Since most CCS projects are experimental there is very little data available to calculate the risk of CCS-induced earthquakes. But there are similarities between CCS and wastewater injection, which is used to dispose of waste fluids from oil fields. James Verdon from the University of Bristol, UK, analysed 11 wastewater injection case studies, each associated with some seismicity, to see what he could learn about the risk of induced seismicity from CCS.

For each case study, Verdon analysed the spatial distribution of the seismicity, and considered the influence of the local geology and the way in which the fluid was being injected. He found that 99% of seismic events occurred within 20 km of the injection well, and that the majority of events took place below the target reservoir.

"We're still not completely clear as to why most events occur below the injection level," said Verdon, who published his findings in Environmental Research Letters (ERL). "However, it is a good thing because events above the injection level would raise concerns that carbon dioxide was leaking out of the target formations and into overlying layers, or to the surface. Events occurring well below the reservoir pose no such concerns."

So what makes some wells more prone to seismicity than others? Verdon showed that seismicity can only occur when there is a nearby fault, which is close to critical stress.

"This [fluid injection] process is not creating earthquakes from scratch," he said. "There has to be a fault that is fairly near to triggering, and the injection activity pushes it over the edge."

Verdon also found that faster fluid injection is more likely to lead to higher pore pressures, which increase the risk of seismicity being triggered. Finally, he noted that injection into harder rocks runs the risk of triggering larger seismic events than injection into softer sediments. "Harder rocks can support higher stresses, which leads to larger events when they fail," he said.

Verdon calculates that the worst-case scenario would be for CCS injection to trigger an earthquake of around magnitude 5. "This is assuming that all the volume change put into the ground is released as a single large event," he said. "In reality, most of the time, most injection into the ground produces little or no seismicity."

Based on his findings, Verdon has two main recommendations for reducing the risk of induced seismicity at CCS sites. The first is to carry out thorough site characterization and geological appraisal prior to injection, to map any faults that might be re-activated and calculate the kind of stresses necessary to trigger these faults to move. And the second recommendation is to monitor the area during the injection process.

"If injection begins to trigger a fault it might be manifest in smaller events that can be detected by sensitive monitoring apparatus," he said. "This information could be used to change the injection programme, perhaps by reducing the injection rate or drilling a new well into a different part of the reservoir."

Verdon thinks that it is important that the issue of induced seismicity is taken seriously and communicated honestly to the public, but he also thinks that this has to be balanced against the risks associated with climate change.

"There will always be a risk that CCS triggers seismicity and we should do the very best we can to appraise and monitor sites to prevent this happening," he said. "But the risks of not pushing ahead with CCS far outweigh the risks of doing it."

Related links

• Significance for secure CO2 storage of earthquakes induced by fluid injection, James P Verdon 2014 Environ. Res. Lett. 9 064022
• James Verdon

Related stories

•  What is carbon capture and storage? James Verdon explains in less than 100 seconds
• A world of trouble
• Carbon capture technology could be vital for climate targets