Mapping Coastlines (Was: Re: Boundary removal.)

classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|

Mapping Coastlines (Was: Re: Boundary removal.)

Andrew Harvey-3
On Thu, Feb 2, 2012 at 11:22 AM, Nick Hocking <[hidden email]> wrote:
> I tried remapping some coastline down near Eden, from
> Bing imagery. This works well buit is very time comsuming
> and the coastline is BIG and probably Bing imagery does
> not cover it all at the necessary resolution.
>
> I suggest we get back our old state natrional boundaries
> from before ABS and then improve them over the next year
> or so, at our leisure.

I'm curious about how to map the coastline from imagery? The coastline
is meant to be mean high tide, but with all those waves coming in and
only a rough guess of where abouts in the 12 hr cycle the imagery was
taken how do you know where to put the way?

_______________________________________________
Talk-au mailing list
[hidden email]
http://lists.openstreetmap.org/listinfo/talk-au
Reply | Threaded
Open this post in threaded view
|

Re: Mapping Coastlines (Was: Re: Boundary removal.)

Ian Sergeant-2
On 2 February 2012 21:46, Andrew Harvey <[hidden email]> wrote:

I'm curious about how to map the coastline from imagery? The coastline
is meant to be mean high tide, but with all those waves coming in and
only a rough guess of where abouts in the 12 hr cycle the imagery was
taken how do you know where to put the way?



If there is a man made seawall or barrier, I use that.  If not, I try to estimate how high the water comes at high tide from the look of the terrain.  If all we have is one image, then that's all we have.

Ian.

_______________________________________________
Talk-au mailing list
[hidden email]
http://lists.openstreetmap.org/listinfo/talk-au
Reply | Threaded
Open this post in threaded view
|

Re: Mapping Coastlines (Was: Re: Boundary removal.)

David Groom
In reply to this post by Andrew Harvey-3


----- Original Message -----
From: "Andrew Harvey" <[hidden email]>
To: "Nick Hocking" <[hidden email]>
Cc: <[hidden email]>
Sent: Thursday, February 02, 2012 10:46 AM
Subject: [talk-au] Mapping Coastlines (Was: Re: Boundary removal.)


> On Thu, Feb 2, 2012 at 11:22 AM, Nick Hocking <[hidden email]>
> wrote:
>> I tried remapping some coastline down near Eden, from
>> Bing imagery. This works well buit is very time comsuming
>> and the coastline is BIG and probably Bing imagery does
>> not cover it all at the necessary resolution.
>>
>> I suggest we get back our old state natrional boundaries
>> from before ABS and then improve them over the next year
>> or so, at our leisure.
>
> I'm curious about how to map the coastline from imagery? The coastline
> is meant to be mean high tide, but with all those waves coming in and
> only a rough guess of where abouts in the 12 hr cycle the imagery was
> taken how do you know where to put the way?
>

With the high resolution imagery its usually quite easy to differentiate
between permanently dry areas, and areas which was been covered by water in
the last 12 hours.

Having said that, tropical regions where there may be large areas of
mangrove etc, it is quite common for the coastline way to be drawn at the
mangrove / water interface rather than the mangrove / land interface  .
This boundary is usually quite visible on even the low resolution imagery.

If all else fails, then you have guess when to put the coastline, someone
with more knowledge can always come along later and correct it.  If its a
choice between no coastline, and inaccurate coastline then I'd always go for
inaccurate.  You could always tag the ways with a fixme if you wanted to
flag them up.

Lastly, if you are redrawing coastline ways then can I make a reminder that
the direction of the way is important.  They must be drawn with the water on
the right hand side.

Regards

David


> _______________________________________________
> Talk-au mailing list
> [hidden email]
> http://lists.openstreetmap.org/listinfo/talk-au
>



_______________________________________________
Talk-au mailing list
[hidden email]
http://lists.openstreetmap.org/listinfo/talk-au
Reply | Threaded
Open this post in threaded view
|

Re: Mapping Coastlines (Was: Re: Boundary removal.)

