Re: [datameet] Identifying urban agglomeration boundaries

[amish sarpotdar]

Hi Aditya and others,

I have often thought of this since my PhD deals with some of the boundary
and territory issues. Some pointers from my own experience:

The GHSL is useful but will be hard to overlay with official boundaries and
is a plumbing fix more than a sustainable solution for figuring out UA's as
the resolution is quite flimsy at lower levels and for smaller UAs.

I think the solution then is to have a tiered approach using Census Town
Codes  that requires a bit of coding and lots of time and cleansing:

1. Gather all the village or town vector boundaries which I think are
already available
2. Check for OG vectors and missing territories.
3. Then Add and combine the vector boundaries for each UA corresponding to
the list in the Census according to town code numbers.

There are a lot of limitations with this approach and it is hard to express
in one single mail. However, the benefits of following the Census town code
approach are massive:


1. Once you have the UA vector file as compiled through census, overlaying
socio-economic and other indicators is a child's play as you just need to
use Census Town codes
2. Time series is possible if done for 2001 and 2021. Note: Between 2001
and 2021 the boundaries have changed but not as massively as between 1991
and 2001.


There are other approaches such as having a tiered composite region
approach, but those are more for time-series of data.

It is useful to acknowledge this is a long drawn process, but I would be
happy to help and volunteer with any data/code/inputs when needed.


Thanks.


Kind Regards,
Amish

On Wed, Oct 14, 2020 at 3:17 PM Aditya Medury  wrote:

> Hi,
>
> I am trying make sense of how to translate the definitions of urban
> agglomerations/million-plus cities in the 2011 census. Census defines urban
> agglomeration as follows:
>
> *An urban agglomeration is a continuous urban spread constituting a town
> and its adjoining outgrowths (OGs), or two or more physically contiguous
> towns together with or without outgrowths of such towns. An Urban
> Agglomeration must consist of at least a statutory town and its total
> population (i.e. all the constituents put together) should not be less than
> 20,000 as per the 2001 Census. In varying local conditions, there were
> similar other combinations which have been treated as urban agglomerations
> satisfying the basic condition of contiguity. Examples: Greater Mumbai UA,
> Delhi UA, etc.*
> *Source: *
> https://censusindia.gov.in/2011-prov-results/paper2/data_files/India2/1.%20Data%20Highlight.pdf
>
> Since they can be a combination of multiple towns/cities, outgrowths,
> municipal corporations, identifying their sociodemographic and built
> environment characteristics from other databases becomes a little
> challenging. For instance, the breakdown of all UAs with more than 1 lakh
> population is specified here:
>
> https://censusindia.gov.in/2011-prov-results/paper2/data_files/India2/Table_4_PR_UAs_1Lakh_and_Above_Appendix.pdf
>
> I can think of a few different ways in which I may be able to approximate
> some boundaries associated with these UAs:
>
>- I could use the ADM3 shapefiles of sub-districts (
>https://earthworks.stanford.edu/catalog/stanford-rj389fh4679) and
>picking locations based on visual inspections/overlap of city points. I
>understand that municipal boundaries may or may not always match the
>administrative divisions (not very clear on this)
>- I could also base it off of pincode layers (
>https://github.com/justinelliotmeyers/INDIA_PINCODES) which might be
>more fine-grained if I need additional flexibility in selecting small
>regions at the periphery.
>- I could go the population grid route and do some clustering to find
>regions that might approximate the UA populations provided in the 2011
>census for some of the bigger cities. But this would get really crude with
>no unique boundary solutions.
>
> Wondering if people have any thoughts on this.
>
> Thanks!
>
> Aditya
>
> --
> 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+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/datameet/da0f619a-9d2a-4161-b3e8-d04ecbd334fbn%40googlegroups.com
> 
> .
>

-- 
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 

Re: [datameet] Identifying urban agglomeration boundaries

[Deepak Sharda]

Hello sir great thinking.

