Just wondering if there is any published material on how to perform spectral decomposition for channel identification from low resolution seismic data.
Dear Brice, your question have two key-words: identification and resolution.
if you are talking about resolution, you are thinking about vertical resolution, concerns the ability to distinguish that more than one reflecting interface; the most common definitions of resolution limit are those attributable to Rayleigh and to Widess. The values of resolvable limit given by these two definitions are, respectively, 1/4 and 1/8 of the dominant wavelength.
How thin is a thin bed? by M. B. Widess, 1973. Geophysics, Vol. 38, N 6, December 1973, p. 1176-1180.
On the other hand, if you are talking about identification, you are thinking about the detection capability in your seismic and your tools, such as spectral decomposition.
Please, check in my profile the following article:
Seismic Attributes Analysis: High frequency and wedge modelling. basically concepts Mario_E_Sigismondi1 2018_all_rights_reserved Mario_E_SIGISMONDIConference PaperFull-text available.
We have been used CWT and HDSF for channel detection in low resolution 3D seismic in structurally complex area at Llanos Foothills of Colombia.
Each seismic dataset may react differently to ( STFT, CWT or MPD ) techniques, and the spectral decomposition methodology does not always work as expected. In CWT for instant the interpreter should experiment iteratively to establish the best wavelet for a particular seismic volume. We have a work under review in both CWT and HDSF. Also, you can check the below link: