I have a feature class that seems to be cutting off geometry. I looked around to see if there is a bounding box applied and found the Domain, Resolution and Tolerance tab in the Feature Class Properties.

The Min / Max values are populated with values that are not appropriate for my region. How do I set / reset the extents to the default value? That is, how do I let the geometry control the extents? Various documents suggest using the "use default extents" checkbox but I don't see that option anywhere.

This feature class is in enterprise SDE - not a local file geodatabase.

  • If you get stuck here, I would attempt to select all and export to a new feature class in a file gdb or shapefile. If that restores all your geometry, you could at least recreate a new feature class to replace the existing one. Commented Aug 27, 2015 at 17:59
  • The max values should be ignored as they're controlled by a combination of the minimum coordinate values, resolution, and the maximum allowed coordinate. Anyway the max values are usually huge. You're going to have create a new feature class (set the coordinate system which will set the domain properly) and import your existing data into it.
    – mkennedy
    Commented Aug 27, 2015 at 18:02
  • The "use default extents" is part of the create feature dataset/class wizard. Also, is there a coordinate system mix-up? That is, you have features with coordinates that are actually in a different coordinate system than the feature class? Maybe edit your question with fc info and some feature coordinates.
    – mkennedy
    Commented Aug 27, 2015 at 18:05

1 Answer 1


There are three sets of properties to be aware of here:

  1. The envelope of horizon of the coordinate system,
  2. The extent of the coordinate reference, and
  3. The envelope of the feature class.

The coordinate system horizon is mathematical space accessible by the coordinate system (e.g., infinite X, -90...+90 Y for GCS_WGS_1984). Esri uses these horizons to create default coordinate references at 0.1-1.0 millimeter precision (which can be overridden).

Esri coordinate references include coordinate system, but also reference a lower-left X/Y origin and X/Y scalefactor, Z and M origins and scalefactors, tolerances, and a precision. The upper-right extent of the coordinate reference is calculated using the lower-left origin, at the maximum number of scalefactor steps at the support precision (HIGH or BASIC, corresponding to 31 and 56 bits). The upper-right corner is rarely a factor, since it gets all the spare bits. Esri has a whitepaper that describes how this works in great detail. It is important to note that increasing the coordinate reference precision greatly in excess of the actual data precision (e.g., sub-millimeter for consumer GPS data) will have a storage and performance impact -- and the defaults used by Esri often err on the side of "too much" precision.

The feature class extent is initialized to empty and maintained during editing (and bulk insert), but can be reset at any time to any value which is within the coordinate reference extent. The feature class extent can impact index use in enterprise geodatabases, since it's assumed that a spatial query that exceeds over 25% of the data space will be suboptimal, forcing the search logic into a full table scan.

You cannot change the lower-left origin and resolution of an enterprise coordinate reference (SRID) once it is created. You also cannot change the coordinate reference of an enterprise feature class once it has been created -- If you specified a Y origin of 0 for a "USPossessions" feature class, then suddenly get some South Pacific data, you must export the existing data, create a new feature class with an origin which will support the new data ({-400,-400} is the safest solution for GCS FCs), then copy the old and new data into it.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.