Indexing geo_shapes with low precision

I am indexing documents that contain a field of geo_shape type. I am
currently indexing with a tree_level of 10 as this is the level of accuracy
that would be needed for geospatial queries using that field in the
future. However, at this point in time, there is no requirement to run
queries against that field so in effect, I am only interested in persisting
the geojson within the index to use for presentation purposes when
retrieving documents.

I am considering using a tree_level of 1 for the field instead of 10. Am I
correct in thinking:

  1. This will reduce the size of the index on disk because fewer geohashes
    will need to be generated and persisted for tree_level 1 compared to 10.

  2. The geo_shape json returned for a given document will be the same for
    tree_level 1 as it would be for 10 i.e. the field source data is unaffected
    but the underlying geohash representation will be.

  3. The accuracy of geo_shape queries on the field will be greatly reduced
    (as noted above, not a concern for the time being).

Any thoughts greatly appreciated!

You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
To view this discussion on the web visit
For more options, visit