Calculate Geometry Tool Not Working Accurately in 3.4

[JohnKellyNeedsGISHelp]

Hi Community,

I am trying to run the Calculate Geometry tool to calculate the acreage of about 5000 small polgyons ranging from 100 acres to as low as .001 acres.  The tool takes about 20 minutes or so to run on my computer, but the problem I have noticed is that it is inaccurate.  When I sort by descending order and compare shape area to acres, it seems that it overestimates several polygons' acreage by a huge amount.  Right now my system is to go through the results and individually re calculate every outlier, but this is proving to be not worth the time and rather difficult.  I've included a picture of the fields in the attribute table to show what the problem is.

Does anyone have any solutions to this?  It seems like the Calculate Geometry tool is skipping some selected polygon, or not processing every selected polygon.

[DanielFox1]

Hi @JohnKellyNeedsGISHelp 

I have linked an article below which might with this but also provide an understanding behind why you see a difference, this is also linked to your projections (including the layer and basemap)

The Shape_Length field in a feature layer’s attribute table displays inaccurate values

"

• The planar calculation method is used to generate the Shape_Length field values. Refer to ArcGIS Pro: Geodesic versus planar distance for more information.
• The Shape_Length field values are generated in a different unit of measurement. Refer to FAQ: What is the unit of measurement used by the Shape_Length field in attribute tables? for more information."

[JohnKellyNeedsGISHelp]

Hi Daniel!  Thank you for your response, that is very interesting. It seems like maybe I should try to choose "Area (Geodesic)" when calculating the acreage maybe?  For projection, should I try to change something in the parameters to match the projection with the basemap?  Thank you.

[MervynLotter]

Hi JohnKelly

If I encounter any odd behavior when working with vector data I usually run the Repair Geometry GP tool to find and fix any errors before further working with the dataset. Not sure if it is applicable to your situation but you may want to consider it.

To calculate size, I always use the Geodesic area calculation unless there is a specific reason to use a planar calculation, like (1) if my organization has a policy to use a certain projection, or if (2) a local projection (and a local datum) is deemed more accurate than the geodesic method with a global datum (like WGS84).

The geodesic area calculation is based on the underlying ellipsoid (model of the Earth's shape) regardless of your chosen projection.  But if your projection uses a different ellipsoid/datum, then you should get different results. 

If you have time, create a point file in Web Mercator, then create a point in Canada and then buffer it, first using planar and then geodesic distances. You will be amazed at the difference in shape (and area). One should never use Web Mercator and a planar calculation method. 

But don't rely on your Shape_Area value and rather use the Claculate Geometry tool to calculate the size of polygons. 

Good luck.