Semi-automatic mapping of pre-census enumeration areas and population sampling frames

Enumeration Areas (EAs) are the operational geographic units for census data collection and frequently form national sampling frames. In many low-income or conflict-affected countries, EA demarcations are incomplete, outdated, or missing, and creating/updating them is time-, cost-, and labor-intensive, especially when balancing population and area constraints. We present a semi-automatic approach to map pre-defined census EAs using high-resolution gridded population and settlement datasets, combined with publicly available natural and administrative boundaries. We demonstrate the approach by generating rural EAs for Somalia, where such mapping did not exist, and compare automated outputs against manually digitized urban EAs in Mogadishu and Hargeysa. The semi-automatically generated EAs align with recognizable ground boundaries, have appropriate sizes and populations for enumerator coverage, and our automated urban EAs have no gaps, in contrast to manually drawn urban EAs. This work shows the time-, labor-, and cost-saving value of automated EA delineation and points to the potential for widely available tools suitable for low-income, data-poor settings and beyond.