StatoilHydro 4-D projects 2006 and 2007 with feathering rating. (Images courtesy of RIL)

When the idea of 4-D seismic as a reservoir imaging and monitoring method was first debated in the mid 1990s, a key concern was the question of repeatability, particularly in the case of surveys carried out using towed streamers. Was it really possible to return to exactly the same offshore location as the baseline survey and be able to accurately replicate all the circumstances of the original acquisition operation?

In the marine environment, currents, tides, weather and different equipment are just some of the variables that have to be considered. Yet if the seismic imaging has to show interpretable differences in the reservoir over time during production, then it was crucial that such changes could be measured accurately without any fear of distortions from extraneous factors.

Today the early skepticism over repeatability of 3-D time-lapse surveys or 4-D has been largely overcome, and towed streamer acquisition of baseline and monitor surveys is the established norm. A lot of the improved confidence in 4-D results stems from the development of effective quality control (QC) strategies which address the repeatability issues, of which two — use of a GIS database and a recently introduced QC system — are covered in this article.

Application of GIS

Making use of a QC system based upon a GIS database has proved to be a surprisingly valuable and relatively straightforward way of helping oil companies, contractors and consultants understand the repeatability challenges and how these can be resolved. For example, the database developed by Reservoir Imaging (RIL) in collaboration with a number of oil companies provides details of numerous 4-D seismic baseline and monitor surveys registered in an accessible form that allows comparisons to be made quickly and easily.

Figure 1 illustrates the top-level view of surveys carried out by StatoilHydro over the last two years, showing the field locations and a measure for each one of the variability of the feathering (currents affecting streamer alignment) experienced in each individual seismic project.

From this top-level view summary attributes can be analyzed and comparisons can be made by area or by acquisition technology, etc. When planning the next year’s campaign, the GIS database becomes a valuable source of information. For instance, it can be used during the contract tender phase to assess the likely success of any particular project proposal and how differing geometries and technologies are likely to impact the final 4-D seismic results based on the legacy information in the GIS database. Over time, as more survey information is added, the database will become increasingly valuable, providing some consistent measure of the factors which impact a 4-D seismic survey and hence the likelihood of success for any planned survey.

As well as looking at high-level data, the database can be used to drill down into each individual area and 4-D survey. At this project level, typical 4-D attributes including source position differences, feather differences, receiver differences, etc., can each be measured from the baseline through to monitor surveys. Other factors that can impact the success of a 4-D project such as topographical issues, tide/current information, local cultural data, field infrastructure, etc., can be introduced. In addition, attributes computed during QC of the acquisition data or in the processing stage may offer further insights into the quality of the 4-D project from a seismic imaging perspective.

Historically this type of data has been stored in separate proprietary systems, which makes the correlation of relevant data difficult and therefore limits its usefulness. However, once integrated within the GIS database, data can be mapped together and be used for statistical analysis and correlations. From a personnel point of view, the integration of all the data makes for a much better collaboration between all the relevant parties (oil company clients, contractors, etc.) needed to make decisions and interpret results from the data.

Taking the collaboration to the next level, most GIS servers have the ability to publish data directly to the Web, allowing secure access to/from a Web browser. This has a practical benefit for the QC process during a survey as data is acquired. Processing can be carried out in the field or remotely from the shore to compute the 4-D repeatability statistics and then be published on a Web page so that all the interested parties in the ongoing survey can monitor the status of the project and get involved in any decisions required to plan infill. Once the survey is complete, the field data can be integrated into the master GIS database.

The GIS-related database is just one approach to the QC of 4-D seismic. Some recent development work has focused solely on the positioning errors likely to affect the survey repeatability needed. Much of the early effort was put into determining how positioning errors impacted 4-D seismic results. Early 4-D seismic survey acquisition experience suggested that much of the predictable noise was being introduced by positioning errors. In other words, without some management of the occurrence of positioning errors, 4-D seismic project results would be compromised, and many fields could not be turned into potential candidates for 4-D survey monitoring.

QC Methods

Positional repeatability is a measure of the positioning differences between the baseline and monitor datasets. The conventional method for measuring relies on an algorithm starting with the position data (P1/90) for both baseline and monitor datasets. Both datasets are binned, and then a bin-by-bin matching algorithm computes the 4-D repeatability attributes for each bin. The results can then be mapped to give a view of repeatability over the entire survey. Figure 2 is an example of a repeatability map showing the difference in receiver position for a 3,300-ft (1,000-m) offset. On the map green indicates areas where there is good repeatability, yellow is marginal, and red is repeatability that has exceeded the normal levels of acceptability.

This type of map gives a good view of the likely impact on the final results based on the positioning repeatability, but there has always been a difficulty using these maps for acquisition QC, which StatoilHydro has been instrumental in addressing. The problem arises from the fact that acquisition QC is focused on ensuring that the survey is being acquired as per the contracted plan. Although this is in principle related to the baseline survey, all sorts of later decisions, such as geometry and other operational changes, can complicate detection of positional errors. One example of how repeatability can be compromised occurs when a decision is made during the design and planning phase of the project to merge two baselines. This is common and tends to happen where there have been reshoots or infill and the merge will reduce the monitor survey line count and hence survey costs. But the effect is to create areas of “designed-in” non-repeatability.

StatoilHydro method

In 2005/2006 StatoilHydro collaborated in a technique to overcome these repeatability issues that impact the overall results. The StatoilHydro method differs from the conventional approach by making comparisons against preplot positions rather than baseline data and by using shot-based results for QC purposes. The benefit of using preplot positions for the analysis is that these are the result of the design and planning phase of the project and have built in all of the decisions made during that process.

The StatoilHydro method also steers away from a bin-based approach to QC and instead analyzes the data in the shot domain. This is a relatively simple method for obtaining a statistical measure of the quality of any particular 4-D shot. It can be applied over a line, then an area, and finally over the entire survey. At the end of the process quality values for each shot have been registered, and these quality values give a simple set of values covering both the source and receiver repeatability. As these measurements are related to the shot position, they are ideal for acquisition QC, giving navigators an uncomplicated view of the quality of the 4-D acquisition without needing to have a detailed knowledge of the design and planning decisions. Figure 3 shows the same data as in Figure 2 but using the StatoilHydro method. Using the new plot it is far simpler to see where the acquisition has deviated from the plan and could benefit from some targeted 4-D infill.