This record outlines a code archive analyzing climate extremes and vegetation responses in South Asia, suggesting valuable insights for future climate adaptation.
This record archives the code and processed outputs supporting a study of hydroclimatic-zone stratified climate-extreme change and vegetation sensitivity across South Asia. Seven hydroclimatic zones were delineated on a 0.10° land grid from ERA5-Land precipitation and temperature diagnostics, elevation and latitude, by consensus of K-means, Ward hierarchical clustering and Gaussian mixture modelling, with a self-organising map as an independent check. Twelve CMIP6 models were evaluated against ERA5-Land for precipitation, maximum temperature and minimum temperature over 1985–2014, screened by variable, bias-corrected at zone level, and combined into annual ensembles under SSP2-4.5 and SSP5-8.5. Annual ETCCDI-style indices (PRCPTOT, RX1day, CDD, TXx, TNn) were analysed with Sen's slope, Mann–Kendall, pre-whitened Mann–Kendall and Pettitt change-point tests under false-discovery-rate control. MODIS NDVI, EVI, GPP and NPP were linked to the observed indices using Spearman correlations and leave-one-year-out validated ridge models, which were then applied to the projections as a confidence-qualified future vegetation-response diagnostic. The record contains two archives. SA-Climate-Veg_code_v1.1.0.zip holds twelve numbered analysis notebooks in execution order, standalone rerun and figure scripts, the data-acquisition notebooks used with Google Earth Engine and Xarray-Earth Engine, workflow and reproducibility documentation, and the Conda environment specification. SA-Climate-Veg_processed-data_v1.1.0.zip holds the consensus zone layers in NetCDF and GeoTIFF form, 116 result tables covering zoning validation, CMIP6 evaluation and screening, bias-correction factors, annual ETCCDI-style zone series and period changes, trend and change-point statistics and vegetation–climate linkage results, together with change and trend raster fields and stage-level methods notes. Source datasets (ERA5-Land, CMIP6, MODIS, SRTM, Natural Earth) are not redistributed and remain available from their original providers; raw and intermediate daily gridded stacks are not archived and can be regenerated with the released code. Coordinate reference system EPSG:4326; analysis grid 0.10°; baseline period 1985–2014. The record is released under CC BY 4.0. The code is additionally available under the MIT Licence, as in the linked GitHub repository.
No takes yet. Share an insight, caveat, or question.
Author names withheld for double-anonymized peer review` (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: