Reconstruction of channelized systems through a conditioned backward-migration method

in: 2016 RING Meeting, ASGA

Abstract

The optimal exploitation of a channelized reservoir requires an accurate knowledge of the geolog- ical heterogeneities impacting flow circulation, such as shale drapes. It may be possible to observe the most recent channel path and the abandoned meanders on seismic images, but smaller-scale structures are generally below seismic resolution. In this paper we propose to reconstruct channelized systems with a stochastic backward simula- tion of the system. To define a realistic parameterization of the simulation, we have studied maps of the last trajectories of the Mississippi river. Curvature and migration vectors have thus been measured. They show the variety of migration patterns that are encountered at half-meander scale. Then, the backward migration method uses the observed correlations between the channel geome- try and its natural migration. Starting from the last stage of the sequence observed on reflection seismic data, the aim is to go back to recover the possible evolution of the river. As abandoned meanders provide clues about the palaeo-locations of the river, they are integrated step by step during the backward process. We applied the method on satellite images of fluvial systems. Each of the different resulting geometries of the system honors most of the conditioning data and presents meandering patterns similar to the observed ones.

Download / Links

BibTeX Reference

@INPROCEEDINGS{,
    author = { Parquer, Marion and Collon, Pauline and Caumon, Guillaume },
     title = { Reconstruction of channelized systems through a conditioned backward-migration method },
 booktitle = { 2016 RING Meeting },
      year = { 2016 },
 publisher = { ASGA },
  abstract = { The optimal exploitation of a channelized reservoir requires an accurate knowledge of the geolog-
ical heterogeneities impacting flow circulation, such as shale drapes. It may be possible to observe
the most recent channel path and the abandoned meanders on seismic images, but smaller-scale
structures are generally below seismic resolution.
In this paper we propose to reconstruct channelized systems with a stochastic backward simula-
tion of the system. To define a realistic parameterization of the simulation, we have studied maps
of the last trajectories of the Mississippi river. Curvature and migration vectors have thus been
measured. They show the variety of migration patterns that are encountered at half-meander scale.
Then, the backward migration method uses the observed correlations between the channel geome-
try and its natural migration. Starting from the last stage of the sequence observed on reflection
seismic data, the aim is to go back to recover the possible evolution of the river. As abandoned
meanders provide clues about the palaeo-locations of the river, they are integrated step by step
during the backward process. We applied the method on satellite images of fluvial systems. Each of
the different resulting geometries of the system honors most of the conditioning data and presents
meandering patterns similar to the observed ones. }
}