Conic projection, but centric (gnomonic)

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
7 messages Options
Reply | Threaded
Open this post in threaded view
|

Conic projection, but centric (gnomonic)

Lukasz Komsta
Dear All,

In late 1970s, Institute of Botany of Jagiellonian University in Krakow (Poland) developed a geobotanical grid. This grid has interesting history, because the details of projection were never officially published. After my reverse engineering followed by further numerical investigation by my colleague it was revealed that this grid is based on something which is called in mapproj package "central conic projection" with one tangent (with quite strange Earth radius, but it is only a detail). The projection is from Earth center, not parallel nor stereographic.

I implemented this grid in various languages and the detailed info is here: https://atpol.sourceforge.io/.

The creators of original grid officially confirmed that this projection is same with their original idea and this caused rising of interest in usage.

However, there is no possibility to set this projection in PROJ4 (as far as i know), which results in incompatibility of many GIS systems with this botanical grid.

The closest solution is

+proj=eqdc +lat_1=52 +lat_2=52 +lat_0=52 +lon_0=19 +axis=esu +a=6390000
+b=6390000 +ellps=sphere +x_0=330000 +y_0=-350000

...but this results in quite significant inaccuracy, as EQDC is not gnomonic (central) projection.

So, my questions are:

1. Is there a possibility to achieve desired projection using PROJ4 ?
2. If not, are there any possibility to implement such a functionality in the future to allow GIS users using this grid?

Thank you very much in advance

Lukasz Komsta

_______________________________________________
Proj mailing list
[hidden email]
http://lists.maptools.org/mailman/listinfo/proj
Reply | Threaded
Open this post in threaded view
|

Re: Conic projection, but centric (gnomonic)

Kristian Evers-2
Hi Lukasz,

> 1. Is there a possibility to achieve desired projection using PROJ4 ?

You can probably answer this better than anyone else, and it would seem you have concluded that it can't.

> 2. If not, are there any possibility to implement such a functionality in the future to allow GIS users using this grid?

You are welcome to submit a pull request on GitHub [0] with the necessary code additions and documentation. If you need help getting started please let us know.

Best regards,
Kristian

[0] https://github.com/OSGeo/proj.4

-----Oprindelig meddelelse-----
Fra: [hidden email] [mailto:[hidden email]] På vegne af Lukasz Komsta
Sendt: 8. november 2017 11:08
Til: [hidden email]
Emne: [Proj] Conic projection, but centric (gnomonic)

Dear All,

In late 1970s, Institute of Botany of Jagiellonian University in Krakow (Poland) developed a geobotanical grid. This grid has interesting history, because the details of projection were never officially published. After my reverse engineering followed by further numerical investigation by my colleague it was revealed that this grid is based on something which is called in mapproj package "central conic projection" with one tangent (with quite strange Earth radius, but it is only a detail). The projection is from Earth center, not parallel nor stereographic.

I implemented this grid in various languages and the detailed info is here: https://atpol.sourceforge.io/.

The creators of original grid officially confirmed that this projection is same with their original idea and this caused rising of interest in usage.

However, there is no possibility to set this projection in PROJ4 (as far as i know), which results in incompatibility of many GIS systems with this botanical grid.

The closest solution is

+proj=eqdc +lat_1=52 +lat_2=52 +lat_0=52 +lon_0=19 +axis=esu +a=6390000
+b=6390000 +ellps=sphere +x_0=330000 +y_0=-350000

...but this results in quite significant inaccuracy, as EQDC is not gnomonic (central) projection.

So, my questions are:

1. Is there a possibility to achieve desired projection using PROJ4 ?
2. If not, are there any possibility to implement such a functionality in the future to allow GIS users using this grid?

Thank you very much in advance

Lukasz Komsta

_______________________________________________
Proj mailing list
[hidden email]
http://lists.maptools.org/mailman/listinfo/proj
_______________________________________________
Proj mailing list
[hidden email]
http://lists.maptools.org/mailman/listinfo/proj
Reply | Threaded
Open this post in threaded view
|

Re: Conic projection, but centric (gnomonic)

stevejking
In reply to this post by Lukasz Komsta
Lukasz Komsta wrote
> The closest solution is
>
> +proj=eqdc +lat_1=52 +lat_2=52 +lat_0=52 +lon_0=19 +axis=esu +a=6390000
> +b=6390000 +ellps=sphere +x_0=330000 +y_0=-350000

Lukasz, if you are trying to transform from this projection to wgs84 or GPS,
you will need to add a +towgs84 or +nadgrids component. I think developing a
nadgrids gridshift file to accompany this would be the best solution, since
that could give the greatest accuracy since it can take account of the
apparent distortions in the original grid (as one of your links indicated),
as well as accommodate the main ellipsoidal latitude and longitude shifts
from your sphere to wgs84.

