HR: 0830h
AN: S31E-0814    [PDF]
TI: 3-D Empirical Travel Times for Global Heterogeneity
AU: * Nicholson, T A
EM: tnicholson@eos.ubc.ca
AF: Department of Earth and Ocean Sciences, University of British Columbia, 6339 Stores Road, Vancouver, BC V6T 1Z4 Canada
AU: Sambridge, M
EM: malcolm@rses.anu.edu.au
AF: Research School of Earth Sciences, Australian National University, Institute of Advanced studies, Canberra, ACT 0200 Australia
AU: Gudmundsson, O
EM: og@dlc.ku.dk
AF: Danish Lithospheric Centre, Oster Voldgade 10, 1350 Copenhagen K,, Copenhagen, K Denmark
AB: Many applications in seismology rely on the accurate prediction of teleseismic travel times. As an alternative to travel time calculation in 1-D or 3-D seismic velocity models, we have developed a new approach based on directly smoothing and interpolating observations from large earthquake catalogues. We call the approach 3-D Empirical Travel Times (ETT). This approach is able to handle both the spatial variability in data density, as well as the many length scales of heterogeneity that influence observed travel times. An important feature of the technique is that after the empirical travel time corrections have been applied the remaining residuals are spatially uncorrelated and, in this sense, all of the signal in the data is accounted for. Travel times predicted with the new approach compare very favourably with those from 1-D and 3-D Earth models. Travel time residuals of nuclear blast data are reduced by 62%, on average, compared to 1-D reference Earth models. Tests with globally distributed earthquake data give an average residual reduction of up to 42% across a range of teleseismic phases. Tests also show that, 3-D empirical travel times reproduce the pattern of regional variation in tele-seismic phases very well. In epicentre determination of 25 nuclear blasts, the use of empirical travel times in place of 3-D global velocity models reduced the average mislocation by over 59%. The new approach can also be used for `denoising travel times from large earthquake databases, giving a new `data source for tomographic studies. Although developed for use teleseismic the empirical travel time approach has the flexibilityto be applied to local or regional seismicity. This model free `database approach to travel time prediction results in a highly accurate and robust method and in a sense a return to the empirical philosophy of Jeffreys and Bullen (1940). It is likely to find applications in a range of seismological problems.
DE: 7215 Earthquake parameters
DE: 7218 Lithosphere and upper mantle
DE: 7219 Nuclear explosion seismology
DE: 7260 Theory and modeling
SC: Seismology [S]
MN: 2003 Fall Meeting