Village boundaries map for Odisha
1,382 views
Skip to first unread message
papes...@gmail.com
Dec 27, 2017, 5:06:52 AM12/27/17
to datameet
Hi Friends,
I need Village boundaries map and Block level map for Odisha (Shape file or geojson). I would be really grateful if some could help me out.
With regards,
Papesh
Research Fellow,
Sociology
Pondicherry University
I need Village boundaries map and Block level map for Odisha (Shape file or geojson). I would be really grateful if some could help me out.
With regards,
Papesh
Research Fellow,
Sociology
Pondicherry University
Thejesh GN
Dec 27, 2017, 5:27:22 AM12/27/17
to data...@googlegroups.com
We are working on it. It will be out sometime early next year.
--
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.
For more options, visit https://groups.google.com/d/optout.
ANANYA BHATIA
Dec 27, 2017, 5:57:47 AM12/27/17
to data...@googlegroups.com
Please check if this helps. and adding to your request here, if anyone can provide with the shapefiles of uttarakhand, latest, village and block level.
--
Ananya Bhatia
Research Analyst
Environics Trust, New Delhi
"Growth for the sake of growth is the ideology of the cancer cell"- Edward Abbey
Sharad Lele
Dec 28, 2017, 1:14:01 AM12/28/17
to datameet
Dear Ananya:
Thanks for sharing this. A quick check indicates that this shapefile is for 2001 census. The field "vid" corresponds to the Census 2001 village code (extended code including state, district, block and village code). Not sure whether the village boundaries match 2001 boundaries, but there is some distinct variation in boundaries w.r.t. 2011 boundaries as given in the District census handbooks of 2011.
Presumably the 2001 codes can be updated by using the correspondence table with 2011 that has been circulated on this group several times before. But the shape mismatch w.r.t. 2011---I don't know how pervasive it is and how it can be corrected...
this is for information of users.
Sharad
Thanks for sharing this. A quick check indicates that this shapefile is for 2001 census. The field "vid" corresponds to the Census 2001 village code (extended code including state, district, block and village code). Not sure whether the village boundaries match 2001 boundaries, but there is some distinct variation in boundaries w.r.t. 2011 boundaries as given in the District census handbooks of 2011.
Presumably the 2001 codes can be updated by using the correspondence table with 2011 that has been circulated on this group several times before. But the shape mismatch w.r.t. 2011---I don't know how pervasive it is and how it can be corrected...
this is for information of users.
Sharad
To unsubscribe from this group and stop receiving emails from it, send an email to datameet+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
--
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.
ANANYA BHATIA
Dec 28, 2017, 1:26:31 AM12/28/17
to data...@googlegroups.com
Yes Even the village codes do not match with census 2011, if you want to tak a common attribute out of the two tables and try to link them. I have been trying for this, but no success. but if you will check and put bhuvan wms layer for odisha villages, it matches. And i do not know what is the problem with the authorities for not sharing the shapefiles, i have tried this for Uttarakhand, but had to do my own "Jugaad". :)
Regards.
To unsubscribe from this group and stop receiving emails from it, send an email to datameet+unsubscribe@googlegroups.com.
"Sharad Lele [शरच्चंद्र लेले]"
Dec 28, 2017, 2:00:40 AM12/28/17
to data...@googlegroups.com
Dear Ananya:
It is possible to match the 2001 code with 2011 codes using the lookup tables that have been circulated earlier on this group.
The fact that bhuvan layer matches only indicates (to me) that Bhuvan is still using 2001 village boundaries...
Sharad
It is possible to match the 2001 code with 2011 codes using the lookup tables that have been circulated earlier on this group.
The fact that bhuvan layer matches only indicates (to me) that Bhuvan is still using 2001 village boundaries...
Sharad
You received this message because you are subscribed to a topic in the Google Groups "datameet" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/datameet/fgDQJybrSk0/unsubscribe.
To unsubscribe from this group and all its topics, send an email to datameet+u...@googlegroups.com.
Dilip Damle
Dec 30, 2017, 2:00:26 PM12/30/17
to datameet
Hello
The Village boundary puzzle will always remain a puzzle.
Recently I have worked on a Creating as good a Village Map as I could do for a Block in Uttarakhand consisting of only 125 Villages.
I had an earlier map that I had traced form the Patwari/Cesuns ,ap 20 years ago. But the areas of that map were in the range of 25% to eeven 200 % of the aera as per revenue department.
The I took a map from Justin (on our group) for that area. It had many missing villages. After several days of retracing on Google Earth I created a better map.
But it was time consuming just for 125 Villages.
