arXiv subjects

D. Sornette

Publications and source records attributed to D. Sornette.

At least 145 records · Page 8Linked to original sources

arXiv2004

Creep failures in heterogeneous materials

We present creep experiments on fiber composite materials with controlled heterogeneity. Recorded strain rates and acoustic emission rates exhibit a power law relaxation in the primary creep regime (Andrade law) followed by a power law acceleration up to rupture over up to four decades in time. We discover that the failure time is proportional to the duration of the primary creep regime, showing the interplay between the two regimes and offering a method of rupture prediction. These experimental results are rationalized by a mean-field model of representative elements with nonlinear visco-elastic rheology and with a large heterogeneity of strengths.

cond-mat.mtrl-sci↗

arXiv2004

Anomalous Power Law Distribution of Total Lifetimes of Branching Processes Relevant to Earthquakes

We consider a branching model of triggered seismicity, the ETAS (epidemic-type aftershock sequence) model which assumes that each earthquake can trigger other earthquakes (``aftershocks''). An aftershock sequence results in this model from the cascade of aftershocks of each past earthquake. Due to the large fluctuations of the number of aftershocks triggered directly by any earthquake (``productivity'' or ``fertility''), there is a large variability of the total number of aftershocks from one sequence to another, for the same mainshock magnitude. We study the regime where the distribution of fertilities $μ$ is characterized by a power law $\sim 1/μ^{1+γ}$ and the bare Omori law for the memory of previous triggering mothers decays slowly as $\sim 1/t^{1+θ}$, with $0 < θ<1$ relevant for earthquakes. Using the tool of generating probability functions and a quasistatic approximation which is shown to be exact asymptotically for large durations, we show that the density distribution of total aftershock lifetimes scales as $\sim 1/t^{1+θ/γ}$ when the average branching ratio is critical ($n=1$). The coefficient $1<γ= b/α<2$ quantifies the interplay between the exponent $b \approx 1$ of the Gutenberg-Richter magnitude distribution $ \sim 10^{-bm}$ and the increase $\sim 10^{αm}$ of the number of aftershocks with the mainshock magnitude $m$ (productivity) with $α\approx 0.8$. More generally, our results apply to any stochastic branching process with a power-law distribution of offsprings per mother and a long memory.

physics.geo-ph↗

arXiv2004

Andrade and Critical Time-to-Failure Laws in Fiber-Matrix Composites: Experiments and Model

We present creep experiments on fiber composite materials. Recorded strain rates and acoustic emission (AE) rates exhibit both a power law relaxation in the primary creep regime and a power-law acceleration before global failure. In particular, we observe time-to-failure power laws in the tertiary regime for acoustic emissions over four decades in time. We also discover correlations between some characteristics of the primary creep (exponent of the power-law and duration) and the time to failure of the samples. This result indicates that the tertiary regime is dependent on the relaxation and damage processes that occur in the primary regime and suggests a method of prediction of the time to failure based on the early time recording of the strain rate or AE rate. We consider a simple model of representative elements, interacting via democratic load sharing, with a large heterogeneity of strengths. Each element consists of a non-linear dashpot in parallel with a spring. This model recovers the experimental observations of the strain rate as a function of time.

cond-mat.mtrl-sci↗

arXiv2004

Bubble, Critical Zone and the Crash of Royal Ahold

Our analysis of financial data, in terms of super-exponential growth, suggests that the seed of the 2002/03 crisis of the Dutch supermarket giant AHOLD was planted in 1996. It became quite visible in 1999 when the post-bubble destabilization regime was well-developed and acted as the precursor of an inevitable collapse fueled by raising expectations of investors to maintain strong herding pressures. We have adapted Weidlich's theory of opinion formation to describe the formation of buy or sell decisions among investors, based on a competition between the mechanisms of herding and of personal opinion opposing the herd. Among four typical patterns of stock price evolution, we have identified a ``critical zone'' in the model characterized by a strong sensitivity of the price trajectory on the herding and personal inclination parameters. The critical zone describes the maturation of a systemic instability forewarning of an inevitable crash. Classification and recognition of the spontaneous emergence of patterns of stock market evolution based on Weidlich's theory of complex systems, and in particular our discovery of the post-bubble destabilization regime which acts as a precursor to a subsequent crash or antibubble, not only presents the possibility of developing early warning signals but also suggests to top management ways of dealing with the coming crisis.

