[UW-GIS-L] watershed question

[Luke Rogers]

Try looking up SETCELL and SETWINDOW in GRID and make sure all your grids
have the same spatial extent and cell size as your DEM or FLOWDIRECTION
GRID. Do the vector > grid conversion using GRID or ARC and not ArcGIS.

-Luke

******************************************************
Luke Rogers, Rural Technology Initiative
Senior Geographic Information Research Scientist
Box 352100 | Seattle WA, 98195 | Tel:(206)543-7418
lwrogers at u.washington.edu | http://www.ruraltech.org 
University of Washington - College of Forest Resources
******************************************************


________________________________

From: uw-gis-l-bounces at mailman1.u.washington.edu
[mailto:uw-gis-l-bounces at mailman1.u.washington.edu] On Behalf Of raytimm
Sent: Thursday, January 05, 2006 12:23 PM
To: 'UW - GIS Discussion & Support'
Subject: [UW-GIS-L] watershed question



Hi All,

I have a problem that has been vexing me all morning and I'm sure (hopeful)
there is a simple solution.  

I'm trying to calculate "watershed" for a set of points that have been
snapped to a stream network - all good so far.

I failed several times to successfully calculate watersheds in both ArcGIS
and ArcCatalog.  So I went back to the old standby, Workstation GRID - which
requires that the source data are all in grids - still good so far.

 

Here's my problem - or one of them:

My pourpoint grid does not lie perfectly on my stream network grid even
though my pour point coverage does lie exactly on my stream network coverage
(they were snapped together originally).  When I look at my pout points,
they look to be perfectly centered within stream network grid cells.  But,
my pour point grid cells are apparently randomly incoherent with my stream
network grid.  They may perfectly overlay, they may partially overlay, they
may be adjacent but completely miss.  So, when I calculate watershed, my
contributing upslope areas are not all associated with the stream grid cell
locations.  

 

BTW, my stream network grid does lie at the intersection of flow direction
patches.  So, that looks good.  

 

What I suspect is happening somewhere, somehow, is that there is some
embedded geographic difference between the pourpoints and the stream network
that isn't apparent because they were snapped together.  But, when I convert
to a grid, the "true" location of the pour points is used.  Also, some of
the pourpoints just seem to be missed in the calculation.  Is there a limit
to the number of points you can use in this kind of calculation?  I don't
get any weird errors - just weird results.

 

Thanks for your help.

 

Raymond Timm

UW School of Aquatic and Fishery Sciences

Box 355020

1122 NE Boat Street
Seattle, WA 98195-5020

206-221-5403

206-685-7471 (fax)