OpenStreetMap

What's up with the Rann of Kutch?

Posted by Alan on 3 June 2016 in English (English).

I started looking at the Rann of Kutch in India, and it doesn’t look very well mapped. I also can’t find any mentions of it on the wiki, or Googling for “Rann of Kutch” in conjunction with OpenStreetMap.

Christoph Hormann’s post is the only thing I can find: http://blog.imagico.de/new-images-for-mapping-in-osm/

Does anyone know if there’s been any previous discussion (perhaps on the talk-in list?) about how to tag it? Or does the local community prefer it this way?

Here’s what it looks like now: screen shot 2016-06-03 at 3 jun 1 20 37

Which looks very different from the image in Christoph’s blog: rann

He notes in another post that it’s the largest area of incorrect coastline in the world, according to his metrics: http://blog.imagico.de/osm-coastline-and-glacier-data-quality-reports/ His map shows that most of the coastline in that area hasn’t been touched in OSM since 2007: cl_date_512

According to Wikipedia, this area is only submerged during the monsoon season, and is dry the rest of the year. So it seems like the correct tagging would be to move the natural=coastline further out, and most of the currently wet areas in OSM should be retagged as natural=wetland and wetland=saltmarsh (see wetland=saltmarsh) on the wiki. But I don’t have any local knowledge, so I’d want to be sure about that.

Here’s an overpass turbo query for natural=wetland in the area:

screen shot 2016-06-03 at 3 jun 12 55 17

There are four large polygons (one is a multipolygon relation, actually), all four of which have natural=wetland but no additional tags… no wetland=saltmarsh that I was hoping for. And mostly these features cover the current OSM land, not the incorrect water areas anyway.

I also did an overpass query for wetland=* just in case there were any polygons that had wetland=saltmarsh (or similar) and which might have accidentally left off the natural=wetland tag. But no luck, there’s basically nothing using wetland=* that’s useful:

screen shot 2016-06-03 at 3 jun 1 01 35

As a last resort, to see if there are any OSM features we can work with (perhaps someone is using non-standard tags), I zoomed into a bit of the India/Pakistan border near the coastline. There is a line of barrier islands that probably mark the true edge of the Great Rann of Kutch, which is where we should see any hidden features, if there are any. Here’s the data overlay layer: http://www.openstreetmap.org/#map=12/23.7720/68.2019&layers=D

screen shot 2016-06-03 at 3 jun 1 10 51

Pretty much all of these islands (like this one) are from the PGS import back around 2007 or 2008. screen shot 2016-06-03 at 3 jun 1 11 43

Finally, if I zoom in enough to edit, we see that the satellite imagery is good enough that someone could trace the correct edges of the salt marsh… but it would take them a bit of time to do it well.

screen shot 2016-06-03 at 3 jun 1 16 22

Does anyone have any thoughts or experience with this area?

Location: India

Comment from BushmanK on 3 June 2016 at 22:29

According to different sources and tidal models, northern part of Indian ocean has some significant tidal range (difference between high and low tide). General convention for shoreline mapping is to use margin between land and water at mean high tide level. Keeping in mind that area you are talking about is basically a tidal flat, it is close to impossible to trace land accurately in accordance with that convention, since satellite image does not represent “medium high tide” level of water. Another thing is that it’s not just some tidal flat, but mud flat with non-uniform currents, so these “islands” are nothing more than temporary higher spots of constantly moving mud, which makes them “temporary” and “non-verifiable” features in terms of OSM core principles. It is easy to see, how those features are changing, using Google Earth historical imagery mode (we can’t trace it, but I can use it to proof my statement). So, as a bottom line, I wouldn’t map any minor mud islands in this area.

Comment from imagico on 3 June 2016 at 22:49

Much of this is not really a wetland in the sense that it is waterlogged soil for most of the year but an area of seasonal water cover - water during and after monsoon, dry during the dry season and wetland only during a short time in between. The false color infrared satellite images i put up are meant to approximately catch the average water cover - the area that could probably be tagged natural=water + seasonal=yes. It seems except for the tidal zones these are largely not or only very sparsely vegetated so i am not sure if wetland=saltmarsh really applies.

The last larger coastline edits there were made by Ben Discoe.

What you show later is the eastern end of the Indus River delta with its mangrove forests (which are less dense than typical tropical mangroves in this climate and in serious decline in the last decades - see wikipedia) Significant parts of this are above the high water line.

The whole area is pretty well covered with high resolution images but muddy water and bare ground - both wet and dry - are often quite difficult to distinguish in these.

Comment from bdiscoe on 4 June 2016 at 00:44

Alan, I’m the main guy who probably did the most work on the Rann of Kutch, including carefully putting the wetland tag over all the land area which, judging from Bing imagery, is sometimes underwater (the very definition of wetland). It’s a hard problem but I’m proud of what I managed to get done.

As for “Here’s what it looks like now”, the pictures you show are of mapnik’s zoom level 9, and that rendering doesn’t even show wetlands at all until zoom level 10. You don’t have to do a overpass turbo query to see the wetlands, just zoom in one more step.

