[talk-au] Mangroves and coastline

[David Groom]

> ----- Original Message ----- 
> From: "Brent Easton" <b.easton at exemail.com.au>
> To: "David Groom" <reviews at pacific-rim.net>; "Stephen Hope" 
> <slhope at gmail.com>; "OSM Australian Talk List" <talk-au at openstreetmap.org>
> Sent: Friday, December 28, 2007 9:08 AM
> Subject: Re: [talk-au] Mangroves and coastline
>
>
> natural=marsh will be rendered eventually.
>
> I have the skills to do the Osmarender version. How do we want it to look? 
> I did a test version a while back with a proper little marsh symbol (lines 
> and grass) like on topo maps. It looked pretty good.

Sounds good.  One important thing is not only how it looks at high zoom, but 
also when at low zoom the general colour is different enough from other 
currently rendered fills.

here's the svg symbol I used in the images for Cairns & Darwin, but I think 
yours is probably going to be better.

        <svg:pattern id="marsh-pattern" x="0" y="0" width="10" height="10" 
patternUnits="userSpaceOnUse" patternTransform="scale(0.25)">
           <svg:rect fill="#f5ffeb" stroke="none" width="10" height="10" 
x="0" y="0" />
           <svg:path fill="none" stroke="#1e1efd" stroke-width="0.5" d="M 
2.25,2.37 C 2.67,1.4 3,1.35 3.5,2 C 4,2.65 4.47,2.4 4.9,1.5 " />
           <svg:path fill="none" stroke="#1e1efd" stroke-width="0.5" d="M 
7,7 C 7.5,6 8,6.4 8.4,6.8 C 8.8,7.25 9.3,7.25 9.7,6.35 " />
        </svg:pattern>


David




>
> Cheers,
> Brent.
>
> *********** REPLY SEPARATOR  ***********
>
> On 27/12/2007 at 12:40 PM David Groom  wrote:
>
>>----- Original Message ----- 
>>From: "Stephen Hope" <slhope at gmail.com>
>>To: "OSM Australian Talk List" <talk-au at openstreetmap.org>
>>Sent: Thursday, December 27, 2007 12:05 PM
>>Subject: [talk-au] Mangroves and coastline
>>
>>
>>>I was looking at the top of Cape York, where the current coastline in
>>> OSM ends.  I was comparing the PGS coastal data with the current OSM
>>> data and a yahoo image of the current coastline, and notices that the
>>> data seems to be out by several kilometres.  Finally I realised that
>>> the PGS data is following the dry land border, whereas the current
>>> data in OSM is following the outer edge of the mangrove swamps.
>>> Normally this is not a major difference, but here there were
>>> differences of up to 12 km.
>>>
>>> Are there any guidelines as to which we should use? I guess it depends
>>> on your definition of the coast. I can see arguments both ways. If we
>>> decide to use the outer mangrove edge, then we should check the PGS
>>> data in Qld even more than normal, because of the reefs and large
>>> mangrove beds behind them in places.
>>
>>
>>In Cairns & Darwin areas, I have marked the coastline as the outer edge of
>>the mangrove swamps. (ie at the mangrove / clear water interface), but
>>kept
>>the PGS outline at the land / mangrove interface removed the tag
>>natural=coastline, and tagged an area as natural= marsh.  (maybe not the
>>best tag, but currently there is no natural=mangrove_swamp, so it appeared
>>to be the best compromise).
>>
>>Sadly neither the Mapnik or osmarender layers currently render
>>natural=marsh
>>(i have put in request for both to do so, but to no avail), but you can
>>see
>>what the rendering would look like at
>>
>>http://wiki.openstreetmap.org/index.php/Image:Cairns.png
>>
>>http://wiki.openstreetmap.org/index.php/Image:Darwin.png
>>
>>David
>>
>>
>>>
>>> While I'm thinking of it - are we interested in mapping the Great
>>> Barrier Reef? (or any other reefs for that matter?)  Permanently dry
>>> islands, sure.  But what about the low-tide only areas, or those that
>>> are just below water level all the time?
>>>
>>> Stephen Hope
>>>
>