A question about projection when creating a new shapefile

[WilliamsonElizabeth]

My data frame projection is WGS 84 (taken from the several ESRI basemaps loaded in the data frame). I brought in some GPS points which were taken in NAD27.  They were projected on the fly to match the data frame projection. I need to create a polygon based on these GPS points. I'll eventually need the polygon information to be in NAD27. My question is: because I'm creating a shapefile based on the GPS points in WGS84, do I need to set the new shapefile projection to WGS84? Or can I set the projection to NAD27, then, as I create the polygons, they will be shown correctly in the data frame, and still truly be NAD27?

Thanks!

[DanPatterson_Retired]

your gps is collecting in decimal degrees, wgs84... it may be set to show nad27, but that is purely for convenience.  My recommendation would be to produce the polygons in Dec Deg, WGS84 datum, then use the Project tool to get it into your desired projection/coordinate system.  In that way, your gps data is kept clean both inside the gps and once brought into arcmap.  The projection after is done as a separate step.

[WilliamsonElizabeth]

Ok will do. Thank you for your speedy help!

[JoeBorgione]

Just have to ask; why would you take a step backward in datums?  Seems like NAD27 is so, well, 1920s.....

1927 Model A, somehwere in Kansas....

[WilliamsonElizabeth]

lol! Well, the repository for all of the archaeological site data in Georgia requires site UTMs to be NAD27. It's what's been used since the Georgia Archaeological Site File was established way back when. I guess they feel it's too late to change. Also, archaeologists like old stuff

[JoeBorgione]

...Archaeologists like old stuff...

So it seems!

[NeilAyres]

As a GIS person I always advise clients when I see a business (mines, utilities, many others), having their main db in a coordinate system based on a classical datum, to bite the bullet, and transform the data now. Otherwise the problems just keep mounting up. Every new data capture project (aerial imagery, lidar, ground GPS etc etc), has to be transformed by some means on to the old datum. Question is, how is that transformation achieved. This part of the workflow is rarely, if ever, documented properly (usually because most people don't understand the implications). So you end up with a db which contains a whole bunch of historical stuff and more modern layers which never, ever, quite line up. And it is impossible to ever get everything sorted out.