As for “doesn’t look very well mapped”, I think your issue is with the default mapnik rendering. It is pretty well mapped.

As for wetland=saltmarsh, I’m not sure it adds much information over just natural=wetland, but that sounds fine by me as an additional tag, feel free to add it.

Comment from Alan on 4 June 2016 at 04:09

Hi Ben! I didn’t mean to disparage your work! The parts you’ve added look really good!

Here’s all the extra detail at zoom 10, as you mention: screen shot 2016-06-03 at 3 jun 9 01 13

But all the water areas to the northwest and northeast of your edits really feel like they shouldn’t be water. They’re almost certainly above the high tide mark. So maybe it’s not that it’s poorly mapped, it’s just there’s a lot more left to do!

Thanks everyone for the comments about appropriate tags. Yes, my zoomed in images were more from the Indus delta area, so they probably require different tags than the salty desert further inland. Maybe wetland=saltmarsh isn’t quite the right tag for either of these zones.

Comment from imagico on 4 June 2016 at 12:37

I added a small demostration of how the eastern Indus river delta area can be mapped in http://www.openstreetmap.org/changeset/39796363. Requires a coastline update to be properly rendered of course.

For reliable mapping of the extent of the tidal area you need imagery taken at low and high tidal levels - for example

for low tide:

http://earthexplorer.usgs.gov/metadata/4923/LC81510442016033LGN00/

for high tide:

http://earthexplorer.usgs.gov/metadata/4923/LC81510442016113LGN00/

In addition you have in this area extensive seasonal flooding during summer storms that goes much further:

http://earthexplorer.usgs.gov/metadata/4923/LC81510442013232LGN00/

The effects of this on the landscape can be easily misinterpreted as tidal effects leading to a wrong estimation of the coastline (which is what happened in the PGS data).

Comment from PlaneMad on 4 June 2016 at 13:11

Having set foot in the Rann, its essentially a very large mudflat crusted with an inch of salt, atleast when its not submerged. There is a very clear boundary where land ends and the saltmarsh begins, but its unclear where the marsh ends and the coastline begins.

The Rann during monsoon and when its dry. NASA

Have tried cleaning up this area before, but it just seemed unclear how to proceed. Nobody has brought it up for discussion in the Indian community either.

Maps in India usually depict it like this:

Comment from Alan on 9 June 2016 at 17:48

I see this related (and closed) issue on the openstreetmap-carto style (#1995) which requests rendering of salt flats, but it seems like the tagging is not well established. Also, that request refers to this feature, which seems to be a year-round salt desert, not a seasonally flooded one like the Rann.

Imagico, you argue that it shouldn’t be a saltmarsh or a wetland because it doesn’t have typical marsh-like vegetation? I note that Wikipedia calls it a salt marsh, whether that helps or not.

It seems to me we have two issues here:

  1. How do we tag the Rann? Should it be wetland=yes, seasonal=yes, (possibly with wetland=saltmarsh. Or just natural=water, seasonal=yes? It seems like in either case, seasonal=yes would be a useful addition.

  2. Is the coastline in the right place? Since almost all of the Rann is above the high tide line, it seems that the coastline should be moved. Where we move it will be tricky to determine, but anything is better than where it is now, because the current situation is obviously wrong, based on the high tide rule.

Comment from imagico on 10 June 2016 at 10:57

  1. I would tag as natural=water + seasonal=yes/intermittent=yes areas that are water covered on a regular basis for a significant part of the year. Areas with flooding only for a short time (a few weeks after heavy monsoon rain or during storms near the coast) should probably not be tagged this way. I would tag as wetland other areas that are neither flooded (standing water) nor dry for a significant time of the year. If they are also dry for parts of the year seasonal=yes seems appropriate (that is probably always the case in this area except for the tidal wetlands). I would tag as wetland=saltmarsh those of these wetland areas which have at least some vegetation cover (i.e. not completely bare mud flats) and which are a salty environment during the time they are wet. Lacking local knowledge in the area i cannot really say to what extent this applies but it seems to me there is significant local differentiation here. In lack of further information i would probably tag what appears in bright cyan in my false color images natural=water + seasonal=yes. I am unsure if it makes sense to have combined mapping (natural=water + seasonal=yes overlapping natural=wetland + seasonal=yes) but in principle that would be a valid statement.

  2. Coastline in the north as Ben mapped it is quite OK i think. In the south the coast should probably be closed a few km above the NH27 bridge around here

I am unsure how far the seawater reaches into the Rann during monsoon storms - however for coastline mapping this is not important since this only considers regular tidal variations and not seasonal storms.

Comment from Alan on 10 June 2016 at 18:11

I made some small test edits here out in the river delta, just to get a sense of how much effort it would take to do the whole thing: http://www.openstreetmap.org/changeset/39915515

Comment from Alan on 10 June 2016 at 19:43

I also added some wetland=saltmarsh tags after confirming that the areas I mapped are above high tide. http://www.openstreetmap.org/changeset/39939704

Also thanks to imagico for a much more detailed test area here: https://www.openstreetmap.org/changeset/39796363


Login to leave a comment