Will enhanced oil recovery be an oil supply savior?
To read ExxonMobil Corporation's website one might get the impression that the world's largest oil and gas company has begun only recently to employ enhanced oil recovery (EOR) techniques. If that were true, this industry bellwether might have been able to say that these techniques will have a substantial effect on the future flow of oil. After all, the claim for EOR is that it could potentially double the amount of oil we can get out of the Earth--from the current one-third to two-thirds or so of the original oil in place. The implication is that not only will future wells yield more of their oil than previous ones, but that far more oil can now be harvested from existing wells.
The big problem with this thesis is that EOR is already being widely applied--so much so that the Oil & Gas Journal will sell you its most recent worldwide survey of EOR projects for only $330. You can get the full historical database all the way back to 1986 for a mere $1,100. (Hint: Both contain more than a few entries.)
The three main types of EOR are gas injection, steam (both cyclic stimulation and flooding), and chemical injection, and they've been around for a long time. The poster child for EOR among the oil optimists is the Kern River Oil Field near Bakersfield, California. Kern was discovered in 1899. As production waned, steamflooding was introduced in 1964. In 1961 production was about 19,000 barrels per day. By 1966 it had risen to 53,000 barrels per day. Production reached its peak at 141,000 barrels per day in 1985. Production continues today at around 80,000 barrels per day.
The Kern River steamflood has proven how well EOR can work in some situations. But as any reader will deduce, the results are already reflected in current production and reserve estimates. Steamflooding has been in use for a very long time.
Natural gas injection is also an old technique used to maintain reservoir pressures. It has been used continuously, for example, at Alaska's Prudhoe Bay Oil Field which began shipping oil in 1977.
Nitrogen injection is newer, but has been used, for example, since 2000 on the huge Cantarell Field in the Mexican portion of the Gulf of Mexico. Results were excellent at first. But the subsequent crash of production at Cantarell has called into question whether this form of EOR merely hastened production without increasing ultimate recovery.
The same issue has been raised by another technique called maximum recovery contact wells, which is often grouped with EOR. The technique worked superlatively for a while in Oman's largest oil field only to lead to a precipitous crash in production later.
Carbon dioxide injection has also been used successfully, but supplies are expensive if they are not near the field. The newest of the major EOR techniques involves inoculating reservoirs with microbes that will make the oil flow more freely. It looks promising.
Oil company sources tell me that indeed these techniques are used wherever practical. Limitations include high capital and operating costs. For example, Cantarell's nitrogen injection system cost $6 billion to build. Other limits may result from the existing infrastructure. For instance, will that infrastructure be able to handle additional production? And, if not, what would it cost to upgrade it?
High costs almost always mean high energy inputs. Even if the capital and expertise is available, it may cost more energy to implement an EOR program than will be gained from the extra oil. Inevitably, the energy return on investment for oil obtained using EOR will be lower, often far lower, than oil obtained using standard methods.
Oil supply optimists often talk about doubling average recovery from oil wells. But B. J. Doyle, vice president of operations for a small Houston-based oil and natural gas exploration company, cautions against such talk. Every reservoir is unique. This means that 1) many reservoirs will simply not benefit from EOR and 2) the increase in recovery when EOR is applied can vary widely.
In addition, Doyle says, there are fields where the techniques won't be applied until small operators take over. After large oil companies get the majority of oil out of a field, they often find it's not worth their while to continue pumping. Frequently, they sell their interests to smaller operators who have the willingness and patience to squeeze out the final barrels using a variety of techniques, some of which wouldn't necessarily qualify as EOR.
In places such as the United States and Canada this happens as a matter of course. That's why these countries have so many operating wells, many of which pump fewer than 10 barrels a day. But this pattern of exploitation in mature oil fields is only happening in these countries because they allow private ownership of oil rights. In places such as Nigeria, Iran and Saudi Arabia, private companies cannot simply purchase or lease mineral rights from the owner because the owner is the government. In Iran and Saudi Arabia, the government has total control over the oil industry. There is no entrepreneurial class of small operators able to take over largely exhausted fields and revive them. That means that in many of the most oil-rich places in the world, EOR on the scale practiced in the United States and Canada will probably never become a reality.
Energy writer Chris Nelder gives us perspective on what we can expect from EOR. He writes that history shows us that EOR "does not affect the date of the peak, nor the peak rate of production. It typically just extends the 'tail' on the back end of the curve and increases the ultimately recoverable total."
Even with all these caveats, I think we can concede that EOR will add to proven oil reserves in the future and likely cushion the decline in oil production after world output peaks. But given the historical record of EOR, it is unlikely these techniques will prove to be the savior of oil production rates that the oil supply optimists would have us believe.
What do you think? Leave a comment below.
Sign up for regular Resilience bulletins direct to your email.