Reproduction package (data and code) for "Measuring PM2.5 within built-up areas: contrasting drivers of urban air pollution across China"

Data and code to reproduce all numbers, tables and figures of the paper "Measuring PM2.5 within built-up areas: contrasting drivers of urban air pollution across China" (Will W. Qiang and Harry F. Lee; Discover Environment, under review). The package computes satellite-derived PM2.5 (Surface PM2.5 V5.GL.03) averaged over built-up land (GAIA) and over the whole municipal districts of 283 Chinese prefecture-level cities, 2008–2018, identifies persistent air-pollution agglomerations with local Moran's I, and estimates two-way fixed-effects models with meteorological variables from the China Meteorological Forcing Dataset. Contents: README.md (instructions, data sources, definitions, data dictionary), LICENSE, run_all.py, scripts for steps 01–10, inputs, and all outputs, including the city-year panel (output/08/panel.csv). Running "python run_all.py" downloads the public inputs and reruns all steps (Python 3.12); "python run_all.py --offline" reruns the agglomerations, models and tables from the included outputs in about 3 minutes. Licences: code under MIT, data compiled by the authors under CC BY 4.0. The yearbook values are taken from the China City Statistical Yearbook (China Statistics Press) and keep the publisher's terms; the workbooks themselves are not included.