Tropical Pacific SST anomalies generated by a Non-Gaussian Linear Inverse Model (NG-LIM)

Tropical Pacific (20S–20N; 120E–50W) sea surface temperature anomalies (ssta) generated by a Non-Gaussian Linear Inverse Model (Martinez-Villalobos et al., 2025; https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2025GL118649). The files are provided in two bundles: 1. Observational bundle (ERSSTv5-based): Tropical_Pacific_SSTA_ERSSTv5.nc. Sea surface temperature anomalies calculated from the NOAA Extended Reconstruction SST v5 reanalysis (1948–2022) (Huang et al., 2017). Tropical_Pacific_SSTA_ERSSTv5_10PCs.nc. Same as the previous dataset but filtered to retain only the first 10 SSTA Principal Components and Empirical Orthogonal Functions, accounting for ~90% of the SSTA variance. Tropical_Pacific_SSTA_LIM.nc. 10,000-yr integration of a standard LIM (Penland and Sardeshmukh, 1995) trained on dataset 2. Tropical_Pacific_SSTA_NGLIM.nc. 10,000-yr integration of a Non-Gaussian LIM (Martinez-Villalobos et al., 2025) trained on dataset 2. 2. CMIP6 model emulations ([Model name].zip):One .zip file for each of 30 CMIP6 historical models. Inside each .zip file, four files are provided in direct analogy to the observational bundle above: Tropical_Pacific_SSTA_[MODEL].nc. Historical SST anomalies from the given CMIP6 model. Tropical_Pacific_SSTA_[MODEL]_10PCs.nc. Same anomalies, reduced to the first 10 PCs/EOFs (variance explained depends on the model, but on average around 75%). Tropical_Pacific_SSTA_[MODEL]_LIM.nc. 10,000-yr LIM integration trained on the 10-PC dataset. Tropical_Pacific_SSTA_[MODEL]_NGLIM.nc. 10,000-yr NG-LIM integration trained on the 10-PC dataset. In addition, the basic code used to calculate and integrate an NG-LIM (and a standard LIM) is provided (NG-LIM_Basic_Code.ipynb), together with full field SST data that can be used to test the code (SST from the NOAA Extended Reconstruction SST v5 reanalysis, regridded bilinearly to a 2.5x2.5 horizontal resolution) When using the data please cite https://doi.org/10.5281/zenodo.14775714 (the data) and Martinez-Villalobos et al., 2025 (for the methodology). Any question, please contact Cristian Martinez-Villalobos at [e-mail]. References Huang, B. et al. (2017). Extended Reconstructed Sea Surface Temperature, Version 5 (ERSSTv5): Upgrades, Validations, and Intercomparisons. Journal of Climate, 30, 8179–8205. Penland, C., & Sardeshmukh, P. D. (1995). The Optimal Growth of Tropical Sea Surface Temperature Anomalies. Journal of Climate, 8, 1999–2024. Martinez-Villalobos, C., Capotondi, A., Deser, C., Dewitte, B., Holbrook, N. J., Newman, M., et al. (2025). A low-order data-driven model of ENSO diversity. Geophysical Research Letters, 52, e2025GL118649. https://doi.org/10.1029/2025GL118649