Full-wavefield focusing in seismic imaging – Concepts, applications, and examples

Retrieving images of geologic features buried deep beneath complex overburdens that are meaningful in terms of structure and, more importantly, in terms of quantitative physical properties remains as a great and elusive challenge to the science of seismic imaging. Over the years, new seismic imaging and velocity model estimation methods often have relied on the concepts of wavefield focusing. As such, the principles and physics of focusing have become central to the theory and practice of seismic imaging. In this lecture, we will review the physics and the formalism of focusing in its most general forms, by means reciprocity relations and language of seismic interferometry. This, in turn, will allow us to build a complete and intuitive understanding of seismic imaging by understanding how source- and receiver-side focusing come together in the retrieval of depth images. Through this exercise, we will not only cover the concept of depth images but also of depth-domain image gathers, i.e. extended images, as a natural consequence of understanding imaging in terms of focusing concepts. With several examples, this exercise will allow us to also discuss the role of primary and multiple reflections in imaging and set a context for current imaging practices using primaries versus novel advances seeking to retrieve geologic information encoded in the multiples at depth. In the second part of the lecture we will cover the recently introduced method of Marchenko redatuming in the context of depth imaging, building on the focusing framework from the first part of the lecture. Here, we will discuss the physical concepts of the Marchenko scheme using the language and physics of focusing from the first part of the lecture. We will discover how this physical framework translates into practical redatuming schemes which rely on focusing to improve existing imaging results as well as to offer new imaging tools. We will cover the latest developments and examples of this approach (with synthetic and field data) while also pointing out its current shortcomings and opportunities for further advances.