1. An information-theoretic approach to extracting climate signals from deep polar ice cores
- Author
-
Elizabeth Bradley, Tyler R. Jones, Joshua Garland, Michael Neuder, and James W. C. White
- Subjects
geography ,geography.geographical_feature_category ,010504 meteorology & atmospheric sciences ,Applied Mathematics ,Geothermal heating ,General Physics and Astronomy ,Statistical and Nonlinear Physics ,Information theory ,01 natural sciences ,Earth system science ,Ice core ,13. Climate action ,Climatology ,0103 physical sciences ,Paleoclimatology ,Polar ,Entropy (information theory) ,Ice sheet ,010306 general physics ,Mathematical Physics ,Geology ,0105 earth and related environmental sciences - Abstract
Paleoclimate records are rich sources of information about the past history of the Earth system. Information theory provides a new means for studying these records. We demonstrate that weighted permutation entropy of water-isotope data from the West Antarctica Ice Sheet (WAIS) Divide ice core reveals meaningful climate signals in this record. We find that this measure correlates with accumulation (meters of ice equivalent per year) and may record the influence of geothermal heating effects in the deepest parts of the core. Dansgaard-Oeschger and Antarctic Isotope Maxima events, however, do not appear to leave strong signatures in the information record, suggesting that these abrupt warming events may actually be predictable features of the climate's dynamics. While the potential power of information theory in paleoclimatology is significant, the associated methods require well-dated and high-resolution data. The WAIS Divide core is the first paleoclimate record that can support this kind of analysis. As more high-resolution records become available, information theory could become a powerful forensic tool in paleoclimate science.
- Published
- 2019
- Full Text
- View/download PDF