Seismic data processing collected from arctic sea ice is often challenging. Large amplitudes of coherent surface waves interfere with the reflected seismic waves and can make the reflection data virtually… Click to show full abstract
Seismic data processing collected from arctic sea ice is often challenging. Large amplitudes of coherent surface waves interfere with the reflected seismic waves and can make the reflection data virtually unusable. In order to understand this noise, forward modelling of the full wavefield is a useful approach. I present and test a forward modelling workflow by comparing the modelling results with common numerical approaches. The main differences between numerical curves and modelling results depend on simplifications in the assumptions of the numerical approaches. Forward modelling is less dependent on rough assumptions and can give a more accurate understanding of the surface wave propagation and its interaction with reflected waves. Furthermore, a thin elastic layer with high velocity located between two low velocity layers, like sea ice, generates guided waves. The numerical approaches focus only on the flexural waves and ignore other wave types. The presented results demonstrate that forward modelling can provide improved detail in the generated wavefield. In addition, the forward modelling results in this work show that leaky Rayleigh and Scholte waves can exist simultaneously.
               
Click one of the above tabs to view related content.