sand vs shoal in tidal range

[Mahabarata]

Hi,

in the tidal range (outside the coastline), we may use natural=sand or natural=shoal (and some people use natural=beach) : the first case should be used for sand in contact with land area (for example a beach at low tide), the second case should be used for unconsolidated material not in contact with lands.

A lot of people use the value beach in the tidal range but, according to this wiki's page, beach should be used only for the land area, and sand for the tidal range. But it is not the problem here.

Because it is the same material in the same area (water at sea), natural=sand (with or without surface=sand), natural=shoal+surface=sand and natural=beach+surface=sand should be, I think, rendered in the same way.

Buy it doesn't work : Here an example with natural=shoal+surface=sand : sand is visible, not a lot but I'm ok with that.

Here an example with natural=beach+surface=sand : same rendering as above.

And here an example with natural=sand (+surface=sand) : nothing is visible.

Could you fix that ?

Best regards

edit : I just see that we have the same problem with beach+gravel, shoal+gravel and shingle (no rendering outside the coastline)

[imagico]

For a bit of background - someone has recently massively edited the wiki pages of coastal features to reflect a very special view how things should be tagged that is not widely represented in practical use of tags, namely that tidalflats should only be tagged as wetland=tidalflat if they are mud based and that every other area of loose material outside the coastline independent of its geomorphological characteristics should be tagged natural=sand or natural=shoal.

The practically dominant use of tags we try to reflect in our rendering is that beaches (i.e. wave formed coastal stretches, which can be grain size from fine sand to boulders) are tagged natural=beach both above and below the coastline, tidal flats (i.e. flat areas of loose material in the tidal zone predominantly shaped by tides) are tagged wetland=tidalflat and distinct protrusions from the surrounding ground in the tidal range (like sandbars) are tagged natural=shoal. And that natural=sand is not widely or consistently used for tidal sand areas (and its adoption for such purpose would remove existing consistent geomorphological distinctions).

There are issues with our current rendering (like #3707, #3840) which we hope to solve by implementing #3854, which in turn depends (both technically and design wise) on getting #4128 settled. In the ac-style i have shown how this could ultimately look like - see https://github.com/gravitystorm/openstreetmap-carto/issues/3854#issuecomment-526128595 or here:

[Mahabarata]

Hi,

The drawing to show the proposal seems great to me. Except for one thing.

I'm not agree with tidal flat : the definition of a tidal flat is clear according to wikipedia = it is another name for mudflat. It means that there is no sand in this areas. In french, the translation is clear too, "vasière" = only mud, not sand. Same thing in german, the translation is "watt". You will never see birds in sand areas (except seagulls) but a lot in mudflats : the seagulls find food in the water, the other birds find food in the mudflat itself because it is a rich area of small animals (small shrimps, small crabs). So to distinguish sand areas and tidalflat areas is very important.

If today we have a lot of sand areas with the bad tags tidalflat, it is, I think, because when the wiki page for tidalflat was created, someone add "sand" in the definition.

It was an error. If someone define a dog like an animal which can be sometimes wild (in some countries you have a lot of dogs without master), does that mean that a wolf is a dog ?

To use a tag in osm which doesn't fit the official definition of the word should be discouraged, not encouraged : for me, tidalflat sand should be rendered with nothing (grey) to let know to users that these tags are wrong and to encourage them to use more appropriate ones.

Best regards

[imagico]

Please no tagging discussion here. We look at what tags are actually consistently used for and we don't care if that use is wrong by someone's view of what the strings used for key and value are supposed to be used for in their culture specific understanding of language.

And wetland=tidalflat is consistently used for tidally dominated flat areas in the tidal zone consisting of loose materials of all grain size without vegetation. At the North Sea coast in the Netherlands, in Germany and in Denmark the majority of the tidal zone is sand covered (or mixed of sand and finer material) and almost all of that is tagged with wetland=tidalflat. Same applies for large tidalflat areas elsewhere on Earth - in particular in Argentina and Alaska. We will not endorse an initiative to degrade this semantically meaningful and diligently applied tagging well documenting the geomorphological characteristics of these areas into a generic there is sand here equally applied to beaches, desert dunes and tidalflats.

Not to mention of course that while you can from low tide satellite imagery well distinguish between tide dominated (tidalflat) and wave dominated settings (beaches) with some experience, you have no chance to make a verifiable cutoff regarding grain size without local inspection on the ground. So the whole endeavor to limit wetland=tidalflat to muddy areas is destined to fail inevitably anyway - the only thing those who push for redefining tags here can hope for is another landuse=forest vs. natural=wood disaster: Diluting all of the semantic meaning mappers try to record diligently until only the smallest common denominator remains (which likely would be: This is a tidal zone). :facepalm:.

We have a well established secondary tag (surface=*, used on ~10 percent of all wetland=tidalflat so far) to document the material of the tidal flat area when known to the mapper. As my demonstration shows we could use that to differentiate wetland=tidalflat rendering in the future.

And by the way - the German term Watt you mention is distinctly not tied to small grain size material - we have distinct subtypes (Sandwatt, Mischwatt, Schlickwatt) reflecting different grain size compositions.