Google and Harvard University have just announced new research in seismology using artificial intelligence. Researches have "trained" computers using a database of information on earthquakes to help the computers form predictions about future events. The focus of the research is on aftershocks which can cause at least as much damage as the original earthquake. The recent earthquake in Indonesia, for example, demonstrates the potential value of such innovations.
The study of earthquakes interested people for centuries, but it was only in the 19th century that an Irish-born engineer applied the term “seismology” to such work.
Robert Mallet was born in 1801 in Dublin. His family owned an ironmonger where he worked after completing school and university. The firm produced a wide range of cast-iron parts, including bridge components. Mallet considered himself an engineer and applied for the chair of engineering at Trinity College (which he did not get).
Lack of a university appointment did not stop Mallet from conducting scientific work, and he developed a particular interest in the study of earthquakes. His paper to the Royal Irish Academy in 1846 claimed to be "the first attempt to bring the phenomena of the earthquake within the range of exact science".
We now accept that the damage of an earthquake is caused by seismic waves travelling from the site of a fault through the ground. The passage of these waves causes the ground to shake and the shaking ground causes things to move, break, and fall.
In the 19th century there was no fully developed explanation, and observations of the aftermath of the earthquake were confusing. Geologists such as Charles Lyell had explained the apparent twisting of objects during an earthquake by reference to vortices. He cited the example of a pair of obelisks in Calabria that had not fallen but instead appeared to have turned on an axis.
Earth’s surface
Mallet proposed a simpler type of motion to account for the same results. He suggested that “rectilinear motion” (rather than any vortices) could account for all the damage seen from earthquakes. He suggested that earthquake was the result of a wave or waves emanating from a geological disturbance and then spreading outwards through the different materials of the earth’s surface. He called these waves “elastic waves of compression”.
According to Mallet, what happened to a building experiencing this motion depended on the relationship between its centre of gravity and its “centre of adherence” (essentially how it was anchored to the earth). A high centre of gravity and the building would simply fall over. For those where the centre of gravity was low and aligned with the “centre of adherence” as well as with the direction of the wave, the base and the top would appear to have moved in opposite directions. However, if all three were not aligned the object would appear to have twisted as parts of it moved in multiple different directions.
Not content to explain his theory of the motion of an earthquake, Mallet outlined at length a whole programme of the study of earthquakes. He proposed a distribution of earthquake observation centres, testing of the effects of compression waves on a wide variety of substances, mapping the locations of known earthquakes (something he later did himself) and much more.
In the same year he delivered a paper on a machine for registering motion of the earth’s surface during even “minute” earthquakes. The machine used the now familiar tracing of a waveform on graph paper although it is not the same as the modern day seismogram.
Broken walls
In 1858 he had the opportunity to put his ideas to further use when he was nominated by the Royal Society to travel to Naples and report on the effects of a 1857 earthquake. He used careful observation of fallen and broken walls, collapsed buildings, landslides and reoriented rivers to help him determine the focal point or what we now call the epicentre.
He claimed that “every displaced object is in fact a seismometer”, helping him to determine the height and velocity of the compression waves and to pinpoint the central fault that had caused the quake.
Mallet’s interest in ballistics also foreshadowed the relationship between war and seismology. Mallet had designed mortars for the Crimean War, and set up experiments to observed the effects of gunpowder explosions. During the Cold War the US department of defence became an important patron of seismology, hoping to use it to detect underground nuclear testing.
Dr Juliana Adelman lectures in history at Dublin City University