Joint Optimization of Vertical Component Gravity and Seismic P-wave First Arrivals by Simulated Annealing

Loading...
Thumbnail Image

Authors

Basler-Reeder, Kyle J.

Issue Date

2015

Type

Thesis

Language

Keywords

geothermal , gravity , imaging , optimization , seismic , velocity

Research Projects

Organizational Units

Journal Issue

Alternative Title

Abstract

Simultaneous joint seismic-gravity optimization improves P-wave velocity models in areas with sharp lateral velocity contrasts. Optimization is achieved using simulated annealing, a metaheuristic global optimization algorithm that does not require an accurate initial model. Balancing the seismic-gravity objective function is accomplished by a novel approach based on analysis of Pareto charts. Gravity modeling uses a newly developed convolution model, while seismic modeling utilizes the highly efficient Vidale eikonal equation traveltime generation technique. Synthetic tests show that joint optimization improves velocity model accuracy and provides velocity control below the deepest headwave raypath. Restricted offset range migration analysis provides insights into both pre-critical and gradient reflections in the dataset.Detailed first arrival picking followed by trial velocity modeling remediates inconsistent data. We use a set of highly refined first arrival picks to compare results of a convergent joint seismic-gravity optimization to the Plotrefa and SeisOpt Pro velocity modeling softwares. Plotrefa uses a nonlinear least squares approach that is initial model dependent and produces shallow velocity artifacts. SeisOpt Pro utilizes the simulated annealing algorithm, also produces shallow velocity artifacts, and is limited to depths above the deepest raypath. Joint optimization increases the depth of constrained velocities, improving reflector coherency at depth. Kirchoff prestack depth migrations reveal that joint optimization ameliorates shallow velocity artifacts. Seismic and gravity data from the San Emidio Geothermal field of the northwest Basin and Range province demonstrate that joint optimization changes interpretation outcomes. The prior shallow valley interpretation gives way to a deep valley model, while shallow antiformal reflectors that could have been interpreted as antiformal folds are flattened. Furthermore, joint optimization provides a more clear picture of the rangefront fault. This technique can readily be applied to existing datasets and could replace the existing strategy of forward modeling to match gravity data.

Description

Citation

Publisher

License

In Copyright(All Rights Reserved)

Journal

Volume

Issue

PubMed ID

DOI

ISSN

EISSN