CWP Seminars  2014 Spring
CWP seminars discuss topics pertaining to our broad areas of research interests. These seminars are led by CWP faculty, students and, on occasion, by guest presenters. CWP seminars are held every Monday at 4 p.m. in the Green Center on the Colorado School of Mines campus. Click here to see previous CWP Seminars.
Note: To view weekly seminar schedules of individual CWP teams, click a link below:
CTeam seminars 
Steam seminars 
Fall 2013 CWP seminars
Date  Speaker(s)  Title  Abstract 
4/14  Guillaume Caumon, Université de Lorraine 
Geological uncertainties and inverse problems: Progresses and challenges 
3D subsurface models are very useful in a number of natural resource management and risk assessment tasks. However, coming up with a model compatible with available geophysical and geological observations can be a difficult challenge. In this talk, I will present recent results which tend to move away from the classical deterministic geological modeling, and rather aim at generating a set of possible models representing geological uncertainties. I will also discuss possible strategies to reduce these uncertainties by solving inverse problems and illustrate these using reservoir history matching examples. 
4/7  Tips for making figures in Madagascar 
Tips for making Project Review presentation figures using different software in use at CWP. 

Tips for making figures in Matlab 

Tips for making figures in Mines Java Toolkit (JTK) 

3/31  3D seismic image processing for unconformities 
I will first talk about a 3D seismic unconformity attribute to detect complete unconformities, highlighting both their termination areas and correlative conformities. As applications, I will then discuss how to use these detected unconformities as constraints to more accurately estimate seismic normal vectors and better flatten seismic reflectors at unconformities. We will still have CWP seminar today at 4pm. We will be hearing from Allison, Farhad and Esteban on tips for making figures using different software. This will all be in regard to preparing ourselves for our CWP presentations. Below are their names and what they will present on Esteban: Madagascar 

Shotbased velocity analysis using dynamic warping 
Typical seismic surveys provide redundant subsurface structural information, and this redundancy can be exploited to optimize models of subsurface properties. I present a method for velocity analysis that computes a velocity model by minimizing the shifts between migration images computed for pairs of closely located shots. I use dynamic warping to estimate vertical depth shifts between migration images, and I optimize the velocity by seeking the model for which the shifts between all pairs of images are zero. 

3/24  Waveform inversion for parameters of microseismic sources in VTI media 
Waveform inversion (WI) can be used to estimate earthquake source parameters, namely source location, moment tensor, and origin time. In this talk, I will present the results of applying WI to microseismic synthetic data. I perform adjoint simulations to calculate the gradient with respect to the source parameters and then use a nondimensionalization approach for simultaneous model updating for all the parameter classes. I will also discuss some of the complications when trying to invert for the location and the origin time simultaneously, and also with an incorrect velocity model. The examples show that it is possible to get good inversion results provided we are close to the actual solution for all three parameters and that the velocity model is known. 

3/17 

Locating weak temporal with diffusive waves 
Abstract unavailable. 
3/10 
CSM Spring Break  no CWP Seminar  
3/3  3D Navigation and imaging of asteroids 
In this presentation, I design the 3D navigation acquisition and do 3D imaging. The idea of geodesic dome has been implemented. As part of our acquisition and data collection design, the geodesic dome is a tool to collect the source and receiver locations for determining the required time for the explorational survey. With the dualorbiter acquisition, we can perform shotrecord migration by rearranging the multioffset data into common shot gathers. With the geodesic dome, the shot locations are evenly distributed on the 3D spherical surface. During the acquisition time, we can visualize the incremental data collection by observing the gradual filling of color in the triangles (representing the source and receiver locations) on the geodesic dome. When the acquisition is complete, we can perform shotrecord migration using the receiver coordinates from the receiver(fararray) geodesic dome. In the ideal acquisition, each triangular patch in the geodesic dome of the source has its own corresponding dense cluster of receivers. These corresponding receivers are used in the migration step as we separate them into reflection and transmission receivers. We perform migration using the reflection receivers and will perform tomography using transmission receivers. 

2/24  Seismic depthconversion: Application of simultaneous multiplewell ties 
One application of simultaneous multiplewell ties is seismic depthconversion. Using these well ties, we compute a 3D timedepth function to convert a 3D timemigrated seismic image to depth. To compute this timedepth function, we relate time shifts computed from well ties to average velocities, and interpolate these average velocities between wells. I will discuss the advantages and pitfalls of our method. 

2/17 
Presidents Day  no CWP Seminar  
2/10  Radiation pattern for elastic VTI media 
When performing inversion for multiple parameters, appropriate parameterization strongly influences the outcome of the experiment. Since there generally are tradeoffs between the parameters, we often have to resort to sequential inversion. One way of choosing an optimal set of parameters or deciding the sequence of inversion is to look at the sensitivity of the parameters for different sets of data. I will talk about 'radiation pattern', which essentially looks at scattering patterns for different parameters at a scatterer as a function of incidence and/or scattering angles. 

2/3  If we cannot see it, can we image it? 
The free surface is the strongest reflector in the subsurface. This reflector gives rise to freesurface multiples. In some cases, freesurface multiples mask or even completely cancel primary events. In these situations can we still correctly image the subsurface? In this seminar I will discuss the way we deal with freesurface multiples to obtain "accurate" images of the subsurface. I will compare conventional imaging methods, that require surfacerelated multiple elimination, to autofocusing imaging. 

Experimental study: Improved focusing using deconvolution 
Time reversal techniques are used in ocean acoustics, medical imaging and nondestructive evaluation to backpropagate recorded signals to the source of origin. We demonstrate experimentally a technique which improves the temporal focus achieved at the source location compared to time reversal. The experiment consists of propagating a source function from a transducer within a concrete block to a single receiver on the surface, and then applying time reversal or deconvolution to focus the energy back at the source location. The proposed method is simple and proven to be robust. Additionally, it's costs are negligible due to deconvolution being a preprocessing step to the recorded data. 

1/27  Evaluation of the Qfactor in the inner core using Fiji earthquakes recorded at European seismological networks 
A study case where I compute the attenuation for the inner core using PKP waves recorded at European seismological networks using the spectral ratio method. I will present the method, how I worked on it, my results, my interpretations, as well as a comparison to other papers. 

Angle gather construction in 3D 
Common image gathers are a useful tool for migration velocity analysis and amplitude variation with angle and azimuth studies. There are different methods to construct angle domain common image gathers. I will discuss the theory behind each method and compare two methods to construct angle gathers from waveequation migration methods. 

1/20  Horizon extraction with constraints  To extract or construct a single horizon from a 3D seismic image, one usually first picks a reference point or seed. This seed then grows to a horizon surface by manually or automatically tracking seismic reflectors along seismic amplitude peaks or troughs. Here, I propose a different method that uses at least one control point to initialize a complete horizontal surface and then iterative updates this surface to a seismic horizon until the surface normal vectors are identical to the seismic normal vectors on the surface. This method allows to specify, perhaps interactively during interpretation, a small number of control points that may be scattered throughout a 3D seismic image. Examples show that control points enable the accurate extraction of horizons from seismic images in which noise, unconformities, and faults are apparent. These points represent constraints that we implement simply as preconditioners in the conjugate gradient method used to construct a horizon. 

1/13  1st CWP seminar of the Spring 2014 semester (CWP administrative topics) 
Previous CWP Seminars
2013 

2012 