Night Lights Data for Delhi

65 views
Skip to first unread message

Anisha Garg

unread,
May 20, 2023, 1:47:05 PM5/20/23
to datameet
Hello, 

I am a PhD student in Economics at the University of Warwick. For my research, I need the night lights data for Delhi at the ward level, as specified in the census. I know that the SHRUG data set is available, but it is at the Delhi level, and not segregated by wards. It would be really helpful if someone could point me to a datasource, or guide me on how I ca leverage the SHRUG data set.

Thanks a lot! :)

VAIBHAV PURI

unread,
May 20, 2023, 2:32:01 PM5/20/23
to data...@googlegroups.com
Hi Anisha


But I doubt it would be of much use given the spatial resolution of 463.83 meters.


Regards,
Vaibhav Puri
Assistant Professor
Department of Economics
Sri Guru Gobind Singh College of Commerce
University of Delhi

--
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+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/datameet/be650d09-a102-485c-b7fc-0a75b2770241n%40googlegroups.com.

Shiv Hastawala

unread,
May 20, 2023, 2:32:26 PM5/20/23
to data...@googlegroups.com
Hi Anisha

I don;t think the SHRUG data set can be broken down any further in terms of geographic boundaries. What you are looking for is not readily available but you can create that data for yourself using vector maps of two things: granular night light data and ward boundaries for Delhi. But you would need to be careful about what year each of the maps correspond to because, for instance, if you want night light data from 2018 then you don't want to merge it with ward boundaries data from, say, 2004. It is possible that ward boundaries keep changing (I'm not aware of it since I haven't worked with ward boundaries in particular, but I know that district and subdistrict and panchayat boundaries keep changing so I'm suspecting that the same thing would happen to ward boundaries too), so you need to make sure that the match between the two types of data makes sense.

 With that caveat, you can do this in free software like QGIS or R, or if you use Stata then you can also do it using that (since we are both econ PhD students, I'm assuming you aren't into Python, although I thin it should be possible with that too).

You can find the requisite data on the links below.

Delhi Ward Boundaries Maps:

Granular Night Time Maps:


Hope this helps.

Thanks and regards.


Yours sincerely

Shiv Hastawala

(He/His/Him)
Doctoral Student
Department of Economics
Binghamton University (State University of New York)

Zoom ID: 201 717 2613



--

chris mathen

unread,
May 20, 2023, 3:00:51 PM5/20/23
to data...@googlegroups.com
It is also possible with python. I don't think the ward boundaries are time invariant. If vector maps are available, then extracting nightlight data is possible. 

Regards, 
Christopher Kuruvilla Mathen
Research Scholar 
IIT Kharagpur

To view this discussion on the web visit https://groups.google.com/d/msgid/datameet/CAGiG-Ay066G8qfVZvyEYj9YVku%2BXygbUMRVFojHfr4u2aKDBHA%40mail.gmail.com.

Anisha Garg

unread,
May 31, 2023, 1:18:27 PM5/31/23
to datameet
Thanks a lot for your replies! :)

This is very helpful. Basically, I have divided Delhi based on my custom boundaries for the year of 2016. What I need now are the metrics like population, income, etc. So, I am using the ward boundaries to super-impose them on my custom boundaries and get the required metrics. 

From https://sec.delhi.gov.in/sec/delimitation-wards-municipal-corporation-delhi, I understand that delimitation took place before 2008, in 2017 and now in 2022. Is this correct or am I missing something here?

Once again, thanks a lot, everyone! :)

Reply all
Reply to author
Forward
0 new messages