[PROJ] CSRS Epoch 2017.50 (NAD83)

[Greg Troxel]

Javier Jimenez Shaw via PROJ <proj at lists.osgeo.org> writes:

> I discovered today that there is a datum in California called "CSRS Epoch
> 2017.50 (NAD83)"
> http://geoweb99.ucsd.edu/index.php/epoch2017/
> http://sopac-csrc.ucsd.edu/
>
> "CSRS Epoch 2017.50(NAD83) replaces the previous “CSRS Epoch 2011.00
> ITRF2005 NAD83(NSRS2007)” datum that included coordinates for 830 CSRN
> stations."
>
> I have not seen it before, so I guess there is no CRS using it in EPSG. (I
> do not find that datum in EPSG)

I haven't heard of this before, but I live in a stable part of the
country :-)

I read about this a bit, and I'm not quite sure what's going on.  As I
understand things, the official federal datum is NAD83(2011) epoch
2010.0 even in regions of California on the Pacific plate.  However, you
can't really make measurements in that frame because the assumption of
small relative motion between your point and the reference stations
doesn't apply, because of not only crossing the plate boundary but the
large deformations.

I think the point is to, while avoiding a full-blown dynamic datum like
"just use ITRF" or the NATRF's intra-frame velocity model, have a series
of static datums that, if a recent one is used, have much lower errors.
I would think that if one examined station coordinates over time, you'd
see PA plate motion and deformation, and probably the eastern edge would
line up with NAD83.

I wonder if the need for this is driven by widespread use of RTK
networks.  The CA RTN situation is unclear to me:
  https://dot.ca.gov/caltrans-near-me/district-6/district-6-programs/d6-land-surveys/d6-rtn-gps

and I'd expect that the base stations configure their epoch 2017
coordinates and thus only the differential values of present-2017 end up
as error sources for users trying to obtain epcoh 2017 coordinates.

My state's RTN, "MaCORS", configures NAD83(2011) epoch 2010.0
coordinates, even though in 14 years at 2 mm/y (since NAD83(2011) is not
quite crust fixed), that's 2.4 cm.   But users want epoch 2010.0
coordinates, not current epoch, and basically all points in
Massachusetts have the same relative motion.

> (Is it normal that some states use their own datum in the US? It is not
> enough with hundred of State Planes ;)

It is absolutely not normal, but also not surprising that California
would do so.  The only other state that has such deformation is perhaps
Alaska, and they might take the approach that if the road hasn't buckled
it's all good.  I'm also not sure about WA.  But I have not heard of
them having datums.

I know you know this, but for others reading, State Plane Coordinate
Systems are projections of existing datums.  Generally, they are
projected from "NAD83(2011) epoch 2010.0".  So that's a much lesser
complication.