TY - THES

T1 - Deconvolution of seismic data from mine production blasts

AU - Reiner, Franziska

N1 - no embargo

PY - 2018

Y1 - 2018

N2 - This study is about the deconvolution of production blasts in a mine. The data for this study was recorded by an array of 124 seismic stations recording continuously during three weeks in November 2016 at Mt. Erzberg mine, Austria. In that time, 10 production blasts were recorded, ranging from 2-6 t explosives in 5-28 boreholes, each. Single blasts were detonated with a nominal delay time of 33-42 ms, however, the actual delay is poorly controlled and can deviate several ms. This results in a complex, unknown source-time function of up to 0.6 s length. First, all traces were aligned at the P-arrival and stacked to obtain a mastertrace for each shotpoint. Then, the Wiener Shaping Filter method was used to obtain a filter which was applied to each seismic section. This worked relatively well for some shots and the S-wave could be detected for shotpoint 27 for example. In a second step, attempts were made to separate each source-time-function into an average wavelet of a single blast and the blast sequence, respectively. This did not work, suggesting that the individual blasts might not be similar enough, at least not considering the full bandwidth.

AB - This study is about the deconvolution of production blasts in a mine. The data for this study was recorded by an array of 124 seismic stations recording continuously during three weeks in November 2016 at Mt. Erzberg mine, Austria. In that time, 10 production blasts were recorded, ranging from 2-6 t explosives in 5-28 boreholes, each. Single blasts were detonated with a nominal delay time of 33-42 ms, however, the actual delay is poorly controlled and can deviate several ms. This results in a complex, unknown source-time function of up to 0.6 s length. First, all traces were aligned at the P-arrival and stacked to obtain a mastertrace for each shotpoint. Then, the Wiener Shaping Filter method was used to obtain a filter which was applied to each seismic section. This worked relatively well for some shots and the S-wave could be detected for shotpoint 27 for example. In a second step, attempts were made to separate each source-time-function into an average wavelet of a single blast and the blast sequence, respectively. This did not work, suggesting that the individual blasts might not be similar enough, at least not considering the full bandwidth.

KW - deconvolution

KW - seismic data

KW - production blasts

KW - Mt. Erzberg mine

KW - Dekonvolution

KW - seismische Daten

KW - Produktionssprengungen

KW - Erzberg

M3 - Master's Thesis

ER -