In Saudi Arabia a wadi is a mostly dry riverbed that
carries water very infrequently (maybe a couple of times year).
Would these tags be ok for a: “sandy bottomed wadi;
Greg, I think you have the tags right but I'd rather see the
waterway and highway mapped as separate ways, even if they
superimpose. While they may be in the same place, they are different
things. If the wadi is like what we see here in Oz, its probably
pretty wide and may be best mapped as an area rather than a line,
that way, the highway way would be more distinguishable. And we need
be realistic about their position and accept they move.
If there is verifiable intermittent waterflow you should tag it
waterway=stream + intermittent=yes or waterway=river + intermittent=yes,
possibly also seasonal=no if waterflow occurs sporadically and not on a
regular seasonal basis.
If you are unsure if there has been continuous waterflow in recently
times at all you should not tag it with the waterway key at all.
If there is a visible track used to drive on a regular basis this should
be mapped separately as highway=track - the track will not be identical
to the streambed center everywhere.
Note not every route that has been used by a vehicle is to be tagged
highway=track, it has to be used on a regular basis and the route needs
to be geometrically defined (i.e. cars use the same route as the ones
before). For example these are not hundreds of highway=track:
I would add that, according to the wiki (https://wiki.openstreetmap.org/wiki/Tag:waterway%3Dwadi), waterway=wadi has been deprecated and should be replaced with waterway=stream or waterway=river, anyway with intermittent=yes. Apart from that, I agree with 61sundowner: the track and the waterway should be mapped separately, each with their own path.
Be careful. Imagine the wadi or intermittent stream floods suddenly. If you know that you can think with two ways (track and stream) you can use the track...but the reality may be different and can be flooded by the "intermittent water".To avoid this I would tag in the same line ,also I would add the tag ford=yes, because instead is not a ford itselfs it has the same conditions.
> On Sep 5, 2016, at 1:48 PM, Martin Koppenhoefer <[hidden email]> wrote:
> sent from a phone
>> Il giorno 05 set 2016, alle ore 20:00, yo paseopor <[hidden email]> ha scritto:
>> also I would add the tag ford=yes, because instead is not a ford itselfs it has the same conditions.
> ford=yes is for fords, not for conditions like a ford
> you should use 2 different ways because the waterway and the highway are different objects with partly different properties.
There are places in the desert southwest of the United States where the place you drive is exactly the water course. And these can extend for miles. Saying that one feature on the ground needs to OSM objects because they have different properties is bogus: It is one object, you drive on it 99.9% of the time, and it carries water 0.1% of the time (percentages arbitrary here and will vary from one instance to the next).
“ford=yes is for fords, not conditions like a ford”? Huh? If is “like a ford” in all aspects, how is it not a ford?
FWIW, I have not noticed it in this thread but the reason brought up a while a go to deprecate “wadi” was because apparently the word “wadi” as different meanings in different areas and in some in means the entire valley not just the intermittent/ephemeral stream location.
Tod (who once lived and still visits places that have features like those being discussed here)
On 06/09/16 08:08, Tod Fitch wrote:
>>> There are places in the desert southwest of the United States where the place you drive is exactly the water course. And these can extend for miles. Saying that one feature on the ground needs to OSM objects because they have different properties is bogus: It is one object, you drive on it 99.9% of the time, and it carries water 0.1% of the time (percentages arbitrary here and will vary from one instance to the next).
"....needs to be separate OSM objects .... bogus...." ? No, sorry Tod,
I disagree. The water course is made by a rain event. The road is
something we have made, in practice just by driving along there in this
case. Both the road and the water course will change over time but
possibly independently of each other. A example close to my heart is
the road to the Finke Gorge National Park, N.T. AU. About 15 to 20Km
from memory. When I first used it, it was almost all through the dry
river bed. Over time, some parts are now along the banks and around
obstacles. The road is referred to on maps and in National Park rules
(etc), the waterway is defined by water flow, governed by the occasional
rain event. They have a different history, a different use and a
> On Sep 5, 2016, at 7:24 AM, Warin <[hidden email]> wrote:
> In an ideal world;
> the wadi would be a way along the lowest path - where water would first flow.
> the track would be a separate way
While there (of course) are some wadis that are roughly the size of a car tracks, the wadis (washes) in Southern California require a Wadi at least 2-3x car width to be able to be passable by a car - rocks, smoke trees, irregularly shaped walls, and other obstacles in a wadi would quickly render such a narrow wadi impassable by car - so a car track is made in a much wider wadi where you can go around these obstacles, including irregularities in the river bed left by a storm.
That is the important part.
Every time there is a flash flood, the sand and obstacles move; the "lowest point" in terms of a mappable way line, may move, and because a larger width wadi was needed, we are talking a structure that is tens to hundreds of meters across.
A structure that subtly changes its terrain and obstacles with each rain.
Since there are known points along the sides of wadis (markers on hills) as well as places where wadis meet, those markers and intersections become the navigation aides, and since the tracks in the riverbed were completely erased, the next person to drive through uses their instinct and desire to reach a far navigation point - reacting solely to the new placement of obstacles to literally make a "new" track. The hard sand left by dried water means you can drive a street car in the wadi at that time - I have made new tracks in a wash in a New Beetle a few times in the past. This makes mapping the exact course of the track impossible but uncecessary; you are too busy looking for rocks sticking up in your path to worry about a 10m deviation from a course. Navigation aides like rough intersections and names of other wadis, springs, caves, landmarks, camp sites, survey markers, and entrances to the wadi area are much more important to spatial navigation.
Here is the area I used to go, with a point dropped on "Hollywood& Vine" - a fake street sign put on a hill to make it easier to find the area's rare survey point. The wadis in the area are larger than most motorway systems.
So, to me, large wadis are a geographic area - similar to how we would map a riverbank. The only time there is water is when the whole thing is an angry torrent.
Tracks meander across the wadi depending on obstacles, crossing centerline only when it is narrow and forced due to a pinch point.
Wadis also spend most of their life as a dry feature - like the storm spillway of a dam - it is a flood hazard that is normally dry and used by people only when dry - it is a feature used and enjoyed because it was shaped by water, like the Grand Canyon, or the Hudson River valley, but we would not call those valleys "water features" - wadis are where we cross into water features, so perhaps people interested in mapping them and dealing with their tracks think of them mainly as a canyon or valley that floods once or twice a year, rather than a river that is dry 99% of the time. Mapping a 200m wide "river" in the desert with some blue lines is disingenuous and dangerous - there sure as hell no water to drink unless it is there to kill you.
Narrow wadis will have a track right on top of the "low point", and that reflects reality - the track cannot be anywhere but the wadi because of terrain.