Pipeline data for the "Top lights" replication package: raw satellite and gridded inputs

Raw satellite, gridded and boundary inputs for the replication package of Bluhm and Krause (2022), "Top lights: Bright cities and their contribution to economic development", Journal of Development Economics 157, 102880.

This record exists so that the whole pipeline can be rebuilt from scratch. Most replicators do not need it: the replication package itself carries the corrected lights, the analysis datasets and every derived panel behind the paper's tables and figures.

Each file is a tar of one source and unpacks to data/input/public/<name> under the repository root. code/setup/download_pipeline_data.sh in the replication package fetches, verifies and unpacks them for you, and takes --only to fetch just part of the record.

Contents: DMSP-OLS stable lights v4 (34 satellite-years, 1992-2013) and radiance-calibrated composites, both from the Earth Observation Group; the VIIRS 2015 annual composite; the authors' de-blurred DMSP lights for Africa; GPW v4 land and water area (CIESIN); gas-flare vectors (NOAA/EOG); Natural Earth populated places; Africapolis 2018; a frozen World Development Indicators snapshot; four OpenStreetMap city extracts; and the German and US benchmark sources.

Attribution travels with the data: the EOG notice and credit line for the DMSP and VIIRS products, the GPW citation, Abrahams, Oram and Lozano-Gracia (2018) for the de-blurring method, dl-de/by-2-0 for the German layers, and ODbL for the OpenStreetMap extracts. Per-source licences are listed in LICENSE.txt in the replication package.