oa Closing-the-Loop Between Reservoir Models and Seismic Gains Through the Pains

Starting in 2007, Sarawak Shell has made a concerted effort to include 3D Close-the-Loop (3D CtL) into our reservoir modeling workflow. The idea behind 3D CtL is to compare the static and dynamic model, in time, to the seismic interpretation and seismic data to ensure consistency. At the foundation of this workflow is an understanding of the rock and fluid properties required to transform reservoir properties into acoustic properties, which are used to convert the depth model to time and generate its seismic response. We have sought to make 3D CtL an integral part of the modeling workflow – resulting in robust models that are consistent with all subsurface data, including the seismic. The expectation is that 3D CtL would be applied iteratively while building the static model (framework and properties) in order to correct obvious mismatches between model and seismic. We have gained experience applying 3D CtL for both carbonate and clastic fields. These gains have not been without pains along the way. One of the challenges has been the acceptance of 3D CtL. Some see it as “extra work” that slows down the project and makes it more difficult to meet project milestones under the continual pressure to do things faster and faster. Those who recognized the value have accepted CtL readily, quickly becoming self-sufficient in its application. Currently, most new static model builds are planning for CtL in the work, as including it early prevents more rework later when a model is near completion. An early hurdle to surmount was defining a methodology to quickly “digest” the visual comparison of the synthetic and seismic volumes to determine meaningful updates to the static model. What should one be looking for? Most of our static models have stacked reservoirs or a thick reservoir interval, which encompass many seismic loops. It is very easy to get lost in the details. First it is important to determine which aspects of the seismic are the most reliable, and where, then only focus on these when comparing synthetic and seismic. A step-wise methodology has been defined to focus our analysis and link observations to specific changes to make in the model. The idea is that one would only proceed once each criterion was satisfied. The steps are: 1. Address inconsistencies at well locations using criteria 2 and 3 (before moving away). 2. Compare the predicted model surfaces in time with the equivalent interpreted time events that were used to build model. The time-thickness of the model “zones” should be consistent with the time interpretation. Mismatches are directly related to rock volume. Larger mismatches are most likely due to layer thickness. 3. Examine loop-to-loop alignment (peaks to peaks and troughs to troughs) within the reservoir packages – the character tie. Misalignment can occur due to inconsistency in the thickness of seismically resolvable internal layers, or the vertical distribution of properties. 4. If seismic is of high quality, extract amplitude maps from synthetic and seismic and compare. The trends should be similar and maps can be used to determine if the lateral property variations are honoring the seismic information. 5. The process of “digesting” the differences provides the opportunity for dialogue between seismologist and geologist to more completely discuss how the model uses all the inputs – from seismic interpretation to usage of seismic inversion results, and linkage to geology and petrophysical evaluation. One of the surprising outcomes of 3D CtL has been an independent check of the petrophysical<br>evaluation of porosity, Vshale (net-to-gross), and saturations. These properties are used in the static model – and are, therefore, the basis for determining the rock property regressions needed to transform reservoir properties into acoustic properties. On several occasions, the process of determining rock properties identified inconsistent petrophysical evaluation between wells, which was corrected before being used in model. Examples and lessons learnt along our CtL journey will be shared to illustrate key points.