Proposal talk:Natural=wadi

What this proposal mostly (if i ignore the usual weasel words - like: typically, many, may be) describes is the geomorphological setting of an alluvial plain. This would, in principle, be a viable thing to map and have a distinct tag for. The proposal so far, however, is not providing much of a guidance to mappers how to identify and delineate alluvial plains. Generally speaking OSM has, so far, very few geomorphologically defined tags in wider use, which is partly because they are not as straight away to identify on the ground as tags characterizing surface material - like natural=sand, especially in settings with vegetation, human influence or later non-fluvial modifications of the surface (like through aeolian processes).

It might also be a good idea to think about filling the gap in bare ground surface mapping that currently exists for unsorted, unconsolidated material (consisting mostly of alluvial deposits and till) independent of the geomorphological structure this is part of. Locally mapping the bare ground material is much easier (and practically also much more popular among mappers) than mapping the geomorphological setting. Unfortulately so far mappers often engross unvegetated unsorted alluvial and glacial deposits under natural=sand/natural=scree/natural=shingle because of the lack of more fitting options in tagging. --Imagico (talk) 09:22, 23 April 2024 (UTC)

I completely agree that it would be good to fill in the gap in tagging for unconsolidated material. We have natural=sand which is widely used and misused. But there are few instances of natural=gravel, no instances of natural=cobble, and scant use of natural=boulder_field. That's something I think about frequently since many of my local trails are in unconsolidated marine terraces which would best be described as cobble. It might be nice to have tags describing unconsolidated surfaces for the entire range of ISO classification. --B1tw153 (talk) 14:42, 23 April 2024 (UTC)

My comment was not about missing grain size ranges, it was about missing tagging for unsorted material compositions in bare ground surfaces with components ranging from commonly finer than sand (silt, clay) to coarser than sand. These are a common occurrence in alluvial and glacial deposits including a large percentage of alluvial plains in arid/semi-arid regions that this proposal seems to be concerned with. All the photos shown in the proposal feature this kind of material. --Imagico (talk) 19:59, 23 April 2024 (UTC)

That's an interesting distinction. I'm definitely in favor of having richer tags to be able to describe and map natural features. How would you consider the case where the material in the wash is essentially the same composition as the banks of the wash, just more recently disturbed and not as settled? Here's an example. It might not be completely apparent in the picture, but if you put a shovel into one of the banks, what you'd get would be the same unsorted, unconsolidated material as in the wash.

--B1tw153 (talk) 20:33, 23 April 2024 (UTC)

Without knowing the specific setting it is hard to tell for sure what the local situation is. The foreground in the valley this is quite clearly the kind of unsorted alluvial deposit i was talking about. The hill sides, however are not the same - you can clearly see substantial exposed bedrock here but you can also see substantial vegetation and therefore likely significant (though thin and patchy) amounts of soil with considerable organic components. It might even be substantially greener during wet season. Overall, this would IMO qualify as a sparse scrubland, in micro-mapping mixed with some bare_rock and scree, but not bare ground overall. All of this refers to mapping local surface characteristics, not the geomorphology or landforms. --Imagico (talk) 21:17, 23 April 2024 (UTC)

Are there issues here that we still need to resolve before the proposal goes to a vote? --B1tw153 (talk) 13:34, 11 June 2024 (UTC)

The proposal does not appear to have changed much since i made the comment so it still applies in my eyes. What you make of this is your decision. As is, with the widespread use of weasel words and the use of undefined concepts like the 'banks' and the 'bed' of a wadi the practical use of the tag will likely be mostly based on the subjective idea of individual mappers on what a wadi/wash is and how it is delineated and less on some overall geomorphological definition.

I don't think it would be a huge amount of work to develop some guidance for mappers (with some pictures and explanations of the geomorphological context to allow them to accurately identify and delineate an alluvial plain) and make this a well defined geomorphological tag. But as said - geomorphologically defined tags are rare in OSM so there is no guarantee that this would practically work. --Imagico (talk) 14:54, 11 June 2024 (UTC)

I can't say that I understand your comments because they seem to be tangential to the proposal. The proposal does not relate to alluvial plains which are a different type of feature. You may contrast this with the Wikipedia articles on wadis, arroyos, or dry washes which relate to this proposal. If you have comments that are specific to the subject of the proposal, I would be happy to address them.

