This deposit contains the analysis code and a sample of the input data for a study quantifying how grassland greening (changes in Leaf Area Index, LAI) affects land-surface temperature (LST) and soil temperature on the Tibetan Plateau (2003–2023).The core method computes the local sensitivity of temperature to LAI using a spatial moving-window Theil–Sen regression, comparing only physically similar pixels (same vegetation class and permafrost type, similar elevation and sub-grid class fractions). Additional scripts perform: machine-learning (Random Forest) bias correction of ERA5-Land soil temperature against in-situ stations; aggregation of monthly sensitivities into climatologies and seasonal means; a surface-energy-balance decomposition of the LST sensitivity into albedo, latent-heat, sensible-heat, ground-heat and emissivity pathways; and auxiliary analyses of snow-cover and lake implication.Because the full inputs span 2003–2023 and are tens of gigabytes (and are openly available from their original providers), the bundled sample contains the ORIGINAL input files at their native resolution, projection and full Tibetan-Plateau extent for a 2-year period (2003–2004, ~6.5 GB). Only the time range is reduced. The file README.txt gives full instructions, software requirements, expected outputs, and the list of original data sources.
Kanon Guédet Guédé (Tue,) studied this question.