Hello, Sorry if this crossposts. I received many replies concerning my original post which is below. I would like to thank Terri Gorham, Richard Duncan, Sarah North, David Lounsbury, Jared MacLachlan, Matt Wilkie, Neeraj Singh, Joe Miller, Conrad Heatwole, Bernard Stroebl, Kevin Lawton, Fridjof Schmidt, Sean Vaughn, and Ivan Petras for replying. The main suggestions for the problem of very small watershed delineation were to fill the DEM first and check the search tolerance for the SnapPourPoint request. I had filled the DEM so that was not the problem. The second suggestion was the problem. I had made a mistake in the search tolerance and it was too small. By increasing it to about two cell sizes (flow accumulation) it works much better. The post which I had read awhile back was by Ivan Petras and he sent me the post (entitled TIP - DEM and Hydrological Analysis). It was helpful. I don't have the link to it on the internet and it is pretty long so I have not tacked it on here. But I have it as a word document so if anyone would like it email me. Other suggested links: http://www.crwr.utexas.edu/gis/gishyd98/GisHyd98.htm http://www.wiley.com/Corporate/Website/Objects/Products/0,9049,38388,00.html http://www.ce.utexas.edu/prof/maidment/giswr98/ex298/prepro.htm http://www.gishydro.umd.edu/ http://gis.esri.com/arcscripts/details.cfm?CFGRIDKEY=951497255 http://cres.anu.edu.au/outputs/anudem.html http://gis.esri.com.arcscripts/scripts.cfm Original Post: Hello, I have set up a watershed delineation tool in a project which uses the Watershed request to delineate watersheds based on the flow direction. The script is somewhat based on the example that is given in the ArcView help with the Watershed request. My question is about why sometimes when I use the delineation tool it seems to delineate a quite reasonable watershed or catchment while other times it only delineates a few cells. I believe it has to do with the DEM and the fact that in the flow accumulation grid that was developed from the DEM there are numerous cells that have 0 flow accumulation values. These cells are not just the ridges that define the catchment but sporadically throughout the grid. I was wondering if anyone knows of any good references on the web or elsewhere that discuss this problem. I remember once reading a long post to one of these lists about issues dealing with watershed delineation but I can't find it again. Thanks, John DeGroote GIS and Data Services The Macaulay Institute Craigiebuckler Aberdeen Scotland UK AB15 8QH Tel: +(44) (0) 1224 498200 Fax: +(44) (0) 1224 311556 email: j.degroote@macaulay.ac.uk http://macaulay.ac.uk