Special rendering for intermittent water areas #996
Comments
|
@imagico - can you propose a rendering? How Australia map show this type of terrain? |
|
Possibilities frequently used in classical maps are:
Of course these could also be combined. |
|
Are tidal planes also intemittent water areas? Would it make sense to have a similar rendering for both? |
|
Most water areas tagged intermittent=yes are seasonal or ephemeral, tidal flats on the other hand are flooded daily. This should definitely be rendered differently. Most common mapping for tidal areas is natural=wetland + wetland=tidalflat/saltmarsh/mangrove, see also #387 |
|
See also #709. |
|
The intermittent bodies of water are also often rendered with blue stripes and a dashed outline, at least in France. I don't know if it could be a better rendering, but it's a possibility, and it would match the current rendering of intermittent waterways. |
|
Seasonality is a specific sort of intermittency - variation is according to seasons instead of more randomly weather-related. I came here to ask for waterways with seasonal=yes to be rendered the same way as intermittent=yes - water bodies have a similar problem. |
|
sent from a phone
I agree and generally support this, even if it might have to wait till we add hstore (if ever) |
Because we already have hstore and are able to use non-standard tags too, we can do it quite easily now. I just wanted to ask if this is still the same (just adding seasonal=*, without "no" probably, the same as we use intermittent=yes) as I understand original proposition or something has changed in the meanwhile? Currently seasonal has 108 828 uses together with waterway=* and 16 696 with natural=water, which makes it very popular tag. It's also growing fast in the last years:
|
|
Seems relevant to render both tags this way, as |
|
Are there some things blocking this? Would be great to render intermittency. |
|
Probably the missing part is someone to make the code. Could you try? |
|
Oh, dear… I will try, as I have no real knowledge about CartoCSS, but, hey, I won't have any knowledge if I don't try. |
|
That's how I got into this whole thing. 😄 I still have only a basic understanding of CartoCSS, but dealing with SQL seems to be more important in osm-carto (even if I also don't know it too much). This case shouldn't be too hard, since you want to just copy intermittent tag behavior, not to create completely new code. |
Well, we'll see… (insert snapping noise of latex gloves here) |
|
How about this way? It is an inner member of the meadow multipolygon, which should explain the grey lines. |
|
I guess very light blue background would work better for me. For me it's enough that stripes are wide. |
|
Do you mean a lighter blue between water-coloured stripes? I let it transparent to leave the eventual other landuse or natural area visible, as an intermittent water area is often used for something else when not flooded. That way, the stripes allow to see both usual landuse and flooded area. |
|
@kocio-pl : could you explain what you meant? I don't understand. Other opinions about possible improvements are welcomed. |
|
Yes, I meant lighter blue instead of a land color. Could you give the examples when the area has a different use? I can't imagine such examples, so I tend to think my solution would be better. Reality check would help us to make decision. |
|
@kocio-pl : here is one: Not as legible as I thought: the borders disappear way too notably between the stripes, and that is even more present as you zoom in. Maybe with a dashed outline? I'll test that. I don't reject your proposal, but, IMO, a plain color would mask another possible landuse, I frequently saw that, so I prefer a way to display it if it exists. |
|
Great, thanks for testing! |
|
@kocio-pl : and with the dashed outline, the same as the one for intermittent streams (4,3): Maybe with longer dashes (8,4): |
|
Why not use the Danger pattern with (or without) dashed stroke ? Very large dash may increase focus on intermittent areas. On french map, dashes of intermittent area are very thin: |
|
2018-03-05 16:21 GMT+01:00 Penegal <[email protected]>:
@kocio-pl <https://github.com/kocio-pl> : here is one:
[image: image]
<https://user-images.githubusercontent.com/13777824/36982649-a2cdbdfa-2090-11e8-9288-c28313ea9cc7.png>
I think the approach is OK, but the details should be adjusted. If you look
at the topographic map in this thread (5/3/2018, 19:02) that one works
nicely because the lines and distances are much finer (i.e. can show much
more detail). Also 0 degress hatching angles is more pleasing than the 60
we have here.
