Presentations

A Bayesian transdimensional approach to estimate temporal changes in the b-value distribution without truncating catalogs

March 07, 2024

Poster, European Geophysical Union 2024, Vienna, Austria

The frequency/magnitude distribution of earthquakes can be approximated by an exponential law whose exponent is the so-called b-value. The b-value is routinely used for probabilistic seismic hazard assessment. In this context we propose to estimate the temporal variations of the b-value together with its uncertainties. The b-value is commonly estimated using the frequentist approach of Aki (1965), but biases may arise from the choice of completeness magnitude (Mc), the magnitude below which the exponential law is no longer valid. Here we propose to describe the full frequency-magnitude distribution of earthquakes by the product of an exponential law with a detection law. The latter is characterized by two parameters, μ and σ, that we jointly estimate with b-value within a Bayesian framework. In this way, we use all the available data to recover the joint probability distribution for b-value, μ and σ. Then, we extend this approach for recovering temporal variations of the three parameters. To that aim, we randomly explore with a Markov chain Monte Carlo (McMC) method in a transdimensional framework a large number of time variation configurations of the 3 parameters. This provides posterior probability distributions of the temporal variations in b-value, μ and σ. For an application to a seismic catalog of far-western Nepal, we show that the probability distribution of the b-value remains stable with larger uncertainties during the monsoon period when the detectability decreases significantly . This confirms that we can see variations in the b-value that are independent of variations in detectability. Our results can be compared with the results and interpretations obtained using the b-positive approach. We hope that further applications to real and experimental data can provide statistical constraints on the b-value variations and help to better understand the physical meaning behind these variations.

A Bayesian Transdimensional Framework to Recover Temporal Changes in the b-value Distribution of Non-truncated Seismic Catalog

December 14, 2023

Talk, American Geophysical Union 2023, San Francisco, US

The b-value is a key parameter which is used to describe the characteristic behavior of earthquakes in a specific region or within clusters. Additionally, it plays a significant role in assessing probabilistic seismic hazard which thus need a robust evaluation of its uncertainty. Conventionally, the b-value is determined for a truncated seismic catalog using the frequentist approach of Aki. However, this method has limitations, as the resulting b-value is influenced by the arbitrary choice of the completeness magnitude, magnitude binning, and maximum magnitude of the catalog. These factors can introduce bias in the analysis of temporal variations. To address this issue, we use a transdimensional inversion that enables the recovery of temporal changes in the entire frequency-magnitude distribution of earthquakes, independent of the completeness magnitude. The Bayesian framework provides the full probability distribution on every temporal segments for the b-value and the two other parameters which are related to the earthquake detection ability. By employing the transdimensional approach, statistically significant changes in any of the three inverted parameters can be automatically identified and associated with a probability. Its future application to real or experimental data can help to describe the temporal evolution of the detection ability of a network as well as relating b-value changes to physical processes in the crustal field or on the fault.

Modélisation Bayésienne des Variations Temporelles de la Distribution Fréquence-Magnitude des Séismes à partir de Catalogues Non Tronqués

November 07, 2023

Poster, Rencontres EPOS France, Saint Jean Cap Ferrat, France

La distribution en fréquence des magnitudes des séismes suit généralement une loi exponentielle décroissante dont l’exposant est le paramètre b de la loi de Gutenberg-Richter qui semble varier à la fois spatialement et temporellement. L'estimation précise de la « b-value » et de son incertitude est cruciale dans l'évaluation du risque sismique. Cependant la méthode traditionnelle d’évaluation de b, donnée par la formule d’Aki, renseigne mal les incertitudes et présente des limitations importantes notamment liées au choix souvent arbitraire de la complétude du catalogue. Ces facteurs peuvent introduire des biais importants dans l'estimation de la b-value, ce qui affecte son interprétation ainsi que l'analyse de ses variations spatio-temporelles. Nous proposons une approche bayésienne qui permet de : (1) modéliser le catalogue complet en convoluant la distribution de fréquences de magnitudes par une fonction de probabilité de détection des évènements; (2) calculer précisément les incertitudes en modélisant la densité de probabilité totale de la b-value et (3) qui détecte les changements temporels dans la distribution fréquence-magnitude des séismes. Pour valider cette méthode, des tests synthétiques ont été réalisés ainsi qu’une application sur un jeu de données du Grand Ouest Népal. Cette approche permet une meilleure compréhension des variations temporelles de b et des autres paramètres de détection, offrant ainsi des perspectives pour les calculs d'aléa sismique.

