In this thesis, an imaging technique that utilizes sparsely sampled, multi-component geophone data and a dense surface distributed acoustic sensor (DAS) acquisition is proposed. The PoroTomo survey at Brady's Natural Lab consisted of 238 multi-component geophones that are spaced anywhere from 60 meters to 150 meters apart. This proves to be a difficult migration problem with such sparse spacing. Fortunately, the PoroTomo survey also included 9 km of surface DAS fiber placed in a variety of orientations. DAS, however, can only record particle motion in the direction that it is oriented. After a broad literature review, it has been found that previous surface DAS surveys have come to be inconclusive regarding the feasibility of using the fiber by itself. These studies, however, only utilize a vertical source in short offsets. Assuming a flat-layered Earth, a P-wave reflection will not show data on a DAS fiber as the particle motion is not polarized properly. The PoroTomo survey utilized a 3-C source that allowed for the testing and proof of this hypothesis. Both 2-D and 3-D numerical experiments are performed to test the feasibility of using multi-component geophone and DAS data together. In 2-D, a reflectivity model is created from the local fault model in the PoroTomo Survey. This provided a variety of structural dips to test the imaging technique. It was found that using an S-source rather than a P-source with these models produced a much sharper resulting image. A quantitative analysis is further performed to provide an unbiased perspective on the results. The quantitative analysis utilized both energy norm image filtering and a convolutional neural network to prove that distributed sensors add value to imaging efforts with sparsely-sampled, multi-component geophones. The 2-D example is an idealized experiment. A more extreme example is performed in 3-D to confirm the conclusions made in 2-D. A methodology to model DAS data in 3-D is presented prior to showing examples of utilizing the two data types together for imaging. The resulting images in 3-D are low frequency due to the velocity model and stability limitations. Quantitative analysis is also required for an unbiased perspective on the results. The quantitative analysis utilized only the energy norm image filtering technique in 3-D as the machine learning algorithm is not able to achieve a reasonable cross-validation accuracy. The results from energy norm image filtering show that utilizing DAS in surface surveys with a sparse multi-component geophone acquisition proves to be useful in reducing the number of false positives by a small fraction. This experiment, however, is still considered inconclusive in regards to identifying if DAS can add value to sparsely sampled geophone data because the geometry of the DAS acquisition is so unique. A more regular experiment must be performed prior to making such conclusions, so 2-D lines of fiber were utilized instead of the PoroTomo acquisition geometry. The 2-D DAS acquisition increases identifying the true positives significantly.
Copyright of the original work is retained by the author.
The export option will allow you to export the current search results of the entered query to a file. Different
formats are available for download. To export the items, click on the button corresponding with the preferred download format.
By default, clicking on the export buttons will result in a download of the allowed maximum amount of items.
To select a subset of the search results, click "Selective Export" button and make a selection of the items you want to export.
The amount of items that can be exported at once is similarly restricted as the full export.
After making a selection, click one of the export format buttons. The amount of items that will be exported is indicated in the bubble next to export format.