Re: [talk-au] Correcting inland water features

[Ian Sergeant]

 On 19/05/18 11:38, Ian Sergeant wrote:
> flow=ephemeral, maybe.  water-presence=ephemeral?

On 19 May 2018 at 11:44, Warin <61sundow...@gmail.com> wrote:

> 'ephemeral=yes' matches the present use of 'intermittent=yes'. I like at
> least some consistency in the tagging.

I think you picked the wrong mapping project :-)

The issue here of course, is that the next tag will something=yes.
Ugliness.  Best to fix it now.

Ian.

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Correcting inland water features

[Warin]

On 19/05/18 11:38, Ian Sergeant wrote:

On 19 May 2018 at 11:34, Warin <61sundow...@gmail.com> wrote:


Some are using stream=ephemeral ... low usage.
I'd rather go with ephemeral=yes as that then can be used on 'lakes' and
other things.

flow=ephemeral, maybe.  water-presence=ephemeral?

Ian.



'ephemeral=yes' matches the present use of 'intermittent=yes'. I like at 
least some consistency in the tagging.



___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Correcting inland water features

[Ian Sergeant]

On 19 May 2018 at 11:34, Warin <61sundow...@gmail.com> wrote:

> Some are using stream=ephemeral ... low usage.
> I'd rather go with ephemeral=yes as that then can be used on 'lakes' and
> other things.

flow=ephemeral, maybe.  water-presence=ephemeral?

Ian.

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Correcting inland water features

[Warin]

On 19/05/18 11:02, Andrew Davidson wrote:

On 16/05/18 07:35, Warin wrote:
Many inland waters in Australia are 'intermittent' meaning they only 
flow when there is rain and that rain may only occur every 5 years or 
so on average.


The vast majority of Australian stream are non-perennial. The WSJ has 
a cool slider that shows this:


http://graphics.wsj.com/documents/wsj_sliders14/AustralianStreams

There are a least two problems for Australian mapping:

1. There is no OSM tag for ephemeral streams. On a stream length basis 
the majority of Australian streams are ephemeral.


Some are using stream=ephemeral ... low usage.
I'd rather go with ephemeral=yes as that then can be used on 'lakes' and 
other things.




2. There is no OSM tag for the end of a watercourse. This causes 
problems for validators because they can't tell if a stream finishes 
because the mapper gave up or the stream itself just peters out.


I did strike one stream that flowed in to a river .. at both ends!! That 
had me puzzled. I tracked it back using topo information to find it went 
over a saddle, so I mapped it as flowing away from the saddle on both 
sides ..


As for indicating an 'end' ... maybe exit=underground on the last node ???


___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Correcting inland water features

[Andrew Davidson]

On 16/05/18 23:26, Andrew Harvey wrote:


GA also has a Surface Hydrology dataset [2] [3] [4] which could be used 
as well.


+1

This is a good source as it covers all of the country and in effect has 
the datasets from the states we already have permission from. It's also 
useful for getting the name of the stream and they flag streams as being 
perennial or not.


The downside is that the data comes from a range of mapping some of 
which includes the 1:250,000 NatMap series, so the streams can be quite 
inaccurate in places.


___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Correcting inland water features

[Andrew Davidson]

On 16/05/18 08:51, Ewen Hill wrote:

There are also a number of areas I think in SA that have been inundated by
streams from a data source and these could possibly be trimmed back. Flood
water on flat deserts will make its own mind up next time. I would be really
keen to hear more and thank you for your time in improving the map.


My personal view is that there is not much point in mapping unnamed 
streams. By only bothering with the named streams you are in effect 
applying something like the Wikipedia notability test.


I know the bit of SA you are referring to and it is a bit of a mess 
because only the SA part of the streams have been imported and they all 
get cut off at the NT border.


___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Correcting inland water features

[Andrew Davidson]

On 16/05/18 07:35, Warin wrote:
Many inland waters in Australia are 'intermittent' meaning they only 
flow when there is rain and that rain may only occur every 5 years or so 
on average.


The vast majority of Australian stream are non-perennial. The WSJ has a 
cool slider that shows this:


http://graphics.wsj.com/documents/wsj_sliders14/AustralianStreams

There are a least two problems for Australian mapping:

1. There is no OSM tag for ephemeral streams. On a stream length basis 
the majority of Australian streams are ephemeral.


