The Earth's coda correlation wavefield: Rise of the new paradigm and recent advances

Seismology has come a long way in providing insights into Earth's internal structure and dynamics. Among many forward and inverse geophysical techniques developed, full waveform modelling, seismic tomography and receiver-based studies enabled detailed imaging of Earth's subsurface. The invention of ambient noise tomography within the last two decades revolutionized studies of Earth's subsurface based on the cross-correlation of the Earth's ambient noise. That method, in particular, enabled imaging of Earth structure in places where earthquakes or receivers do not exist. At the same time, progress in imaging the Earth's deepest shells has been impeded by the lack of geometric coverage of body waves due to uneven global distribution of seismic sources and receivers and the fact that the ambient noise studies cannot reach deeper than the uppermost Earth's shells, near its surface. In seeking the ways forward, global seismologists started experimenting with cross-correlating the part of the seismograms recorded many hours after the first arrivals of body-waves, the so-called earthquake coda. As in many science disciplines, initial work on this topic resulted in controversies but also led to new realisations and discoveries that all contributed to the rise of a new paradigm – the earthquake coda-correlation wavefield, which is the focus of this review paper. We do not attempt here to provide a review of ambient-noise correlation, although we use some familiar concepts to introduce coda-correlation. Our main goal is to review theoretical and observational work to date that resulted in a better understanding of the coda-correlation wavefield, both as a phenomenon and a powerful method. The features in global correlograms exist due to many cross-terms of reverberating body-waves, a principle fundamentally different from the reconstruction of surface waves in the ambient-noise correlograms. Once the coda-correlation wavefield is fully understood through theoretical developments, the method becomes a powerful way to study Earth's deep structure. Apart from providing a review of the most important results to date, we scrutinise the process of making global correlograms and analyse their characteristics while taking into consideration various aspects, such as the time after the earthquake, the source-receiver geometries, the level of seismicity and the type of earthquake mechanism. Furthermore, we provide practical examples on how to build correlograms and interpret correlogram features using programming language Python. Our review seeks to promote the topic of coda correlation among already experienced researchers as well as students who embark on this interesting research, which may play a central role in global and planetary seismology in the coming decades.