I suggest to do it with LULC data. We have color coded raster data of area
(Red for urbanised or populated area).

we can extract polygon from LULC  data for Boundaries.

regards



On Wed, 14 Oct, 2020, 7:47 pm Aditya Medury,  wrote:

> Hi,
>
> I am trying make sense of how to translate the definitions of urban
> agglomerations/million-plus cities in the 2011 census. Census defines urban
> agglomeration as follows:
>
> *An urban agglomeration is a continuous urban spread constituting a town
> and its adjoining outgrowths (OGs), or two or more physically contiguous
> towns together with or without outgrowths of such towns. An Urban
> Agglomeration must consist of at least a statutory town and its total
> population (i.e. all the constituents put together) should not be less than
> 20,000 as per the 2001 Census. In varying local conditions, there were
> similar other combinations which have been treated as urban agglomerations
> satisfying the basic condition of contiguity. Examples: Greater Mumbai UA,
> Delhi UA, etc.*
> *Source: *
> https://censusindia.gov.in/2011-prov-results/paper2/data_files/India2/1.%20Data%20Highlight.pdf
>
> Since they can be a combination of multiple towns/cities, outgrowths,
> municipal corporations, identifying their sociodemographic and built
> environment characteristics from other databases becomes a little
> challenging. For instance, the breakdown of all UAs with more than 1 lakh
> population is specified here:
>
> https://censusindia.gov.in/2011-prov-results/paper2/data_files/India2/Table_4_PR_UAs_1Lakh_and_Above_Appendix.pdf
>
> I can think of a few different ways in which I may be able to approximate
> some boundaries associated with these UAs:
>
>- I could use the ADM3 shapefiles of sub-districts (
>https://earthworks.stanford.edu/catalog/stanford-rj389fh4679) and
>picking locations based on visual inspections/overlap of city points. I
>understand that municipal boundaries may or may not always match the
>administrative divisions (not very clear on this)
>- I could also base it off of pincode layers (
>https://github.com/justinelliotmeyers/INDIA_PINCODES) which might be
>more fine-grained if I need additional flexibility in selecting small
>regions at the periphery.
>- I could go the population grid route and do some clustering to find
>regions that might approximate the UA populations provided in the 2011
>census for some of the bigger cities. But this would get really crude with
>no unique boundary solutions.
>
> Wondering if people have any thoughts on this.
>
> Thanks!
>
> Aditya
>
> --
> 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+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/datameet/da0f619a-9d2a-4161-b3e8-d04ecbd334fbn%40googlegroups.com
> 
> .
>

-- 
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+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/datameet/CA%2B4SsQD76Yz0SdmXw5ZFVEhne-5nGZK6NnLAAzR_z12HMRUZiA%40mail.gmail.com.

[datameet] Identifying urban agglomeration boundaries

[Aditya Medury]

Hi,

I am trying make sense of how to translate the definitions of urban 
agglomerations/million-plus cities in the 2011 census. Census defines urban 
agglomeration as follows:

*An urban agglomeration is a continuous urban spread constituting a town 
and its adjoining outgrowths (OGs), or two or more physically contiguous 
towns together with or without outgrowths of such towns. An Urban 
Agglomeration must consist of at least a statutory town and its total 
population (i.e. all the constituents put together) should not be less than 
20,000 as per the 2001 Census. In varying local conditions, there were 
similar other combinations which have been treated as urban agglomerations 
satisfying the basic condition of contiguity. Examples: Greater Mumbai UA, 
Delhi UA, etc.*
*Source: *
https://censusindia.gov.in/2011-prov-results/paper2/data_files/India2/1.%20Data%20Highlight.pdf

Since they can be a combination of multiple towns/cities, outgrowths, 
municipal corporations, identifying their sociodemographic and built 
environment characteristics from other databases becomes a little 
challenging. For instance, the breakdown of all UAs with more than 1 lakh 
population is specified here:
https://censusindia.gov.in/2011-prov-results/paper2/data_files/India2/Table_4_PR_UAs_1Lakh_and_Above_Appendix.pdf

I can think of a few different ways in which I may be able to approximate 
some boundaries associated with these UAs:

   - I could use the ADM3 shapefiles of sub-districts (
   https://earthworks.stanford.edu/catalog/stanford-rj389fh4679) and 
   picking locations based on visual inspections/overlap of city points. I 
   understand that municipal boundaries may or may not always match the 
   administrative divisions (not very clear on this)
   - I could also base it off of pincode layers (
   https://github.com/justinelliotmeyers/INDIA_PINCODES) which might be 
   more fine-grained if I need additional flexibility in selecting small 
   regions at the periphery. 
   - I could go the population grid route and do some clustering to find 
   regions that might approximate the UA populations provided in the 2011 
   census for some of the bigger cities. But this would get really crude with 
   no unique boundary solutions.

Wondering if people have any thoughts on this.

Thanks!

Aditya

-- 
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+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/datameet/da0f619a-9d2a-4161-b3e8-d04ecbd334fbn%40googlegroups.com.