|
Works for me. |
|
Then here it comes with horizontal 2px lines and dashed border: And without the dashed border: The dashed border seems pretty useless with thin stripes, so I'll get rid of it. Is it OK, now? Anything to improve? |
|
For me that's very convincing and the border is not needed. Could you prepare a PR? |
|
@kocio-pl : before a PR shouldn't I wait for other advices, or is there a clear enough consensus? |
|
Looks good to me too! Go ahead I’d say. |
|
I've learned that waiting too much does not help. We're discussing it for many days already and it will get more attention once the PR is ready. |
|
2018-03-06 11:26 GMT+01:00 Matthijs Melissen <[email protected]>:
Looks good to me too! Go ahead I’d say.
+1, what might eventually be worth a consideration: how does this relate to
natural=wetland, which one is "wetter"?
|
|
For info: The horizontal hatching in French topographic maps indicates Zone inondable - which means area subject to flooding which is not the same as intermittent waterbodies in OSM - see http://www.viewranger.com/help/FranceLegend.pdf Might be good to keep in mind regarding mapper feedback, in particular for French mappers. |
|
Good to know, however it's inevitable that we will use similar visual elements in the other meaning or context than some other maps. It might be more interesting if we have similar type of object so we could differentiate them - do we have "area subject to flooding" tagging scheme? |
|
2018-03-06 12:21 GMT+01:00 Christoph Hormann <[email protected]>:
For info: The horizontal hatching in French topographic maps indicates *Zone
inondable* - which means *area subject to flooding* which is not the same
as intermittent waterbodies in OSM - see http://www.viewranger.com/
help/FranceLegend.pdf
Is this about the difference between "generally not a waterbody, but
exceptionally flooded" and "generally a waterbody, but exceptionally dried
out"? Will this be verified statistically?
The wiki currently says about intermittent: "*a water
<https://wiki.openstreetmap.org/wiki/Key:water>=* body is not permanently
filled with water ... either regularly depending on seasons or based on
other conditions"*
Maybe this requires further specification?
|
|
The Zone inondable (area subject to flooding) feature in French map is not related to risk such as river flooding. It's restricted to body of water (pond, small lake) where the water level is susceptible to greatly fluctuate (e.g. flooding). In the example of the topographic map figured above, it's typically a seasonal small lake. It's dry during a long period because it's located in a karstic system. You won't find such hatching along rivers for example. |
|
For better understanding - Zone inondable has a fairly specific meaning in French - see https://fr.wikipedia.org/wiki/Zone_inondable - that is strongly tied to French climate and the flooding patterns that are common in France. A few examples of use of this in the classic IGN 25k map: |
|
Being French, I can tell you that what is mapped as zone inondable is any area which is intermitently covered by water; that includes both floodable areas and intermittent water bodies. In fact, most areas mapped as zone inondable are only intermittent water bodies, the rest being only the most flood-prone areas – not all of them, by far –, so you don't have to bother with the distinction between floodable areas and intermittent water bodies just because the french IGN confuses them on its maps. Edit: after a look at @imagico's examples, I correct: there are more flood-prone areas than I thought which are mapped under zone inondable, but it also includes salterns and intertidal zones, so one should really not understand zone inondable as floodplain, as a naive translation could translate it. |
|
BTW - what do you think about #709? Maybe we could fix it in a similar way? |
|
Another interesting thing to test: #1167. |
|
One more similar idea: #1547. |
@imagico proposed in #805 to render specially intermittent water areas.
intermittent tag is in the dabase, seasonal needs style file change.
quote from #805 (comment)
This should be extended to water areas, for example the current appearance of Australia is quite irritating:
http://www.openstreetmap.org/#map=6/-24.377/129.749
since nearly all of the water areas you can see there are intermittent.
Both intermittent=yes and seasonal=* should be considered.
Current use according to taginfo:
The text was updated successfully, but these errors were encountered: