Percentage area overlap between years

[GaiusWilson]

I have GPS locations for 5 animals, over two years (so over 6000 points). Using the minimum bounding geometry tool, I have been able to create polygons (MCP) for each month for each individual over the two years. What I want to do is calculate the area overlap for each month from different years. So for example animal ID1, I want to calculate the percentage of area overlap in January 2017 and January 2018, Feb 2017 and Feb 2018…etc., In some cases there will be no overlap

Using the INTERSECT tool, and to calculate for one or two months or individuals, it can be done manually. How would I do it for numerous animals over numerous years? What tool do I need to use and how would I specify this so that ArcMap (or QGIS) understands to take Jan 2017 and compare it with Jan 2018 and give me the area overlap (percentage)? I’ve tried model builder too but I don’t see how I can specify which 2 fields (month and year of one and compare with same month but different year) / features to take into consideration for the analysis. I also can’t figure out how to calculate the percentage. 

Any advice would be most appreciated. Thanks. 

[DanPatterson_Retired]

It is hard to make a firm statement about what overlay method would be best, but you have looked at Intersect...

what about

Union—Help | ArcGIS Desktop 

Union retains the attribute information from all the stacked layers.

Depending on your table configuration, (ie separate Area fields... not Shape_Area) you could retain the area information between steps.

I wouldn't recommend stacking (unioning) everything until you see whether 2 layers yield any new insights.  If it does, an incremental approach may work

On the other hand, Tabulate Area in the spatial analyst extension is another possibility, but you would have to deal with the potential loss of precision associated with the representation of vector data as raster

[GaiusWilson]

Thank you for reply Dan. I am not sure the "union" tool would work because it would simply join the polygons (and all of them, rather than just Jan 2017 and Jan 2018, Feb 2017 and Feb 2018.....). I have separated the 12 polygons (corresponding to each month) for 2017 from the 12 polygons for 2018 as 2 separate layers. I have tried the tabulate intersection and that works to a certain extent but then there is manual separation and calculation of percentage as each month is tabulated against every other month. I need to look at the Tabulate area that you suggest. 

[GaiusWilson]

Btw, this is my first post, so I am not even sure if my question is widely visible. I am sure there are many people who have done this kind of analysis before, and I am sure there is a simpler way to calculate percentage overlap for specific months/areas. Thanks...

[XanderBakker]

I would problably create a python script to do this. If you can share the data I am able to help you with that.

[XanderBakker]

In addition to my last comment a couple of things more:

• You posted this question in the discussion part of the GeoNet Resource Hub, It might be better to share this with Managing Data and/or Python place to get more views
• If you created the influence zone using the MBG there can be areas that might overlap although there was never any animal you could also create a line from the sequence of migration points and apply a buffer to that line to get the area.

[GaiusWilson]

Thank you Xander for your reply, thank you for your willingness to help. I do not know how to use Python as I have never used it before. If the script is written, does it still work through ArcMap? 

[XanderBakker]

A python script can be run in the python window of ArcMap or as a standalone script, but in case you are planning to repeat a process and want to share that with others you normally would create a toolbox with it that enables an easy interface.