cond-mat.stat-mech↗

arXiv2004

Power Law Distributions of Offspring and Generation Numbers in Branching Models of Earthquake Triggering

We consider a general stochastic branching process, which is relevant to earthquakes as well as to many other systems, and we study the distributions of the total number of offsprings (direct and indirect aftershocks in seismicity) and of the total number of generations before extinction. We apply our results to a branching model of triggered seismicity, the ETAS (epidemic-type aftershock sequence) model. The ETAS model assumes that each earthquake can trigger other earthquakes (``aftershocks''). An aftershock sequence results in this model from the cascade of aftershocks of each past earthquake. Due to the large fluctuations of the number of aftershocks triggered directly by any earthquake (``fertility''), there is a large variability of the total number of aftershocks from one sequence to another, for the same mainshock magnitude. We study the regime where the distribution of fertilities mu is characterized by a power law ~1/μ^(1+gamma). For earthquakes, we expect such a power-law distribution of fertilities with gamma = b/alpha based on the Gutenberg-Richter magnitude distribution ~10^(-bm) and on the increase ~10^(alpha m) of the number of aftershocks with the mainshock magnitude m. We derive the asymptotic distributions p_r(r) and p_g(g) of the total number r of offsprings and of the total number g of generations until extinction following a mainshock. In the regime γ<2 relevant for earhquakes, for which the distribution of fertilities has an infinite variance, we find p_r(r)~1/r^(1+1/gamma) and p_g(g)~1/g^(1+1/(gamma -1)). These predictions are checked by numerical simulations.

cond-mat.stat-mech↗

arXiv2004

New statistic for financial return distributions: power-law or exponential?

We introduce a new statistical tool (the TP-statistic and TE-statistic) designed specifically to compare the behavior of the sample tail of distributions with power-law and exponential tails as a function of the lower threshold u. One important property of these statistics is that they converge to zero for power laws or for exponentials correspondingly, regardless of the value of the exponent or of the form parameter. This is particularly useful for testing the structure of a distribution (power law or not, exponential or not) independently of the possibility of quantifying the values of the parameters. We apply these statistics to the distribution of returns of one century of daily data for the Dow Jones Industrial Average and over one year of 5-minutes data of the Nasdaq Composite index. Our analysis confirms previous works showing the tendency for the tails to resemble more and more a power law for the highest quantiles but we can detect clear deviations that suggest that the structure of the tails of the distributions of returns is more complex than usually assumed; it is clearly more complex that just a power law.

physics.comp-ph↗

arXiv2004

Discrete Hierarchical Organization of Social Group Sizes

The ``social brain hypothesis'' for the evolution of large brains in primates has led to evidence for the coevolution of neocortical size and social group sizes. Extrapolation of these findings to modern humans indicated that the equivalent group size for our species should be approximately 150 (essentially the number of people known personally as individuals). Here, we combine data on human grouping in a comprehensive and systematic study. Using fractal analysis, we identify with high statistical confidence a discrete hierarchy of group sizes with a preferred scaling ratio close to 3: rather than a single or a continuous spectrum of group sizes, humans spontaneously form groups of preferred sizes organized in a geometrical series approximating 3, 9, 27,... Such discrete scale invariance (DSI) could be related to that identified in signatures of herding behavior in financial markets and might reflect a hierarchical processing of social nearness by human brains.

cond-mat.stat-mech↗

arXiv2004

Response Functions to Critical Shocks in Social Sciences: An Empirical and Numerical Study

We show that, provided one focuses on properly selected episodes, one can apply to the social sciences the same observational strategy that has proved successful in natural sciences such as astrophysics or geodynamics. For instance, in order to probe the cohesion of a policy, one can, in different countries, study the reactions to some huge and sudden exogenous shocks, which we call Dirac shocks. This approach naturally leads to the notion of structural (as opposed or complementary to temporal) forecast. Although structural predictions are by far the most common way to test theories in the natural sciences, they have been much less used in the social sciences. The Dirac shock approach opens the way to testing structural predictions in the social sciences. The examples reported here suggest that critical events are able to reveal pre-existing ``cracks'' because they probe the social cohesion which is an indicator and predictor of future evolution of the system, and in some cases foreshadows a bifurcation. We complement our empirical work with numerical simulations of the response function (``damage spreading'') to Dirac shocks in the Sznajd model of consensus build-up. We quantify the slow relaxation of the difference between perturbed and unperturbed systems, the conditions under which the consensus is modified by the shock and the large variability from one realization to another.

