Skip to Main content Skip to Navigation

Improving earthquake forecast models for PSHA with geodetic data, applied on Ecuador

Abstract : Probabilistic Seismic Hazard Assessment (PSHA) relies on long-term earthquake forecasts, and ground-motion models. Up to now, geodetic data has been rather under-used in PSHA, although it provides unique and unprecedented information on the deformation rates of tectonic structures from local to regional scales. The aim of this PhD thesis is to improve earthquake recurrence models by quantitatively including the information derived from geodetic measurements, with an application to Ecuador, a country exposed both to shallow crustal earthquakes and megathrust subduction events. The second chapter presents the building of a probabilistic seismic hazard model for Ecuador, using historical and contemporary seismicity, recent knowledges about active tectonics, geodynamics, and geodesy. I contributed to this collective effort in two ways: 1) the building of earthquake catalogs from global seismic datasets; 2) the establishment of average slip rates on a set of simplified crustal faults, from GPS velocities. The hazard calculations led at the country scale indicate that uncertainties are largest for sites on the northern coast and along the faults in the Cordillera. The second chapter of this PhD focuses on the determination of the seismic potential of the Quito fault system. Quito city lies on the hanging wall of this ∼60-km-long reverse active fault, representing significant risks due to the high population density. We constrain the present-day strain accumulation associated with the fault system with GPS data and Persistent Scatterer Interferometric Synthetic Aperture Radar (PS-InSAR) analysis. 3-D spatially variable locking models show that a large part of the fault is presently experiencing shallow creep, hence reducing the energy available for future earthquakes, which has a significant impact for hazard calculation. In the third part of this PhD, we evaluate the ability of geodetic data to constrain earthquake recurrence models for the subduction zone in northern Ecuador. We quantify the annual rate of moment deficit accumulation at the interface using interseismic coupling models, and identify the uncertainties related to the conversion in terms of total seismic moment release. Based on a newly-developed earthquake catalog, we propose to establish recurrence models that match both the catalog-based seismicity rates and the geodetic moment budget. We set up a logic tree for exploring the uncertainties on the seismic rates and on the geodetic moment budget to be released in earthquakes. The exploration of the logic tree leads to a distribution of possible maximal magnitudes Mmax bounding the earthquake recurrence model; we extract only those models that lead to Mmax values compatible with the extent of the interface segment according to earthquake scaling laws. This new method allows 1) to identify which magnitude-frequency form is adapted for the Ecuadorian subduction; 2) to generate a distribution of moment-balanced recurrence models representative of uncertainties and propagate this uncertainty up to the uniform hazard spectra; and 3) to evaluate a range for the aseismic component of the slip on the interface. Considering the recent availability of massive quantity of geodetic data, this new approach could be used in other regions of the world to develop recurrence models consistent both with past seismicity and measured tectonic deformation.
Document type :
Complete list of metadata
Contributor : Abes Star :  Contact
Submitted on : Friday, July 2, 2021 - 1:08:12 AM
Last modification on : Saturday, July 3, 2021 - 3:27:13 AM


Version validated by the jury (STAR)


  • HAL Id : tel-03276307, version 1



Judith Mariniere. Improving earthquake forecast models for PSHA with geodetic data, applied on Ecuador. Earth Sciences. Université Grenoble Alpes [2020-..], 2020. English. ⟨NNT : 2020GRALU019⟩. ⟨tel-03276307⟩



Record views


Files downloads