Tamas Nemeth

ADVERTISEMENT
From SEG Wiki
Revision as of 16:03, 12 April 2017 by JohnWStockwellJr (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search
Tamas Nemeth
Tamas Nemeth headshot.png
PhD university University of Utah


Tamas Nemeth was honored with the J. Clarence Karcher Award for his exceptional work on the imaging of sparse, irregularly sampled seismic data. Through sophisticated mathematical analyses and skillful implementations of algorithms, he has made significant contributions to both surface seismic prospecting and borehole geophysics, and has in particular enhanced the profession’s understanding of migration, velocity analysis, Radon transforms, and cross-well imaging.

Biography Citation for the J. Clarence Karcher Award

Contributed by Bob Langan

It is always a pleasure to observe a bright young researcher tackle difficult geophysical problems with enthusiasm, a fresh perspective, stubbornness, and creativity. These qualities describe Tamas Nemeth as he has sought to fuse mathematical concepts and algorithm development with a variety of seismic data types.

Tamas undertook studies in geophysics because it combined his interests in geography and mathematics. To him it was a least-squares (LS) solution to possible career choices, an approach that is a continuing theme in his life. He entered St. Petersburg School of Mines in 1984 and obtained the Engineer in Geophysics Degree in 1989. Following graduation he joined the Eötvös Institute in Hungary as a field geophysicist on a vibroseis crew.

Tamas began graduate studies in geophysics under Jerry Schuster at the University of Utah in 1992. In his first project he implemented a dynamic damping approach to the regularization of tomographic inversion problems that allows one to obtain velocity models which best fit observed traveltimes.

While attempting to migrate reflection energy in some cross-well and VSP data sets, Tamas decided to include constraints consistent with the acquisition geometry and geologic structure, and he chose an LS optimization formulation for the migration. At the time, attempts to migrate cross-well seismic data were often plagued by artifacts that rendered them difficult to interpret. A variant of the VSP-CDP map for reflection imaging has been popular because it eliminates many of these artifacts, but these mappings have their own set of limitations. Tamas applied his new migration algorithm to a west Texas cross-well data set provided by Chevron and he obtained a crisp image that captured critical fractures in the reservoir.

Tamas spent a summer at Amoco’s research facility in Tulsa, Oklahoma in 1995. Amoco had acquired a very large (225 000 traces) cross-well data set at Hugoton Field in Kansas, in a joint project with Texaco and Conoco, and was processing the data. In collaboration with Anthony Vassiliou, Tamas developed an anisotropic migration algorithm. They obtained an impressive reflection image of their objective, a subtle sand channel. This was a major advance because the well spacing was relatively great (2000 ft) for the acquisition technology of the time, and the signal-to-noise ratio of the data was relatively poor.

After returning to Utah, Tamas pursued two research topics with the hope that one would mature into his thesis (an LS solution?). The first was migration velocity estimation. He derived a general relationship between traveltime perturbations and perturbations in both the raypath and slowness, and this relationship yielded as special cases previously published formulas for migration velocity analysis, depth-focusing analysis, and reflection tomography. Of course the code was cast in LS form, but he lacked a suitable data set on which to test his ideas.

He did have a suitable data set for his second topic, LS (what else?) migration. He extended LS migration into filtering, with the migration operator treated as a type of transform. One could then design different operators for each wave mode. This novel approach achieved improved wavefield separation and resolution in the presence of noise, gaps, and overlapping data, and holds promise for processing multicomponent data.

Tamas joined Chevron Petroleum Technology Company in late 1996 as a research geophysicist. At Chevron he has worked on model-based imaging and construction of velocity models for Chevron’s proprietary depth migration. He has continued his earlier work on the constrained migration of borehole seismic data by developing a 3-D algorithm for VSP data and multiple cross-well profiles. He is now diversifying into non-LS solutions as well.