Over the last 1.5 weeks, I tried to fix all of the bugs in the DENISE-BE SH module. DENISE-BE should now deliver the same results as the DENISE-SH code. I currently only checked the SH-FWI for the Ainos-SH field dataset: https://danielkoehnsite.wordpress.com/blog/ngp-course-seismic-full-waveform-inversion/ Due to time constraints, I could only implement the SH-FWI using FD_ORDER=2 operators. For higher order fd operators the SH-FWI will fail.
Over the last 1.5 weeks, I tried to fix all of the bugs in the DENISE-BE SH module. DENISE-BE should now deliver the same results as the DENISE-SH code. I currently only checked the SH-FWI for the Ainos-SH field dataset: https://danielkoehnsite.wordpress.com/blog/ngp-course-seismic-full-waveform-inversion/ Due to time constraints, I could only implement the SH-FWI using FD_ORDER=2 operators. For higher order fd operators the SH-FWI will fail.