Supplements for Do coupled megathrusts rupture

codebase_coupling.zip Analysis code; Fig2.py is a worked example 11 published coupling models (NetCDF lon/lat grids) Earthquake source models: SRCMOD, USGS finite-fault, plus 1960 Chile / 1964 Alaska / Maule 2010 Updated code with examples: https://github.com/baroryan/slip-coupling Updated coupling models: the Coupling Cloud, https://couplingcloud.ucsd.edu coupling_with_noise.zip 12 .npz, one per margin, each an array raw_noise of shape (n_lat, n_lon, 500) 500 coupling realizations per margin, from GNSS uncertainty propagated through TDE Green's functions Full coupling fields, not perturbations; not clipped to [0,1] samples_mean_null.zip k_3p5_v2.parquet — coupling sampled over observed ruptures (12.5 M rows) epi_center_only.parquet — same, epicentres only (35 k rows) null_per_event/ — null ensemble, each rupture relocated 500× (13 files, 1.245 B rows) Columns: coupling, Mw, label, eventTag, long/lat, riseTime, weight samples_noise.zip sptial_mean_k_0p7.parquet — observed rupture samples (2,499,655 rows) sptial_noise_k_0p7.npz — coupling for each sample under all 500 realizations (2,499,655 × 500) HBI_strong_dw.nc and HBI_moderate_dw.nc 3-D quasi-dynamic rate-and-state cycle simulations (HBI), strong vs. moderate dynamic weakening Variables: slip (m), vel (m/s), tau, on x/y (km) × time; plate rate 1e-9 m/s These dataset files are split into multiple files combine them using cat (for example: cat HB_strong_dw_segment_* > HBI_strong_dw.nc)