Districts geometry in India

[Cristina Vrinceanu]

Hi all, 

I am doing an analysis where I map various socio-economic datasets to Indian districts. However, since these are linked to the year, I need to first curate the geometries for each to link to the name and area.

After doing a quality analysis, I decided the OSM/Overture districts seem to be falling best on the current boundaries, but I have some missing geometry gaps (topology errors) in RJ, AS, MP and GJ. I cannot seem to find a comparison dataset for the prior year so I could correctly fill in these gaps and assign them to the right district. 

In RJ, for example, the gap covers and area where the Uidapur district was split last year in August and now some parts of the gap are still in Uidapur, some belong to the newly formed Salumbar etc. The official RJ data is not up to date (State GIS Portal), neither is Google Maps. For 2023, the only source I found is the ESRI Living Atlas districts which are apparently from Survey of India have very fragmented (many multipart separate geometries) that are not in the OSM dataset, in some places they seem to have been simplified or digitised from a lower scale and in some places they are completely shifted from the OSM boundary. 

Does anyone know any reliable source to compare with for 2023/2024? Or has any suggestion on how to assign missing bits?

At the moment I am manually tracing the district tehsil movements and morphing (based on info online) and comparing with villages outline (all gap areas are just villages), although the latter is very annoying since villages coincide in name within the same state.

Any suggestions are appreciated. Thanks!

[Gunngunn]

You may try creating unique identifier either use village code or concatenate state name, dist name,block name, panchayat name and village name... Either of them may work 

--
Datameet is a community of Data Science enthusiasts in India. Know more about us by visiting http://datameet.org
---
You received this message because you are subscribed to the Google Groups "datameet" group.
To unsubscribe from this group and stop receiving emails from it, send an email to datameet+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/datameet/30a45ff0-4655-422c-8ce6-bad672a5e978n%40googlegroups.com.

[Cristina Vrinceanu]

Thanks Gunngunn, 

I ended up matching with the ESRI ones and filling them but still need to do a quality check. Some are missing tehsils, some are wildlife sanctuaries, some are just random areas. FYI: ESRI look the way they look because they seem to be a hybrid between official SOI and digitised Census Handbook maps, so can't be used per se. I believe the fill may be correct though, missing parts seem to be falling on the right outline but I have no information on the timestamp for SOI boundaries used by ESRI (assuming 2024). In the absence of proper official data, I will take this.

To unsubscribe from this group and stop receiving emails from it, send an email to datameet+u...@googlegroups.com.

[Digvijay Bendrikar Shinde]

As part of our work with the Public Health Department of Rajasthan, We at Khushi Baby worked on creating 50 new district boundaries for Rajasthan. 

as of June 2024, the official map of the new district boundaries of Rajasthan is not available in the public domain.

somehow, Wikipedia has the map! :D

Regards,

Digvijay

To view this discussion on the web visit https://groups.google.com/d/msgid/datameet/b66d0db5-90b8-43b8-9803-bf42d1d5c191n%40googlegroups.com.