Add color fill for mangrove, reedbed and saltmarsh wetlands

[jeisenbe]

Fixes #2013
Fixes #2025
Fixes #3799

Changes proposed in this pull request:

• Add @scrub color fill for wetland=mangrove
• Add @grass color fill for wetland=reedbed
• Add @grass color fill with a white dot pattern for wetland=saltmarsh

1) Mangrove @scrub fill - #c8d7ab

z13 Potowai

Before
https://www.openstreetmap.org/#map=13/-4.2983/134.9890

After

z15 mangroves by tidalflat

https://www.openstreetmap.org/#map=15/-4.4014/135.0908
Before

After

2) Saltmarsh with white dot pattern and @grass fill - #cdebb0

• Pattern from http://www.imagico.de/map/jsdotpattern.php#x,128,jdp68152;g,5,32,32;rx,25,2,32,32;rx,25,2,32,32;rx,25,2,32,32;rd,0,0,0,dot,0.125,10,9,0,jdp48264,ffffff,cdebb0;

A. Salt marsh (left) and marsh (right) - before

https://www.openstreetmap.org/#map=13/51.3590/4.1393

After

B. Before - Small salt marshes: one over sea water, one over land in database

https://www.openstreetmap.org/#map=15/51.4552/3.6555

After

C. Before - saltmarsh (left) next to marsh (center), mouth of the Eel river, California

After

3) Reedbed @grass fill #cdebb0

z18 reedbeds vs marshes before

https://www.openstreetmap.org/#map=18/51.73020/3.72768

After

z16 reedbed farm before

https://www.openstreetmap.org/#map=16/51.4514/3.9198

After

[imagico]

Looks good and works fine overall.

I think it would be better to only render the grain pattern for saltmarsh on the higher zoom levels where wetlands are also differentiated through the overlay (that is z13 and above).

It might also be worth trying the reef pattern (in bright color obviously) as alternative to the pattern you designed. I could imagine it might look a bit less noisy.

[imagico]

By the way @jeisenbe - i sent you two emails about two weeks ago and i have not received a reply - could you check your spam filter?

[jeisenbe]

By the way @jeisenbe - i sent you two emails about two weeks ago and i have not received a reply - could you check your spam filter?

I never got those emails @imagico, even after checking the spam filter. Did you every try resending them?

[jeisenbe]

I've made the suggested change to limit the pattern to z13 and above only.

And I've also tested using a pattern based on reef.svg but in white with partial opacity as suggested:

0. Current commit
   

1. Reef dots - 100% opacity white
   

2. Reef dots - 75% opacity
   

3. Reef dots - 50% opacity
   

The 100% opacity version is ok, but the others seem too subtle to me.

[imagico]

I never got those emails @imagico, even after checking the spam filter. Did you every try resending them?

I sent them again on June 20 - from two different mail clients. That Google has an arrogant attitude in terms of what email they regard as legitimate and what not is known but i have never seen this in such extreme form with both mail from my gmx.de and from imagico.de being silently ignored (which is not ok by any measure - they'd either have to reject it or deliver it).

I tried using the OSM messaging system now - which should work no matter how much google tries to prevent it from working.

[jeisenbe]

Here are tests at https://www.openstreetmap.org/#map=#14/40.6192/-124.2910 with salt marsh and marsh in the same view:

0. Current commit
   

1. 100% opacity reef pattern in white
   

2. 75%
   

3. 50%
   

[jeisenbe]

@imagico thoughts on the current pattern vs the more subtle reef.svg white dot pattern?

[imagico]

The reef pattern seems a bit weak as is - you could try to make it heavier but ultimately the current version is fine as well.

[jeisenbe]

OK. I prefer having the slightly stronger pattern in the current commit, if it's not a problem.

…

On 8/21/19, Christoph Hormann ***@***.***> wrote: The reef pattern seems a bit weak as is - you could try to make it heavier but ultimately the current version is fine as well. -- You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub: #3807 (comment)

[imagico]

Ok, i am fine with this now.

[jeisenbe]

I'm going to merge this now, since it's been approved by another maintainer and there were no other concerns. Just checking: is it best to "squash and merge" the 5 commits into one, or just rebase and merge the 5 separate commits?

It seems like it would help keep the development history more concise to use squash and merge.

[imagico]

I don't really mind but others have indicated to prefer squash and merge.

[kocio-pl]

The motivation was that squash and merge keeps the code changes clean, so people outside the project can see what changes, not the inner technical details of the change. It also makes writing Changelog easier.

[jeisenbe]

Should I update the changelog when merging this PR, or is that done later?