Frank Foster Evison (1922–2005)
Geophysics – earthquake prediction
Frank possessed a strong belief that scientists have a duty to society and that reliable earthquake prediction would help minimise loss of life and suffering.
Frank Evison was a pioneer in the field of earthquake prediction, yet when he was born in 1922, we were still 40 years away from an understanding of plate tectonics. During his lifetime, our understanding of earthquakes improved dramatically, and international optimism about their prediction peaked in the 1970s before eventually waning in the 1990s. Despite this, Frank’s commitment never faltered. He believed passionately that, as a scientist, he had a duty to society, and his dedication to producing a reliable method of earthquake forecasting continued until his death in 2005.
Frank was born in Christchurch, and he lived there with his parents and siblings until he was 15. His mother was a trained teacher, and before Frank started school at 6, she taught him to read, write and do basic maths. Frank continued to do well academically once at school and was often top of his class.
In 1937, Frank’s family moved to Wellington, and he started at Wellington College. Frank stayed in Wellington for his university studies and graduated from Victoria University of Wellington with a BSc in physics in 1944 and a MA with Honours in mathematics in 1946. Although his studies focused on science, Frank was also very interested in economics, English and history and completed papers in these subjects.
After the war, during which Frank served on several coastal radar stations, he worked his passage to the UK as a donkeyman on a ship. It was in London that Frank met his wife Joan and started to specialise in geophysics, gaining his PhD from the University of London.
On his return to New Zealand, Frank spent a number of years as a government scientist with the Department of Scientific and Industrial Research (DSIR). During this time, Frank made one of his most well known discoveries – coal-seam guided S waves, eventually renamed ‘Evison waves’.
In 1967, Frank began an academic career and was appointed inaugural Professor of Geophysics at Victoria University. Shortly after this, Frank began his research into earthquake forecasting. Many aspects of his upbringing, personality and academic background were evident in the way he tackled his research:
- Respect for data: Frank did not belong to the computer generation, and although he did use computers when necessary, he had learned to carry out data analysis by hand and was able to find precursory swarms by examining printed earthquake catalogues. As a result, he believed strongly in an empirical approach to test earthquake prediction models and was pioneering in his rigorous testing of models.
- A questioning nature: Frank was very articulate and loved to argue and debate. He pushed others to defend their ideas and question the prevailing view of science. Frank argued passionately for his ideas but was also willing to admit when he was wrong.
- Belief in collaboration: Frank travelled extensively as part of his research and was very well known overseas. He believed strongly in true international collaboration and also served on a number of international committees. His role in establishing the Institute of Geophysics at Victoria University is also evidence of his belief in the importance of collaborative research.
Towards the end of his life, Frank was frustrated that he had not made more progress towards his goal. However, true to his nature, Frank remained focused on the future and remained committed to the idea that, at some stage, accurate earthquake prediction would be possible. Frank’s legacy continues in the scholarship that was established in his honour in 2006 and the international symposium in 2008 that brought together leading scientists in earthquake forecasting.
Robert Cross, VUW Image Services
The Wellesley Club Inc
Changing scientific ideas
Each specialised field of science has key ideas and ways of doing things. Over time, these ideas and techniques can be revised or replaced in the light of new research. Most changes to key science ideas are only accepted gradually, tested through research by many people.
Advances in science and technology
All scientists build their research and theories on the knowledge of earlier scientists, and their work will inform other scientists in the future. A scientist may publish hundreds of scientific reports, but only a few are mentioned here.
This part of the timeline outlines just a few events in the personal life of the featured person, some of which influenced their work as a scientist.
Before the 1960s, it was thought that continents were set in the same position forever. The realisation that the Earth’s plates are dynamic revolutionised the study of earthquakes.
Significant improvements in technology (particularly in communication and travel) make earthquake data much more uniform and readily available. This makes it much easier to look for patterns and leads to an increase in forecasting efforts.
Earthquakes can be predicted
The predominant scientific view in the 1970s is that earthquake prediction is possible.
Earthquakes cannot be predicted
A less optimistic view prevails, and the international research focus starts to shift from earthquake prediction to damage mitigation.
Increasing public demand for information
The rise of the internet and mobile phone technology increases public demand for information, especially following a large earthquake. This increases pressure on scientists to provide accurate short-term forecasts.
New technology renews optimism
The prevailing view is that earthquake forecasting methods will gradually improve due to new and better data streams (enabled by modern technology) combined with improved understanding of the physics of earthquake generation.
The first seismograph in New Zealand is installed in Wellington. Seismographs measure and record information during earthquakes.
Image acknowledgement: Te Ara
Elastic rebound theory
After the San Francisco earthquake, HF Reid develops a theory that earthquakes result from the sudden elastic rebound the sides of a fault driven by of previously stored energy. This theory underpins many long-term forecasts in the years to come.
Creeping faults don’t tend to have large earthquakes. Frank organises the building of a wall across the Alpine Fault to see if it’s creeping.
Image acknowledgement: Peter Knoop, Creative Commons Attribution ShareAlike 3.0
Japanese prediction plan
A 5-year plan with the goal of accurate earthquake forecasting is launched in Japan. Methods to be explored include observation of tides, crustal deformation and seismic activity as well as rock testing.
Expansion of seismograph network
As director of the Geophysics Division of DSIR, Frank organises major upgrade and expansion of New Zealand seismograph network.
Image acknowledgement: Te Ara
Interest in precursors
Frank uses first portable seismographs in NZ to compare mechanisms of main shock and the aftershocks of the Inangahua earthquake.
Image acknowledgement: GNS Science Limited, Lloyd Homer
Tsuneji Rikitake publishes key paper suggesting use of a variety of geophysical precursors as a strategy to predict earthquakes.
