We use cookies to ensure that we give you the best experience on our website. You can change your cookie settings at any time. Otherwise, we'll assume you're OK to continue.

Department of Mathematical Sciences

Seminar Archives

On this page you can find information about seminars in this and previous academic years, where available on the database.

Statistics Seminars: Challenges for Statistical Applications in Near-Surface Geophysics

Presented by Bernd Kulessa,

18 February 2004 14:00 in CM221

"Geophysical methods have been widely used for deep-earth studies since the early decades of the 20th century, such as e.g. earthquake seismology and hydrocarbon reservoir exploration. With the advent of affordable, sophisticated, and highly mobile field equipment especially since the late 1980s, combined with ever increasing computational power and software resources, geophysical technology is becoming increasingly popular for shallow investigations (i.e. within the first few tens of metres below the ground surface). The latter are multi-facetted, including e.g. environmental, engineering, archaeological, glaciological, military, and forensic applications. Conventional methods of shallow ground investigation typically involve either borehole drilling or excavation, both of which are financially and logistically expensive, as well as limited in their aerial coverage. In contrast, geophysical methods are advantageous because they are largely non-invasive, and now routinely support spatially continuous coverage in 2-D or even 3-D for a fraction of the cost of conventional methods. Given such exciting capabilities for rapid collection of large, multi-dimensional data sets using a variety of geophysical techniques, each of which is sensitive to a particular physical ground property, new challenges have emerged in recent years, which are conveniently formulated in terms of two key questions:
1. How can different types of geophysical data (such as e.g. electrical conductivity, seismic velocity, radar velocity, etc.) be quantitatively amalgamated to produce one common diagnostic variable that reflects the desired ground phenomenon (e.g. hydrocarbon contamination) particularly well, and at particularly enhanced spatial resolution?
2. How can the spatial scales of geophysical methods (typically several decimetres to metres) be matched to those of ‘ground-truthing’ borehole data (typically several millimetres to centimetres)? Suitable scaling approaches would strongly support the use of spatially continuous geophysical data as tools for interpolation of the targeted ground property between sampling stations, thus building up validated, multi-dimensional images of this property. It is anticipated that statistical methods are particularly well-suited to meeting these key challenges, as underlined by recent encouraging developments, which are nonetheless still sparse. This seminar presentation is specifically designed to highlight these key challenges, and hopes to stimulate discussions regarding potential solutions based on statistical methods. The first part of the presentation will focus on introducing the concepts of the most important shallow geophysical methods, as well as environmental targets of particular current focus. In the second part a variety of particularly relevant geophysical case studies will be presented, finally leading to a series of open questions and key challenges. "

Queen's University Belfast

Contact for more information