Talk:Tag:waterway=brook

From OpenStreetMap Wiki
Jump to: navigation, search

Problems of the waterway=brook concept

This tag has a number of serious issues with its definition and its position within the framework of waterway mapping in OSM. These make use of this tag rather counterproductive for the usability of waterway data in OSM. Specifically:

  • This tag breaks with the basic convention of waterway tagging that distinction between natural and artificial waterways is a primary criterion. This distinction is of fundamental importance for any more sophisticated analysis and use of the waterway network since artificial waterways do not generally follow the basic rules of natural waterways (like water always taking the steepest path downhill).
  • The tags definition is centered on topographical aspects (depression in the ground) and not on water flow as the waterway=* key implies. See also natural=gully and barrier=ditch. Water is only part of the definition insofar that water has to be present at least some part of the year. Water flow OTOH is completely optional.
  • The size criteria are vague, especially since it is unclear how depth and width are measured if water is normally absent.

Actual use of the tag reflects these issues, it is used for a wide range of features, essentially a catchall tag for waterway=stream, waterway=drain, waterway=ditch, natural=gully and barrier=ditch.

--Imagico (talk) 09:45, 22 April 2015 (UTC)

I agree on your points. This tag should make clear where it fits compared to other waterways. --Jojo4u (talk) 17:32, 4 August 2015 (UTC)
I also agree, and find it questionable whether there is a need for this tag in the waterway family at all. --Tordanik 19:50, 4 August 2015 (UTC)

Since the tag continues to be used quite inconsistently and no arguments have been brought forward for it or improvements made to the tag definition i would like to change the wiki page in a similar way as it has been done for waterway=wadi to discourage further use. If anyone has objections please say so (but also address the issues mentioned then please). --Imagico (talk) 16:20, 27 October 2016 (UTC)

Basically, this "waterway=brook" tag should be replaced by "waterway=stream"+"intermittent=yes".
Note that direction of water flows may be questionable, difficult to determine unless there's a natural incline, just like with "waterway=ditch"
However a ditch is by nature artificial, while a brook (a very small stream which may easily become invisible or quite difficult to distinguish in wet lands) is by nature natural (or should be).
Most of the time the brook will look like a series of small wet land, and vegetation is essentially the same os on surrounding lands. While a stream has a much clearer path and is cleaner with water-specific species.
It may also be difficult to distinguish from unpaved tracks in forests, or could be very old unmaintained tracks, left by the passage of animals, which may be occasionnally flooded just like lands aound.
A brook specifically has no clear waterbed, unlike natural "streams", or artificial "ditches" (created by humans but still made with earth), or drains (the waterbed is artificial, generally made of concrete or earth has been removed to exhibit the rock, and drains may be underground through "tunnel=culvert"+"layer=-1").
So we eliminate drains as replacements for brooks, but the only remaining distinction is with intermittent streams: the nature of the soil and vegetation compared to surrounding lands makes clearly the difference.
So I'm not conveinced at all by the arguments exposed above: a distinction is still clear and there's no contradiction. Brooks and wadis could be essentially the same as intermittent streams (when they are dried) but they behave very differently when they are flowings: wadis will have evident rapid flows and will erode the soils and clear it with just sands and rocks, and a clear direction (that's why wadis are clearly intermittent streams, but brooks are not: water does not significantly modify their bed and its direction is not clear, a brook is just a particular type of wetland more frequently flooded than surrounding lands, and brooks still capture waters from these lands which will dry first; also a brook will not necessarily join a continuous stream, it may go nowhere, with water reinfiltrating soils around and pumped by the vegetation)
As well a brook is clearly not a gully (that has a clearly eroded bed). In fact brooks are not even a barrier for walkers or vehicles: you can step into them without any problem and most often don't need to jump over it, unlike ditches, drains, gully and streams that are clearly difficult to cross or exit if you enter their bed. You'll only need to jump over it or could have a vehicle locked inside when it's rarely flooded, but if this is the case, it will be also difficult to walk or drive on surrounding fields.
So the arguments based on size or depth for qualifying brooks are not the correct ones (they are not really measurable), the good arguments is their intermittence, and the nature of soils and the ease to cross through them without additional equipement (of course depth should remain under the length of a human leg, and their width will not be more than a couple of human steps, so you don't need any jump unless it is currently flooded, but jumping over them will still be relatively easy). — Verdy_p (talk) 12:28, 11 March 2017 (UTC)
Note: a French translation for "(a) brook" would clearly be "(un) ru" (meaning a smaller type of "(un) ruisseau" in French, which translates "(a) stream" in English). — Verdy_p (talk) 12:38, 11 March 2017 (UTC)