I might have misunderstood the aim of this proposal. If this proposal is not meant to introduce tagging for a geomorphologically defined feature of physical geography but rather for tagging the culture specific classification of toponyms under these three terms (Wadi, Arroyo, Wash) then my comment is indeed off-topic. However, if that is the case the proposal is very sparse on discussing the meanings of these three terms in their respective cultural contexts (which differ strongly and - at least for Wadi - is much broader than the proposal indicates). --Imagico (talk) 16:09, 11 June 2024 (UTC)

We did indeed have a detailed discussion about the terms "wadi," "arroyo," and "wash" in their cultural and regional contexts in the Community Forum thread that led to the proposal. I didn't want to repeat that in the proposal, nor did I think a summary would do justice to other people's views. So there is a link to the thread in the proposal. --B1tw153 (talk) 01:18, 12 June 2024 (UTC)

Boundaries of a wadi

What is on my mind is that many of the wadis in the Sinai and Dead Sea area do not actually have distinct banks other than the slopes of their valley on much of their length. Also, way beyond what is proposed, the term "wadi" there does not just denote the area between the banks but includes the banks of the intermittent flow (or the flanks of the v-shaped valley as high as the stream may come to reach). What's more, often "wadi" is used to denote the complete valley, effectively meaning up to the watershed to the adjacent valley. Correspondingly, in these wadis, a large majority of the surface of what is termed “wadi” is not of loose sediment as proposed but solid rock.

The proposal confines the meaning of "wadi" in OSM-speak to only "the area between the banks of a wadi", which is much narrower than common use of the term at least in the Middle East. This does not seem ideal.

Also, the proposed definition is circular and self-referencing, defining a wadi as part of a wadi, which is better avoided. --ChillyDL (talk) 20:16, 11 June 2024 (UTC)

It is relatively common for tags used in OSM to have slightly different meanings than the colloquial terms that they're derived from. For example, the OSM definitions of the tags natural=reef and natural=shoal differ from the colloquial use of these terms in order to make distinctions that the common terms do not make. You could say the same thing about the proposed natural=wadi tag. It does not necessarily correspond to the colloquial use of the term, but it does provide a tag that can be used to map the broad, flat area of a normally dry watercourse -- something that many mappers have attempted to do over the years which has resulted in the misuse of existing tags. The proposed natural=wadi tag makes it possible to map these areas consistently, with a tag that is specific to the feature. As for the other aspects of colloquial "wadis," OSM already has well defined tags that can be applied. For example, it would be perfectly fine to map a v-shaped valley as natural=valley + name=Grand Wadi, and even to map areas within the valley with natural=bare_rock, while mapping the area of the dry watercourse within the valley as natural=wadi. --B1tw153 (talk) 01:18, 12 June 2024 (UTC)

Wash or Wadi?

I think my apprehension is with the terminology: “Wadi” does not normally denote what it is supposed to in this proposal, i.e. just the area between the banks.

What about natural=wash? Collins’ dictionary gives this definition of a wash: “land that is habitually washed by tidal or river waters”. More specifically, e.g. the glossary of mojavedesert.net: “WASH - A narrow, constricting dry bed of an intermittent stream, as at the bottom of a canyon, typically dry but subject to rapid flow during flash flooding.”[1]

To me, natural=wash thus would seem better fitting to tag the area between the banks as proposed. --ChillyDL (talk) 12:03, 12 June 2024 (UTC)

Well, natural=wash would have been my first pick as well. However, in the discussion in the OSM Community Forum, it was pointed out that this term is only used locally within the American Southwest and that outside that area the term "wash" would not be associated with this type of landform and could be confusing. Given the preference for UK English usage in OSM tags, the discussion settled on "wadi" as the more likely term for this type of landform. So, the proposal is for natural=wadi as this was the consensus from the forum discussion. --B1tw153 (talk) 23:36, 17 June 2024 (UTC)