Enhancing in search of Milankovitch cycles from stratigraphic record using convex optimization algorithm

Area of Science:

• Geosciences
• Stratigraphy
• Paleoclimatology

Background:

• Identifying Milankovitch cycles is crucial for precise geochronology in cyclostratigraphy.
• Sedimentary noise often obscures these important orbital signals, challenging conventional spectral analysis.

Purpose of the Study:

• To develop and validate a novel algorithm for enhancing Milankovitch cycle resolution in stratigraphic records.
• To improve stratigraphic precision by effectively distinguishing orbital periodicities from noise.

Main Methods:

• Developed a new algorithm employing convex optimization within spectral analysis.
• Applied the algorithm to diverse stratigraphic sections (Miocene Java, Mahakam Delta, Pleistocene Flores, Towuti Lake).

Main Results:

• Successfully identified all targeted Milankovitch signals with over 99% confidence.
• The algorithm effectively rejected the null hypothesis of noise or signal absence at a 1% significance level.
• Observed no deviations from identified periodicities, confirming signal robustness.

Conclusions:

• The convex optimization algorithm significantly enhances the detection of Milankovitch cycles in complex stratigraphic data.
• This method offers a robust tool for improving geochronological accuracy in cyclostratigraphic studies.
• Accurate parameter selection (z and λ) is key for optimal algorithm performance.