Richter distribution produces AMR behavior closest to the observed AMR behavior. The proviso is that the magnitude parameters b and mmax are such that a sequence of large events that suppresses activity for several centuries is unlikely to occur. Repeated simulation from the stochastic model using such distributions produces 30-year hazard estimates at various magnitudes, which are compared with the estimates from the 2003 Working Group on California Earthquake Probabilities. [ABSTRACT FROM AUTHOR]

In 1976 Frank Evison identified the first examples of earthquake swarms as long-term precursors of main-shock events, and thereby discovered the predictive scaling relations of long-term seismogenesis. From this time on, forecasting became the main focus of his research. After learning from an early attempt to communicate forecasts confidentially to government, he recognised the importance of hypothesis testing, and the precursory swarm hypothesis was cast in a form similar to a regional likelihood model. Tests of its performance relative to a stationary Poisson model at M ≥ 5.8 in New Zealand were begun in 1977. The initial hypothesis was that of a 1–1 relation between swarms and main-shock events. Following a study of the Japan catalogue, the generalised swarm hypothesis, in which multiple swarms were precursory to multiple main-shock events, was formulated. Tests of this form of the hypothesis at M ≥ 6.8 were initiated in a region of surveillance east of Japan in 1983. Eventually the generalised hypothesis was adopted in New Zealand also. In 1999, tests were begun in a region of Greece. In 1994–1995, several main-shock events favourable to the swarm hypothesis occurred, however four main-shock events near Arthur’s Pass, New Zealand, occurred without precursory swarms. Subsequent analysis showed that events called “quarms”, which were similar to swarms but more protracted in time, had preceded these events. This led to the proposal of a qualitative physical process to account for swarms, quarms and the predictive relations: A three-stage faulting process, in which a major crack induces aftercracks in its neighbourhood, just as a main shock induces aftershocks. An inference from this process was that the most general long-term precursor should be an increase of seismicity at similar magnitudes to the eventual aftershocks. It turned out that such a precursory scale increase nearly always occurs before major earthquakes and conforms to the predictive scaling relations. Setting aside the problem of identifying the scale increase before the major earthquake, the EEPAS (Every Earthquake a Precursor According to Scale) forecasting model was formulated. The success of this relatively weak model in forecasting major events in New Zealand, California, Japan and Greece shows that the predictive scaling relations are ubiquitous in earthquake catalogues. Although none of the formal tests of the swarm hypothesis were successful in their own terms, they were beneficial in identifying shortcomings in its formulation, thereby leading to improved understanding of long-term seismogenesis and a better forecasting model. Some puzzling aspects of the scaling relations are whether they vary regionally, and why the precursor area and aftershock area scale differently with magnitude. A more practical question is whether the EEPAS model can be strengthened, by making use of the clustering of some precursors in swarms and quarms, to bring us nearer to the original goal of forecasting individual major earthquakes. [ABSTRACT FROM AUTHOR]

We construct a single hazard function from multiple predictive parameters independently developed for moderate earthquakes in Kanto, Japan, during a learning period from 1990 to 1999, and applied to a testing period from 2000 to 2005. Here, we consider as predictive parameters the a and b values of the Gutenberg–Richter relation, the ν value (change in b value), and the Every Earthquake a Precursor According to Scale (EEPAS) model rate. To study the correlations among the parameters, we prepare two groups of space–time coordinate sets for assessment, namely the background and conditional groups selected from the learning period. The background group contains ten thousand sets of coordinates randomly selected from the space–time volume of our study. The conditional group contains 33 sets of space–time coordinates corresponding to the epicenters of the target earthquakes ( M ≥ 5.0) just before their times of occurrence. Each parameter for the background group is transformed so that its distribution conforms to the standard Normal function. The mean and variance of the conditional distribution is then estimated after applying the same transformation to the conditional group. Using the means and variances of b values, ν values and EEPAS rates and the correlation matrices in the background and conditional distributions, we construct a combined hazard function following the procedure developed for normally distributed parameters. The information gain per event ( IGpe) of the new hazard function is 0.26 and 0.3 units larger than that of the EEPAS rate for the learning and testing period, respectively. The R-test confirms the statistical significance of the difference in the IGpe value for the testing period. [ABSTRACT FROM AUTHOR]

