Despite the high volume of material out there about induced earthquakes, it can be hard to separate fact from opinion. To help explain what induced seismicity is, how it is caused and what the risks are, a group of researchers from UCL Department of Chemical Engineering and Institute for Risk and Disaster Reduction have published “Addressing the risks of induced seismicity in subsurface energy operations”.

What causes “induced seismicity”? Induced earthquakes, those mainly caused by human action, can invoke strong feelings towards the processes that cause them, the most widely known among these is hydraulic fracturing (less favourably known as “fracking”). But hydraulic fracturing for shale gas extraction is not the only cause of induced earthquakes – several industrial activities are capable of inducing or triggering earthquakes, including mining, dams, conventional oil and gas operations, groundwater extraction, CO2 Capture and Storage (CCS), underground waste fluid disposal and the creation of geothermal energy systems. Rightly or wrongly, negative public perception and local public opposition to induced seismicity has led to numerous international projections having been suspended, delayed or curtailed.

How does industrial operation induce earthquakes? The Earth’s crust is believed to be in a state where it is critically stressed and only small stress changes in the right direction can cause an earthquake. Industrial action can alter the stress field in the most shallow part of the Earth’s crust, inducing a seismic event.

Are these events getting more likely? The number of documented cases of man-made earthquakes in different industrial activities is rapidly increasing: In 2017 alone, there were two reported record-breaking magnitude induced seismic events. One of the more well-known areas of induced seismicity is in the United States: The notable increase in seismicity within the last decade in the previously seismically quiet State of Oklahoma has been widely attributed to large scale waste water injection wells connected to the hydrocarbon production industry.

How big are induced earthquakes? Most induced earthquakes are low in magnitude (typically less than magnitude 4). However, even these small events are capable of causing structural damage to properties and evoking widespread fear and anxiety. We say most are small, but there are some examples where large magnitude earthquakes have been alleged to be caused by human activities. For example, in China in 2008, a dam was built that filled a reservoir behind it. A short time later, a magnitude 8 earthquake occurred in the region. Some scientists proposed this large earthquake was caused by the mass loading of the water in the dam and its penetration into rock, affecting the subsurface pressure in an underlying fault line and possibly setting off a series of ruptures that led to the earthquake.

So what is being done about it? Minimising seismic risk should be a high priority for industrial operators. All fluid injection processes should require detailed seismic hazard assessments for imaging and characterising faults prior to operations, with dedicated monitoring systems in addition to existing national seismic monitoring facilities. For assessing the risks, monitoring the operations, and designing mitigation strategies using predictive models that can characterise the spatiotemporal evolution of induced seismicity would be extremely helpful. Examples of best practice approaches show that maintaining a transparent dialogue between operator and the public, while adhering to the regulatory processes can allow safe operations in an atmosphere of public acceptance.

Where can I find out more? With all the controversy around such events, we need to understand what are the risks of induced earthquakes and how can we model them. In the published article in Wiley Interdisciplinary Reviews, Richard Porter, Alberto Atriolo, Haroun Mahgerefteh and Joanna Faure Walker provide a review of several alleged induced seismicity case studies that have occurred in the last 15 years covering a variety of causal mechanisms. We discuss issues relating to public perception and procedures and strategies that could be implemented to help prevent and mitigate future occurrences.

The above work was funded by Horizon 2020 research and innovation programme, Grant/Award Number: 640979

The Fault2SHA ESC (European Seismological Commission) Working Group hosted a session on Wednesday 5th September at the ESC 2018 Meeting held in Valletta, Malta. Oona Scotti represented the group in her keynote on the opening day of the conference, in which she addressed “Modelling fault systems in PSHA: Challenges Ahead”. The Fault2SHA Working Group, for which I am on the Executive Committee, links different researchers working on faults and seismic hazard assessment (SHA) in Europe and beyond. This collaboration has brought together field geologists, fault-modellers and probabilistic seismic hazard modellers. The group provides a forum in which data, results, modelling capabilities, and improvements in scientific understanding can be shared. If you want more information, and to join, see Fault2SHA. The next Fault2SHA workshop will be in Kaust, Saudi Arabia, in November 2018 and the next meeting will run on 3rd-5th June 2019 in Barcelona, Spain.

I lead the Fault2SHA Central Apennines Laboratory. Our team comprises researchers from Italy (Paolo Boncio, Bruno Pace, Laura Peruzza, Francesco Visini), France (Lucilla Benedetti, Ooona Scotti) and the UK (Joanna Faure Walker, Gerald Roberts). At ESC in Malta, I introduced the central Apennines Laboratory and our current activities to the wider working group. The Central Apennines, as well as being a beautiful place to conduct fieldwork with the opportunity to obtain detailed datasets, suffers from large magnitude earthquakes. Indeed, earthquakes in the Central Apennines have featured widely in the UK press due hosting the two deadliest earthquakes in Europe of the last ten years: the 2009 L’Aquila sequence and the 2016 Amatrice-Norcia sequence.

The Fault2SHA Central Apennines Laboratory, which formed in January 2018, held an in-person meeting in July at the University of Chieti-Pescara, Italy. The photograph shows (from left to right) Oona Scotti, Francesco Visini, Joanna Faure Walker, Bruno Pace, Laura Peruzzi, Lucilla Benedetti, and Paolo Boncio.

During the Fault2SHA ESC session, I presented a second talk and a poster about my research investigating the importance of incorporating detailed fault geometry for understanding seismic hazard. The oral presentation demonstrated the importance of incorporating detailed fault geometry and loading on faults between earthquakes in Coulomb Stress Transfer modelling, a process that causes the stress on faults to change in response to an earthquake on a neighbouring fault. This was based on work carried out by Zoe Mildon (former IRDR PhD student, now a lecturer at the University of Plymouth) in collaboration with Gerald Roberts, Shinji Toda and myself (see Midon et al. 2016 and Mildon et al. submitted preprint). The poster displayed the importance of detailed fault geometry and slip-rate data for calculating earthquake probabilities and ground shaking intensities. I further represented Zoe for her poster within the session on earthquakes in regions of distributed deformation, that showed surface ruptures from the 1997 Colfiorito Earthquake in the central Apennines was due to primary earthquake slip (see Mildon et al., 2016 for details).

I thank all those at the conference with whom I had interesting discussions and I look forward to seeing all of our research progress.