Introduction

Having recently arrived as a Ph.D. student of the SEP as of a few days ago, I will now take the liberty of introducing myself by discussing some problems which interest me and some thoughts I have regarding their solution.

My interests reside generally in the realm of migration/inversion (m/i) of seismic reflection data, with a particular emphasis on subsurface target characterization. By m/i, I mean hybridizing separate but related concepts from migration and inversion camps, in order to produce high-resolution depth images of relative changes in a few different elastic properties, such as P and S impedance (see Lumley and Beydoun, 1991, this report). Of course, before one can attempt a respectable migration or inversion, the problem of velocity analysis needs to be addressed. I will be interested to consider the 3-D interval velocity analysis problem, and perhaps utilizing SEP's Connection Machine as the appropriate tool. After velocity analysis and m/i, quantitative appraisal of results would be highly useful. By appraisal, I mean making some quantitative measures of confidence in inversion results (velocity and elastic parameters), either relatively within a given image, or more absolutely from image to image. Finally, the m/i results and their corresponding confidence images may be calibrated and combined to yield comprehensive characterizations of subsurface target zones. The characterizations may lead to direct detection of interesting features, such as hydrocarbon presence (e.g., Parsons, 1986), or indirect correlations with secondary physical parameters, such as porosity or sand/shale content, through geostatistical techniques (e.g., Doyen, 1988).

Ideally, all the above could be addressed as a comprehensive project, given a suitable data set. Such a data set would probably require a small 3-D survey over a producing reservoir, coincident 2-D seismic for tying purposes, and multiple well control with V_p, V_s, and density logs through the reservoir. Of course, it would be nice if the structure was not too complicated to begin with, and any extras such as crosshole data exhibiting reflection events across the target zone would be extraordinary. I am in contact with ARCO regarding a data set, but am open to all suggestions.

Finally, it is interesting to me to note that development of the above technology could have equally important applications and ramifications in the fields of environmental, near-surface, or crustal geophysics, in addition to the more obvious utility in hydrocarbon exploration and reservoir management.