During the mid 1980s, I helped Jon Nyquist, a Phd student in Geology and fellow chess player at the University of Wisconsin to estimate the location and depth of the Atlantic Mid Continental Rift from ultra-sound data using broken and unbroken plate models, I proposed using a conditional Laplacian/ modified profile likelihood method I'd published as a comment in JASA 1982. Jon has become very eminent since.
FLEXURAL MODELING OF THE MIDCONTINENTAL RIFT
by Jonathan Nyquist and Herbert Wang
While I was teaching Statistics 431: Applied Categorical Data Analysis at Wisconsin, I helped a Geology graduate student called Dennis Kerr to develop new methodology for analyzing geological layers using suitably normalised quasi-independent contingency tables to estimate the transition matrices in Markov chains.
During the late 1990s I worked with Professor Ian Main and my University of Wisconsin STATLAB colleague Orestis Papasouliotis on various statistical applications in Geophysics.
Our first paper used the Bayesian Information Criterion BIC to detect
significant breaks in geophysical data
Our second paper concerned the estimation of frequency size distributions for global earthquakes, star-shapes and fault lengths.
In our second paper, also with Kes Heffer, we proposed a simple predictive algorithm for determining good pairings of injector and producer wells in hydrocarbon reservoirs
These methodologies were developed further in another paper
For our international patent please see