Compatibility of D8 flow directions with TauDEM's

[PaulRGit]

Hi!

I come up with another issue. I am currently trying to move from a full TauDEM workflow to a full python TopoToolbox one. However I am lacking some functionalities of TauDEM in the toolbox, namely:

• "streamnet" catchments delineations (topotoolbox does not divide a whole catchment into small ones, or it should be done manually one sub-catchment at a time which is in my opinion a bit painfull)
• "streamnet" meta outputs such as stream-wise strahler order or a stream dependence tree
• "dinfdistdown" function to compute Height Above Nearest Drainage raster

I have therefore chosen a mixed workflow, by delegating computation-heavy tasks to the TopoToolbox, such as flow directions, and bindings the outputs to the above cited TauDEM functionalities.

However this does not work since the D8 algorithm used by TauDEM differs from the one used in the toolbox.

Is there any possibility to force the use of taudem algo?

I have been looking at the libtopotoolbox C code and I have found a single implementation of the D8 algorithm, so my guess is that it is currently not possible.
Furthermore I am guessing that what makes the toolbox so efficient is the actually the choices made for the D8 algorithm, so I suppose it would be a nonsense to go "back" to the TauDEM one.

I would be pleased to get your insights on the feasibility of my project!

Paul

[wkearn]

You are correct that there is essentially one flow routing algorithm currently available in pytopotoolbox, which is the default D8 with morphological carving algorithm available in the MATLAB version of TopoToolbox. The core D8 algorithm should be more or less the same as what we do without the hydrological conditioning, and we should be able to more clearly expose that for users who already have a conditioned DEM.