[Tagging] Coastline for rivers, estuaries and mangroves?

David Groom reviews at pacific-rim.net
Sun Sep 9 21:35:24 UTC 2018

------ Original Message ------
From: "Joseph Eisenberg" <joseph.eisenberg at gmail.com>
To: tagging at openstreetmap.org
Sent: 07/09/2018 04:02:26
Subject: Re: [Tagging] Coastline for rivers, estuaries and mangroves?

>I've now edited the coastline in the area mentioned. I have now added 
>natural=coastline along all the ways forming the edge of the mangroves 
>and open water. 

I have to say that to me this seems wrong. Coastline tags are now on 
ways forming channels 40m wide and 30km from open ocean.  I just don't 
see that these are "coastlines" .


>Further west, I moved the administrative boundary off of the coastline 
>of internal waterways, positioning it near the low water line / 
>baseline, because I believe this is closer to the official Indonesian 
>definition for Kabupaten (admin level 6) boundaries, and it no longer 
>creates separate polygons around each patch of mangroves. 
>This brought up another issue. I did not want to delete the 
>natural=water areas, so I changed them to multipolygons (since I had to 
>break the closed ways to make a proper coastline) and marked salt=yes, 
>removing natural=river from the areas that might better be described as 
>tidal channels. I considered using water=tidal or water=salt, but both 
>of these tags seem to have limited use and an unclear definition; JOSM 
>suggested salt=yes.
>But I am uncertain what to do with the waterway=river in the case of 
>tidal channels and the complex connections between rivers in these 
>mangrove areas. A search of taginfo did not find an alternative tag, 
>although river=tidal is in use. I think there should still be a 
>waterway midline for the large tidal channels in the mangroves which 
>can be used by boats or even ships, to help navigation software. (Many 
>of these channels were actually created by flowing river water; the 
>rivers in this area meander strongly and often change the location of 
>the mouth, as can be seen by comparing the current situation to 100 
>year old Dutch maps)
>Perhaps waterway=river with tidal=yes or river=tidal is the best option 
>to prevent tag fragmentation? Or is river=tidal_channel preferable? The 
>problem is determining the direction of water flow when two channels 
>connect. Besides tidal, is there a better tag to imply two-way water 
>flow depending on the tidal cycle?
>>Message: 3
>>Date: Wed, 5 Sep 2018 18:03:36 +0200
>>From: Christoph Hormann <osm at imagico.de>
>>To: "Tag discussion, strategy and related tools"
>>         <tagging at openstreetmap.org>
>>Subject: Re: [Tagging] Coastline for rivers, estuaries and mangroves?
>>Message-ID: <201809051803.36467.osm at imagico.de>
>>Content-Type: text/plain;  charset="utf-8"
>>On Wednesday 05 September 2018, Joseph Eisenberg wrote:
>> > Specific examples:
>> >
>> > 1) This changeset on the River Dart in southwest England was the
>> > source of the Help site question:
>> > https://www.openstreetmap.org/changeset/61959067
>>The coastline closure there:
>>is both below the lower limit of the proposal and below the the range 
>>can imagine a meaningful coastline closure rule to allow.
>>I would however be interested in hearing any universal rule that would
>>allow this kind of placement based on physically observable criteria
>>and that would maintain the coastline as a meaningful geometry on its
>> > It looks like quite a large estuary, much wider than the non-tidal
>> > part of the river upstream.
>>That is largely not really an estuary but more of a ria.  I have no 
>>for this at hand but you can likely see an abrupt change in the
>>elevation profile near Totnes where the submerged section of the 
>>river valley starts.  So in this case it would make a lot of sense to
>>place the coastline closure near the upper end of the tidal section
>>because this is much better defined in terms of physical geography.
>> > 2) The estuaries and mangrove tidal channels in this area:
>> > https://www.openstreetmap.org/#map=11/-4.8806/136.9339
>>Here i likewise see no meaningful motivation for the current coastline
>>placement - like here:
>>Poor image quality in the available sources makes identifying the 
>>of the mangroves difficult, you really need to make use of available
>>lower resolution open data images in the area for proper maiing here.
>>But you can conclude a few thing from the structure of the network of
>>channels.  For example
>>is quite clearly not a river but a tidal channel (there is no river
>>feeding it, it is just draining seawater that has entered during
>>raising tide).
>> > I previously changed the coastline to be closer to the river mouths
>> > in another section of coast to the southeast, but perhaps I should
>> > change it back? The whole idea of coastline around mangrove swamps 
>> > most confusing. I don't think the mangroves should be outside of the
>> > coastline, but where then should it be?
>>Common practice is to place the coastline at the outer end of the
>>mangrove forest.  This is a pragmatic solution because placing it
>>inside the mangrove would be non-verifiable.  Of course mapping the
>>mangrove is important for the data to be meaningful in this case.
>>Christoph Hormann
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/tagging/attachments/20180909/e7c2e38f/attachment.html>

More information about the Tagging mailing list