Networking in GRASS

[Malcolm Williamson]

Hello, Evaristo
For one of the only raster-based networking discussions that I have seen, 
refer to C. Dana Tomlin's _Geographic_Information_Systems_and_
Cartographic_Modeling_ (Prentice-Hall, Inc., 1990). A very good raster 
analysis reference, in case you're not familiar with it.
        -Malcolm Williamson
--
Malcolm D. Williamson - GIS Specialist           E-mail: malcolm at cast.uark.edu
Center for Advanced Spatial Technologies      Telephone: (501) 575-6159
Ozark Rm. 12                                        Fax: (501) 575-5218 
University of Arkansas              
Fayetteville, AR 72701

On Thu, 25 Jan 1996, 
Evaristo QUIROGA wrote:

>       Hello GrassGurus:
> 
> We have been working for more than 3 years with GRASS. Until we have used it
> as a graphic database with plotting aim and simple analysis of overlay
> operations, proximity functions, spread functions and seek functions.
> Lately we have developed some models with the combination of the map algebra
> and unix scripts.
> 
> In this line of progressive complication we have gone into the network
> problematic (transport of water through a system of irrigation channels,
> ordering of a river network, etc) and its resolution in GRASS environment. 
> That's my question: How is treated the problems that implicate networking in 
> GRASS? Have they to be solved only by means of combination of r.mapcalc
> instruccions in  UNIX scripts loops? Have they to accomplish any intermediate
> step? Do the misterious, for me, cost maps and this asociate functions play
> any role in this point? And, would the run times of the network functions 
> allow real time applications?
> 
> 
> 
> As a first example of this problematic and other similars, here you are a
> case:
> 
> We desire to improve the quality of our DEMs by means of incorporating 
> complementary information to the contour lines, in the sites file to
> interpolate. This new information is of points type (3d vertex points) and
> linear type (stream networks, ridges, etc), which have been digitized from th
e
> same maps as the contour lines. We want that this linear information 
> to be reflected in the created DEMs. The trouble is that vectorial
> information is 2D, it hasn't elevation. Here begin the problem:
>       The stream and contour line vector maps are converted to raster.
> >From this maps, by means of the map algebra (r.mapcalc), we create a new map
,
> in which all the cells of the stream map that have a value different of 0 in 
> the contour map take this value in the new map. Now we have a result map
> which have the elevation in the intersection of the streams with the contour
> lines. Later we have to fill in with elevation values the stream path in all
> its length. For this, in all the stream cells, we have to follow its path in
> the two senses, until the two nearest cells which have elevation value are
> found. The elevation value of the target cell is equal to the lineal 
> interpolation along the river path of this two nearest cells with elevation 
> value. The problem is how to find from a stream cell the two nearest cell 
> with elevation value following the stream path in both senses. Could it 
> be made with neighbouring loops? With networks functions? 
> 
>       Thanks in advance.
>                                       Evaristo.
> 
> ------------------------------------------------------------------------
>        __                                                      __
>       / /          Evaristo QUIROGA RAIMUNDEZ                 / /
>   ___/ /                                                  ___/ /
>  /    /                                                  /    /
> /    /          UNIVERSIDAD AUTONOMA DE BARCELONA       /    /
> \__/\\    Gabinete de Analisis Territorial y Ambiental  \__/\\
>      \\__ Unidad de Geodinamica Externa y Hidrogeologia      \\__
>      /  /\                 Edificio CS                       /  /\
>      \ / /\       08193 - Bellaterra (BARCELONA)             \ / /\
>       \ /  \               SPAIN - EU                         \ /  \
>        \___/        email: evaristo at cataw.uab.es               \___/
> ------------------------------------------------------------------------
>