2004Near Surface GeophysicsRequires access

Ultra‐shallow seismic reflection imaging in a region characterized by high source‐generated noise

Cédric Schmelzbach, A. G. Green, H. Horstmeyer

Open publisher page 38 citations

Abstract

ABSTRACT Coincident 30 Hz and 100 Hz ultra‐shallow seismic data sets acquired across a sedimentary sequence in northern Switzerland are characterized by unusually high levels of source‐generated noise (i.e. guided and surface waves). To image reflections in these noisy data it is necessary to apply an extensive processing sequence that includes semi‐automatic first‐break picking, amplitude scaling, multichannel spiking deconvolution, bandpass‐frequency filtering, careful selection and application of top mute functions, iterative velocity and residual static correction analyses, application of NMO corrections and appropriate stretch mutes, linear τ–p filtering and stacking. After processing, nearly identical reflection patterns are observed on the resultant 30 Hz and 100 Hz stacked seismic sections. In particular, an ultra‐shallow reflector is mapped beneath the entire length of the approximately 50.3 m long profiles at 1.4–2.3 m depth; differences in depth between the two stacked sections are less than 0.15 m. Since this reflector is practically coincident with an ultra‐shallow refractor across which the velocity increases from about 275 m/s to over 2100 m/s, it is interpreted as the groundwater table. Such an interpretation is consistent with the groundwater level observed in nearby boreholes. Underlying the ultra‐shallow reflection is a quasi‐continuous event that originates from 1.9–10.0 m depth. This deeper reflection probably represents the boundary between glaciofluvial and underlying glaciolacustrine sedimentary units. We conclude that state‐of‐the‐art processing techniques may allow ultra‐shallow reflections to be extracted from seismic data dominated by source‐generated noise. As shown in earlier studies, the ultra‐shallow seismic reflection method is capable of supplying high‐resolution subsurface images at locations where the geo‐radar technique is limited by lack of depth penetration.

About this research paper

What this paper is about

ABSTRACT Coincident 30 Hz and 100 Hz ultra‐shallow seismic data sets acquired across a sedimentary sequence in northern Switzerland are characterized by unusually high levels of source‐generated noise (i.e. guided and surface waves). To image reflections in these noisy data it is necessary to apply an extensive processing sequence that includes semi‐automatic first‐break picking, amplitude scaling, multichannel spiking deconvolution, bandpass‐frequency filtering, careful selection and application of top mute functions, iterative velocity and residual static correction analyses, application of NMO corrections and appropriate stretch mutes, linear τ–p filtering and stacking. After processing, nearly identical reflection patterns are observed on the resultant 30 Hz and 100 Hz stacked seismic sections. In particular, an ultra‐shallow reflector is mapped beneath the entire length of the approximately 50.3 m long profiles at 1.4–2.3 m depth; differences in depth between the two stacked sections are less than 0.15 m. Since this reflector is practically coincident with an ultra‐shallow refractor across which the velocity increases from about 275 m/s to over 2100 m/s, it is interpreted as the groundwater table. Such an interpretation is consistent with the groundwater level observed in nearby boreholes. Underlying the ultra‐shallow reflection is a quasi‐continuous event that originates from 1.9–10.0 m depth. This deeper reflection probably represents the boundary between glaciofluvial and underlying glaciolacustrine sedimentary units. We conclude that state‐of‐the‐art processing techniques may allow ultra‐shallow reflections to be extracted from seismic data dominated by source‐generated noise. As shown in earlier studies, the ultra‐shallow seismic reflection method is capable of supplying high‐resolution subsurface images at locations where the geo‐radar technique is limited by lack of depth penetration.

Why it matters

OpenAlex reports 38 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

ABSTRACT Coincident 30 Hz and 100 Hz ultra‐shallow seismic data sets acquired across a sedimentary sequence in northern Switzerland are characterized by unusually high levels of source‐generated noise (i.e. guided and surface waves). To image reflections in these noisy data it is necessary to apply an extensive processing sequence that includes semi‐automatic first‐break picking, amplitude scaling, multichannel spiking deconvolution, bandpass‐frequency filtering, careful selection and application of top mute functions, iterative velocity and residual static correction analyses, application of NMO corrections and appropriate stretch mutes, linear τ–p filtering and stacking. After processing, nearly identical reflection patterns are observed on the resultant 30 Hz and 100 Hz stacked seismic sections. In particular, an ultra‐shallow reflector is mapped beneath the entire length of the approximately 50.3 m long profiles at 1.4–2.3 m depth; differences in depth between the two stacked sections are less than 0.15 m. Since this reflector is practically coincident with an ultra‐shallow refractor across which the velocity increases from about 275 m/s to over 2100 m/s, it is interpreted as the groundwater table. Such an interpretation is consistent with the groundwater level observed in nearby boreholes. Underlying the ultra‐shallow reflection is a quasi‐continuous event that originates from 1.9–10.0 m depth. This deeper reflection probably represents the boundary between glaciofluvial and underlying glaciolacustrine sedimentary units. We conclude that state‐of‐the‐art processing techniques may allow ultra‐shallow reflections to be extracted from seismic data dominated by source‐generated noise. As shown in earlier studies, the ultra‐shallow seismic reflection method is capable of supplying high‐resolution subsurface images at locations where the geo‐radar technique is limited by lack of depth penetration.

Key concepts: Geology, Reflection (computer programming), Seismology, Deconvolution, Noise (video), Reflector (photography), Regional geology, Geodesy

Related papers

Back to paper searchBrowse research topicsOriginal source
Ultra‐shallow seismic reflection imaging in a region characterized by high source‐generated noise — Research Paper | ScholarLens