[Qgis-user] raster layer combination

[chris hermansen]

Milena and list,

On Sat, Mar 27, 2021 at 11:20 AM Milena Muehler <milena.muehler at gmx.de>
wrote:

> Dear community,
>
> as an absolute beginner I try to explain my issue and intension.
>
> So I have several raster layers (population density; livestock density;
> mammal density; etc.) and two text layers (so I've imported two csv files
> with longitudes and latitudes into QGIS and now I have points about the
> location of large airports and location points where there was a zoonotic
> infectious disease outbreak).
>

Please be aware that it is difficult or impossible to defend the claim that
some data (like human population density for example) has any kind of
Gaussian-like spatial distribution.  With data like this, turning it into a
raster and thinking of the interpolated values as suitable for combining
with other supposedly continuous data may be very difficult to justify.

This conversation
https://www.researchgate.net/post/Which-are-the-sufficient-conditions-for-Kriging-estimation
has a nice compact discussion of the kinds of assumptions that underlie the
application of kriging as a spatial interpolation technique.

>
> My intention is to „combine“ all layers and to get all the values under
> this specific point of an outbreak. So that I have the exact value of the
> population density, livestock density ans so on where the outbreak
> occurred. I guess I have to make the calculation within a certain vicinity
> though.
>

You should probably first carry out some spatial statistics on the data
that resulted in your  raster layers to make sure any "combination" of the
layers actually explains something.

Oops I see Nicolas Cadieux also had something to say about this.



-- 
Chris Hermansen · clhermansen "at" gmail "dot" com

C'est ma façon de parler.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/qgis-user/attachments/20210327/864374c1/attachment.html>