Data and Code Supporting Local-Scale Evaluation of Satellite-Driven Elevation Models in Low-Elevation Delta Terrains in Bangladesh

This dataset supports the manuscript “Local-Scale Evaluation of Satellite-Driven Elevation Models in Low-Elevation Delta Terrains in Bangladesh”. The study evaluates four open-access digital elevation model (DEM) products SRTM, NASADEM, Copernicus GLO-30, and ALOS AW3D30 over the low-relief Jamuna floodplain of Sirajganj District, Bangladesh. The analysis uses a dual-domain validation design, separating water-surface validation using ICESat-1 GLAS altimetry from land-surface validation using Survey of Bangladesh spot-height reference data. Because the candidate DEM products do not share the same native vertical reference system, the workflow includes vertical-datum harmonization between EGM96- and EGM2008-referenced heights before computing final accuracy metrics. The uploaded package contains derived project materials needed to support the reported analysis, including vertical-datum harmonization outputs, local EGM96–EGM2008 geoid-difference summaries, DEM validation tables, error metrics, bootstrap confidence intervals, paired statistical comparison outputs, MCDA ranking outputs, Monte Carlo rank-stability summaries, figure-generation materials, selected figures, scripts, notebooks, metadata files, and package manifests. This upload is not intended to redistribute all raw source datasets. Publicly available DEM and satellite products remain available from their original providers. Third-party or restricted reference data, including Survey of Bangladesh source materials, are not redistributed unless redistribution rights are confirmed. Where needed, derived summaries are provided to support reproducibility without exposing restricted source data. The package is intended to document the analytical workflow and provide the derived materials necessary to verify the reported DEM ranking, vertical-datum harmonization, water-domain and land-domain validation results, MCDA scores, bootstrap intervals, and pairwise statistical tests.