The following text is for questions 39-43

    Scientists used artificial intelligence (AI) to study the spatial relationship between main earthquakes and their aftershocks. In tests, AI predicted the aftershock locations better than the traditional methods that many seismologists use.

    In 1992, a series of earthquakes prompted an outbreak of interest among seismologists. They were trying to map out where exactly an aftershock might occur based on how a mainshock might shift stresses on other faults. After 1992, researchers began trying to refine the complicated stress change patterns using different criteria. The most use criterion, the Coulomb failure stress change, depends on fault orientations. However, stresses can push on the faults from many directions at once. Consequently, fault orientations in the subsurface can be complicated.

    Using AI, the data included more than locations and magnitudes. The data considered different measures of changes in stress on the faults from the quakes. The AI learned from the data to determine how likely an aftershock was to occur in a specific place. The team tested how precise the system could pinpoint aftershock locations using data from another 30,000 mainshock-aftershock pairs. The AI consistently predicted aftershock locations much better than the Coulomb failure criterion.

    However, the study focuses just on permanent shifts in stress due to a quake. Aftershocks may also be triggered by a more momentary source of stress. A quake's rumbling through the ground could produce this kind of stress. Another question is whether Al-based forecast system could leap into action quickly enough after a quake. The predictions in the new study benefited from a lot of information about which faults slipped and by how much. In the immediate aftermath of a big quake, such data wouldn't be available for at least a day.

In paragraph 2, why does the author include information about the earthquakes happening in 1992?