Arc Hydro Fill Depression

[ArminSchulz]

Hi esri Community, 

Thanks to Arc Hydro I could identify my depressions in my DEM. Now I would like to fill my depressions with a given volume depending on my DrainArea, Precipitation event and the characteristic of my DrainArea which I can calculate on base of the attribute table of the calculated depression layer. 

Do you have an idea how I can fill my depressions with these volumes to get than a raster layer with the water depths?

thanks for consideration!

[DeanDjokic]

Not completely sure what you are trying to accomplish.  In a "mechanical" way, once you have identified which depressions you want to fill, you can copy all OTHER sink polygons into a separate feature class (these are the "deranged polygons" that you do NOT want to fill) and then use Arc Hydro "Fill Sinks" function and provide the deranged polygon feature class as one of the inputs.  All other depressions will be filled.  That will give you the "hydro DEM" that you can compare with the original DEM and get specific statistics.  Also, if you look at the "Sink Poly" feature class resulting from the "Sink Evaluation" function you can see some metrics (FillArea, FillVolume, FillDepth) that might help you with your sink selection.

A more comprehensive discussion is provided in Arc Hydro document "Arc Hydro - Identifying and Managing Sinks.pdf" https://community.esri.com/t5/water-resources-documents/arc-hydro-identifying-and-managing-sinks-pdf... .

[ArminSchulz]

Thank you Dean Djokic for the quick answer.

I'm sorry I couldn't explicate my question understandable, so I try it again to make my question hopefully comprehensible.

From my DEM I could identify all depressions (Depression Evaluation) as polygon layer with all necessary information like FillArea, FillVolume, DrainArea etc. . I also calculated a raster layer with the depth of these depressions by subtracting the DEM from the filled DEM. With the raster layer I can now represent the depths of each depression. 

The FillVolume, which I got from the Depression Evaluation, is the maximum volume/the potential retention volume of each depression. So if the rainfall rate on the drainage area is less than the potential retention, the water can be stored in the depression. If the rainfall rate on the drainage area is higher than the potential retention, the water will overflow the depression. 

Now I want to do it the other way around. I have a rainfall event with a rainfall rate of 60 mm/h. With this rainfall rate, the runoff coefficient and the drainage area I can calculate a volume for each depression which will be filled. On base of this rainfall event/volume I would like to get a new polygon layer for each depression with the area of the depression which got filled. This is what I want to get..

I hope it is now understandable.

Thank you again

[DeanDjokic · Accepted solution]

Thanks for the problem elaboration. As always, there are many ways to do it, but I think the easiest would be to use "Drainage Area Characterization" Arc Hydro tool (in Terrain Morphology -> Drainage Boundary Processing toolset). This tool "slices" through each drainage area from its bottom up and calculates elevation-area-volume (EAV) curves and stores that in the resulting EAV table.
So your workflow is then to calculate for each drainage area the water volume coming in based on the P-E-I (precip, evaporation, infiltration), from that volume look up the depth from EAV table and add that to the minimal elevation in each drainage area. That gives you the "water surface elevation" (WSE) in each drainage area after being "filled" with water. Calculate WSE-DEM and where positive you have "flood" (ignore the rest). Convert that into a polygon and that is your result.

[ArminSchulz]

alla grande!

Thank you Dean Djokic, it sounds like a very good solution. I already figured out a similar solution routine which needed to be programmed, but in this way it should be much easier. 

[SergeyMyagkov]

Hello. I have the same question. If you solved this problem, could you explain how it was solved?