[Qgis-user] Calculating affected regions of high-water situations

[Werner Macho]

Hi Andreas!
> I thought that this is a common use case in GIS - as many communities
> have to do the "Gefahrenkarten" - risk maps. But probably this work has
> been outsourced in most cases to real hydrology experts. Then these
> hydrology experts probably do not want to make the process too easy and
> available to the average GIS users - otherwise they would loose their
> jobs/contracts ;-(
It IS a use case for GIS - but in this case (I think i mentioned it
twice that I worked in this area for about 7 years) I have to say that I
would love to see an easy way to do the analysis but there is a lot of
knowledge about flowing processes necessary to 1. create the model
correct 2. run the hydraulic analysis the right way and 3. interpret the
outcome correct.
Ever since I started in this area long time ago I always wished there
would be a one-click tool but I think I have to disapoint you here.
Every river I started to analyse ("Gefahrenkarte, Risikozonen") was a
new experience and sometimes reacted completely different to any river I
have had before.
Beside the fact that you need deeper knowledge in hydrology to make the
correct input assumptions (for e.g. roughness.. so just having a DEM is
definitely not enough) you also need to have a very accurate model (2m
DEM is probably also not enough) to get correct results (especially for
smaller rivers 3-10m width).
>From my experience all I can say is that it is not enough to throw a DEM
into some analysis tools and expect an outcome with cm accuracy. In my
last work it was a toolchain of 5-8 programs to get the correct output
(including GIS). And even then it was sometimes not more than a rough
assumption. (You'll never know what the next rain will bring and what
influence all these newly built buildings will have to the riverstream).
I'd just say the process now is unfortunately far away from being easy.
But as I already said .. It always depends on what you are expecting. If
an accuracy of 1-2m is enough than it's probably really just a matter of
one click in GRASS. But 1-2m could mean death or severe damage in
populated areas.

So as a conclusion: You are surely able to do calculations with GIS (see
the GRASS modules) but that will give you just an overview where water
"might" be.
If you want "waterproof" results you have to go a long way and use GIS
in combination with other tools to be able to make real
riskzone-management-maps.

But you see some people are working to integrate such tools into a QGIS
chain (We'd like to call it aQuaGIS (c) by Pete Wells) maybe some time
we end up having a one-click tool but I don't expect that in the near
future.

If you are really interested in this topic we can talk about it at our
next HF in Valmiera or probably in Zurich at New years eve :)

kind regards
Werner