Right way to analyze lage area DEM

[xwq0371]

Thank you for the great tool in advance.Recently, I ran into this issue and I would like to get some suggestions.

Right now I am doing a analysis on a large area of dem, like 4km x 4km, I find out that there could be multiple watershed thus multiple outlet points is generated. I am wondering if applying analysis to each watershed with different outlet point,then patch them into one is the right way to go or not.Since I tried to just use one of the outlet points to do the analysis regarlessly and the result is not that differentiate from the separate analysis I mentioned above, for my study case, the losses is accpetable.

[lrntct]

Sorry for the late answer.
If you can separate the watersheds, it's the easiest way to do it, yes.
If they are in parallel, there are no real issues, the results should be the exact same.

If the watersheds are in series, you can set a type 4 boundary at the exit point, save the boundary flow results, and use the records as an input for the downstream watershed. Be aware that you will lose information between the two, and the results will be slightly different between a large watershed and two separate ones.

[xwq0371]

Thanks for the reply.I still have some questions about the boundary condition.

For type 2, I read on the google forum it has some bugs and need more test.Is it ready to use for version 25.8?

The reason I was thinking about use a type 2 is because I wish I can run the simulation without define every water outlet point. Since my study area is either in urban and some mountain area, I wish the water just flow out of the dem area automatically and don't show that much deep stacked water near the boarder.

[lrntct]

Type 2 should work properly in the last version. However, it is applied only at the very edge of the rectangular domain, not inside.