cond-mat.stat-mech↗

arXiv2003

Properties of Foreshocks and Aftershocks of the Non-Conservative SOC Olami-Feder-Christensen Model: Triggered or Critical Earthquakes?

Following Hergarten and Neugebauer [2002] who discovered aftershock and foreshock sequences in the Olami-Feder-Christensen (OFC) discrete block-spring earthquake model, we investigate to what degree the simple toppling mechanism of this model is sufficient to account for the properties of earthquake clustering in time and space. Our main finding is that synthetic catalogs generated by the OFC model share practically all properties of real seismicity at a qualitative level, with however significant quantitative differences. We find that OFC catalogs can be in large part described by the concept of triggered seismicity but the properties of foreshocks depend on the mainshock magnitude, in qualitative agreement with the critical earthquake model and in disagreement with simple models of triggered seismicity such as the Epidemic Type Aftershock Sequence (ETAS) model [Ogata, 1988]. Many other features of OFC catalogs can be reproduced with the ETAS model with a weaker clustering than real seismicity, i.e. for a very small average number of triggered earthquakes of first generation per mother-earthquake.

cond-mat↗

arXiv2003

Causal Slaving of the U.S. Treasury Bond Yield Antibubble by the Stock Market Antibubble of August 2000

Using the descriptive method of log-periodic power laws (LPPL) based on a theory of behavioral herding, we use a battery of parametric and non-parametric tests to demonstrate the existence of an antibubble in the yields with maturities larger than 1 year since October 2000. The concept of ``antibubble'' describes the existence of a specific LPPL pattern that is thought to reflect collective herding effects. From the dependence of the parameters of the LPPL formula as a function of yield maturities and using lagged cross-correlation calculations between the S&P 500 and bond yields, we find strong evidence for the following causality: Stock Market $\to$ Fed Reserve (Federal funds rate) $\to$ short-term yields $\to$ long-term yields (as well as a direct and instantaneous influence of the stock market on the long-term yields). Our interpretation is that the FRB is ``causally slaved'' to the stock market (at least for the studied period), because the later is (taken as) a proxy for the present and future health of the economy.

cond-mat.stat-mech↗

arXiv2003

Antibubble and Prediction of China's stock market and Real-Estate

We document a well-developed log-periodic power-law antibubble in China's stock market, which started in August 2001. We argue that the current stock market antibubble is sustained by a contemporary active unsustainable real-estate bubble in China. The characteristic parameters of the antibubble have exhibited remarkable stability over one year (Oct. 2002-Oct. 2003). Many tests, including predictability over different horizons and time periods, confirm the high significance of the antibubble detection. We predict that the Chinese stock market will stop its negative trend around the end of 2003 and start going up, appreciating by at least 25% in the following 6 months. Notwithstanding the immature nature of the Chinese equity market and the strong influence of government policy, we have found maybe even stronger imprints of herding than in other mature markets. This is maybe due indeed to the immaturity of the Chinese market which seems to attract short-term investors more interested in fast gains than in long-term investments, thus promoting speculative herding.

cond-mat.stat-mech↗

arXiv2003

Deviations of the distributions of seismic energies from the Gutenberg-Richter law

