[MDAL-Developer] Volume intrepretation

[Paul Harwood]

Hi Martin and thanks for responding.

I should have said that I had studied that doc - and it does not actually
answer my question. I also looked at the UGRID CF definitions but they are
deliberately vague about this subject as well - suggesting two possible
interpretations.

This problem only really comes to light when you start thinking about
creating a 3D visualisation of the data.

The problem is that the doc has two interpretations, which are not
compatible and neither is properly represented by the data - viz:

[image: v2.png]  [image: v1.png]

I have chosen to interpret the data set as something more like this

[image: v3.png]

I think that is correct but it is not clear from the docs and seems a
little strange as a data structure

On Tue, 24 Aug 2021 at 10:22, Martin Dobias <wonder.sk at gmail.com> wrote:

> Hi Paul
>
> Probably the best description how the vertical levels in meshes are meant
> to work is this QGIS enhancement proposal:
> https://github.com/qgis/QGIS-Enhancement-Proposals/issues/158
>
> In short, within one face, if you have N vertical levels, there will be
> N-1 volumes in between those levels for that particular face. There are two
> important considerations that may not be obvious:
> 1. for each face, the vertical levels may be at different elevations/depths
> 2. for each face, there could be a different number of vertical levels
> MDAL_D_volumesCount() returns _total_ number of volumes, not per-face
> number of volumes. Based on all this, the total number of vertical level
> entries therefore is "number of volumes" + "number of faces" because for
> each face we need that one extra vertical level.
>
> Regards
> Martin
>
>
> On Mon, Aug 23, 2021 at 9:23 PM Paul Harwood <runette at gmail.com> wrote:
>
>> Hello all
>>
>> Another question for you. I am looking at an app that converts from
>> MDAL DataOnVolume dataset to a voxel-based representation.
>>
>> That should not be particularly difficult. However, it has made apparent
>> that it is not particularly clear what the "Vertical Levels" data in
>> MDAL actually means.
>>
>> My first understanding - that since the data values are volumetric the
>> "vertical levels" represent the centroid of the volume being measured - is
>> slightly undermined by the fact that there are (face count + volume count)
>> values in the array which is too many for that interpretation.
>>
>> The size of the array suggests that the interpretation should be one of
>> layered meshes and the volumes are the spaces between the layers. That is
>> fair enough but there is only one value per extruded face cell - when it is
>> clear that the extruded face cell is not necessarily parallel to anything
>> else. Is this vale then the vertical level of the centroid extruded face
>> cell?
>>
>> Does anyone have a definitive answer?
>> _______________________________________________
>> MDAL-Developer mailing list
>> MDAL-Developer at lists.osgeo.org
>> https://lists.osgeo.org/mailman/listinfo/mdal-developer
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/mdal-developer/attachments/20210824/fbe7ad0f/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: v2.png
Type: image/png
Size: 4006 bytes
Desc: not available
URL: <http://lists.osgeo.org/pipermail/mdal-developer/attachments/20210824/fbe7ad0f/attachment-0003.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: v1.png
Type: image/png
Size: 9304 bytes
Desc: not available
URL: <http://lists.osgeo.org/pipermail/mdal-developer/attachments/20210824/fbe7ad0f/attachment-0004.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: v3.png
Type: image/png
Size: 9054 bytes
Desc: not available
URL: <http://lists.osgeo.org/pipermail/mdal-developer/attachments/20210824/fbe7ad0f/attachment-0005.png>