Enhancing spectral fidelity in cyclostratigraphic studies using superlets

Cyclostratigraphy relies on spectral analysis to decode the imprint of astronomical cycles in stratigraphic proxy data. Traditional methods, such as Fourier-based techniques and the continuous wavelet transform (CWT), are constrained by a fundamental trade-off between temporal and frequency resolution. These limitations constrain the ability of these spectral techniques to track changes in astronomical periods with stratigraphic depth and to separate cycles with closely spaced frequencies simultaneously. The recently developed superlet transform overcomes these classical limitations by combining multiple wavelets at the same central frequency, each with a different number of cycles controlling its Gaussian envelope width. By computing the geometric mean of the wavelet responses, superlets achieve enhanced frequency resolution while maintaining temporal precision, yielding a sharper time–frequency representation than the conventional CWT. Here, we present a new suite of functions in the WaverideR R package that applies the superlet transform to cyclostratigraphic datasets. The implementation is specifically tailored to the characteristics of (cyclo)stratigraphic proxy records, incorporates log2-period scaling, supports analysis in both the depth and time domains, and employs FFT-based convolution to improve computational efficiency. Using these tools, users can generate superlet scalograms, identify and track the periods of astronomical cycles, and construct cyclostratigraphic age models. The superlet transform also enables the study of amplitude-modulation patterns and the discrimination of closely spaced cycles, such as those comprising short eccentricity or precession signals, a task that the classical CWT struggles with. Tests on both synthetic and real stratigraphic datasets demonstrate that the superlet transform substantially improves spectral fidelity compared to traditional wavelet- and Fourier-based methods, establishing it as a powerful tool for analysing and interpreting astronomical signals in proxy records.