Director's Message

Colleagues,

The week ahead is a big one for the School.

A busy schedule has been arranged around Professor Carol Frost’s visit to present the inaugural Chappell Lecture which will be held at 4.30–6.30pm Thursday in the RN Robertson Theatre in the Research School of Biology. Carol’s schedule includes research presentations in addition to Chappell Lecture on Thursday, and various opportunities for staff and students to liaise with Carol. Please make Carol welcome. 

In other news, It is with great pleasure that I announce  Penny King, Andy Hogg and Hrjove Tkalčić have been promoted to Level E , Rhodri Davies to Level D and Katherine Grant to Level B in the 2018 round. Andrew Berry and Caroline Eakin will be awarded the CoS prize for excellence in education. This reflects the outstanding student engagement and learning in EMSC1008 ‘Earth’ achieved over the past 5+ years. Andrew has asked me to make special mention and thanks to RSES teaching assistants (Kate, Brendan, Huijan, and Ulli Proske) and the demonstrators whose efforts and contributions have made this possible, as well as past lecturers.Please join me in congratulating each of them on this great achievement.

And finally a reminder to all those who knew and worked with Ian MacDougall about his funeral service which will be held at 3:30 pm today (Monday 19th Nov) in the Great Hall at University House.

Best wishes from Bangalore, where Greg, Simon and myself have commenced a journey of vegetarian gourmet fulfilment (so far so good) as part of the School and College of Science’s engagement with India strategy.

Kind regards, 

Steve

Australian National Geodesy Meeting 2018

Talk on *Joint categorical/elastic inversion from seismic data*

Dr James Gunning (CSIRO Energy) will visit us next week. He will give a talk on Wednesday (21/11) at 11:30am in the Hales Room:

In oil and gas applications of seismic data, mapping rock and fluid types (facies) is very desirable for hydrocarbon field planning and forecasting. Well log data commonly show both compaction and significant mixture character of elastic parameters, corresponding to different rock types and fluids. This inspires the construction of mixed categorical/continuous distributions for use as priors in Bayesian inverse frameworks. The most straightforward route to integrate seismic data is then using "AVO" style convolutional forward models in the inversion, i.e. a cascaded inversion based on true-amplitude migrated seismic data. The most powerful general framework for these mixed discrete/continuous optimisation problems are varieties of the EM algorithm, which can also be shown to be applicable also to a (vastly heavier) full waveform inversion formulation of the problem.  Particular and useful cases of the AVO problem can be reduced to pure discrete inversion problems, which make for revealing connections to global optimisation problems in operations research, including the powerful ideas behind semidefinite programming.

We hope to see many of you there. If anyone wishes to meet with James during his visit, please let Andrew Valentine know via email at andrew.valentine@anu.edu.au. 

Reminder

2018 Inaugural Chappell Public Lecture is on Thursday 22nd Nov 4.30 - 5.30 pm, with refreshments to follow.

Featuring guest speaker Dr Carol D. Frost from the University of Wyoming. The lecture series is in honour of Professor Bruce Chappell. 

If you haven't registered please register here by 19th Nov.

Abstract for Cliff Thurber’s talk on 3rd December 2018

A major field project was carried out in March 2016 at the Brady geothermal field in west-central Nevada, as part of a Department of Energy project nicknamed PoroTomo.  A vibroseis truck was used as a repeatable seismic source, and both nodal seismic instruments and a distributed acoustic sensing (DAS) system were deployed in the study area.  The main goal of PoroTomo is to demonstrate the potential for developing a fine-scale model of the physical properties of a portion of the geothermal field, in this case the volume around the injection wells.  The field work was scheduled to overlap the time period of a planned shut-down of the geothermal power plant for annual maintenance, allowing for detection of possible temporal change in the subsurface structure as fluid injection and extraction rates were varied.

My presentation will consist of three parts.  First, I will present a brief comparison of data from the DAS system and the nodal array.  After making corrections to convert DAS data from strain rate to ground velocity and convert geophone array data from ground velocity to strain rate, we find excellent agreement between the two data types for a regional earthquake.  Next, I will show the results from a tomographic inversion of P-wave travel time data from the vibroseis source on a combined node and DAS dataset.  Finally, I will present the results of the analysis of earthquake activity that occurred during the 2016 plant shutdown and other shutdown periods before and since 2016.  We find clear evidence that cessation of fluid pumping triggers earthquakes at Brady.

Clifford Thurber, University of Wisconsin-Madison