2. There is no OSM tag for the end of a watercourse. This causes 
problems for validators because they can't tell if a stream finishes 
because the mapper gave up or the stream itself just peters out.


___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au

Re: [talk-au] Correcting inland water features

[Phil (The Geek) Wyatt]

Hi folks,

 

Bay or river closing lines are defined here

 

http://www.ga.gov.au/scientific-topics/marine/jurisdiction/maritime-boundary-definitions

 

There is also the GA Coastline which may help

 

http://services.ga.gov.au/site_11/rest/services/Global_Map_Project_Australian_Base_Map/MapServer/6

 

 

Cheers - Phil

 

 

From: Andrew Wiseman [mailto:andrew_wise...@apple.com] 
Sent: Friday, May 18, 2018 10:07 AM
To: OSM Australian Talk List
Subject: Re: [talk-au] Correcting inland water features

 

Thanks for the feedback and suggestions everyone, that’s very helpful. We have 
also added AGRI to our list of sources.

 

Where the inland water meets the coastlines, we have seen a few different 
styles of modeling so I wanted to see if you had a local policy about it. 

 

In some cases we’ve seen things like this, where the coastline goes very far 
upstream: https://www.openstreetmap.org/way/14763476#map=11/-12.4279/132.4683 

 

In other cases, the coastline is farther downstream, like where the Pine River 
stops here: https://www.openstreetmap.org/way/159813122#map=14/-27.2842/153.0761

 

We were thinking the second example is better. If that’s the case for the 
Alligator River we would put the coastline here (which is also where the GA 
Surface Hydrology polygon has the river starting) 
https://www.openstreetmap.org/?mlat=-12.2296501875 

 =132.402678198#map=15/-12.2293/132.4083 and change the existing coastline 
to riverbank.

 

Please let me know what you think.

 

Thanks,

 

Andrew


Andrew Wiseman |  Maps | andrew_wise...@apple.com

 

APPLE CONFIDENTIAL

This message (including attachments if any) is for the private use of the 
addressee only and may contain confidential or privileged information. If you 
have received this message by mistake please notify the sender by return e-mail 
and delete this message and any attachments from your system. Any unauthorized 
use or dissemination of this message, and any attachments in whole or in part 
is strictly prohibited.

 

 





On May 16, 2018, at 9:26 AM, Andrew Harvey  wrote:

 

Agreed with Warin and Ewen.

 

I'd like to point out the AGRI imagery [1], yes it's old (2006-2011), black and 
white, and missing parts of the imagery due to cloud and seams, but it may help 
as an extra source to compare. It's good enough to make out rivers and it 
should be accurate positionally since it was validated with ground surveys. CC 
BY 4.0 and usable in OSM from the blanket Geoscience Australia wavier[5].

 

GA also has a Surface Hydrology dataset [2] [3] [4] which could be used as well.

 

Some features are better tagged as a wetland 
https://wiki.openstreetmap.org/wiki/Tag:natural%3Dwetland. Local knowledge is 
always prefered, but so long as features mapped from local knowledge are not 
impacted, something mapped from other sources is better than nothing, which can 
be improved further by local knowledge.

 

I'm interested to know more about the alignment issues when they meet the 
coast, is there a specific example of where this needs work at the moment in 
OSM?

 

PS. Microsoft's Open Maps team has similar tasks at 
https://github.com/Microsoft/Open-Maps/issues/14 and 
https://github.com/Microsoft/Open-Maps/issues/11

 

[1] 
https://github.com/osmlab/editor-layer-index/blob/gh-pages/sources/australia/au/AGRIblack-and-white25m.geojson?short_path=4ca9ae8

[2] 
https://ecat.ga.gov.au/geonetwork/srv/eng/catalog.search?node=srv#/metadata/1186e898-14b5-812e-e053-10a3070a76f0

[3] 
https://ecat.ga.gov.au/geonetwork/srv/eng/catalog.search?node=srv#/metadata/12777e32-ec4f-055a-e053-10a3070a2ce2

[4] 
https://ecat.ga.gov.au/geonetwork/srv/eng/catalog.search?node=srv#/metadata/123f4803-04d7-32d1-e053-12a3070a99ac

[5] 
https://wiki.openstreetmap.org/wiki/File:GeoscienceAustralia_CCBY_Waiver_EmailAcceptance.pdf

 

___
Talk-au mailing list
Talk-au@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-au