Persistent segmentation of the seismicity across seismic cycle at the downdip end of the seismogenic zone: the 2015 Gorkha earthquake in Nepal

December 14, 2021

Poster, American Geophysical Union 2021, New Orleans, US

Understanding along strike variations of the seismicity at the down-dip end of the seismogenic zone remains a challenge, partly because little is known about the heterogeneities of frictional properties and fault zone geometry there. Here we look specifically at the seismicity variations along the rupture of the 2015 Gorkha earthquake, Nepal.We first relocate the clusters of aftershocks that occurred during the 5 years following the earthquake. We compare them with the background seismicity that happened during the 4 decades that preceded the event. The heterogeneities show similarities and fall at short distances from heterogeneities in the coseismic slip and the location of high frequency seismic bursts generated during the rupture. These persistent spatial variations of the seismic heterogeneities are intriguing, and could be related with variations of the frictional properties or geometry of the Main Himalayan Thrust fault zone at depth. They suggest also that the seismicity occurring during the interseismic period along the Himalayas may reveal the seismo-geological segmentation, which influences both the coseismic rupture and the postseismic relaxation.

Teleseismic depth determination techniques and uncertainties : an Himalayan case study

June 25, 2021

Talk, CTBT: Science and Technology Conference, Vienna, Austria

Accurate estimates of the depth of seismic events allow determining whether they are associated to a given tectonic structure. It is also a good discriminator between earthquakes and explosions. However, automatic depth determination at teleseismic distance remains a challenge: the depth phases (pP, sP), reflected on the free surface, are sometime difficult to pick in the teleseismic signal. This is particularly true when the events are intermediate magnitudes (M<5), and fall at shallow depths in complex tectonic environments. To overcome that challenge, we implement two teleseismic depth estimation methods : (1) a cepstral analysis allowing to extract the pP-sP reflected waves in a the P-coda from their phase's similarity with the direct P wave and (2) an envelope stacking procedure aiming to highlight these secondary arrivals from their energetical contents. These two complementary methods allow improving signal over noise ratios and automatically identifying coherent depth phases among the thousands of teleseismic stations and arrays available from global teleseismic networks, including those of the International Monitoring System monitored by CTBTO. We confront our results to a set of well determined regional depths determined from a dense temporary network deployed in the Nepalese Himalayas, a region of high-topography and relatively shallow seismicity.

Seismicity in far-western Nepal reveals flats and ramps along the Main Himalayan Thrust

December 14, 2019

Talk, American Geophysical Union 2019, San Francisco, US

Unravelling relations between lateral variations of mid-crustal seismicity and the geometry of the Main Himalayan Thrust system at depth is a key issue in seismotectonic studies of the Himalayan range. Indeed, these relations may reveal along strike changes in the behaviour of the fault at depth related to fluids, the basal friction, the thermal structure or the local geometry of the flat and ramp system and more generally of the stress build up along the fault. Some of these variations may control the intermediate, large or great earthquakes ruptures extension and are therefore crucial to document for better assessing the regional seismic hazard. Far Western Nepal is particularly exposed to large and great earthquakes, the last of which dates back from 1505 CE. The region is also associated to striking lateral spatio-temporal variations of the midcrustal seismicity monitored by the Regional Seismic Network of Surkhet-Birendranagar. This network was complemented between 2014 and 2016 by 15 temporary stations deployed above the main seismic clusters giving new potential to regional studies. Absolute, relative locations and focal mechanisms were determined to gain insight on the fault behaviour at depth. We found that the microseismicity is clustered in three belts parallel to the front of the Himalayan range and confined between the Main Boundary Thrust and the Main Central Thrust. Finest locations reveal close relationships between seismic clusters and fault segments at depth among which two mid-crustal ramps and reactivated tectonic slivers. Interpreted fault planes of computed focal mechanisms are consistent with compressive stress and regional deformation inferred from geodetic studies. Our results support a geometry of the Main Himalayan Thrust involving several fault patches at depth separated by 3 to 6 km-high ramps and tear faults. This has an impact on the assessment of the geometry of the coseismic ruptures breaking partially or totally the locked fault zone in the future.

Marine