Haicheng prediction in China
Using a sequence of foreshocks, scientists predict the Haicheng earthquake and evacuate the city, saving thousands of lives. Some scientists do not view this as a true prediction – rather, a very lucky coincidence.
Precursory swarm hypothesis
Frank’s first attempt at a forecasting model based on idea that swarms of earthquakes act as precursors to main-shock events. Frank sees predictive potential of these swarms and begins to work with statistician David Rhoades.
Predicting earthquakes in the USA
The National Earthquake Hazards Reduction Program (NEHRP) is launched in the USA with a focus on earthquake prediction techniques.
Seismic gap theory – McCann et al.
A seismic gap is a period of inactivity along a fault that has been seismically active in the past. Many scientists (including McCann et al.) theorise that the likelihood of an earthquake increases with the length of seismic gap.
Generalised precursory swarm hypothesis
Based on a study of Japanese earthquakes, Frank develops a more complex version of his first prediction model. He hypothesises that clusters of precursory swarms of earthquakes are followed by clusters of main-shock events.
Code of conduct for scientists
Frank is involved in the drafting of an international code of conduct for scientists involved in earthquake prediction and becomes even more committed to rigorous testing of prediction models.
Parkfield prediction experiment
Scientists Bakun and Lindh predict that a moderate-size earthquake will occur at Parkfield, California, between 1985 and 1993. (A large earthquake did occur but not until 2004.)
Earthquakes cannot be predicted
Geller et al. publish a paper in Science claiming that earthquakes cannot be accurately predicted. They urge investment in earthquake-resistant structures and tsunami warning systems rather than earthquake prediction.
‘Tail wags the dog’ method
Vladimir Keilis-Borok and his team at UCLA claim to have successfully predicted two earthquakes in the USA and Japan. A subsequent publicly announced prediction of a large earthquake in California proves to be a false alarm.
Precursory scale increase phenomenon
Frank and David Rhoades publish their work on the precursory scale increase phenomenon. They provide 47 examples of an increase in seismicity before large earthquakes in California, Greece, Turkey, Japan and New Zealand.
EEPAS forecasting model
David and Frank develop the EEPAS (every earthquake a precursor according to scale) forecasting model based on the precursory scale increase phenomenon. The model is tested and later used in operational forecasting in New Zealand.
Formation of CSEP
CSEP (Collaboratory for the Study of Earthquake Predictability) is established and promotes renewed international collaboration and rigorous computer testing of earthquake prediction models.
Work continues on the EEPAS model
David Rhoades continues to apply the EEPAS model to catalogues of earthquakes around the world with the goal of increasing the strength of this model.
Scientists on trial in Italy
Six Italian scientists and one government official put on trial in Italy for manslaughter after failing to predict the 6.3 magnitude earthquake in April 2009 that caused the deaths of 309 people in the Italian city of L’Aquila.
Frank Foster Evison is born in Christchurch where he lives with his family until they move to Wellington in 1937.
Image caption: Roger, Frank and Harry Evison, Christmas 1927
Image acknowledgement: Harry Evison
Life in Wellington
Attends Wellington College. A love of tramping, skiing and mountaineering develops.
Image caption: Rusty Rawlings and Frank Evison (on right) on Mount Duff 1951
Image acknowledgement: Harry Evison
Graduates from Victoria University of Wellington with a BSc in physics in 1944 and a MA with Honours in mathematics in 1946.
World War II
Serves in the Royal New Zealand Air Force as the commanding officer of the radar station in Wellington for part of WW II.
Travels to Britain
Initially works at Cambridge and then gains a Diploma from the Imperial College of Science and Technology and a PhD in geophysics from the University of London.
Marries Joan Alpers. They go on to have three children – David, Margaret and Rosemary. Family holidays often involve trips to out-of-the-way seismographs!
Joins Geophysics Division of the Department of Scientific and Industrial Research (DSIR) and works in exploration geophysics. Discovers coal-seam guided S waves in 1955 – renamed ‘Evison waves’ in 1985.
Research gains recognition
Gains a Nuffield Fellowship in 1957 and a Fulbright Award in 1963.
Image acknowledgement: NIWA
Inaugural Professor of Geophysics
Appointed inaugural Professor of Geophysics at Victoria University of Wellington.
Establishes Institute of Geophysics
Has a vision for an interdepartmental institute with members from geology, physics, chemistry, mathematics and geography departments in addition to members outside the university.
Begins research into earthquake forecasting
Possesses a strong belief that scientists have a duty to society and that reliable earthquake prediction would help minimise loss of life and suffering. Works passionately towards this goal until his death.
Collaboration with David Rhoades
Begins work with David Rhoades, now a statistician at GNS Science. This successful partnership continues until Frank’s death.
Image acknowledgement: GNS Science
Heads up a UNESCO conference on earthquake prediction in Paris. Helps to formulate a code of practice for earthquake prediction and chairs the Commission of Earthquake Prediction.
Retires as chair of the Geophysics Institute and continues as Emeritus Professor.
Image acknowledegment: Robert Cross, VUW Image Services
Joan had been Frank’s wife for 41 years.
Services to seismology (OBE)
Appointed as an Officer of the Most Excellent Order of the British Empire for services to seismology.
Passes away in his home in Wellington at the age of 82.
Image acknowledgement: The Wellesley Club
Frank Evison Research Scholarship in Geophysics established through donations from Frank’s family, the Earthquake Commission, GNS Science, the New Zealand Geophysical Society and a range of private donors.
Evison Symposium in Wellington
Evison Symposium on Seismogenesis and Earthquake Forecasting attended by national and international scientists. Two special journals are published to honour Frank’s interest in earthquake generation and forecasting.