I am experiencing a repetitive delta between my GCP’s and my point cloud prior to data alignment. I’m ok with aligning my data to my GCP’s, but I want to make sure everyone else is seeing a similar delta. Below is a list of the details of my scenario.
- GCP’s and RTK achieved with local ROCK network
- Baseline < 1 mile
- GCP’s collected in NAD83 (ft) and NAVD88 Geoid 18 (ft)
- LiDAR processed on ROCK Desktop with obs file from the base and typical CRS (UTM 15/Ellip.) confirmed with Cloud upload
- LiDAR data then re-projected into NAD83/NAVD 88 Geoid 18 (ft)
- GCP’s then added to project
- GCP’s are roughly 3’ high in elevation and 5’ shifted east compared to the data set.
This is repeated over and over. I’ve also used Leica Smart Net to get my GCP’s and the delta was repeated.
ROCK Support rep explained that this was a reasonable delta due to converting from the WGS84 system to the NAD83.
Can anyone else confirm that they are seeing similar deltas due to reprojection?
1 Like
Did you enter a base location in WGS84? If not then the RINEX header location will be used which is an approximate (non-corrected) location from when you start logging. Another source of difference is if you are using an NTRIP with NAD83(2011) reference your GCPs may be showing you NAD83 locations, which is quite different in some locations than WGS84. Other differences can be GNSS epochs and geoid models. Best to have good GCPs where you know epoch and geoid model and always align to those. Also tag a benchmark and set permanent control!
1 Like
My GCPs are always “high” due to ellipsoidal and orthmetric differences. This is common, however, your x,y shift is owing to coordinate mismatch.
As noted prior to this post, make sure all your coordinates match.
You got this!
1 Like
Entering a base location could be one issue. I have assumed the RINEX file I’m downloading from my ROCK base has a corrected location since the base is permanent. That location is provided in NAD83(2011). What’s your go-to tool for transformation to WGS84?
You are correct that I am getting my GCP’s in NAD83. I re-project my LIDAR to NAD83 and then add the GCP’s as ROCK has suggested in their learn section. I have made sure that my GCP’s match my reprojection CRS. I am very confident in my GCP’s as I have done that side of the job. It’s my LiDAR data that seems a little further away than expected.
What’s a typical z delta you see?
Our typical delta is around 15-17m (+/-50’), owing to the ellipsoidal/orthometric height separation in my part of the world.
Once its all processed, and then transformed to my chosen flavor of coordinate system, then I move the “z” to where I want it.
1 Like
i believe that this is not the right way to do it. we did this at first, you can get away with it on small sites, but because of the undulation difference in ellipsoidal to ortho you can have problems on a large site. let s say a corridor survey over a couple km. depending on the area even under 1km will like show a difference in separation across the site.
i would suggest that you should also process the vertical reprojection in the rock or other software to get from ellipsoidal to your desired vertical datum.
and i agree make sure that your starting coordinate is correct and in wgs84. AND make sure your elevation is at phase center on the base head. qgis is free and does many transformations. but it is a little finicky.
personally i wish you could enter your base coordinate in any coordinate system. and specify your output coordinate system at the start of processing, AND IN DESKTOP this would save a ton of time and frustrations. if everyone asks maybe this will get changed.