Published April 9, 2025 | Version v1
Conference paper Open

Stratifying settlement by building type: assessing the accuracy of the GHS-BUILT-S dataset

  • 1. WorldPop, School of Geography and Environmental Science, University of Southampton

Description

Global high-resolution datasets on buildings and settlement have become increasingly available over the past decade but have lacked attributes on building type or use. The recent release of the GHS- BUILT-S dataset enables stratification of residential and non-residential built settlement globally for the first time. In this study we compare GHS-BUILT-S against national residential household datasets for Germany and Zambia, assessing its accuracy in different built landscapes. Our results show that a greater geographic extent is identified as residential in GHS-BUILT-S, including most mapped residential households in the census datasets. However, areas of non-residential buildings are likely underestimated in GHS-BUILT-S.

Files

GISRUK2025_Paper_Chamberlain_et_al_2025_proceedings_submission.pdf

Files (520.4 kB)