There are several issues
1. No one actually has correct georeferecned accurate enough boundaries.
2. Whatever is available is just the correct (to an extent) juxtaposition of villages.
The Village boundary puzzle will always remain a puzzle.
Recently I have worked on a Creating as good a Village Map as I could do for a Block in Uttarakhand consisting of only 125 Villages.
I had an earlier map that I had traced form the Patwari/Cesuns ,ap 20 years ago. But the areas of that map were in the range of 25% to eeven 200 % of the aera as per revenue department.
The I took a map from Justin (on our group) for that area. It had many missing villages. After several days of retracing on Google Earth I created a better map.
But it was time consuming just for 125 Villages.
There are several issues
1. No one actually has correct georeferecned accurate enough boundaries.
2. Whatever is available is just the correct (to an extent) juxtaposition of villages.
3. Sometimes the Uninhabited villages are missing as in 2001 census Data.
4. It sis not necessary that the Habitation of a particular villages is within the Revenue boundary of that village.
5. Sometimes the habitations merge between two villages to form an apparently single village.
Better and more accurate village maps can only be created IMO by having at least ONE POINT inside a Khasara or a survey number 100% accurately mapped and them merging them to get a Revenue village map.
For large units of Survey numbers we will need More than one points.
I am pessimistic about we ever achieving that accuracy in India.
Because I have seen that even in a Specific Project based PAID Survey assignment for Engineering applications some surveyors FUDGE the data.
There could be Good quality work in some pockets but when you mix Good Quality and Bad quality then the end result is bad quality and not average quality.
We need to realise that.
For large units of Survey numbers we will need More than one points.
I am pessimistic about we ever achieving that accuracy in India.
Because I have seen that even in a Specific Project based PAID Survey assignment for Engineering applications some surveyors FUDGE the data.
There could be Good quality work in some pockets but when you mix Good Quality and Bad quality then the end result is bad quality and not average quality.
We need to realise that.
"Sharad Lele [शरच्चंद्र लेले]"
Jan 1, 2018, 12:54:57 AM1/1/18
to data...@googlegroups.com
Dilip:
Having banged my head against the same brick wall, I agree broadly with what you have said. I have a few additional points, though.
1. If one wants a single village or few villages to be mapped, one can try to use the village cadastral maps. It is laborious, because again the maps are not georeferenced, but feasible with some ground data. Even here, errors are significant: 100s of meters often (especially for common lands/forest lands on the edges of villages, and especially in hilly areas)
2. The only publicly available source of village boundary maps (a full sheet of polygons) is the district census handbook, which contain 1 page each showing all the villages in a taluka (or sometimes CD block). They in turn have only sourced this from the directorate of land records and survey settlement in each state. We spoke to the directorate in karnataka, and the way they have created these taluka-level maps was to lay down all the individual village cadastral maps on the floor and piece them together like a jigsaw puzzle. of course they could not do this very precisely--they were left with gaps and overlaps, and they simply 'adjusted' them as best as possible. These 'puzzles' covered the whole floor of a big hall (because a taluka may contain 100 villages) and then they 'reduced' the hall size map to a page using analog drawing techniques. So the census/DLS maps are quite crude. As Dilip says: adequate to let you know which village is where, and what is its approximate shape, but area of the polygon can differ from cadastral map areas by upto 50%. To add to the problem: census has been very inconsistent in the way they deal with uninhabited villages: sometimes showing them, sometimes not, etc.
3. I don't quite agree with Dilip about habitation: It is true that folks owning land in village A may actually live in village B. As an extreme version of this, it can happen that village A may have zero population, but non-zero cultivated land, because all the cultivators live in village B. But the 'settlement' of a village by definition has to lie within its boundary, because settlement refers by definition to the people living in the village (not all people owning land in the village).
4. So also I don't understand Dilip's point 5 about habitations merging. Sure: settlement be side by side and physically contiguous across a revenue village boundary, but that to me is not an issue: census, if properly done, will put the two sets of households in their respective villages.
5. Frankly, at this point, I would be happy even if NRSC/Bhuvan and others released a polygon layer of 2011 and 2001 village polygons AS IT IS, but with consistent and complete numbering of polygons, and georeferenced to the best of their ability (with a statement about how much error is expected).
