I am wondering about WGS84 as we move through time.
My understanding is that any given spot on the earth's surface latitude and longitude will change over time in this system. There are two sources of change:
1) NGA (National Geospatial-Intelligence Agency) will periodically shift the center of the WGS84 ellipsoid to better align with their current best guess for the center of the earth's mass (I think they are using a geoid to find the center of mass). At which time the will release a new version of the WGS84 and note the GPS weeks in the name, e.g. WGS84 (G1150).
2) WGS84 is not locked to any land mass but to the stars. So, as the tectonic plates drift around on the earth's mantle a location on the earth's surfaces can change over time.
Does that sound right?
So, when I select WGS84 as my GCS what version of WGS84 am I getting?
WGS84 (G873)
WGS84 (G1150)
WGS84 (G1674)
WGS84 (G1762)
Something else?
Thanks!
Forest
Solved! Go to Solution.
Yes, that's correct. You can also have movement due to subsidence, glacial rebound, tectonic motion (earthquakes), etc. US control points have a setting for "stability" even.
We're just now adding the WGS84 realizations to 10.5.1, and Pro 2.0. However, we have no transformations between them, as none have been published. There are a few ITRFyyyy-WGS84(Gxxxx) transformations, but none are official.
Melita
Yes, that's correct. You can also have movement due to subsidence, glacial rebound, tectonic motion (earthquakes), etc. US control points have a setting for "stability" even.
We're just now adding the WGS84 realizations to 10.5.1, and Pro 2.0. However, we have no transformations between them, as none have been published. There are a few ITRFyyyy-WGS84(Gxxxx) transformations, but none are official.
Melita
Thanks, Melita. So, in 10.4 and older, we are getting WGS84 (G873)?
Hi Forest,
Technically, it's a non-specific WGS84 definition. That is, it could be used for any or all versions of WGS84.
I would like ArcGIS to support a time field in coordinate transformations, so that you can do more accurate transformations (e.g. 14 parameter time-dependent helmert transformation from ETRS89 to WGS84).
Hi Chris,
We're working on it! It's currently working at the projection engine level. We need to make a new geoprocessing tool and support an "Epoch" parameter on a GCS.
Melita
This is good news, but will I be able to select a time field to use in the transformation as opposed to choosing between discreet epochs?
The 14/15 parameter time-enabled transformation usually converts between two reference frames at a particular epoch date. That'll be round 1. To change epochs within a reference frame, we need the velocity model for the area and/or seismic event models. We're also working on that, but that may or may not get into the software in the initial install.
Melita