From Individual Gathers to Source-Composite Inversion in Active MASW: A paired field comparison of 8kg hammer and gun acquisition protocols
Year: 2026
Type: preprint
Abstract: In multichannel analysis of surface waves (MASW), the suitability of an active source is evaluated by its ability to generate coherent and continuous surface-wave dispersion over a usable frequency and wavelength range, as these characteristics influence the reliability of mode identification and the attainable depth of investigation. We present a hierarchical paired comparison of hammer and gun acquisition protocols using 13 matched source positions and identical 12-receiver geometry (5m spacing, 55m aperture and 30m nearest offset). All 26 SEG-2 gathers were processed individually using the same phase-shift imaging and automated ridge-extraction criteria, followed by independent manual fundamental-mode picking to verify the automatic interpretation before acceptance and inversion. The verified individual gathers were then combined through normalized common-offset trace stacking and normalized dispersion-image stacking for each source. Every gather yielded an accepted fundamental-mode path. At the individual-gather level, mean maximum usable wavelength was 30.14m for the hammer and 29.45m for the gun (gun minus hammer = −0.69m; 95% paired bootstrap interval −3.81 to 1.50m). The hammer contained a greater proportional 4.5–10Hz spectral fraction (0.967% versus 0.672%), but this exploratory contrast did not remain significant after Holm correction [14]. Source-composite curves spanned 21.0–46.0Hz for the hammer and 20.5–43.5Hz for the gun, with maximum wavelengths of 32.14 and 33.41m, respectively. The stacked gun-minus-hammer wavelength difference was 1.27m (paired spatial-bootstrap 95% interval −1.39 to 2.50m), equivalent to upper screening depths of 16.07 and 16.71m under a 0.5 wavelength convention. Fixed-layer conditional inversions produced overlapping shear-wave velocity intervals in all four model layers; gun-minus-hammer differences ranged from −17.6 to +14.1m/s and all 95% intervals included zero. Agreement between independently extracted trace-stack and image-stack curves was 7.16m/s for the hammer and 4.15m/s for the gun. These results show that source ranking is endpoint- and threshold-dependent: the hammer supplied more relative low-frequency energy, whereas the gun stack gave a slightly longer wavelength, but neither a reliable depth advantage nor a distinct velocity structure was resolved. The proposed individual-to-stack consistency framework, strengthened by manual verification of automated modal picks, provides a reproducible way to prevent source-energy observations from becoming unsupported claims of deeper MASW investigation
DOI: 10.31223/x5rj8s
Institutions: Ghana Atomic Energy Commission, University of Ghana
Language: English
Topic: Seismic Waves and Analysis
Subfield: Geophysics
Field: Earth and Planetary Sciences
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7225455429