Andrew Harvey-3
On Thu, Feb 2, 2012 at 11:24 PM, Ian Sergeant <[hidden email]> wrote:
> If there is a man made seawall or barrier, I use that.  If not, I try to
> estimate how high the water comes at high tide from the look of the
> terrain.  If all we have is one image, then that's all we have.

If there was a seawall that the water reached once a day that would
make it easy as that is your mean high tide mark, but few places have
such a wall.

On Thu, Feb 2, 2012 at 11:43 PM, David Groom <[hidden email]> wrote:
> With the high resolution imagery its usually quite easy to differentiate
> between permanently dry areas, and areas which was been covered by water in
> the last 12 hours.

I'm not an expert but for areas like a coastal beach which have waves
coming in won't the peak point where the water comes to be higher that
mean high tide?

For a lake with no upstream influence that would work well, but I was
thinking about where you have beach with waves coming in.

If anyone has some expertise and knows if this peak water point caused
by the waves is generally close to mean high tide or not that would be
good to know as then we can just trace/measure that mark.

>
> Having said that, tropical regions where there may be large areas of
> mangrove etc, it is quite common for the coastline way to be drawn at the
> mangrove / water interface rather than the mangrove / land interface  . This
> boundary is usually quite visible on even the low resolution imagery.

I think I'm going against what I said in an earlier thread on this
list, but I think now that the tide mark should be mapped
independently of what plant life is growing in that area of
land/water.

> If all else fails, then you have guess when to put the coastline, someone
> with more knowledge can always come along later and correct it.  If its a
> choice between no coastline, and inaccurate coastline then I'd always go for
> inaccurate.  You could always tag the ways with a fixme if you wanted to
> flag them up.

There are some lakes which I've observed over the full cycle (but even
that isn't really a mean, but just a sample of one day) and mapped
those more accurately, but it isn't so easy when you have waves coming
in.

>
> Lastly, if you are redrawing coastline ways then can I make a reminder that
> the direction of the way is important.  They must be drawn with the water on
> the right hand side.

Yep, defiantly.

_______________________________________________
Talk-au mailing list
[hidden email]
http://lists.openstreetmap.org/listinfo/talk-au
Reply | Threaded
Open this post in threaded view
|

Re: Mapping Coastlines (Was: Re: Boundary removal.)

David Groom
> ----- Original Message -----
> From: "Andrew Harvey" <[hidden email]>
> To: "OSM Australian Talk List" <[hidden email]>
> Sent: Friday, February 03, 2012 10:20 AM
> Subject: Re: [talk-au] Mapping Coastlines (Was: Re: Boundary removal.)
>
>
> On Thu, Feb 2, 2012 at 11:24 PM, Ian Sergeant <[hidden email]>
> wrote:
>> If there is a man made seawall or barrier, I use that.  If not, I try to
>> estimate how high the water comes at high tide from the look of the
>> terrain.  If all we have is one image, then that's all we have.
>
> If there was a seawall that the water reached once a day that would
> make it easy as that is your mean high tide mark, but few places have
> such a wall.
>
> On Thu, Feb 2, 2012 at 11:43 PM, David Groom <[hidden email]>
> wrote:
>> With the high resolution imagery its usually quite easy to differentiate
>> between permanently dry areas, and areas which was been covered by water
>> in
>> the last 12 hours.
>
> I'm not an expert but for areas like a coastal beach which have waves
> coming in won't the peak point where the water comes to be higher that
> mean high tide?
>
> For a lake with no upstream influence that would work well, but I was
> thinking about where you have beach with waves coming in.
>
> If anyone has some expertise and knows if this peak water point caused
> by the waves is generally close to mean high tide or not that would be
> good to know as then we can just trace/measure that mark.

That will depend on the gradient of the land between the highest high water
spring tide, and the lowest high water spring tide, and what location you
are in, since the tidal variation between spring and neap tides varies
enormously depending whereabouts you are.

I think you are possibly seeking a higher degree of accuracy than is
required.  Also then next question would then be to concern ourselves with
whether we took the peak water point of the largest expected wave, or the
average wave.

