Intelligent Technical Benchmarking of In-Situ Oilsands Projects
Benchmarking is a frequent exercise in oil and gas analysis. Operators and investors want to know how projects compare to peers, which projects present good investment opportunities, and what performance expectations to expect for future development.
No two oilsands projects are the same. Being able to compare well pads and projects to each other, can be a challenging task. Luckily, there’s visual analytics. In this post, I will demonstrate how understanding the physics of the process, and making some intelligent adjustments to production plots, can improve performance analysis.
Steam Oil Ratio
The most popular metric for evaluating in-situ oilsands projects is Steam Oil Ratio (SOR). In the simplest terms, a thermal in-situ oilsands project consists of: generating steam, injecting said steam into the reservoir to heat and thereby lower the viscosity of bitumen, followed by producing the warmed bitumen. With SOR defined as the volume of steam injected into the reservoir divided by the oil produced, SOR is a clear, reliable metric of efficiency. As shown in prior blogs, Greenhouse Gas (GHG) emissions intensity is directly proportional to SOR. Almost anything else you do to improve the process is a rounding error. Thus, if you care about GHG emissions, SOR is also the best metric for comparison.
But should SOR be the only technical benchmark used to compare projects? Depending on your purposes, SOR may only tell part of the story.
Making Lemonade from Lemons
By far, the most important consideration for predicting technical in situ oilsands performance is Geology. Using the Lindrain equation from gravity drainage theory, every factory that impacts the prediction of performance is dictated by what is left underground.
- The basic reservoir parameters (thickness, porosity, permeability, saturation, and thermal diffusivity) are fundamentally a function of geology.
- There is some control over viscosity via on injection pressure and temperature; or using a different process such as solvent. That said, maximum operating pressure is still dictated by the fracture pressure of the overburden and pressure of gas and/or water zones. The density and viscosity at reservoir temperature, is not a choice.
- Even well length is dictated by the geology. There is flexibility in where and how long an operator drills their wells, but the reservoir architecture still ultimately dictates well placement.
Thus, comparing raw SOR performance between operators is like comparing basketball skills between people of different heights.
“Who has the best performance?” might be the wrong question, when we could be asking “Who is making the most of the reservoir they have?” Another important question is, which pads are under or over performing? If the two pads are of unequal reservoir quality, simply comparing rate and SOR is insufficient; we cannot tell if the performance is due to better reservoir, or better operations.
How to Account for Geology when Benchmarking
To properly compare projects, patterns and well pairs, you need a database of well pairs and geological parameters, as well as the processes in place to combine quantitative petrophysical data, core data, production data and qualitative reservoir quality assessments. Fortunately, GLJ can leverage over two decades of experience in the in-situ oil sands space.
Building the data pipelines to combine these different data sources into a digestible format is not a trivial task. Like many oil and gas professionals, I had been doing it manually in excel for years! However, with advancements in automation and data visualization, combining 20 years of data gathering is no longer such a daunting task.
Simply integrating the original bitumen in place, into the production performance analysis will vastly improve the ability to compare different pads and projects. For example, the following is a plot of production, normalized as rate per well pair, and to the first month of steam injection.
From this plot it looks like we have got a mix of good and poor performers; though it is difficult to benchmark based on rate alone. Not all the wells are the same length, so we could normalize to rate/well length, but we can kill multiple birds with one stone.
One of the easiest ways to improve this analysis is to include volumetric data. By plotting recovery factor versus the pore volume of steam injected. In doing this small adjustment, we now have normalized multiple parameters — such as, porosity, saturation, thickness in addition to drainage area (well length and inter well spacing).
Using this plot, we can see that most of the pads are tracking each other closely, with two of the pads historically underperforming. Because of the normalization, we know that the performance is due to something other than average petrophysical parameters.
We can also compare different projects, using the same metric. This allows for an instant comparison of performance, that is automatically adjusted for petrophysical parameters and drainage area. With this analysis in hand, we can now start to investigate what else is different between the different pads and/or projects.
The ability to rapidly normalize performance analysis for different input parameters, vastly improves the ability to benchmark different in situ oilsands projects. The example above shows how something as simple as grouping wells by drainage pattern and augmenting the production/injection data with some strategic normalization, can improve your analysis.
Feel free to reach out to GLJ if you are interested in a more in-depth and personalized analysis.
Continue receiving industry news, forecasts, updates and general communications.Sign me up
Interested in hearing more about who we are, what we do, and how we can help – we’d love to chat more.Get in touch