Geostatistical Seismic Inversion with Self-Updating of Local Probability Distributions
- PDF / 5,395,193 Bytes
- 21 Pages / 439.37 x 666.142 pts Page_size
- 54 Downloads / 189 Views
Geostatistical Seismic Inversion with Self-Updating of Local Probability Distributions Leonardo Azevedo1 · João Narciso1 Rúben Nunes1 · Amílcar Soares1
·
Received: 30 December 2019 / Accepted: 19 September 2020 © International Association for Mathematical Geosciences 2020
Abstract Three-dimensional subsurface elastic models inverted from seismic reflection data are the basis of the geo-modeling workflow. These models are often used to predict the spatial distribution of reservoir rock properties such as porosity, volume of minerals and fluid saturations. Stochastic seismic inversion methods are important modeling tools, as they allow one to infer high-resolution subsurface models and assess uncertainties related to the spatial distribution of the inverted petro-elastic properties. Within this framework, iterative geostatistical seismic inversion methods use stochastic sequential simulation and co-simulation as a model generation and perturbation technique based on the mismatch between synthetic and real seismic data. This work proposes an alternative approach of iterative geostatistical seismic inversion based on the concept of self-updating of local probability distributions of the elastic property of interest to be inverted. The model perturbation is conditioned by local probability distribution functions, which are iteratively updated based on the data misfit at previous iterations. This approach allows for better exploration of the model parameter space, avoiding local fast convergence at early steps of the inversion, and wider exploration of the model parameter space. The method is applied to a two-dimensional nonstationary synthetic dataset and to a three-dimensional real case example with a blind well test.
B
Leonardo Azevedo [email protected] João Narciso [email protected] Rúben Nunes [email protected] Amílcar Soares [email protected]
1
CERENA/DECivil, Pavilhão de Minas, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1049-001 Lisbon, Portugal
123
Math Geosci
Keywords Geostatistical seismic inversion · Local uncertainties · Direct sequential simulation · Self-learning
1 Introduction Given the recent technological advances in geophysical survey methods, seismic reservoir characterization has become a central step in the geo-modeling workflow. It allows one to infer the spatial distribution of subsurface petro-elastic properties such as acoustic and/or elastic impedance, porosity and fluid saturation (e.g., Doyen 2007; Soares et al. 2017). At this stage, the available dataset normally comprises well and seismic reflection data. Well-log data represent direct measurements of the subsurface properties with low uncertainty and high vertical resolution. However, they are scarce and sparsely located within the region of interest. On the other hand, seismic reflection data are indirect measurements of the subsurface properties of interest with small vertical resolution but extensively covering the spatial extent of the target ar
Data Loading...
