Hi Run, I am pretty sure that the coherence in MSNoise is computed as described by Clarke et al. (2011) (https://doi.org/10.1111/j.1365-246X.2011.05074.x). It is defined in Appendix A of the paper. Not sure on the details exactly on how matlab and scipy differ from this, but I've certainly seen the definition given by Clarke et al. elsewhere also. Perhaps this is helpful to you? Cheers, Alec PhD Student at ISTerre, France ----- Mail original ----- De: "Kohtaro R. Araragi" <kararagi(a)eri.u-tokyo.ac.jp> À: msnoise(a)mailman-as.oma.be, "Thomas Lecocq" <thomas.lecocq(a)seismology.be> Envoyé: Jeudi 21 Juillet 2022 15:27:24 Objet: [MSNoise] source of coherence calculation Dear Thomas (c.c. MSNoise community), I'm Run, a former student of ERI who has been working with Florent, (Yosuke) Aoki-san, (Takao ) Ohminato-san. I hope you are doing well. I am recently checking the processing of MWCS method and noticed that the algorithm of coherence ( functions of "smooth(x, window='boxcar', half_win=3):" and "getCoherence(dcs, ds1, ds2)" at move2obspy.py) is different from matlab (mscohere) or python (scipy.signal.coherence) though it works very well. Could you let me know if there were some original papers or documents that explain the background about it? I appreciate if someone tell me about relevant information how to obtain the coherence in detail. Thanks in advance, Run -- ____________________________________________________ K. Run Araragi Research student/M.Sc. Earthquake Research Institute University of Tokyo 1-1-1 Yayoi, Bunkyo-ku Tokyo, Japan 113-0032 Skype name: kalessinlord Tel: 81-3-5841-5697 Cel: 81-90-1166-3091 ____________________________________________________ _______________________________________________ MSNoise mailing list MSNoise(a)mailman-as.oma.be http://mailman-as.oma.be/mailman/listinfo/msnoise