Quasi-Experimental Evaluation of the Association between Payments for Ecosystem Services and Satellite-Based Vegetation Dynamics in Cundinamarca, Colombia

This repository contains the data, analytical scripts, and supplementary materials supporting the study entitled “Quasi-Experimental Evaluation of the Relationship between Payments for Ecosystem Services and Vegetation Condition Using NDVI in Cundinamarca, Colombia.” The study evaluates the relationship between participation in the “Yo Protejo ¡Agua para Todos!” Payment for Ecosystem Services program and changes in vegetation condition in Cundinamarca, Colombia, during the 2015–2024 period. The empirical strategy combines panel-data difference-in-differences models, event-study specifications, propensity-score-based inverse probability of treatment weighting, alternative spatial comparison groups, and multiple robustness analyses. The repository includes: Google Earth Engine scripts used to process Landsat 8 and Landsat 9 Surface Reflectance imagery and calculate annual vegetation indices, including the Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), and Soil-Adjusted Vegetation Index (SAVI). Stata do-files used for data preparation, descriptive statistics, pre-treatment trend analyses, placebo tests, difference-in-differences estimations, event-study models, propensity score estimation, inverse probability of treatment weighting, and robustness checks. Analytical datasets containing annual vegetation-index observations for treated properties participating in the Payment for Ecosystem Services program and spatial comparison units located in neighboring non-treated areas. Processed model outputs, tables, and figures used to prepare the results reported in the manuscript and its supplementary materials. The analytical period is divided into a pre-intervention phase (2015–2017), a program implementation phase (2018–2020), and a post-implementation phase (2021–2024). The unit of analysis corresponds to georeferenced 30 × 30 m satellite-derived observation points, organized as a longitudinal panel. The materials are provided to facilitate transparency, reproducibility, verification, and future research on the environmental effectiveness of Payment for Ecosystem Services programs. Geographic identifiers and variables that could disclose confidential information about individual beneficiaries or private properties have been removed, anonymized, or spatially generalized where necessary. Users of these materials should cite both the associated manuscript and this Zenodo repository.