David

>
>>
>> Having said that, tropical regions where there may be large areas of
>> mangrove etc, it is quite common for the coastline way to be drawn at the
>> mangrove / water interface rather than the mangrove / land interface .
>> This
>> boundary is usually quite visible on even the low resolution imagery.
>
> I think I'm going against what I said in an earlier thread on this
> list, but I think now that the tide mark should be mapped
> independently of what plant life is growing in that area of
> land/water.
>
>> If all else fails, then you have guess when to put the coastline, someone
>> with more knowledge can always come along later and correct it. If its a
>> choice between no coastline, and inaccurate coastline then I'd always go
>> for
>> inaccurate. You could always tag the ways with a fixme if you wanted to
>> flag them up.
>
> There are some lakes which I've observed over the full cycle (but even
> that isn't really a mean, but just a sample of one day) and mapped
> those more accurately, but it isn't so easy when you have waves coming
> in.
>
>>
>> Lastly, if you are redrawing coastline ways then can I make a reminder
>> that
>> the direction of the way is important. They must be drawn with the water
>> on
>> the right hand side.
>
> Yep, defiantly.



_______________________________________________
Talk-au mailing list
[hidden email]
http://lists.openstreetmap.org/listinfo/talk-au
Reply | Threaded
Open this post in threaded view
|

Re: Mapping Coastlines (Was: Re: Boundary removal.)

Mick
In reply to this post by Andrew Harvey-3
Is there not an Official Datum for tide heights, I vaguely remember reference to it in a project I was involved in back in the early 1990's.

Failing that, are the GeoScience Australia Topo 250k series maps an acceptable source? I have the complete digital set here as either ESRI shape files or MapInfo tabs. From memory the mapping ranges from 1973 to 2001 and publication dates to 2006.

According to the website ( https://www.ga.gov.au/products/servlet/controller?event=DEFINE_PRODUCTS ) the data is released under Creative Commons Attribution 3.0 Australia Licence.

mick

_______________________________________________
Talk-au mailing list
[hidden email]
http://lists.openstreetmap.org/listinfo/talk-au
Reply | Threaded
Open this post in threaded view
|

Re: Mapping Coastlines (Was: Re: Boundary removal.)

Ian Sergeant-2
On 4 February 2012 12:45, mick <[hidden email]> wrote:
 
Failing that, are the GeoScience Australia Topo 250k series maps an acceptable source? I have the complete digital set here as either ESRI shape files or MapInfo tabs. From memory the mapping ranges from 1973 to 2001 and publication dates to 2006.

I think the 250k GA map is going to be considerably less accurate than any of the techniques we are discussing.

Ian.

_______________________________________________
Talk-au mailing list
[hidden email]
http://lists.openstreetmap.org/listinfo/talk-au
Reply | Threaded
Open this post in threaded view
|

Re: Mapping Coastlines (Was: Re: Boundary removal.)

Andrew Harvey-3
In reply to this post by David Groom
On Sat, Feb 4, 2012 at 4:00 AM, David Groom <[hidden email]> wrote:
> That will depend on the gradient of the land between the highest high water
> spring tide, and the lowest high water spring tide, and what location you
> are in, since the tidal variation between spring and neap tides varies
> enormously depending whereabouts you are.
>
> I think you are possibly seeking a higher degree of accuracy than is
> required.  Also then next question would then be to concern ourselves with
> whether we took the peak water point of the largest expected wave, or the
> average wave.

I think you are right, I'm probably trying to get too much detail.

The point I was trying to make was I thought mean high tide was
supposed to ignore the effects of waves. Perhaps though for OSM in
lure of more accurate details, the highest wave would do as a
coastline...

On Sat, Feb 4, 2012 at 12:45 PM, mick <[hidden email]> wrote:
> Is there not an Official Datum for tide heights, I vaguely remember reference to it in a project I was involved in back in the early 1990's.

Then we would still need the contour for that height.

_______________________________________________
Talk-au mailing list
[hidden email]
http://lists.openstreetmap.org/listinfo/talk-au