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

[Werner Macho]

On 12/10/2012 01:01 PM, Andreas Neumann wrote:
> Hi all,
>
> I am looking for tools for calculating the affected regions when a river
> level rises 1, 2 or 3m. I have DTM data (2m resolution) and cadastral
> data with the areas of the flowing water bodies at hand.
>
> Are there any tools that can help me get the affected raster cells in
> the DTM if the river level rises? With a lake, it would be a bit easier,
> but the river also has a slope ...
>
> I am ok with any OS GIS solutions (GRASS, QGIS, Taudem or Saga, etc.).
Hi Andreas,

That's the area I worked in for the last 7 years. To be honest (and you
already mentioned it) it's easier in GIS with a larger area (lake).
I expect you already know tools like r.terraflow in GRASS GIS (Probably
not - that would be the first point to look at)
Due to the nature of rivers it is not so easy to rise river levels
(depending on the width of the river).
There are also r.inund.fluv and r.hazard.flood so it seems with just GIS
you have to use GRASS here - though I don't know about the reliability
of this tools.

As I studied hydrology and hydraulics there are a lot of proprietary
tools out there that can calculate what you need (and a lot more) like
TUFLOW and Hydro_AS-2D (with Hydro_AS I used to work for the last 7
years). But I think that would be a bit too sophisticated for your
needs. In my work I needed tools that can do such predictions with an
accuracy on a cm base so probably all my coming answers are too much
work for your needs.

With Opensource it can maybe be done with Taudem or ANUGA but I did not
have the time until now to take a closer look at this opensource tools.
That would be the point where Pete Wells from Lutraconsulting could jump
in :).
The only tool I have experience with is (free) BASEMENT which is being
developed at the ETH Zurich but preparing the model and calculating it
with BASEMENT is nearly as complex as doing it with Hydro_AS (due to the
fact that both programs are using the same type of input models). But I
know that a group at the ETH is already working on a QGIS Plugin to make
it easier to create the input models for BASEMENT (As Pete already wrote
the Crayfish plugin to use QGIS as a Postprocessor for this Models). I
hope that ETH will release their plugin as GPLed (though I am not sure
now) and that work in this field will continue.

I already started a wiki page some time ago on this topic
(http://hub.qgis.org/wiki/quantum-gis/Hydrology_and_Hydraulic_modelling)
and it seems that a lot of people are either working or are at least
interested in this topic.

As I still believe that you need something more simple for your solution
I hope that probably someone else can jump in here and explain an easy
way to just do sinkfilling in small rivers (maybe r.hazard.flood is
enough for you?) but personally I have not seen such an (easy) tool. But
as already said, it depens on how reliable you would like to have the
results. If it exists I would also be interested howto use it.

For the other programs you can always come and ask me howto do it but be
warned that creating such an input model is really time consuming.

Hope that answers at least some parts of your question. (And I know that
you'll surely find a lot more on the given WIKI Page - thanks to all
contributors to the page!)

kind regards
Werner