Interpolate Shape drops M values

[KimOllivier]

Problem: To create a graph profile with a featureclass you need M and Z values.
Therefore use a route featureclass and overlay over a DTM to add elevations using Interpolate Shape tool.
But then the M values have disappeared, so I have to rebuild the M values from somewhere.

Could the M measures be retained?

[ClaytonCrawford]

There's an enhancement request in on this already. By default, InterpolateShape introduces new vertices in the line geometry. Would you want M's interpolated for them based on existing M's or not? Or, would you be using the 'interpolate vertices only option' and not run into this (risking the omission of 3D info along the route)?

[TimWhiteaker]

Nearly 10 years later, I find this issue still exists!  What happened with the enhancement request?  I suppose a workaround could be, once you've got Z, is there a way to take the original line-with-M and interpolate the M onto the line-with-Z? 

> Would you want M's interpolated for [new vertices] based on existing M's or not?

Yes please.

[DataProcessor]

I have just discovered this bug myself! How strange it still hasn't been fixed.... Can you introduce the option to interpolate M for new vertices based on existing M's? It's critical for creating graph profiles.

[MK13]

@ClaytonCrawford is there an update on this issue?

[ClaytonCrawford]

There wasn't an intent to preserve M's to start with. There's the question of how to handle M's for vertices that get introduced through densification/interpolation. It's considered an enhancement to the tool rather than a bug to fix and in the decades of the tool's existence only a handful of users have asked for this. Is there not a way to add M's after interpolation rather than before? Consider the length of a 3D line is greater than a 2D line. Would this be a reason to change the M's?

[TimWhiteaker]

We don't have a good way to add M after interpolation. Though the length of a 3D line is greater than 2D, we wouldn't change the M values.  Whatever M value was at a given vertex in the original line, we would want that M value at the same vertex in the newly interpolated line, and we would want M values for any new vertices to be interpolated from that.

[ClaytonCrawford]

What should happen when a portion of a line falls off the surface? How should Ms be estimated for the introduced vertices at the edge of the surface where a portion of the line has been clipped away?

[TimWhiteaker]

Our surface is guaranteed to cover the line, so that isn't an issue for us. Someone else (@MK13 ?) should chime in if that fits their use case.

[MK13]

@ClaytonCrawford I too use interpolate vertices only so I don't know how you'd address the stated scenario. Either way, it's frustrating to lose Ms when attempting to add Zs to create an polylineZM. It would be a nice enhancement to have on the tool.