[HOT] [Talk-ca] Flooding in Richelieu River, Quebec, Canada :Follow-up(Complement of information)

[Daniel Begin]

Hi all,

 

I've tried to "put my boots on the ground" without  going there !-).  I
found a pretty clear satellite image of the river at its max level (30-31m)
- which doesn't really changed since ...

 

http://earthobservatory.nasa.gov/NaturalHazards/view.php?id=50577

 

I made few geometric adjustments to the image and compare the result with
30m GeoBase and SRTM contours. Geobase contours gives better results between
St-Jean and L'Île aux Noix but from there to the US boundary, neither
GeoBase nor SRTM is right.

 

I'm trying to produce a flooded area from DEM data while I'm looking at an
image that contains what I'm looking for - the flooded area!  Why not
mapping from it? I understand the image can be used -
http://earthobservatory.nasa.gov/ImageUse/

 

Any idea on how I can use this GeoTiff image in JOSM?

 

Cheers,

Daniel

 

 

  _____  

From: Daniel Begin [mailto:jfd553 at hotmail.com] 
Sent: May-27-11 15:18
To: 'Brent Fraser'
Cc: 'HOT Openstreetmap'; 'talk-ca'
Subject: Re: [Talk-ca] [HOT] Flooding in Richelieu River, Quebec,Canada
:Follow-up(Complement of information)

 

Bonjour Brent,

Which one is better in this situation, SRTM or GeoBase?  I would prefer SRTM
for the following reasons...

 

SRTM has 90m cell size but the data in each cell are "real elevation" -
including roof tops and crop height !, 

Geobase has 30m cell size but the data in each cell is an interpolation
between adjacent contour interval from 50K map.

 

So, in flat/low hills areas, SRTM will give a much better idea of the height
than Geobase. For example, a 8m hill, standing between elevation 31m and 39m
won't be visible in Geobase data. Same reasoning for a steep cliff, and so
on...

 

Furthermore, I suspect that using the 30m contour from SRTM might be valid
for St-Jean-Sur-Richelieu area  but the 31m should probably be used near
Lake Champlain.

 

In this case I would quote James Ewen, You need to "Put your boots on the
ground" to decide which one is valid !-)

 

I'm also looking at converting this into .gpx format as proposed Richard

Cheers,

Daniel

 

  _____  

From: Brent Fraser [mailto:bfraser at geoanalytic.com] 
Sent: May-27-11 14:19
To: Daniel Begin
Cc: 'Pierre Béland'; 'HOT Openstreetmap'; 'talk-ca'
Subject: Re: [Talk-ca] [HOT] Flooding in Richelieu River, Quebec, Canada
:Follow-up(Complement of information)

 

Daniel,

  The SRTM data has 90m cell size, while the CDED (from the Geobase site)
has 30m cells (and 1m height resolution) which might rendered better
contours.

Best Regards,
Brent Fraser


On 5/27/2011 11:52 AM, Daniel Begin wrote: 

Bonjour tous le monde,

 

I have generated a 30m and 31m contour lines for Richelieu river and lake
Champlain (using SRTM data). It fits the 30m contour provided by
Jean-Guilhem but doesn't seem to fit pretty well the flooded wetland area
provided by Pierre.

 

Any idea if this data can be used (usgs licence point of view)?

And if it can be usefull?

 

Daniel

 

  _____  

From: Pierre Béland [mailto:infosbelas-gps at yahoo.fr] 
Sent: May-27-11 12:40
To: HOT Openstreetmap
Cc: talk-ca
Subject: Re: [HOT] [Talk-ca] Flooding in Richelieu River, Quebec,Canada
:Follow-up(Complement of information)

 

Jean-Guilhem Cailton  wrote on 2011-05-27


> According to the shapefile data, Lake Champlain, and hence
Venise-en-Québec are above the 30 m elevation.

> The shapefile contains punctual elevations of 31 m in this area (Plage
Missisquoi, for example).

> The next contour line would be the 40 m one, but it does not look like it
would be very useful for this.

This is exact. The 40 meter contour line is not usefull for us.

 

Thanks Jean-Guilhem.

 

Pierre Béland

 
 
_______________________________________________
Talk-ca mailing list
Talk-ca at openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-ca
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/hot/attachments/20110527/067a17eb/attachment.html>