Opened 5 years ago

Closed 3 months ago

#92 closed enhancement (fixed)

Add oblique mercator projection

Reported by: mcginnis Owned by: davidhassell
Priority: medium Milestone:
Component: cf-conventions Version:
Keywords: Cc:

Description

The Oblique Mercator projection is used by at least one regional climate model, RegCM3, which is part of the NARCCAP climate modeling program. Currently we record its map projection information using the transverse_mercator projection, which I have learned is very similar but not quite the same. I propose to add this map projection so we can get it right.

Proposed text:


Oblique Mercator

grid_mapping_name = oblique_mercator

Map parameters:

  • azimuth
  • latitude_of_projection_origin
  • longitude_of_projection_origin
  • scale_factor_at_projection_origin
  • false_easting
  • false_northing

Map coordinates:

The x (abscissa) and y (ordinate) rectangular coordinates are identified by the standard_name attribute value projection_x_coordinate and projection_y_coordinate respectively.

Notes:

Notes on using the PROJ.4 software package for computing the mapping may be found at http://www.remotesensing.org/geotiff/proj_list/hotine_oblique_mercator.html . The Rotated Mercator projection is an Oblique Mercator projection with azimuth = +90.


If adding a new attribute for azimuth is problematic, this proposal could be modified to add the rotated_mercator projection instead, which is a special case of Oblique Mercator with azimuth = 90.

Note that apparently there is a subtle technical difference between an Oblique Mercator projection and a Hotine Oblique Mercator projection that depends on when the rectification from skew grid to map grid is applied. Since most mapping packages don't support a rectified grid angle parameter at all (effectively giving it a default value of 90 degrees, such that it has no effect), to avoid unnecessary proliferation of attributes I propose to omit this parameter and elide this distinction until such time as it proves necessary.

My knowledge of this topic is quite limited; I have made this proposal based on what understanding I have gleaned from the geotiff website and communications with colleagues working with our RegCM3 output in GIS. Commentary from experts would be very welcome.

Change History (14)

comment:1 Changed 5 years ago by jonathan

Dear Seth

Thanks for this proposal. I too am not an expert but it looks sensible to me, there is clearly a use-case for it, and I support it. It is fine to add an extra parameter of azimuth. That needs to be defined in Table F.1. Please could you propose a new draft entry for that table as well?

Best wishes

Jonathan

comment:2 follow-up: Changed 5 years ago by mcginnis

Jonathan--

Happy to.

--Seth

Azimuth entry for Table F.1:

Attribute: azimuth Type: N Description: Specifies a horizontal angle measured in degrees clockwise from North. Used by certain projections (e.g., Oblique Mercator) to define the orientation of the map projection relative to a reference direction.

comment:3 in reply to: ↑ 2 Changed 5 years ago by pbentley

Replying to mcginnis:

Hi Seth,

Azimuth entry for Table F.1:

Attribute: azimuth Type: N Description: Specifies a horizontal angle measured in degrees clockwise from North. Used by certain projections (e.g., Oblique Mercator) to define the orientation of the map projection relative to a reference direction.

Since there are potentially many different ways in which azimuth values are used in the CRS world, I'm wondering if a more specific attribute name might be worth considering here.

Searching the online EPSG registry of geodetic parameters it looks like the analogous coordinate parameter is called "azimuth of initial line", which might suggest an attribute name of azimuth_of_initial_line, for example. But I could well be wrong (it wouldn't be the first time ;-).

For information, the corresponding parameter description in the EPSG registry is:

"The azimuthal direction (north zero, east of north being positive) of the great circle which is the centre line of an oblique projection. The azimuth is given at the projection centre."

Regards,

Phil

comment:4 follow-up: Changed 5 years ago by mcginnis

I hadn't thought about other CRS uses for azimuth, so that makes sense. Thanks for the suggestion, Phil.

Searching around the web, I think that for this projection it's most commonly referred to as the "center line", so I will revise the entry for table F.1 as follows:

`Attribute: azimuth_of_center_line

Type: N

Description: Specifies the horizontal angle, measured in degrees clockwise from North, of the great circle used as the center line of the Oblique Mercator projection.`

--Seth

comment:5 Changed 5 years ago by caron

Hi Seth:

Do you have a sample file we could use for testing? thanks.

John

comment:6 Changed 5 years ago by mcginnis

Hi John,

I've placed a sample file here:

http://www.narccap.ucar.edu/temp/oblique/orog_RCM3.nc

--Seth

comment:7 in reply to: ↑ 4 Changed 5 years ago by pbentley

Replying to mcginnis:

`Attribute: azimuth_of_center_line

With an alias of azimuth_of_centre_line, presumably, for us old-world users (only kidding :-)

Although, thinking about it, as an alternative you might want to mirror some of the existing grid mapping attributes (which use 'central') and go with azimuth_of_central_line.

Phil

comment:8 follow-up: Changed 4 years ago by mcginnis

Hi Phil,

Google says usage of "central line" vs "center line" is about 50/50, so if it'll make it more consistent with other grid mappings, let's go with azimuth_of_central_line.

Thanks for the suggestion!

Cheers,

--Seth

comment:9 Changed 4 years ago by jonathan

Dear Seth

Thanks for investigating that. Does anyone have any more comments on this proposal as it now stands?

Cheers

Jonathan

comment:10 in reply to: ↑ 8 Changed 4 years ago by pbentley

Hi Seth,

Replying to mcginnis:

Google says usage of "central line" vs "center line" is about 50/50, so if it'll make it more consistent with other grid mappings, let's go with azimuth_of_central_line.

Thanks for the suggestion!

+1. Good to go for v1.7, I reckon.

Phil

comment:11 Changed 4 years ago by caron

Looks ok to me.

comment:12 Changed 4 years ago by jonathan

Enough support has been expressed for this proposal to be accepted according to the rules. If there are no objections before 20 March, it will be accepted. Thanks to all for their contributions.

Jonathan

comment:13 Changed 5 months ago by davidhassell

  • Owner changed from cf-conventions@… to davidhassell
  • Status changed from new to accepted

comment:14 Changed 3 months ago by painter1

  • Resolution set to fixed
  • Status changed from accepted to closed
Note: See TracTickets for help on using tickets.