A new non-parametric statistic is introduced for the characterization of deviations from power laws. It is tested on the distribution of seismic energies given by the Gutenberg-Richter law. Based on the two first statistical log-moments, it evaluates quantitatively the deviations of the distribution of scalar seismic moments from a power-like (Pareto) law. This statistic is close to zero for the Pareto law with arbitrary power index, and deviates from zero for any non-Pareto distribution. A version of this statistic for discrete distribution of quantified magnitudes is also given. A methodology based on this statistics consisting in scanning the lower threshold for earthquake energies provides an explicit visualization of deviations from the Pareto law, surpassing in sensitivity the standard Hill estimator or other known techniques. This new statistical technique has been applied to shallow earthquakes (h < 70 km) both in subduction zones and in mid-ocean ridge zones (using the Harvard catalog of seismic moments, 1977-2000), and to several regional catalogs of magnitudes (California, Japan, Italy, Greece). We discover evidence for log-periodicity and thus for a discrete hierarchy of scales for low-angle dipping, low-strain subduction zones with a preferred scaling ratio g=7+-1 for seismic moments, compatible with a preferred scaling ratio of 2 for linear rupture sizes, and consistent with previous reports. We propose a possible mechanism in terms of cascades of fault competitions.

physics.data-an↗

arXiv2003

Andrade, Omori and Time-to-failure Laws from Thermal Noise in Material Rupture

Using the simplest possible ingredients of a rupture model with thermal fluctuations, we provide an analytical theory of three ubiquitous empirical observations obtained in creep (constant applied stress) experiments: the initial Andrade-like and Omori-like $1/t$ decay of the rate of deformation and of fiber ruptures and the $1/(t_c-t)$ critical time-to-failure behavior of acoustic emissions just prior to the macroscopic rupture. The lifetime of the material is controlled by a thermally activated Arrhenius nucleation process, describing the cross-over between these two regimes. Our results give further credit to the idea proposed by Ciliberto et al. that the tiny thermal fluctuations may actually play an essential role in macroscopic deformation and rupture processes at room temperature. We discover a new re-entrant effect of the lifetime as a function of quenched disorder amplitude.

cond-mat.mtrl-sci↗

arXiv2003

Long-range Static Directional Stress Transfer in a Cracked, Nonlinear Elastic Crust

Seeing the Earth crust as crisscrossed by faults filled with fluid at close to lithostatic pressures, we develop a model in which its elastic modulii are different in net tension versus compression. In constrast with standard nonlinear effects, this ``threshold nonlinearity'' is non-perturbative and occurs for infinitesimal perturbations around the lithostatic pressure taken as the reference. For a given earthquake source, such nonlinear elasticity is shown to (i) rotate, widen or narrow the different lobes of stress transfer, (ii) to modify the $1/r^2$ 2D-decay of elastic stress Green functions into the generalized power law $1/r^γ$ where $γ$ depends on the azimuth and on the amplitude of the modulii asymmetry. Using reasonable estimates, this implies an enhancement of the range of interaction between earthquakes by a factor up to 5-10 at distances of several tens of rupture length. This may explain certain long-range earthquake triggering and hydrological anomalies in wells and suggest to revisit the standard stress transfer calculations which use linear elasticity. We also show that the standard double-couple of forces representing an earthquake source leads to an opening of the corresponding fault plane, which suggests a mechanism for the non-zero isotropic component of the seismic moment tensor observed for some events.

physics.geo-ph↗

arXiv2003

Sub-critical and Super-critical Regimes in Epidemic Models of Earthquake Aftershocks

We present an analytical solution and numerical tests of the epidemic-type aftershock (ETAS) model for aftershocks, which describes foreshocks, aftershocks and mainshocks on the same footing. The occurrence rate of aftershocks triggered by a single mainshock decreases with the time from the mainshock according to the modified Omori law K/(t+c)^p with p=1+theta. A mainshock at time t=0 triggers aftershocks according to the local Omori law, that in turn trigger their own aftershocks and so on. The effective branching parameter n, defined as the mean aftershock number triggered per event, controls the transition between a sub-critical regime n<1 to a super-critical regime n>1. In the sub-critical regime, we recover and document the crossover from an Omori exponent 1-theta for t 1 and theta>0, we find a novel transition from an Omori decay law with exponent 1-theta fot t t*. The case theta<0 yields an infinite n-value. In this case, we find another characteristic time tau controlling the crossover from an Omori law with exponent 1-theta for t<tau, similar to the local law, to an exponential increase at large times. These results can rationalize many of the stylized facts reported for aftershock and foreshock sequences, such as (i) the suggestion that a small p-value may be a precursor of a large earthquake, (ii) the relative seismic quiescence sometimes observed before large aftershocks, (iii) the positive correlation between b and p-values, (iv) the observation that great earthquakes are sometimes preceded by a decrease of b-value and (v) the acceleration of the seismicity preceding great earthquakes.

