Paul Cupillard

Joint mineral physics and seismic wave traveltime analysis of upper mantle temperature

We employ a new thermodynamic method for self-consistent computation of compositional and thermal effects on phase transition depths, density, and seismic velocities. Using these profi les, we compare theoretical and observed differential traveltimes between P410s and P (T 410 ) and between P600s and P410s (T 660‐410 ) that are affected only by seismic structure in the upper mantle. The anticorrelation between T 410 and T 660‐410 suggests that variations in T 410 and T 660‐410 of ~8 s are due to lateral temperature variations in the upper mantle transition zone of ~400 K. If the mantle is a mechanical mixture of basaltic and harzburgitic components, our traveltime data suggest that the mantle has an average temperature of 1600 ± 50 K, in agreement with temperature estimates from magma compositions of mid-ocean ridge basalts. We infer a 100 K hotter mantle if we assume the mantle to have a homogeneous pyrolitic composition. The transition-zone temperature beneath hotspots and within subduction zones is relatively high and low, respectively. However, the largest variability in T 410 and T 660‐410 is recorded by global stations far from subduction zones and hotspots. This indicates that the 400 K variation in upper mantle temperature is complicated by tilted upwellings, slab fl attening and accumulation, ancient subduction, and processes unrelated to present-day subduction and plume ascent.

The Collaborative Seismic Earth Model: Generation 1

Abstract We present a general concept for evolutionary, collaborative, multiscale inversion of geophysical data, specifically applied to the construction of a first‐generation Collaborative Seismic Earth Model. This is intended to address the limited resources of individual researchers and the often limited use of previously accumulated knowledge. Model evolution rests on a Bayesian updating scheme, simplified into a deterministic method that honors todays computational restrictions. The scheme is able to harness distributed human and computing power. It furthermore handles conflicting updates, as well as variable parameterizations of different model refinements or different inversion techniques. The first‐generation Collaborative Seismic Earth Model comprises 12 refinements from full seismic waveform inversion, ranging from regional crustal‐ to continental‐scale models. A global full‐waveform inversion ensures that regional refinements translate into whole‐Earth structure.

Stress estimation in reservoirs using an integrated inverse method

Estimating the stress in reservoirs and their surroundings prior to the production is a key issue for reservoir management planning. In this study, we propose an integrated inverse method to estimate such initial stress state. The 3D stress state is constructed with the displacement-based finite element method assuming linear isotropic elasticity and small perturbations in the current geometry of the geological structures. The Neumann boundary conditions are defined as piecewise linear functions of depth. The discontinuous functions are determined with the CMA-ES (Covariance Matrix Adaptation Evolution Strategy) optimization algorithm to fit wellbore stress data deduced from leak-off tests and breakouts. The disregard of the geological history and the simplified rheological assumptions mean that only the stress field, statically admissible and matching the wellbore data should be exploited. The spatial domain of validity of this statement is assessed by comparing the stress estimations for a synthetic folded structure of finite amplitude with a history constructed assuming a viscous response.

Integrated Inverse Method to Estimate Virgin Stress State in Reservoirs and Overburden

Summary Stress estimation in reservoirs and overbuden has become a key point during the exploration and the exploitation of the oil ans gas fields. We propose in this abstract an integrated method to compute a physically admissible (i.e. satisfying the equilibrium equations) 3D stress field in whole geological models. Stress field is computed using a simple elastic behavior and it is constrained to the wellbore data using an inverse approach. Forward problem is solved using a Finite Element Analysis. The model parameters to invert are the Neumann conditions which are assumed to be piecewise linear functions along the vertical direction. The data parameters are stress observations which came from the hydraulic fracturing and the borehole breakouts. Misfit between the computed stress and the observed stress is minimized using the CMA-ES algorithm. The method is tested with a synthetic case by taking a stress field computed with the Limit Analysis method as reference. The inversion results show that the method is able to well retrieve the stress variation in the geological model.

Appraising Structural Interpretations Using Seismic Data Misfit Functionals

Structural interpretation can be challenging because of complex wave interactions and limited seismic bandwidth. A single seismic image can lead to multiple structural interpretations, reflecting structural interpretation uncertainties. Typically, this uncertainty is captured by generating several possible structural geometries. However, a quantitative assessment of the different possible structural interpretations is often difficult. In this paper we propose a methodology for assessing structural interpretations using seismic data misfit functions. We first develop a conceptual framework for solving such a problem before applying the method to a carefully designed synthetic study. Our results suggest that it is possible to appraise structural interpretation using seismic data if an appropriate misfit function is used.

Simulating Micro-seismic Activity with a Discrete Geomechanical Model

Acoustic emissions are powerful data for characterizing the deformation of geological media. Nevertheless, there is still a lack of knowledge to understand the link between the mechanical stimulation of the medium and the recorded acoustic emissions. In this paper, we open the path for a tool that extracts and analyzes acoustic emissions from geomechanical simulations performed with a discrete element method. We discuss the capability of our tool by comparing statistics of our synthetic seismic activity with classical scaling laws observed on natural seismic catalogs. Analyses are performed on intact rock samples. We retrieve consistent spatial correlations of micro-seismic events and consistent

Multiscale full waveform inversion

b-values

The deep structure of the North Anatolian Fault Zone

. However, fitting Omoris law is more challenging.