Code and data for "Emergence of a hemispheric dipole in global mountain local precipitation"

Overview This repository contains the source code, sample data, and the extracted Mountain Local Precipitation (MLP) datasets used in the manuscript. The contents are organized into three main directories: code/: Contains the core algorithms for MLP identification (both the event-based framework and the daily-scale adaptation), as well as scripts for processing other environmental variables. sample_data/: Contains examples of intermediate data generated during the MLP processing, alongside the World Bank socioeconomic data used in the study. result/: Contains the complete, finalized MLP datasets used in the manuscript, including time-series data and spatial shapefiles with calculated slopes and multi-year averages for each basin. Directory Details 1. code/ All scripts are provided as Jupyter Notebooks (.ipynb). While there are slight syntax variations depending on the specific precipitation dataset being processed, the core methodology remains consistent across all products. This directory includes: Core MLP Identification: Scripts for the two primary extraction methods: Event-based framework (hourly scale) Daily-scale adaptation Variable Processing: Scripts used to process and extract other environmental and climatic variables, such as air temperature and aerosols. 2. sample_data/ This directory provides examples of the intermediate data used during the MLP extraction process, as well as the World Bank national income classification data. Note on Data Volume: The raw precipitation data and intermediate processing files span a long temporal period (1980–2024) at high spatiotemporal resolutions (hourly, 0.1°). Because the total volume of these files exceeds 20 TB, it is not feasible to host them here. Therefore, this repository provides representative sample data for the intermediate steps, while the fully processed final results are provided in the result/ directory. 3. result/ This directory contains the complete MLP extraction results derived from all precipitation datasets analyzed in the study. Time-series Data (.csv): The extracted MLP time series are saved as plain text CSV files, categorized by individual basins and different spatial area thresholds. These can be directly opened and analyzed in Windows (e.g., via Excel) or any standard data analysis software. Spatial Data (.shp): Basin-level spatial data are provided as Shapefiles. These files contain the multi-year averages and trend slopes of MLP across all precipitation datasets, integrated with basin-level population counts and income levels. These files can be opened and visualized using GIS software such as ArcGIS or QGIS. System Requirements All code execution and data processing were performed under the following system environment: Operating System: CentOS 7.7 Python Version: 3.12.9 For further questions regarding the methodology or data, please refer to the Materials and Methods section of the manuscript.