Integrating 4D Seismic to Optimize Production

Abstract

4D or time-lapse seismic methods have been proven by Shell and many other companies to be a valuable tool for reservoir monitoring and management1–4. In this paper, we use case studies to describe best practise. The main example is a mature field in the Northern North Sea, where continuous drilling of infill wells for two years has, so far, confirmed.the interpretation of remaining oil distribution from the 4D seismic data. In mature fields, where infill wells and sidetracks are targetted at small remaining volumes (where partial water flooding may have occurred), it becomes very important to achieve a good integration of 3D and 4D seismic data with the static and dynamic model. This is achieved qualitatively using new visualisation tools, and quantitatively using simple modelling methods such as 4D wedge modelling or more complex methods such as the generation of full field synthetic seismic. The updating of the reservoir simulator is performed in greater detail with the objective that the prediction remains accurate further into the future. This allows more reliable well targetting years after the acquisition of a 4D monitor survey, thereby reducing the necessary frequency of monitor survey acquisition. Impacting the business Infill targets. Shell has demonstrated the value of 4D both in relatively young fields (e.g. Draugen, Norway1), and in more mature fields (e.g. UK Central3 and Northern North Sea4). In a young setting infill well targets can be quite sizeable, whereas in a mature setting, remaining oil targets are generally smaller in volume, and can only be accessed economically using sidetracks or wells with multiple targets. Within many of these fields development is phased, with the most productive high permeability (‘primary’) levels targetted by early wells, and the less productive lower permeability (‘secondary’) levels being targetted by sidetracking of early primary wells which have watered out. This phasing is particularly advantageous for 4D driven development. Due to field maturity, remaining oil pockets at the primary-level are often small and would present uneconomic targets on their own, even for sidetracks. However, while infill wells and sidetracks are planned for an untapped secondary-level target they can often be designed to pass through these small primary-level oil pockets, allowing their production and further improving the ultimate recovery. Most recent production wells drilled now achieve this dual primary-level / secondary-levelpurpose. Re-perforations. Re-perforation is also used for optimising well production as a result of 4D observations. In the example shown in Figure 1 the well was largely watered out, producing approximately 2000 bbls of oil per day. The 4D seismic indicated that the water cut was coming from the middle and toe of the well (beyond a small fault), whereas the heel of the well still lay in unswept oil. As a result of these observations, the middle and toe of this well were packed off, and new perforations were made at the heel of the well. Oil production increased to approximately 5000 bbls per day and the cumulative production increase has exceeded 500 000 bbls - a substantial return for a simple packer/re-perforation job. Although one can also use production logging to find out such information, the 4D can sometimes be much cheaper per well than production logging. 4D also gives a approximate value of the vertical stand-off of the water from the well, and therefore the likelihood of water coning after re-perforation. Injector placement. Most 4D-driven targets are remaining oil pockets, however the improved understanding of fluid flow and connectivity offered by 4D can also be used to improve injector placement or the optimal scheduling of injector well drilling. In the Northern North Sea, we observe not only 4D hardening due to water sweep, but also 4D softening as a result of reservoir inflation. This is due to poorly connected water injectors pressuring up isolated parts of the reservoir (Figures 2 and 3). Infill targets. Shell has demonstrated the value of 4D both in relatively young fields (e.g. Draugen, Norway1), and in more mature fields (e.g. UK Central3 and Northern North Sea4). In a young setting infill well targets can be quite sizeable, whereas in a mature setting, remaining oil targets are generally smaller in volume, and can only be accessed economically using sidetracks or wells with multiple targets. Within many of these fields development is phased, with the most productive high permeability (‘primary’) levels targetted by early wells, and the less productive lower permeability (‘secondary’) levels being targetted by sidetracking of early primary wells which have watered out. This phasing is particularly advantageous for 4D driven development. Due to field maturity, remaining oil pockets at the primary-level are often small and would present uneconomic targets on their own, even for sidetracks. However, while infill wells and sidetracks are planned for an untapped secondary-level target they can often be designed to pass through these small primary-level oil pockets, allowing their production and further improving the ultimate recovery. Most recent production wells drilled now achieve this dual primary-level / secondary-levelpurpose. Re-perforations. Re-perforation is also used for optimising well production as a result of 4D observations. In the example shown in Figure 1 the well was largely watered out, producing approximately 2000 bbls of oil per day. The 4D seismic indicated that the water cut was coming from the middle and toe of the well (beyond a small fault), whereas the heel of the well still lay in unswept oil. As a result of these observations, the middle and toe of this well were packed off, and new perforations were made at the heel of the well. Oil production increased to approximately 5000 bbls per day and the cumulative production increase has exceeded 500 000 bbls - a substantial return for a simple packer/re-perforation job. Although one can also use production logging to find out such information, the 4D can sometimes be much cheaper per well than production logging. 4D also gives a approximate value of the vertical stand-off of the water from the well, and therefore the likelihood of water coning after re-perforation. Injector placement. Most 4D-driven targets are remaining oil pockets, however the improved understanding of fluid flow and connectivity offered by 4D can also be used to improve injector placement or the optimal scheduling of injector well drilling. In the Northern North Sea, we observe not only 4D hardening due to water sweep, but also 4D softening as a result of reservoir inflation. This is due to poorly connected water injectors pressuring up isolated parts of the reservoir (Figures 2 and 3).

Turbidite reservoir compartmentalization and well targeting with 4D seismic and production data: Schiehallion Field, UK

“…We use all the 4D seismic surveys acquired to-date to test and enhance the dynamic reservoir model predictions (cf. Staples et al 2002Staples et al , 2005. 4D seismic (or time-lapse 3D seismic) involves repeat acquisition of 3D seismic after a period of oil, gas or water production and water and/or gas injection.…”

“…This method of conditioning static and dynamic modelling with 4D seismic data, is similar to 4D integrated reservoir modelling (4D IRM) workflows described elsewhere (e.g. Staples et al 2002Staples et al , 2005Lygren et al 2003;Waggoner et al 2003).…”

Seismic History Matching of Nelson Using Time-Lapse Seismic Data: An Investigation of 4D Signature Normalization

“…12-Oil-and water-production rates in matched and forecast periods (light blue) for the best history-matched model using production data only (blue) and using both production and seismic data (green) for the best model of well-based approach using base model. Connolly (1999), Staples et al (2002)]. In such a case, the layer thickness can affect the interpreted change in acoustic impedance.…”

Reservoir Monitoring With Seismic Timeshifts: Geomechanical Modeling For Its Application in Stacked Pay

“…Modeling studies, field cases and operational impact of this technique have been published [see e.g. [1][2][3][4] yet some important rock physics questions remain, notably on the magnitude of acoustic velocity changes during production-induced depletion and fluid injection (see Appendix). The focus of most AI-change models is on the compacting reservoirs, and rock property changes in the bounding non-depleting formations are often not considered.…”