Create Replica with Different Output Coordinate System

[mpboyle]

Now that ArcGIS Pro includes some replication tools at version 2.5, I'm wondering if there will be a workflow similar to the one outlined in this technical article for creating replicas in different co...?

Better yet...could the the environment variables arcpy.env.outputCoordinateSystem and arcpy.env.geographicTransformations be used as substitutes for this manual workflow?

We would like to have all (or as much as possible) of our web services in Web Mercator to avoid any type of on-the-fly projection...but the only method I've found for create a child replica in a different coordinate system is the workflow linked above.

[GrantHopkins1]

@Robert_LeClair , thanks again for the help. I'll give this a shot.

[KellyAlfaro_Haugen]

Are we in 2025 and the only way to replicate into a different coordinate system is still a manual creation and load of all your feature datasets, and feature classes, in the new projection and then create the replica? ArcMap had a one-stop shop to do this process as an easy 3 step process. Now I'm either required to do it manually or create my own model?  

[Robert_LeClair]

@KellyAlfaro_Haugen  - the ArcMap one-stop shop you refer to, is this the Extract Data Wizard that was in ArcMap by chance?  If so, starting at ArcGIS Pro 3.3, there's a new GP tool called Extract Data from Geodatabase that has the same functional workflow I believe.  This could simplify the process.

[KellyAlfaro_Haugen]

That would be fantastic if it does. I am of course stuck at 10.9.1 and 2.9 for now but I will keep that in mind for future upgrades. 

[KellyAlfaro_Haugen]

So I have another question, within this one, and you seem to be the one in the know. 

We have replicas where a layer is in the Feature Dataset but does not get included in the replica. Instead the feature is updated through some other process (usually python). 

Example: Wetlands

In the parent database there is a Feature dataset with multiple layers. One layer "Wetlands"  is updated by staff with an "Archive Field" that is True/False. When they map a new wetland polygon they will keep the old one just marking it Archive = True. 

In the child database (one-way replica) there are two layers "Wetlands" and "WetlandsHistoric". All of the other feature classes are included in replication. These two are not. They are updated via python (truncate/append) where Archive is either Ture or False. So they should not be in the Replica.

The Problem:

In ArcMap when loading schema and data (Data Extract tool) and creating the replica you had the choice to exclude layers from each of these steps.

With the ArcPro Replica tool your only choice seems to be to "Register existing data only". Which is fine for "WetlandsHistoric" because does not appear in the parent database but is a problem for "Wetlands" which does. 

Does this mean that my only options are to not include the "Wetlands" layer in the child database until after the replica is built or to build the replica and then unregister just the layers like this one that I don't want? Because either way means more steps and is not as nuanced as in ArcMap.