Steve


Lukasz Komsta wrote

> Dear All,
>
> In late 1970s, Institute of Botany of Jagiellonian University in Krakow
> (Poland) developed a geobotanical grid. This grid has interesting history,
> because the details of projection were never officially published. After
> my reverse engineering followed by further numerical investigation by my
> colleague it was revealed that this grid is based on something which is
> called in mapproj package "central conic projection" with one tangent
> (with quite strange Earth radius, but it is only a detail). The projection
> is from Earth center, not parallel nor stereographic.
>
> I implemented this grid in various languages and the detailed info is
> here: https://atpol.sourceforge.io/.
>
> The creators of original grid officially confirmed that this projection is
> same with their original idea and this caused rising of interest in usage.
>
> However, there is no possibility to set this projection in PROJ4 (as far
> as i know), which results in incompatibility of many GIS systems with this
> botanical grid.
>
> The closest solution is
>
> +proj=eqdc +lat_1=52 +lat_2=52 +lat_0=52 +lon_0=19 +axis=esu +a=6390000
> +b=6390000 +ellps=sphere +x_0=330000 +y_0=-350000
>
> ...but this results in quite significant inaccuracy, as EQDC is not
> gnomonic (central) projection.
>
> So, my questions are:
>
> 1. Is there a possibility to achieve desired projection using PROJ4 ?
> 2. If not, are there any possibility to implement such a functionality in
> the future to allow GIS users using this grid?
>
> Thank you very much in advance
>
> Lukasz Komsta
>
> _______________________________________________
> Proj mailing list

> Proj@.maptools

> http://lists.maptools.org/mailman/listinfo/proj





--
Sent from: http://osgeo-org.1560.x6.nabble.com/PROJ-4-f3840930.html
_______________________________________________
Proj mailing list
[hidden email]
http://lists.maptools.org/mailman/listinfo/proj
Reply | Threaded
Open this post in threaded view
|

Re: Conic projection, but centric (gnomonic)

Lukasz Komsta
Dear Steve,

> Lukasz, if you are trying to transform from this projection to wgs84 or GPS,
> you will need to add a +towgs84 or +nadgrids component. I think developing a
> nadgrids gridshift file to accompany this would be the best solution, since

Thank you, in my version of cs2cs (and also online calculators), adding
towgs84 would change the result (but increasing error), whereas adding
nadgrids does not change anything at all (it seems that ellps=sphere
forces this behavior).

With regards,

Lukasz

_______________________________________________
Proj mailing list
[hidden email]
http://lists.maptools.org/mailman/listinfo/proj
Reply | Threaded
Open this post in threaded view
|

Re: Conic projection, but centric (gnomonic)

Lukasz Komsta
In reply to this post by Kristian Evers-2
W dniu 2017-11-08 o 15:17, Kristian Evers pisze:

Hi Kristian

>> 1. Is there a possibility to achieve desired projection using PROJ4 ?
>
> You can probably answer this better than anyone else, and it would seem you have concluded that it can't.

To be honest, my experience with PROJ4 and GIS is only basic. I have
browsed all documented projections and did not find such one, but I
asked here to be sure because maybe there is some hack in style "setting
projection xx with parameters yy will give the same results" and so on.

>> 2. If not, are there any possibility to implement such a functionality in the future to allow GIS users using this grid?
>
> You are welcome to submit a pull request on GitHub [0] with the necessary code additions and documentation. If you need help getting started please let us know.

Thank you, I will look at it and start to work.

With best regards

Lukasz

_______________________________________________
Proj mailing list
[hidden email]
http://lists.maptools.org/mailman/listinfo/proj
Reply | Threaded
Open this post in threaded view
|

Re: Conic projection, but centric (gnomonic)

Andre Joost
In reply to this post by Lukasz Komsta
Am 10.11.2017 um 15:07 schrieb Lukasz Komsta:

> Dear Steve,
>
>> Lukasz, if you are trying to transform from this projection to wgs84 or GPS,
>> you will need to add a +towgs84 or +nadgrids component. I think developing a
>> nadgrids gridshift file to accompany this would be the best solution, since
>
> Thank you, in my version of cs2cs (and also online calculators), adding
> towgs84 would change the result (but increasing error), whereas adding
> nadgrids does not change anything at all (it seems that ellps=sphere
> forces this behavior).
>

+nadgrids expects a gridf shift file which you have to create on your
own. It shifts the unprojected coordinates (i.e. degrees), regardless of
the projection. I doubt this is of any use for you.

HTH,
André Joost


_______________________________________________
Proj mailing list
[hidden email]
http://lists.maptools.org/mailman/listinfo/proj
Reply | Threaded
Open this post in threaded view
|

Re: Conic projection, but centric (gnomonic)

Lukasz Komsta
In reply to this post by Kristian Evers-2
W dniu 2017-11-08 o 15:17, Kristian Evers pisze:

>> 2. If not, are there any possibility to implement such a functionality in the future to allow GIS users using this grid?
>
> You are welcome to submit a pull request on GitHub [0] with the necessary code additions and documentation. If you need help getting started please let us know.

Dear Kristian,

Just implemented this functionality in my own fork (Github username
luqqe), tested it carefully and then created a pull request.

Thank you and best regards,

Lukasz

_______________________________________________
Proj mailing list
[hidden email]
http://lists.maptools.org/mailman/listinfo/proj