Supplementary Material II for "Spatially coherent warming and heterogeneous rainfall change in São Paulo State, Brazil"

This record contains Supplementary Material II associated with the manuscript “Spatially coherent warming and heterogeneous rainfall change in São Paulo State, Brazil”. It comprises four Python scripts implementing the complete computational workflow used to process the Brazilian Daily Weather Gridded Data (BR-DWGD), calculate temperature and precipitation indicators, estimate long-term trends and change points, produce spatially aggregated summaries, and generate the principal and supplementary figures presented in the manuscript.

The scripts must be executed sequentially:

01_SP_BRDWGD_Daily_Data_Export_1961_2025.py retrieves daily precipitation, maximum-temperature, and minimum-temperature data from the BR-DWGD collection in Google Earth Engine for the native 0.1° grid cells intersecting São Paulo State. The data are organized by variable and exported as monthly CSV files for the period from 1 January 1961 to 31 December 2025.

02_SP_BRDWGD_ETCCDI_Trends_and_Changepoints_1961_2025.py reconstructs the daily grid-cell time series and calculates 31 temperature and precipitation indicators at annual, seasonal, and monthly timescales. Percentile-based indicators are calculated relative to the 1961–1990 reference period. Trend magnitudes are estimated using Sen’s slope and expressed per decade, while statistical significance is evaluated using the Hamed–Rao modified Mann–Kendall test at p ≤ 0.05. Temporal change points are assessed using the Pettitt, Standard Normal Homogeneity, and Buishand range tests.

03_SP_BRDWGD_Trend_Synthesis_Tables_and_Figure12_1961_2025.py produces area-weighted analytical summaries for São Paulo State and its Water Resources Management Units. It also generates the integrated synthesis of annual, seasonal, and monthly temperature and precipitation trends presented as Figure 12 in the manuscript.

04_SP_BRDWGD_Trend_Maps_1961_2025.py generates the spatial trend maps presented in Figures 2–11 of the main manuscript and Figures S1–S8 of Supplementary Material I. Trend magnitudes are represented by Sen’s slope per decade, and hatching denotes grid cells with statistically significant Hamed–Rao modified Mann–Kendall trends at p ≤ 0.05. The maps use geographic coordinates referenced to the World Geodetic System 1984 (WGS 84; EPSG:4326) and include the boundaries of the Water Resources Management Units of São Paulo State.

The scripts are provided to support methodological transparency, computational reproducibility, and the independent verification of the results reported in the associated manuscript. Their execution requires access to Google Earth Engine and Google Drive, together with the Python packages identified within each script. Users may need to configure their Google Earth Engine project, input and output directories, and computational environment before execution.

The underlying BR-DWGD dataset is not redistributed in this record and remains subject to the terms and conditions established by its original providers. This supplementary material should be used in conjunction with the associated manuscript and Supplementary Material I. Any substantive modification to the deposited scripts will be released as a new Zenodo version.