Data for PlaSIM-LSG AMOC index simulation in large ensemble using the TAMS algorithm

Data description The dataset refers to four different experiments: simulations at constant 500 ppm and 600 ppm CO₂, and simulations following the SSP2-4.5 and SSP5-8.5 scenarios. Each experiment is provided in two parts: “intial” files: contain the 256-member ensemble of free simulations. “final” files: contain the ensemble simulations after applying the TAMS algorithm. File contents "log*.o" files contain all the information during the application of TAMS. In particular the probability estimates given in the last line of the file. Note that this values, for technical reasons, represent an overestimation by a factor (1-25/256). Please refer to the paper for final estimates. "*.tar" files contain the information on AMOC index and scorefunction evolution during the simulation: Open those files with "tar -xf *.tar". In each folder, a "ctrobs" directory contains files of the type "namefile.ctrlobs/score.mmmm.yyyy.txt". Those files contain a single value representing che AMOC-index/score of the mmmm trajectory at time yyyy. "timesereis.py" contains the script necessary to produce timeseries in Fig 2 and Fig 3 "Response Tams.ipynb" contains the script necessary to produce plots in Fig 1 and supplementary. Please note the data is derived from Cini et al, 2023 "TAMS_Matteo_scorefunction.py" contains the core of the TAMS algorithm, here specifically adapted to work in combination with PlaSIM-LSG simulations Please note that "TAMS_amoc_final_ssp585" and "TAMS_amoc_final_ssp245" represent the TAMS simulation with rispectively 2100 and 2150 year as time limit, while their corresponding "intial" simulations extend further, providing the evolution of the AMOC up to the year 2200