How Grivation affects the grid of the map

[martupp]

After geo-referencing a map, OOM finds that Grivation should be -1,5 dg.
I expected OOM to rotate the map clockwise 1,5 dg, but there was no change.
After adding the value for Magnetic Declination 5.04 dg, Grivations value changed to 3.54 dg.
After clicking OK, OOM asked to rotate the map counter-clockwise and suggested 5.04 dg.
As a result of this action my object on the map rotated 5.04 dg counter-clockwise in stead of the 3.54 dg that was to be expected and was also the value in Grivation.

In my opinion shows Grivation only how much the map should be rotated, but does nothing to rotate.
If so, then I would expect not a suggestion for 5.04 dg (value for Magnetic Declination) but rather a suggestion for the value in Grivation (3.54).

But I would prefer OOM to rotate the map as soon as it was geo-referenced (clockwise 1,5 dg) and in the next stage a new rotation (counter-clockwise 5.04 dg) for Magnetic Declination.

I attached screenshots of 3 Omap files in which every stage is visible. I seem not to be able to attache the Omap files here...But I can send them!

with regards.
Mart

Start with a basic map and a map/object *(Track) drawn from the center.






After Geo-referencing the map!






After added Magnetic declination!

I measured the angle. It was 5.04 dg!

[dg0yt]

You can attach ZIP files.

IMO The dialog has two use cases: Initial setup, and updates. Usually I do not need rotation during setup. For updates, the rotation must consider the difference between old and new.

[martupp]

Thanks dg0yt. Here are the maps.

Omap-files.zip

I made these 3 maps to see what happens with a map after it is being geo-referenced and corrected for magnetic declination. In 1_Local map with object.omap I drew a line (from the center) parallel with grid-north, so a map was created.
In 2_Geo-referenced map after grid-converse.omap I expected the angle-fault for grid-conversion should be corrected, but nothing changed in the map. In my opinion is grid-north still not True North.
In 3_Geo-referenced map after magnetic declination.omap the map has rotated 5.04 dg, to much (1,5 dg) to have grid-north lined up with Magnetic North. Or am I missing something?

[dg0yt]

From what I understand from your description that you first setup (and apply) a georeferencing with a declination of 0 dg (resulting in the second map), and then you update the declination to 5.04 dg. This means that the difference between the old value and the new value is 5.04 degree, and that is the rotation which is needed for correction -- assuming that the old value was correct. And this results in the third map.

If you want to move from an incorrect georeferencing to a correct one, the options are:

• either explicitly transform (rotate etc.) the objects to match the unchanged georeferencing values,
• or modify the georeferencing values ot match the unchanged objects.

After geo-referencing a map, OOM finds that Grivation should be -1,5 dg.
I expected OOM to rotate the map clockwise 1,5 dg, but there was no change.

Grivation isn't a primary factor here. We are hardly interested in a specific "grid" (north). We need to know the actual orientation of objects with regard to magnetic north (in those countries, where orienteering maps are oriented to magnetic north), and this is described in the declination.

We can ask online services for the declination for a specific location. But only the user can inform the program how to apply this on existing objects in the map.

[martupp]

Thanks again.
I believe I understand wat you wrote and I agree that the map should be orientated to magnetic north.

I just tried to understand how OOM handled the grid conversion. I did not find a good explanation on Internett and I wanted to see wat happens after OOM corrected the angle fault just for grid conversion. So I thought it would show me with the help of the 3 maps. So I got confused when the line-object from the first map did not rotate after the Grid Conversion....

I had understand that both the geo-refenced objects on the geo-refeferenced map and the georeferenced template would be lined up with the grid-north, which I expected to act as True-North with a angle-fault for Grid Conversion at the coordinates of the Reference-point.
Now I wonder: If OOM does not rotate the map and template, than the map will have a angle-fault to True North (and thus also to Magnetic North after rotating the map with the value for Magnetic Declination).
I cannot see a rotation for Grivation, but it somehow it must be done so for the georefenced objects on the geo-refeferenced map and the geo-referenced template, because otherwise the value for Magnetic Declination would rotate the map to much in my example.
My line/object (and eventually other objects) did not rotate by means of Grivation. As you wrote, these need to be corrected before rotating before adjusting for Magnetic Declination. That is something to think of when one wants to geo-reference an older map.

Still, I don´t quite understand how Grivation influences the total rotation of the map, which in my case should be 3.54 dg instead of 5.04 dg to line the grid up with Magnetic North.

I hope you can/will explain.

with regards.
Mart

[martupp]

Can it be that OOM does not rotate the grid for grid convergence, but projects the geo-referenced data (imported objects and geo-referenced templates) at an angle ( Grivation ) to the grid?

[pkturner]

I had understand that both the geo-referenced objects on the geo-refeferenced map and the georeferenced template would be lined up with the grid-north, which I expected to act as True-North with a angle-fault for Grid Conversion at the coordinates of the Reference-point.

In Mapper (OOM) the positions of the objects on the map are primary, rather than their positions on the grid or their latitude/longitude.

When you change from -local- to UTM georeferencing, Mapper does not understand the relationship between the local grid and the UTM grid. It has no basis to move the map objects, and so leaves their map coordinates unchanged. Mapper leaves the declination unchanged, which is reasonable.

how Grivation influences the total rotation of the map, which in my case should be 3.54 dg instead of 5.04 dg to line the grid up with Magnetic North.

Mapper rarely rotates the map. It uses grivation to calculate the grid coordinates of objects. And also to display the grid - try the "Show grid" button, aligning the visible grid with "grid north".

Can it be that OOM does not rotate the grid for grid convergence, but projects the geo-referenced data (imported objects and geo-referenced templates) at an angle ( Grivation ) to the grid?

Yes. See #2190.

[martupp]

Hi pkturner.
Thanks a lot
Now things are beginning to be more clearly for me. I´ll have to re-read the other case a few times to get the picture really clear. As you wrote, it would be nice if someone would write down how the mechanisme works. I am doing that for myself now and for those who want to read, but I am afraid it would not be good enough for the help-files. I am satisfied with the answers i got from you and dg0yt. Now I am thinking about how to close this case with a comment that answers the subject of it. I am afraid to write something which is not correct, so if you have a suggestion for the comment...

[dg0yt]

NB we don't think/speak about templates in singular ("the georeferenced template"), but in plural. With multiple templates, there may be multiple grids (actual CRS).

There is the single map georeferencing. It establishes a geospatial connection between the drawing and the real objects. And there is one implicit, built-in assumption (based on IOF standards): magnetic north is "up". On paper as on screen.

[pkturner]

Now I am thinking about how to close this case with a comment that answers the subject of it. ..., so if you have a suggestion for the comment...

The subject refers to your expectation that, when you changed the georeferencing declination, the map would rotate by the grivation angle. You could edit the title, if that would clarify what you put in your closing comment.

If you would like to add to the help file, please take a look at my new PR #2501 which expands the Georeferencing help. It needs some editing, and suggestions are welcome. I will see whether I can send you a readable file.

[pkturner]

I put an HTML file version of expanded Georeferencing help into a comment for PR #2501, seeking edits or further comments.

[martupp]

Hi Scott. Thanks for your suggestion and the link to the help file. I was occupied with something else but I´ll have a look at the file. I started to make a document myself in which I will describe in (i hope) simple words how grid-convergence and magnetic declination affects our map. This should explain most for those without a geographic education. Maybe it can be useful as a help file. I´ll let you know when it is ready for reviewing.
For this moment I only change the title of this case and come back to it later.