|
|
| GeoCommunity Mailing List |
| |
| Mailing List Archives |
| Subject: | Re: [gislist] GRID Tables |
| Date: |
11/10/2005 03:55:01 AM |
| From: |
Peter Shary |
|
|
I am working with elevation grids solving similar tasks at my PC.
A simple solution is to sequentally read one line of ELEV, one line
of LANDCOVER, calculate "on-the-fly" ASPECT and SLOPE
(not storing them on your disc), and use these to compute FINAL.
I use a new extended system of 18 basic topographic attributes and
other environmental data (like your LANDCOVER) to produce my
FINALs (for soil, vegetation, gold, oil spills,...), but I store on my
disc only regression coefficients and names of used topographic
attributes, any FINAL grid can be fastly calculated at any time later
using these coefficients and names. Few grids of topographic
attributes computed from an ELEV grid are stored on my disc,
only in the case they are non-local and may need some time for
their calculation. This way I treated very large grids, like GTOPO30
(at 900m resolution) for Eurasia, or joined and re-projected 20 and
more SRTM tiles (of 30 or 90m resolution).
By the way, are you sure that ELEV, ASPECT and SLOPE are
effective topographic attributes? - I've checken this for numerous
landscapes, and my conclusion is that this is not so. Additionally,
ASPECT needs circular statistics (0 and 360 degrees mean the same).
So, if you wish to describe thermal regime of slopes, insolation
is better than ASPECT, because it is 100% for perpendicular to land
surface solar irradiation (0 for shady sides of slopes), and its formula
is very simple:
http://members.fortunecity.com/eco4/research_shary/id3.html
If you wish to describe hydrological terrain features, up-slope
area would be better choice (it predicts all rivers and dry valleys
using only ELEV grid), and so far.
To calculate FINALs using multiple regression, I use only 3 of
basic topographic attributes, but these 3 ones are automatically
selected from total 18 basic ones using the criterium of highest
correlation.
Finally, if you wish to see all your topographic attributes at given
point, the solution is to read a small subgrid of ELEV and calculate
them, this is done immediately. Of course, to see your ASPECT or
SLOPE map images, you must first calculate them, but this is a fast
procedure, and your software might delete these grids when you exit
your program. These solutions were implemented in my software,
Analytical GIS Eco ( http://come.to/eco4 ), its full-featured DEMO
with four terrain examples is free and is available from me via email.
Hope this helps.
Cheers,
Peter
------- -------
Peter A. Shary, PhD,
scientific researcher and GIS developer
142290 Poushchino Moscow region
Institutskaya street, bldg.2
Institute of physical, chemical and biological problems
of the Russian Academy of Sciencies
Russian Federation
Phone: 007 0967 733604
URL (software): http://members.fortunecity.com/eco4/giseco/
URL (research): http://members.fortunecity.com/eco4/research_shary/
------- -------
-----Original Message-----
From: Vassil Vassilev <vasil_gis@yahoo.com>
To: gislist@lists.thinkburst.com
Date: Mon, 7 Nov 2005 05:14:38 -0800 (PST)
Subject: [gislist] GRID Tables
>
> Dear List,
>
> I am making a model for the tortoise distribution
> which takes into account elevation, aspect, slope and
> land cover. For all of these parameters I prepared a
> separate grid which represent a code for each
> variable. Now I have to make a final grid with indexes
> which should be based on a table in which the four
> variables are described, for example if ELEV is ... if
> ASPECT is ... if SLOPE is ... if LANDCOVER is ... than
> FINAL is ... etc. there are 88 different combinations.
>
> I can do this by two or three ways but they are quite
> computer resources consuming.
>
> I will highly appreciate your suggestions on the
> topic.
>
> Thank in advance,
>
> Vassil
>
>
>
>
>
>
> __________________________________
> Yahoo! Mail - PC Magazine Editors' Choice 2005
> http://mail.yahoo.com
> _______________________________________________
> gislist mailing list
> gislist@lists.geocomm.com
> http://lists.geocomm.com/mailman/listinfo/gislist
>
> _________________________________
> This list is brought to you by
> The GeoCommunity
> http://www.geocomm.com/
>
_______________________________________________
gislist mailing list
gislist@lists.geocomm.com
http://lists.geocomm.com/mailman/listinfo/gislist
_________________________________
This list is brought to you by
The GeoCommunity
http://www.geocomm.com/
|
|
Sponsored by:
For information
regarding
advertising rates
Click Here!
|
