Archive Meta Data

Author(s)

Peter Ditlevsen and Susanne Ditlevsen

Title

CorrectionNatureCommunications2023

Description

This is data and UPDATED code for the paper: "Warning of a forthcoming collapse of the Atlantic meridional overturning circulation" published in Nature Communications, 14, Article number: 4254 (2023). 

Unfortunately, we identified two coding errors in the original files. They only affect the file AMOCestimation.Rmd (and thereby the file AMOCestimation.html), which are updated here. The other files are also included, they are not changed compared to the original archive (https://doi.org/10.17894/ucph.b8f99b67-d4e6-4a2e-b518-00bddeed323b). The corrections are:

1) The Strang splitting maximum likelihood estimator had a minor error (thanks to Anders Gantzhorn Kristensen for identifying it). In the last step of the Strang splitting in the pseudo maximum likelihood estimation, the flow should have been evaluated in the observation at time t_i, but was wrongly partially evaluated at time t_(i-1). When corrected, the tipping time estimate changes by 8 years.

2) The code for calculating uniform residuals was wrong. It has no influence on estimates, only on model control. 

The following files are included: 

Supplementary_Information.pdf
Pdf containing supplementary information for the paper.
S1: Maximum likelihood estimation from data following an Ornstein-Uhlenbeck process.
S2: Pseudo-likelihood estimation of the tipping time, based on Strang splitting.

AMOCestimation.html (UPDATED)
File containing R code, explanations and figures from the paper.

AMOCestimation.Rmd (UPDATED)
R Mark down file that produces the AMOCestimation.html file. The file needs access to the files below.

AMOCdata.txt 
Text file with the AMOC fingerprint. Contains the following variables:
"time" : time in years
"AMOC0" : sea surface temperature (SST) anamoli, corrected for seasonal variations
"AMOC1" : AMOC0 subtracted 1 x global mean
"AMOC2" : AMOC0 subtracted 2 x global mean
"GM"       : global mean temperature (GM)

SimulatedTraces.Rdata
R archive containing 3 elements:
xx.data : R data frame containing 1000 simulated traces
nsim : numerical variable containing number of simulated traces, equals 1000
Delta_t : time step in years used in the simulations, equals 1/12

EstimMatrix_1000repetitions.xlsx
Excel file containing the 1000 estimates from simulated data, used for constructing bootstrap confidence intervals.