Two weeks ago, I attended the SPE-CSGM (Society of Petroleum Engineers)-Canadian Society of Gas Migration workshop on “Gas Migration Challenges – Identification and Treatment,” in beautiful Banff, Alberta.
I take Banff for granted as I live only an hour away, but it’s a very nice location to have an event. We even had pretty good weather, but unfortunately we were inside for the workshop for two days.
The workshop was the first on the topic of gas migration and surface casing vents. It brought together industry professionals working on these issues, including regulators, industry experts, service companies, and consultants. The two-day workshop provided extremely informative content and discussions surrounding the often-not-discussed topics of gas migration (also known as stray gas in the USA) and surface casing vent flows (SCVF). Both topics are moving toward increased regulator scrutiny, and require some fundamental approaches to measurement, assessment, diagnosing, and repair.
As part of the conference, CMI's presentation, “Characterizing the source zones for surface casing vent leaks,” summarized fundamental best practices on sample collection (liquid and gas) and analysis, as well as how to use routine and advanced geochemistry to diagnose a SCVF source for effective remediation. The talk was well received and attendees asked many excellent questions. The following questions and answers were the primary focus following the talk.
Q&A from SPE-CSGM
In your presentation, you discussed how samples degrade, thereby, changing the isotopic composition. How often do you run into sample degradation?
Sample degradation depends on a number of factors, such as geographical location, soil types, water chemistry, flow rates, etc., but overall we generally see it in approximately 10-20% of our surface casing vent samples and in more than 90% of our soil gas samples.
If gas migration samples have been isotopically altered, are they still useful for determining the potential thermogenic source of the gas?
A single ‘degraded’ sample will not provide very much insight toward determining the gas source. In geographical areas with a high degree of microbial activity, we typically need to collect more data and develop a geochemical model for proper interpretation. Typically, a gas migration sample will not match a deeper thermogenic source; however, if you characterize the ‘end members’ from potential source zones in the area, you can model the source zone for a degraded sample. In addition, gas migration samples do help determine if the gas is natural or biogenic in origin, but again, this requires a bit of caution.
On slide 15 of the presentation, how did you confirm that the Cretaceous zone was the source of the surface casing vent leak?
The first line of evidence that helped to identify the Cretaceous zone was the water chemistry. The second line of evidence was the isotope signature of the gases, which also implicated a Cretaceous zone. We were able to pinpoint the source zone to within a 100-150 meter interval in the Cretaceous using a mud gas isotope log from a drilling program we conducted in the local area. Methane is more sensitive to isotope fractionation, so we used the isotope signature of ethane from the surface casing vent sample and the mud gas isotope log to identify the source zone.
On slide 20, there is a cross plot of two different gas isotopes for the surface casing vent samples. How do you know they came from the same zone?
We know the two different gas isotopes for the SCV samples came from the same zone because they clustered in the same area of the cross plot. This cross plot was geochemically different compared to the commonly implicated source zone thought to be responsible for the SCV leaks in the area. The geochemical differences between the actual source zone and the implicated source zone were small, therefore, it required very high-quality samples and analyses to characterize the correct source zone. In this case, the source zone was an overlying zone, which is why there had been many failed attempts to stop the leaks.
In your experience, how reliable is soil gas data for interpretation?
Soil gas data are highly variable and dependent upon location in province. If you’re sampling from a dry, highly porous, sandy soil, soil gas data is generally good. Gases are altered as they pass through tight clays and wet soils, where greater residence times in the soil during migration, increases the likelihood that the isotope signatures will be altered before they reach the surface. As mentioned in the presentation, sampling issues can also play a part in making interpretation of soil gas data very difficult.
What kind of data would you need to increase your confidence in soil gas data?
The first step to increasing confidence in soil gas data is to ensure your potential source zone ‘end-member’ data is from the local area. End-member data can be production casing samples, down-hole samples from a targeted zone, or a full mud-gas isotope log from the area. You can model potential source zones and production zone data to de-convolute the changes caused by microbial degradation in the soil gas samples. If model zones are implicated, you can take it a step further and do mixing models to help determine the potential sources.
In the complex case studies you described in your presentation, where was all the data acquired from?
The data for these studies were acquired from several different studies we designed and conducted on the same production unit. This included a mud-gas isotope log (MGIL) that provided gas isotope data, from surface to depth, in the local area. Additionally, we did a geochemical characterization of the production zone across the entire field (both gases and fluids). Then, we characterized overlying deeper saline aquifers with DST sampling approaches from source water wells, and supplemented this data with shallow groundwater monitoring well samples to retrieve gases and liquids from the shallow zones.