[PROJ] About Mexican Geoid

[Greg Troxel]

Javier Jimenez Shaw <j1 at jimenezshaw.com> writes:

> Recently I found that Mexico has a geoid model, used with vcrs NAVD88:
> GGM10 - Geoide gravimétrico mexicano 2010
> http://en.www.inegi.org.mx/temas/geoide/#Downloads
> The file is a text format, lat, long, undulation. It should be easy to
> import.
> The reference is ITRF2008 epoch 2010 (it sounds strange to me not using
> NAD83x, but it is fine, right?... well, epoch 2010 instead of 2005 may be a
> problem)
> Unfortunately I didn't find any transformation in EPSG from ITRF2008 (epoch
> 2010) to NAVD88. We should do a PROJ one for the database (or ask them to
> do in EPSG, but without much hope).

In the US, geoid models are split into three camps, with the first two
defined only in the US (or North America) and the last world-wide:

  NGS (current NSRS): hybrid geoid model that converts from NAD83 HAE to
  NAVD88

  NGS (forthcoming NSRS): gravimetric geoid model that (I think)
  converts from NATRF2022 HAE to heights above the NAPGD2022 level
  surface.

  NGA: gravimetric geoid model that converts from WGGS84 HAE to WGS84
  orthometric height.  Not part of the NSRS, and not used by Federal and
  State governments.

Canada is different; it seems they moved from CGVD28 to CGVD2013 which
uses a gravimetric geoid model that relates NAD83(CSRS) HAE to CGVD2013
heights.  So it seems very much like NAPGD2022 (and I wouldn't be at all
surprised if there is a harmonized W_0).  I am not familiar with Mexican
geodetic practice but wouldn't want to make assumptions based on US/CA.

So I would wonder

  Is the Mexican vertical NSRS (meaning the official legal height system
  used by the government, surely a different acronym) NAVD88?

  Is the Mexican horizontal NSRS NAD83?

  Do they really mean that this geoid model converts ITRF2008 HAE to
  NAVD88?  Or is that just how the lat/lon of the model is expressed?
  If it's really about ITRF2008 HAE, that sounds very unusual, but it's
  just a set of coefficients to convert between two things, so it's
  entirely possible and well defined.  In other words, I wonder if your
  interpretation is due to imprecise language in the downloads page, or
  if you really are correct.

  (Agreed that using ITRF2008 with a different epoch is surprising.)


As for transformation ITRF2008 to NAVD88, it seems obvious at first
glance (dangerous) that the right approach is to transform ITFF2008 to
NAD83(2011) and then to use GEOID18 (NGS) to get NAVD88, but I may be
missing something.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 194 bytes
Desc: not available
URL: <http://lists.osgeo.org/pipermail/proj/attachments/20210719/11ff8c2d/attachment.sig>