Why no need for “fill-sinks” when using r.watershed in GRASS QGIS?

Question

I've seen a few places that when using the r.watershed function in GRASS QGIS, you do not need to use the "fill-sinks" procedure on the DEM before running the function.

According to the documentation: "Option elevation specifies the elevation data on which entire analysis is based. NULL (nodata) cells are ignored, zero and negative values are valid elevation data. Gaps in the elevation map that are located within the area of interest must be filled beforehand, e.g. with r.fillnulls, to avoid distortions. The elevation map need not be sink-filled because the module uses a least-cost algorithm."

How does the least-cost algorithm work, and why is it not necessary to perform a "fill-sinks" procedure when using it?

Answer 1

The GRASS algorithm in r.watershed does not need a filled DEM because it implements multi flow direction (MFD) that allows the procedure to "look ahead" recursively to find where the stream is going, and to route the stream through a sink in a reasonable way. It does not "simulate" filling.

THere's a nice example in this paper. See Ssection 3.2 and Figure 7.

Answer 2

Interesting question, I have not applied that process, but I have read some publications of Mitasova, especially those on simulation. I assume that the same principle applies.

I understand that several processes do not require filling of sinks because the algorithm "simulates" filling during flow and continuous. Imagine that you are filling the sump with water when you reach the top and the flow continues on its way.

A problem of similar algorithms in other programs that when the flow stops they do not find a solution.

An interesting aspect, would be to evaluate how the filling or not of the sinks affects the result, because it uses the route of the minimum cost, and a sink, I suppose, implies a greater cost