Why not ask your question on the new OpenStreetMap Community Forum?

# How to map braided rivers?

 6 From Wikipedia: A braided river consists of a network of small channels separated by small and often temporary islands called braid bars. I wish to map narrow (tens of metres) braided rivers. 1) Firstly, we need a way tagged waterway=river. But there are two small problems: a) the actual way shifts over time and especially after flooding events, b) there can be two or more channels. The imagery will be out of date. 2) My major question is can we map the width of a braided river? It would be incorrect to use tags natural=water, and water=river or waterway=riverbank because most of the width is sediment. Only during floods is it full of water. asked 10 Sep '14, 05:37 stweb 124●4●5●8 accept rate: 0% Thanks, my rivers are relatively small with narrow channels. For now: 1) I will trace the main channels and tag them with waterway=river. 2) For the deeper section near the river mouth, I will additionally trace this area and tag it with natural=water and water=river. I like the Piave example using natural=riverbed, surface=gravel. Maybe later. Excellent support! (11 Sep '14, 07:44) stweb I've added a bit on the wiki for natural=shingle another alternative for mapping parts of riverbeds. (11 Sep '14, 13:25) SK53 ♦

 2 There seem to be at least three approaches: You could just map river channels using waterway=river and not worry about the riverbank (see the Waimakariri River) You could map waterway=riverbank by the "usual flow", and trace major islands (multipolygon inner role) as seems appropriate. (maybe the Yukon River is an example?) 2a. The Piave does this, and adds natural=riverbed, surface=gravel for the surrounding area of irregular flow You could map waterway=riverbank as the maximum flow area, and trace islands, and use water=river for the main channel (see the Tagliamento) I would look at other rivers in the region, and consider how often a river gets that higher flow level. Fundamentally, OSM doesn't handle changes in water level well (we don't map flood plains), so I wouldn't stress about what is "approved" or not. After all, the proposal to get rid of waterway=riverbank was approved three years ago! :) answered 10 Sep '14, 19:29 neuhausr 7.5k●8●70●121 accept rate: 21% then you have several paralel waterways with the same name or only one with a name what is the "usual flow" when you have multiple channels ? 2a. something possible but needs discussions (how to tag the channels, etc) "riverbank as the maximum flow area" is not an option when this "maximum" is only reached once per year (or even less). (11 Sep '14, 12:25) Pieren @Pieren, all good points/questions. As you say, it'd be good topic for someone to bring up on the tagging list--after a week or two, there might be some consensus! :) For now, though, I just wanted to help answer the question, so I described in untechnical terms what I saw when I looked at some of the different braided river examples mentioned in wikipedia. (11 Sep '14, 16:56) neuhausr
 1 This is a common issue with many glacier fed Alaskan rivers. Spring runoff is many times the normal river flow and the result is immense areas that are clean of vegetation. There were imports done long ago that use method 2 above, "map waterway=riverbank by the "usual flow", and trace major islands (multipolygon inner role) as seems appropriate". This is what I've done in the areas I've mapped. It is a ton of work but less than what would be required were you to add additional tags defining the riverbed as gravel, mud, etc.The original imports mapped sandbars as well as forested islands but these change so much that I ignore them. If an island has significant forest or even dwellings and roads, I trace it and assign an "inner" role. When the original import indicates an inner polygon that's no longer visible on Bing, I delete it. If you don't want to spend the considerable amount of time required by the multipolygon approach, just tracing in the main channels is a help. But I can tell you from personal experience that driving along a major Alaskan river that has been rendered with only a thin blue line is underwhelming to say the least. The Susitna, Copper, Matanuska, Tanana, and the mighty Yukon itself would look puny if not for their included riverbanks. answered 11 Sep '14, 02:05 AlaskaDave 5.3k●75●107●164 accept rate: 16%
 0 Map like neuhausr stated, add intermittent=yes, if there is every now and then a river or stream, http://wiki.openstreetmap.org/wiki/Intermittent And let mappers include yourself change the streams after flooding. OSM is a actual map, it changes everyday, like a stream. answered 10 Sep '14, 22:08 Hendrikklaas 9.3k●204●238●386 accept rate: 5%
 2 The only mention about "braid rivers" I found in the wiki are here about an import and mainly there about riverbanks. I don't think we have currently a suitable and widely approved solution for your problem. You say the full width is sediments and is covered only during floods but I think it's not always the case. As wikipedia says, it's also common in mountains e.g. exposed to heavy rains (this is the only real cases I met in my mapping experience about this). I wouldn't discuss here for a solution since we don't have a black or white answer. I would suggest to post your question again to a more appropriate environment for an open discussion like on the forum or even better the tagging mailing list. answered 10 Sep '14, 12:49 Pieren 9.8k●20●83●157 accept rate: 15%
Your answer
 toggle preview community wiki:

### Follow this question

By Email:

Once you sign in you will be able to subscribe for any updates here

By RSS:

Answers

Answers and Comments

Markdown Basics

• *italic* or _italic_
• **bold** or __bold__
• link:[text](http://url.com/ "title")
• image?![alt text](/path/img.jpg "title")
• numbered list: 1. Foo 2. Bar
• to add a line break simply add two spaces to where you would like the new line to be.
• basic HTML tags are also supported

Question tags:

×73
×23
×1

question asked: 10 Sep '14, 05:37

question was seen: 5,933 times

last updated: 11 Sep '14, 16:59

powered by OSQA