cond-mat.stat-mech↗

arXiv2003

Mainshocks are aftershocks of conditional foreshocks: How do foreshock statistical properties emerge from aftershock laws

The inverse Omori law for foreshocks discovered in the 1970s states that the rate of earthquakes prior to a mainshock increases on average as a power law ~ 1/(t_c-t)^p' of the time to the mainshock occurring at t_c. Here, we show that this law results from the direct Omori law for aftershocks describing the power law decay ~ 1/(t-t_c)^p of seismicity after an earthquake, provided that any earthquake can trigger its suit of aftershocks. In this picture, the seismic activity at any time is the sum of the spontaneous tectonic loading and of the activity triggered by all preceding events weighted by their corresponding Omori law. The inverse Omori law then emerges as the expected (in a statistical sense) trajectory of seismicity, conditioned on the fact that it leads to the burst of seismic activity accompanying the mainshock. The often documented apparent decrease of the b-value of the GR law at the approach to the main shock results straightforwardly from the conditioning of the path of seismic activity culminating at the mainshock. In the space domain, we predict that the phenomenon of aftershock diffusion must have its mirror process reflected into an inward migration of foreshocks towards the mainshock. In this model, foreshock sequences are special aftershock sequences which are modified by the condition to end up in a burst of seismicity associated with the mainshock.

cond-mat.stat-mech↗

arXiv2003

Predictability in the ETAS Model of Interacting Triggered Seismicity

As part of an effort to develop a systematic methodology for earthquake forecasting, we use a simple model of seismicity based on interacting events which may trigger a cascade of earthquakes, known as the Epidemic-Type Aftershock Sequence model (ETAS). The ETAS model is constructed on a bare (unrenormalized) Omori law, the Gutenberg-Richter law and the idea that large events trigger more numerous aftershocks. For simplicity, we do not use the information on the spatial location of earthquakes and work only in the time domain. We offer an analytical approach to account for the yet unobserved triggered seismicity adapted to the problem of forecasting future seismic rates at varying horizons from the present. Tests presented on synthetic catalogs validate strongly the importance of taking into account all the cascades of still unobserved triggered events in order to predict correctly the future level of seismicity beyond a few minutes. We find a strong predictability if one accepts to predict only a small fraction of the large-magnitude targets. However, the probability gains degrade fast when one attempts to predict a larger fraction of the targets. This is because a significant fraction of events remain uncorrelated from past seismicity. This delineates the fundamental limits underlying forecasting skills, stemming from an intrinsic stochastic component in these interacting triggered seismicity models.

cond-mat.stat-mech↗

arXiv2003

Diffusion of Earthquake Aftershock Epicenters, Omori's Law and Generalized Continuous-Time Random Walk Models

The epidemic-type aftershock sequence model (ETAS) is a simple stochastic process modeling seismicity, based on the two best-established empirical laws, the Omori law (power law decay ~1/t^{1+θ} of seismicity after an earthquake) and Gutenberg-Richter law (power law distribution of earthquake energies). In order to describe also the space distribution of seismicity, we use in addition a power law distribution ~1/r^{1+μ} of distances between triggered and triggering earthquakes. We present an exact mapping between the ETAS model and a class of CTRW (continuous time random walk) models, based on the identification of their corresponding Master equations. This mapping allows us to use the wealth of results previously obtained on anomalous diffusion of CTRW. We provide a classification of the different regimes of diffusion of seismic activity triggered by a mainshock. Specifically, we derive the relation between the average distance between aftershocks and the mainshock as a function of the time from the mainshock and of the joint probability distribution of the times and locations of the aftershocks. Our predictions are checked by careful numerical simulations. We stress the distinction between the ``bare'' Omori law describing the seismic rate activated directly by a mainshock and the ``renormalized'' Omori law taking into account all possible cascades from mainshocks to aftershocks of aftershock of aftershock, and so on. In particular, we predict that seismic diffusion or sub-diffusion occurs and should be observable only when the observed Omori exponent is less than 1, because this signals the operation of the renormalization of the bare Omori law, also at the origin of seismic diffusion in the ETAS model.

cond-mat.stat-mech↗