6. HOWEVER, THERE IS MORE ACCURATE DATA, IT IS JUST NOT BEING PUT IN THE PUBLIC DOMAIN:
a) most states have land record digitisation which now includes digitization of cadastral maps. But they either don't share it at all, or charge you huge amounts for it, even though it is created with public money. of course, cadastral data are i) not georeferenced, and ii) out of date, iii) one village at a time (which when combined with the non-georeferenced nature make things difficult to put together. But again, something would be better than nothing.
b) NRSC and/or state agencies are working on projects to georeference village boundaries and individual parcels. No one knows how far they have gone in each state, how good their work is, and cerrtainly none of it is in the public domain. But it is there, and we must constantly push to get access to it as a matter of right.
c) Many individual programmes/projects such as watershed development, irrigation, agriculture department projects do their own digitisation and georeferencing. Again, they refuse to share it.
d) Census 2011 had put out a ppt claiming (for towns) they have even mapped individual streets and buildings (I had posted the link to this ppt some years ago), but when we asked them, they denied any knowledge about it!
7) About whether accurate data is every possible: the problem is not technical or even one of quality control in implementation. The problem is inherently political: folks in villages or towns have always opposed rigorous surveying because they think they stand to lose, or might get punished when their encroachments come to light, etc. DLR-Karnataka ran a pilot project which took them 6 months per village because of this kind of opposition.
Sharad
Having banged my head against the same brick wall, I agree broadly with what you have said. I have a few additional points, though.
1. If one wants a single village or few villages to be mapped, one can try to use the village cadastral maps. It is laborious, because again the maps are not georeferenced, but feasible with some ground data. Even here, errors are significant: 100s of meters often (especially for common lands/forest lands on the edges of villages, and especially in hilly areas)
2. The only publicly available source of village boundary maps (a full sheet of polygons) is the district census handbook, which contain 1 page each showing all the villages in a taluka (or sometimes CD block). They in turn have only sourced this from the directorate of land records and survey settlement in each state. We spoke to the directorate in karnataka, and the way they have created these taluka-level maps was to lay down all the individual village cadastral maps on the floor and piece them together like a jigsaw puzzle. of course they could not do this very precisely--they were left with gaps and overlaps, and they simply 'adjusted' them as best as possible. These 'puzzles' covered the whole floor of a big hall (because a taluka may contain 100 villages) and then they 'reduced' the hall size map to a page using analog drawing techniques. So the census/DLS maps are quite crude. As Dilip says: adequate to let you know which village is where, and what is its approximate shape, but area of the polygon can differ from cadastral map areas by upto 50%. To add to the problem: census has been very inconsistent in the way they deal with uninhabited villages: sometimes showing them, sometimes not, etc.
3. I don't quite agree with Dilip about habitation: It is true that folks owning land in village A may actually live in village B. As an extreme version of this, it can happen that village A may have zero population, but non-zero cultivated land, because all the cultivators live in village B. But the 'settlement' of a village by definition has to lie within its boundary, because settlement refers by definition to the people living in the village (not all people owning land in the village).
4. So also I don't understand Dilip's point 5 about habitations merging. Sure: settlement be side by side and physically contiguous across a revenue village boundary, but that to me is not an issue: census, if properly done, will put the two sets of households in their respective villages.
5. Frankly, at this point, I would be happy even if NRSC/Bhuvan and others released a polygon layer of 2011 and 2001 village polygons AS IT IS, but with consistent and complete numbering of polygons, and georeferenced to the best of their ability (with a statement about how much error is expected).
6. HOWEVER, THERE IS MORE ACCURATE DATA, IT IS JUST NOT BEING PUT IN THE PUBLIC DOMAIN:
a) most states have land record digitisation which now includes digitization of cadastral maps. But they either don't share it at all, or charge you huge amounts for it, even though it is created with public money. of course, cadastral data are i) not georeferenced, and ii) out of date, iii) one village at a time (which when combined with the non-georeferenced nature make things difficult to put together. But again, something would be better than nothing.
b) NRSC and/or state agencies are working on projects to georeference village boundaries and individual parcels. No one knows how far they have gone in each state, how good their work is, and cerrtainly none of it is in the public domain. But it is there, and we must constantly push to get access to it as a matter of right.
c) Many individual programmes/projects such as watershed development, irrigation, agriculture department projects do their own digitisation and georeferencing. Again, they refuse to share it.
d) Census 2011 had put out a ppt claiming (for towns) they have even mapped individual streets and buildings (I had posted the link to this ppt some years ago), but when we asked them, they denied any knowledge about it!
7) About whether accurate data is every possible: the problem is not technical or even one of quality control in implementation. The problem is inherently political: folks in villages or towns have always opposed rigorous surveying because they think they stand to lose, or might get punished when their encroachments come to light, etc. DLR-Karnataka ran a pilot project which took them 6 months per village because of this kind of opposition.
Sharad
Reply all
Reply to author
Forward
0 new messages