TY - JOUR
T1 - Inverse stochastic-dynamic models for high-resolution Greenland ice core records
AU - Boers, Niklas
AU - Chekroun, Mickael D.
AU - Liu, Honghu
AU - Kondrashov, Dmitri
AU - Rousseau, Denis Didier
AU - Svensson, Anders
AU - Bigler, Matthias
AU - Ghil, Michael
PY - 2017
Y1 - 2017
N2 - Proxy records from Greenland ice cores have been studied for several decades, yet many open questions remain regarding the climate variability encoded therein. Here, we use a Bayesian framework for inferring inverse, stochastic-dynamic models from 18O and dust records of unprecedented, subdecadal temporal resolution. The records stem from the North Greenland Ice Core Project (NGRIP), and we focus on the time interval 59-22 ka b2k. Our model reproduces the dynamical characteristics of both the 18O and dust proxy records, including the millennial-scale Dansgaard-Oeschger variability, as well as statistical properties such as probability density functions, waiting times and power spectra, with no need for any external forcing. The crucial ingredients for capturing these properties are (i) high-resolution training data, (ii) cubic drift terms, (iii) nonlinear coupling terms between the 18O and dust time series, and (iv) non-Markovian contributions that represent short-term memory effects.
AB - Proxy records from Greenland ice cores have been studied for several decades, yet many open questions remain regarding the climate variability encoded therein. Here, we use a Bayesian framework for inferring inverse, stochastic-dynamic models from 18O and dust records of unprecedented, subdecadal temporal resolution. The records stem from the North Greenland Ice Core Project (NGRIP), and we focus on the time interval 59-22 ka b2k. Our model reproduces the dynamical characteristics of both the 18O and dust proxy records, including the millennial-scale Dansgaard-Oeschger variability, as well as statistical properties such as probability density functions, waiting times and power spectra, with no need for any external forcing. The crucial ingredients for capturing these properties are (i) high-resolution training data, (ii) cubic drift terms, (iii) nonlinear coupling terms between the 18O and dust time series, and (iv) non-Markovian contributions that represent short-term memory effects.
U2 - 10.5194/esd-8-1171-2017
DO - 10.5194/esd-8-1171-2017
M3 - Journal article
AN - SCOPUS:85029618402
VL - 8
SP - 1171
EP - 1190
JO - Earth System Dynamics
JF - Earth System Dynamics
SN - 2190-4979
IS - 4
ER -