[GRASS-user] Re: Importing Dems with r.in.xyz

[Micha Silver]

On 06/05/2010 09:45 AM, Hanlie Pretorius wrote:
> 2010/6/4, Micha Silver<micha at arava.co.il>:
>    
>> On 06/04/2010 04:20 PM, Hanlie Pretorius wrote:
>>      
>>> Hi,
>>>
>>> I've been following the procedure at
>>> http://grass.osgeo.org/wiki/Import_XYZ to import DEMs into GRASS.
>>>
>>> It has worked for two DEMS, but I have a problem with the third one at
>>> the step where one "verifies that the number of rows in the ASCII file
>>> corresponds to the number of cells in the enlarged region".
>>>
>>> At this point, g.region reports 1146474 cells in the region, while I
>>> have 1146370 lines of coordinates in my file.
>>>
>>>
>>>        
>> Assuming you already did the "r.in.xyz -s ..." step to get and set your
>> region to match the input data.
>> So it looks like there are about 100 coordinates missing from the ASCII
>> file. Maybe "holes" in the data?
>> One way to work around this might be to import the point data as a 3D
>> vector, then run the usual v.surf.rst interpolation.
>> v.in.xyz -z in=ascii.txt out=elev_pts z=3 fs=,
>> then
>> g.region vect=elev_pts res=<choose appropriate resolution>
>> v.surf.rst elev_pts elev=dem layer=0
>>      
> Thanks for the suggestion, Micha.
>
>   I've tried the interpolation route before and found that the
> difference between the resulting interpolated surface and the original
> DEM was up to 7m. I want to use the DEMs for a flood application, so
> they need to be as accurate as possible.
>
>    
I wonder if playiing with the "tension" and "smooth" parameters would 
help? Higher tension (>40) means that the interpolation acts more like a 
rubber sheet, so each point influences a maller area, and smooth=0 means 
that the final dem must go exactly thru each point.
> I was thinking perhaps importing the points as vectors, converting
> them to raster and then doing a nearest neighbour or IDW interpolation
> to fill the gaps. At least then I'll be able to see where the gaps are
> and limit the interpolated pixels using a mask?
>    
Worth a try. But as you probably know, nearest neighbor won't take 
trends in the surface into account. IDW the same.  And IDW usually gives 
worse terrain results than RST.
>    
>>      
>>> The output of g.region is:
>>> -----
>>> projection: 99 (Transverse Mercator)
>>> zone:       0
>>> datum:      ** unknown (default: WGS84) **
>>> ellipsoid:  wgs84
>>> north:      -49312.49
>>> south:      -74587.5
>>> west:       -3015862.5
>>> east:       -2987512.5
>>> nsres:      25.00000989
>>> ewres:      25
>>> rows:       1011
>>> cols:       1134
>>> cells:      1146474
>>> -----
>>>
>>> The first three and last three coordinates in my text file are:
>>> -----
>>> -74575.000000,-3015850.000000,1548.830000
>>> -74575.000000,-3015825.000000,1548.330000
>>> -74575.000000,-3015800.000000,1547.500000
>>> .
>>> .
>>> .
>>> -49324.990000,-2987575.010000,1510.980000
>>> -49324.990000,-2987550.010000,1511.240000
>>> -49324.990000,-2987525.010000,1511.470000
>>> -----
>>>
>>> Can someone help me to figure out what's going on?
>>>
>>> Thanks
>>> Hanlie
>>> _______________________________________________
>>> grass-user mailing list
>>> grass-user at lists.osgeo.org
>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>>
>>> This mail was received via Mail-SeCure System.
>>>
>>>
>>>
>>>        
>>
>> --
>> Micha Silver
>> Arava Development Co. +972-52-3665918
>> http://www.surfaces.co.il
>>
>>
>>
>>      
> This mail was received via Mail-SeCure System.
>
>
>    


-- 
Micha Silver
Arava Development Co. +972-52-3665918
http://surfaces.co.il