Intro

[AidanWilliams1]

Hi everyone, I am new to the ESRI community forum and I am a graduate student of Geographic Information Science. I need help correcting some GPS points that were incorrectly collected in the field using Field Maps. 

I'm hoping to connect with someone at ESRI that might be able to explain how we can fix this situation. 

We were using field maps to collect GPS points, but we accidentally collected the points using two different datums. Some points were collected using NAD83 and others were collected using WGS84. The points were stored in one single layer in our ArcGIS Online web map. We then opened the point layer using ArcGIS Pro desktop. We don't know exactly which points belong to which datum. When we open the layer in ArcGIS Pro, the layer is in the WGS84 Web Mercator projection. We need all of the points to be in NAD83, but we realized there is error in the points that were collected in WGS84.

What happens to the GPS points when they go through ArcGIS online? Does the coordinate system stay the same? How can we correct the layer of GPS points so that they are all the same datum?

Any help would be greatly appreciated! Thank you!

Aidan

[MichaelLohr]

Aiden, you may get a better response from someone else, or this could turn into a tech support call. Reprojections and transformations are pretty complex and many others are far better versed than I am .
How many points are we talking about total and how many are in the wrong datum?
I think what you have to do is to come up with a way that you can segregate the two point sets, either based on dates/times collected, unique IDs, or one of the GNSS metadata fields like receiver name. Once you are able to segregate the points, then the incorrect datum set would need to be reprojected to the correct coord system.
If the amount of shift is grossly large, you can sometimes visually see the separate point groups, but this is not always the case. Small shifts are difficult to identify which points are correct and which are not. Depending on how many points are in question, recollecting the points in the correct system might have to be an option if you were not able to come up with a separation technique based on looking at the two raw data sets.
Were the points collected by multiple GPS receivers working in the same area on the same day and uploaded to a master point file via cellular link, like you would do with ArcGIS Online? We have done that many times and are able to segregate points based on something different in the received data, like a different User Name or a different Receiver Name in the GNSS metadata fields.
Hope that helps some, or maybe you have already had those same thoughts. Good Luck.

[AidanWilliams1]

Thank you for the response. We discovered that in the attribute table of the GPS points data the latitude and longitude values collected by the GPS receiver are stored correctly (before datum transformation). This means that all of the coordinates are in the same geographic coordinate system which was NAD83(2011). This saved us from having to separate the GPS points collected using a different iPhone that did not have the correct transformation setting applied. 

Here is where we found that information:

https://fwp-gis.mt.gov/arcgis/help/en/field-maps/android/help/high-accuracy-data-collection.htm#ESRI... 

Supported GPS metadata

If using a single GPS location to set a point's location, the following GPS metadata values are recorded:

• Receiver name
• Latitude—Position received from the GPS receiver before applying data transformations in the location profile
• Longitude—Position received from the GPS receiver before applying data transformations in the location profile
• Altitude—Ellipsoidal height received from the GPS receiver (not orthometric height, as used for z-values)