GPU-MatchLocate1.0 :GPU-based Match and Locate
- The GPU-M&L code is mainly modified from Miao Zhang's M&L (Zhang and Wen, 2015) (https://github.com/Dal-mzhang/MatchLocate2)
- The strategies of GPU-programming are mainly modified from Beaucé et al. [2017] (https://github.com/beridel/fast_matched_filter)
1.The GPU-MatchLocate package contains following main files:
bin: executable file
src: source code
Demo: an example
2.Revision History:
06/08 2018 M. Liu Initial coding
14/08 2019 M. Liu Version 1.0 released
3.References:
1)Zhang, M., and L. Wen (2015), An effective method for small event detection: match and locate (M&L), Geophysical Journal International, 200(3),1523-1537.
2)Beaucé, E., W. B. Frank, and A. Romanenko (2017), Fast matched filter (FMF): An efficient seismic matched‐filter search for both CPU and GPU architectures, Seismological Research Letters, 89(1), 165-172.
3)Liu, M. H. Li, M. Zhang, and T. Wang (2019), Graphics Processing Unit-based Match&Locate (GPU-M&L): An improved Match and Locate technique and its application, Seismological Research Letters, 91 (2A): 1019–1029.
4.Usage:
Please see user guide
5.Demo:
5.1 compile source code
$cd ./src
$make
$cd sacCC
$make
5.2 download continuous data (User-specified)
$cd ../../Demo
$python data_download.py
5.3 download templates and create routine catalog (User-specified)
$python template_download.py
5.4 set common origin time
$cd ./Trace
$perl SACH_O.pl 20190704
5.5 mark P/S wave arrival time and calculate slowness
$cd ../Template
$perl marktaup_p.pl
$perl marktaup_s.pl
5.6 calculate weighting factor for each trace
$bash cal_weights.sh
5.7 detect and locate events
$cd ..
$perl RunprocAll.pl
5.8 generate new catalog
$cd ./MultipleTemplate
$perl MergeEvents.pl 20190704
$perl SelectFinal.pl 2019 07 04 Allevents
5.9 compare waveforms between template and detection
$perl PlotEventWaveform.pl DetectedFinal.dat 1
6.Author:
Min Liu, China University of Geosciences, Beijing ([email protected])
also at Dalhousie University ([email protected])