[GRASS-user] create basins of similar area

[Margherita Di Leo]

Hi Martin,

On Fri, Apr 26, 2019 at 3:35 PM Martin Landa <landa.martin at gmail.com> wrote:

> Hi,
>
> čt 25. 4. 2019 v 20:14 odesílatel Ken Mankoff <mankoff at gmail.com> napsal:
> > Maybe make basins with threshold an order of magnitude smaller. Then
> merge small areas to get your 4-5 km2 areas. Still not totally sure how
> this makes sense hydrologically.
>
> right, it doesn't make sense probably from hydrologic point of view.
> Let me explain reasoning. My colleagues are working on developing
> distributed event-based model for surface and subsurface runoff and
> erosion. We are working on experimental version which enables
> computation parallelization. In initial phase AOI is split into tiles.
> Instead of regular grid of tiles a basin-based tiles are used. This is
> reason why basins (defining tiles) should be of similar area size.
> Martin
>
>
I understand this is case study involves a vast area and editing basins
manually (merging adjacent basins) is out of the question, isn't it? If
this is the case, I would propose to prototype a procedure based on similar
geomorphological characteristics to determine the 'optimal' threshold for
clusters, being optimal the threshold that gives you the desired (range of)
basin size. Just thinking out loud here. The threshold is the cutoff of
upslope area draining in the outlet cell. For a fixed threshold values, you
will obtain basins that are similar in size in areas that have similar
geomorphological characteristics, e.g. slope. I would cluster areas with
similar range of slope values and calibrate a suitable threshold that gives
the desired basin area for each cluster. This looks like a lot of
prototyping work, however, if you make a first run with a tentative
threshold value, display the resulting basins and visually compare with
slope map you should get a rough idea of what to do with the slope ranges
already.

Hope this helps,


-- 
Margherita Di Leo
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/grass-user/attachments/20190426/22f6f58c/attachment.html>