silentguy

>Thanks for your email. In our online FAQ for the DEGURBA methodology (https://population.un.org/wup/faqs), you'll see:

Does DEGURBA combine too many cities into a single conurbation?

In countries where population data were only available at relatively coarse resolution and densities are generally high, these estimates may include conurbation that are too large. For example, the DEGURBA-defined city that includes Hajipur in India appears to aggregate multiple distinct cities and towns. Using finer grained population data or even better a population grid based on a geo-coded census may lead to this conurbation shrinking and turning into a multitude of individual cities and towns.

The following online portal provides you more details: https://human-settlement.emergency.copernicus.eu/raster.php - Based on this methodology, Hajipur corresponds to the whole red urban centre combining multiple contiguous urban units. If you zoom on the interactive map you'll see the name of smaller urban places included. For the most recent periods, the initial units of analysis for the population grid were the 5,967 Sub-districts/Tehsils/Cercles of the 2011 census of India projected using the 2001-2011 censuses. Population distribution is downscaled using estimates of building volume from the GHSL, which you can see here: https://human-settlement.emergency.copernicus.eu/visualisati...

You can see that Patna is the larger/closest urban area, but it is not physically connected to the rest of the urban area due to the river in between. SO Patna appears as distinct urban area.