The debate, which began raging in the early 2000s, about whether wastewater from production that was being injected into subsurface reservoirs caused seismicity – the occurrence or frequency of earthquakes in a specific region – has been largely settled.
“It really started to get settled by 2016 when, for a couple of reasons, the injection rates started to come down in Oklahoma, and the earthquake rates came down, too. Then the same thing was observed in the Fort Worth Basin at about the same time,” said Peter Hennings, research professor and principal investigator for the Center for Injection and Seismicity Research at the Bureau of Economic Geology at the University of Texas’ Jackson School of Geosciences.
And the fact that those earthquake rates came down after injection rates started decreasing was more illuminating than any local relationship observed between injection and specific quakes.
More on that in a moment.
The debate is settled, for the most part, but interest in the topic is still high. Hennings and his team published a series of papers on the relationship between wastewater injection and seismicity in the December issue of the AAPG Bulletin, which caught the attention of numerous publications – both popular science and peer-reviewed – last month.
A Bad Rap for Frac’ing
Before the connection was made between injection and seismicity, many thought it was the actual drilling or hydraulic fracturing that was causing the earthquakes.
Hennings sympathized with the logic.
“It was understandable. And there are cases, pretty well-known cases, of hydraulic fracturing that triggered seismicity in places like the Alberta Basin in Canada, and in Texas, especially in the Eagle Ford region in south Texas. From a lay person’s perspective, that’s an easy thing to understand. You know, drill a well, and then you use this word ‘fracking’ and then that whole thing doesn’t sound all that far away from earthquakes conceptually,” he said.
It was discovered, though, that quakes were more from the aftereffects of drilling rather than anything specifically related to the drilling itself – and even that came with a caveat.
“When injection rates go up, month after month, then year after year, then the earthquake rate goes up,” said Hennings.
That’s compelling evidence, he said, but since there’s generally a lag between the injection and the earthquake activity, the connection is somewhat tenuous.
“But when you see it on the other side of the curves – the downward trends, that’s when it becomes convincing,” he said.
He is talking about when such injection is curtailed – and this was happening both in Oklahoma and Texas, primarily the Fort Worth Basin – that the relationship between the two was more conclusively linked.
“The down trajectory really is what gave the whole thing conclusive evidence,” he said.
He said the lessons learned, in retrospect, came from both the doing – the injection, and the undoing – the curtailing of such activity.
“When you produce fluids from the subsurface you learn things; when you inject fluids into the subsurface you learn much more. Earth is a mechanical system and every action gets a reaction,” Hennings explained.
For quakes triggered by wastewater injection, Hennings adds that it is useful to take the full measure of the problem and consider the induced earthquake system: “the earthquakes induced by a specific set of subsurface operations that interact uniquely with extant geologic elements in an understandable spatiotemporal context.”
Once the link was established in those early areas, regulatory controls were put in place and more recently, declines are now seen in the monthly rate of problematic earthquakes across a number of impacted regions (northern Delaware Basin, central Midland Basin and Midland Basin Eastern Shelf).
The Permian Basin of west Texas and southeast New Mexico, Hennings reminds us, had experienced a huge breadth of induced seismicity that had been linked to a variety of operational causes and influences. Specifically, the rate of earthquakes of local magnitude 3.0 or higher in west Texas and southeast New Mexico increased greatly in 2017 as compared to historic norms. This change in earthquake rate spatiotemporally coincided, in general, with the widespread implementation of horizontal completions of unconventional wells and the attendant and significant increase in the rate of injection of wastewater coproduced with oil from the shale (salt-water disposal).
“Much has been observed, studied and written concerning induced seismicity in the Permian Basin region as the earthquake data and the causal links from public and private sources have evolved. The links between operational activity and spatiotemporal earthquake trends have provided evidence of classes of causation that have become widely accepted,” he said.
Toward a Comprehensive Understanding of Induced Seismicity
While the recent data is welcomed, many operational and scientific questions remain that have implications for the environmental and economic sustainability of Permian Basin production.
Hennings said it will take more integration of geoscience disciplines using copious and quality data to understand the local and specific relationship between injection, earthquake rate and earthquake magnitude.
“In the public and/or academic sphere, it’s generally not easy to assemble the human capital, data and information needed to fully address the issue. Many published papers have worked to address the issue by using the lens of the seismicity itself, perhaps combined with some measures of injection, but this does not address the underlying geologic nature of the system that hosts these earthquakes,” he explained.
One of the big challenges in fully understanding the connection between induced seismology, both its increase and decrease, is that it doesn’t affect all regions in the same way.
“One size does not fit all because each induced seismicity system exists within an ecosystem with a unique geological framework and operational influence. This is not a static thing as the ecosystem evolves as the geology reacts and operational footprint changes.”
There are a number of moving parts.
To that end, the Bureau of Economic Geology has spent its more than 100-year history preparing itself to address these sorts of integrated challenges and why, about 10 years ago, the Bureau, under Scott Tinker, established TexNet, an independent scientific body that monitors earthquake activity, analyzes associated data and distributes findings and data to government, industry and the public for their benefit and the benefit of the State of Texas, and CISR, which conducts leading science and application on the impacts of large-scale injection on the subsurface and surface environment (see accompanying article).
Hennings now said these entities have the people, the data and the experience to dive deeper and more broadly into these sorts of challenges.
Such challenges include finding out exactly what controls seismicity in specific areas.
“Is the seismogenic fault rupture, as triggered by injection, occurring along major faults in the geologic basement where strong rocks are involved and earthquake moment can be larger, or does the rupture occur on smaller faults in the sediments where weaker rocks are involved and the moments are smaller?”
Risk-Benefit Trade-off
There is also the question – not a small one – of how much risk we can tolerate. For instance, none of the cases of seismicity caused by such injections has resulted in the loss of life or significant structural damage.
“What can be said is that with regulatory actions taken by the Texas Railroad Commission, the New Mexico Oil Conservation Commission and the Oklahoma Corporation Division, along with proactive industry cooperation, the rate of injection into the reservoirs that carry the greatest hazard of triggering earthquakes has decreased, as have the rate of larger earthquakes,” said Hennings.
That’s the good news.
The bad?
There will still be seismic activity.
“We should not assume that this will be the last of them, particularly given that the rate of deep injection at high rates continues in some seismically-prone areas,” Hennings said.
Ultimately, he said, the cases of reduction of deep injection in Oklahoma, the Fort Worth Basin and several areas of the Permian Basin and the observed reduction in earthquakes provide powerful lessons and positive results – it just won’t happen overnight.
After injecting these saline fluids into the subsurface, which adds energy to the system by pressurizing the reservoirs, it will take time, Hennings said, “for the elevated pore pressure to diffuse away and lower the earthquake triggering potential.”