Skip to main content
Article

Using earthquake science to inform business decisions

An afternoon with Willis Re and the Willis Research Network

Reinsurance|Property|Risk Control and Claims Advocacy
N/A

By Dr. Rosa Sobradelo , Crescenzo Petrone and Myrto Papaspiliou | July 12, 2019

Deepening the understanding of the science behind earthquakes through an award-winning collaboration of academia and the risk industry.

For over a decade, the Willis Research Network (WRN), an award-winning collaboration between academia and the risk industry, has been working closely with world-renowned earthquake scientists to deepen the understanding of the science behind earthquakes. Through supporting pioneering initiatives such as the Global Earthquake Model (GEM), we can help our clients get better insights into seismic risk and inform risk management and risk transfer decisions.

Last month, we held a Seismic Risk Seminar to summarize our current scientific research. Three successive sessions provided an overview of a particular research topic, and how such insights can inform Willis Re’s view of risk, and be used to support reinsurance, capital and risk underwriting decisions.

The odd couple – aftershocks and clustering. How can we adjust or build models to account for these coupled events?

Following a large earthquake, the transfer of stress can bring neighbouring faults closer or further away from failure. Events like Christchurch (2010-2011), Central Italy (2012, 2016-2017) and Mexico (2017-2018) have highlighted the significant impact of sizeable aftershocks and triggered events.

Willis Research Network Earthquake Seminar: Volkan Sevilgen

Volkan Sevilgen, and Dr. Ross Stein from Temblor.inc outline their methodology to perform Coulomb stress-based forecasts of aftershock probabilities and locations following large or urban mainshocks.

Through the WRN collaboration with Temblor we can incorporate stress transfer and aftershock risk in our standard Catastrophe Modelling. The joint presentation between Volkan Sevilgen, Dr. Ross Stein, Temblor.inc, Dr. Myrto Papaspiliou, at Willis Re, outlined the methodology and framework established for the rapid creation of conditioned event sets that provide a modified view of risk. Following a significant earthquake, this can then inform critical decision-making - for reinsurance decisions such as back-ups, re-assessing vertical and sideways programme capacity and a technical input to pricing and structuring.

Reinsurance and model validation - Willis Re’s use of Global Earthquake Model data in model validation and development

The Global Earthquake Model (GEM) is a public-private partnership which aims to better quantify and manage seismic risk by creating open-source hazard & risk assessment software and tools, as well as the first global earthquake risk model. Dr. Marco Pagani, the GEM’s Hazard Coordinator, presented an overview of the data, products and tools developed over the last few years.

Contact Us
Willis Research Network Earthquake Seminar: Dr. Marco Pagani

Dr. Marco Pagani presents an overview of the data, products and tools developed over the last few years at Global Earthquake Foundation (GEM).

Tim Edwards of Willis Re, then demonstrated a number of case studies benefiting from the use of GEM products. This includes assisting clients’ reinsurance decision-making in countries not well served by vendor models, such as the Middle East and North Africa. He also discussed how we work with GEM products to inform model validation and develop our own view of risk. Finally, these insights can also support original risk underwriting and portfolio optimisation across Europe.

Use of 3D ground motion simulation to reduce the uncertainty in loss estimates

We looked at a particular Reinsurance, Capital and Regulatory challenge in the US and Canada: A large M9.0 earthquake in the Cascadia Region could create a major loss for the insurance and reinsurance markets, but there remains significant uncertainty about the scale of potential impact. Three-dimensional ground motion simulation techniques have been widely used in academia to assess the ground shaking resulting from an earthquake. Our WRN partner, Prof Kim Olsen from San Diego State University, USA, has been one of the pioneers of this methodology to capture phenomena that current catastrophe models poorly represent. Willis Re has worked with Prof Olsen over the last two years implementing his research for earthquake mega-thrust scenarios that drive tail risk.

With support from Prof Olsen, Dr. Crescenzo Petrone, from Willis Re, presented how such techniques can be used in a loss modelling framework, with a specific focus on M9.0 megathrust scenarios in the Cascadia Subduction Zone, Pacific Northwest, USA. He discussed how this leading-edge modelling can help us better understand the range of losses and reduce the uncertainties associated with such a major event.

Summary

The Willis Research Network presents  leading-edge research alongside clear practical applications for risk management decisions. The seminar provided three examples of ongoing seismic research combined with insight from industry professionals. For copies of the presentations and more information on any of these topics, please visit our website or contact us by email.

Authors

Dr. Rosa Sobradelo
Senior Research Manager

Rosa is a Senior Research Manager for the Willis Research Network where she manages the Earth Risk Hub. That is to say research activities in the areas of geological risks such as tsunami, volcanic and earthquake risk with leading academic partners around the world, matching them with business needs to derive tangible outputs to improve our client services. She also works on consultancy projects relating to volcanic hazard and risk. She is a statistician at heart, with a PhD in Applied Statistics and Operations Research from the Polytechnic University of Catalunya, an MSc in Mathematics, Statistics and Operations Research from New York University, and a BSc in Business Administration from University of Santiago de Compostela, and has made a career out of risk building probabilistic models for Health, Finance and Catastrophe Risk.

Crescenzo Petrone
Myrto Papaspiliou

Related content tags, list of links Article Reinsurance Property Risk Control and Claims Advocacy

Related solutions