I'm new to GIS and trying to find the area comprising the catchment for municipal stormwater infrastructure, like culverts. I'm getting incredibly small catchment areas, which by visual inspection aren't correct. In the picture below, my pour point is in the top left black area, and the background is a DEM. By clicking around in the DEM, I know that some cells adjacent to the black cells have higher elevation values, and I can't see to figure out why they're not being picked up.
What I'm looking for is to start at some point, and have a tool search for all cells adjacent with higher elevation.s If the cell adjacent has higher elevation, it gets added to the output. I thought this is what "Watershed" does, but I could be mistakes. I've tried different DEMs for the area, checking my extents, filling sinks, etc. These aren't necessarily in high flow areas corresponding to streams, so I'm thinking that might be part of it, but I'm not 100% sure.
If you fill sinks, the areas in small catchments for individual drainage inlets (catch basins) will be filled in and you will loose the definition. However, if you don't fill sinks you could have local sinks inside the small catchments that would interfere. Have you burned pipes as streams? Burning pipes as streams will avoid filling these local sinks. One problem with burning in the pipes as streams, though, is you do loose definition of the "interior" small catchments that are served by an individual catch basin.
If there is a way to limit how much the sinks are filled, then you might be able to find a happy medium where you fill the minor sinks just enough to make the watershed delineation work for interior small catchments to catch basins.
I've use Arc Hydro for larger watersheds. My process is to burn the pipes into the DEM prior to filling sinks. Then the flow direction is done. After that, other steps such as flow accumulation are used to define catchments. These processes don't define a particular watershed. You use batch points or another method to delineate those.
I think there is now a more complex Arc Hydro process that allows you to define points for catch basins that remain as sinks and then you use another grid or polyline dataset that defines the subsurface pipe system. I would think this requires accurate storm drain pipe and inlet datasets. I'd love to have time to learn this, but it is more refined that I have time (or good data) for.
For Arc Hydro, see possible solutions in this thread: https://community.esri.com/t5/water-resources-questions/archydro-problem-solvers/m-p/499200#M2431