An Annual Consultation on the Likelihood of Earthquakes is regularly undertaken in continental China. It is a special kind of intermediate-term medium-range earthquake forecast and a practical service provided by seismological communities within the capability of modern seismology. One of the problems the Annual Consultation encounters is its false alarm rate, which has been higher than expected and has caused numerous controversies on the methodology and significance of the Annual Consultation. To tackle this problem we use the concept of reverse tracing of precursors (RTP) to analyze the long-term seismic activity in the alarm regions, or regions thought to have increased probability of earthquakes, as identified by the Annual Consultation. We apply the Pattern Informatics (PI) and relative intensity (RI) algorithms to the long-term estimation of seismic hazard in the RTP analysis. A retrospective study of the Sichuan-Yunnan region of southwest China, using the data from 1990 to 2003, shows that, by choosing the threshold proportion of overlapped areas between the long-term estimates and the Annual Consultation, the RTP analysis can correctly remove some of the false-alarm regions from the “black list’ and thus reduce the false-alarm rate of the Annual Consultation, without reducing the hit rate. Remarkably, the RTP approach, although originally proposed for short-term earthquake predictions, seems able to contribute to the Annual Consultation as well. [ABSTRACT FROM AUTHOR]

After reviewing the problems associated with the current implementation of the LURR theory, we suggest that taking account of stress field complexity and stress reorientation may resolve these problems. By introducing the concept of Maximum Faulting Orientation (MFO), we propose a new approach for calculating LURR. Results presented for the case of the Northridge earthquake provide encouragement for the stress-reorientation explanation and the new approach. [ABSTRACT FROM AUTHOR]

The possibility of intermediate-term earthquake prediction at Mt. Vesuvius by means of the CN algorithm is explored. CN was originally designed to identify the Times of Increased Probability (TIPs) for the occurrence of strong tectonic earthquakes, with magnitude M ≥ M 0, within a region a priori delimited. Here the CN algorithm is applied, for the first time, to the analysis of volcanic seismicity. The earthquakes recorded at Mt. Vesuvius during the period from February 1972 to June 2004 are considered, and the magnitude threshold M 0 selecting the events to be predicted is varied within the range: 3.0–3.3. Satisfactory prediction results are obtained, by retrospective analysis, when a time scaling is introduced. In particular, when the length of the time windows is reduced by a factor 2.5–3, with respect to the standard version of CN algorithm, more than 90% of the events with M ≥ M 0 occur within the TIP intervals, with TIPs occupying about 30% of the total time considered. The control experiment ``Seismic History'' demonstrates the stability of the obtained results and indicates that the CN algorithm can be applied to monitor the preparation of impending earthquakes with M ≥ 3.0 at Mt. Vesuvius. [ABSTRACT FROM AUTHOR]

Suppose a forecasting scheme associates a probabilityp* with some observed outcome. The entropy score given to this forecast is then -logp*. This article provides a review of the background to this scoring method, its main properties, and its relationships to concepts such as likelihood, probability gain, and Molchan’s ?-t diagram. It is shown that, in terms of this score, an intrinsic characterization can be given for the predictability of a given statistical forecasting model. Uses of the score are illustrated by applications to the stress release and ETAS models, electrical signals, and M8. [ABSTRACT FROM AUTHOR]

Scaling relations previously derived from examples of the precursory scale increase before major earthquakes show that the precursor is a long-term predictor of the time, magnitude, and location of the major earthquake. These relations are here taken as the basis of a stochastic forecasting model in which every earthquake is regarded as a precursor. The problem of identifying those earthquakes that are actually precursory is thus set aside, at the cost of limiting the strength of the resulting forecast. The contribution of an individual earthquake to the future distribution of hazard in time, magnitude and location is on a scale determined, through the scaling relations, by its magnitude. Provision is made for a contribution to be affected by other earthquakes close in time and location, e.g., an aftershock may be given low weight. Using the New Zealand catalogue, the model has been fitted to the forecasting of shallow earthquakes exceeding magnitude 5.75 over the period 1965–2000. It fits the data much better than a baseline Poisson model with a location distribution based on proximity to the epicenters of past earthquakes. Further, the model has been applied, with unchanged parameters, to the California region over the period 1975–2001. There also, it performs much better than the baseline model fitted to the same region over the period 1951–1974; the likelihood ratio is 1015 in favor of the present model. These results lend credence to the precursory scale increase phenomenon, and show that the scaling relations are pervasive in earthquake catalogues. The forecasting model provides a new baseline model against which future refinements, and other proposed models, can be tested. It may also prove to be useful in practice. Its applicability to other regions has still to be established. [ABSTRACT FROM AUTHOR]