Data files for "Locally stationary Argo ocean heat content estimates: Modeling, validation and uncertainty quantification"

This repository contains output files generated by the pipeline described in the paper "Locally stationary Argo ocean heat content estimates: Modeling, validation and uncertainty quantification" to produce monthly ocean heat content (OHC) estimates and uncertainties for 2004-2022. For the source code and detailed documentation of each pipeline step, please refer to https://github.com/ttsukianto/LocalGP_OHC. An overview of the pipeline steps and corresponding files is provided below. Preprocessing (Data.zip) Files starting with "Argo_data_aggr": Post-quality control Argo profile data (January 2023 GDAC snapshot) Processed folder: Vertically integrated temperature profiles for each vertical section (15-975 and 975-1850 dbar) Monthly folder: Monthly vertically integrated temperature profiles (files starting with "intTemp") and monthly residuals with climatological mean field removed (files starting with "intTempRes") Filtered folder: Monthly vertically integrated temperature profiles filtered by ocean masks in Data/Masks Extended folder: Monthly residuals grouped into 3-month windows for parameter estimation/anomaly mapping/conditional simulation Masking (Data.zip) Masks folder: Data availability masks defining ocean area sufficiently sampled by Argo (files starting with "dataMask"), data mask combined with Roemmich & Gilson bathymetry mask (https://sio-argo.ucsd.edu/RG_Climatology.html; files starting with "combinedMask"), and basin masks Mapping MeanField.zip: Estimated climatological mean field MLE.zip: GP parameter estimates (i.e., temporal median of maximum likelihood estimates) Anomaly.zip: Mapped unscaled OHC anomaly fields (i.e., before scaling by rho0 and cp0) FullField.zip: Mapped unscaled OHC full fields (i.e., climatological mean added to anomalies in above file) LocalCondSim_*.tar.gz: Local conditional simulation realizations by year with nugget effect included (zero-mean and unscaled) This work used Bridges-2 at Pittsburgh Supercomputing Center through allocation MTH230014 from the Advanced Cyberinfrastructure Coordination Ecosystem: Services & Support (ACCESS) program, which is supported by National Science Foundation grants #2138259, #2138286, #2138307, #2137603, and #2138296.