Early warning signal analysis of Dansgaard-Oeschger events across Greenland ice cores

Code for the analysis of EWS for DO events across Greenland ice core records related to

Hummel, C., Boers, N., and Rypdal, M. (2025): Inconclusive early warning signals for Dansgaard-Oeschger events across Greenland ice cores, [preprint], https://doi.org/10.5194/egusphere-2024-3567

Full Changelog: https://github.com/hummelsumm/do_ews_across_greenland/commits/v_zenodo

 δ18O records from Greenland ice cores

• NGRIP with irregular, 5-, 10- and 20-year temporal resolution 
• NEEM with 10-year temporal resolution
• GRIP with 20-year temporal resolution
• GISP2 with 20-year temporal resolution

Early warning indicators

• based on critical slowing down: Variance V, lag-one autocorrelation parameter α , and restoring rate λ
  
  Analyses of V and α of regularly sampled records are based on and adapted from Niklas Boers' code related to Boers, N. (2018): Early-warning signals for Dansgaard-Oeschger events in a high-resolution ice core record.
  
  Analyses of λ of NGRIP in 5-year resolution are based on Niklas Boers' code related to Boers, N. (2021): Observation-based early-warning signals for a collapse of the Atlantic Meridional Overturning Circulation.
  
  Analyses of α of the irregularly sampled NGRIP record are based on Mudelsee, M. (2002): TAUEST: a computer program for estimating persistence in unevenly spaced weather/climate time series, as well as Lenoir, G. and Crucifix, M. (2018): A general theory on frequency and time–frequency analysis of irregularly sampled time series based on projection methods – Part 2: Extension to time–frequency analysis. The corresponding python2 code is provided in the WAVEPAL package, which has been adapted to python3.
  
  
• based on wavelet analysis: scale-averaged wavelet coefficient w , and local Hurst exponent H
  
  Analyses of regularly sampled records are based on Torrence, C. and Compo, G. P. (1998): A Practical Guide to Wavelet Analysis. Wavelet software was provided by C. Torrence and G. Compo, and is available at http://paos.colorado.edu/research/wavelets/''. We use code adapted from Bernd Blasius' implementation of his WaveletAnalysis code
  
  Analyses of irregularly sampled records are based on Lenoir, G. and Crucifix, M. (2018): A general theory on frequency and time–frequency analysis of irregularly sampled time series based on projection methods – Part 2: Extension to time–frequency analysis. The corresponding python2 code is provided in the WAVEPAL package, which has been adapted to python3

Data sources:

• NGRIP: NGRIP members (2004): High-resolution record of Northern Hemisphere climate extending into the last interglacial period, and Gkinis et al. (2014): Water isotope diffusion rates from the North-GRIP ice core for the last 16,000 years – Glaciological and paleoclimatic implications, provided on https://www.iceandclimate.nbi.ku.dk/data/
• NEEM: Gkinis, V., et al. (2020): NEEM ice core High Resolution (0.05 m) Water Isotope Ratios (18O/16O, 2H/1H) covering 8-129 ky b2k
• NGRIP, GRIP, GISP2 (20-year resampling): Seierstad et al. (2014): Consistently dated records from the Greenland GRIP, GISP2 and NGRIP ice cores for the past 104 ka reveal regional millennial-scale δ18O gradients with possible Heinrich event imprint, and Rasmussen et al. (2014): A stratigraphic framework for abrupt climatic changes during the Last Glacial period based on three synchronized Greenland ice-core records: refining and extending the INTIMATE event stratigraphy,  provided on https://www.iceandclimate.nbi.ku.dk/data/
• GS/GI onsets: as given in Table S1 in the supplement of Boers, N. (2018): Early-warning signals for Dansgaard-Oeschger events in a high-resolution ice core record