I am using ArcMap 10.2.2 (Advanced license, no toolbar add ons/no spatial analyst) on a Windows 7 computer (i7 processor, 8GB RAM). I have used ArcMap for other basic analyses, but am fairly new to the program and have only very basic coding skills.

I am trying to calculate the number of mammal species (TerrestMamm_diss on the image below/light red outlines) that are found within the geographic range of a subset of those species (all species in the order Carnivora, or Carns1 on the map/dark green outlines, one of which is selected). My ideal output would look like the attribute table below plus an additional field indicating the count (i.e. FID, Shape, binomial, mammal_count).

screenshot of ArcMap document

There are ~250 features on the Carns1 layer and ~5,300 on the TerrestMamm_diss layer (whose attribute table looks just like that of Carns1 shown in the image, since Carns1 is just a subset of TerrestMamm_diss). Both layers have overlapping polygons.

I do not need to keep the identification/binomials of the mammal species found within each Carnivora range, I just need the total count of species. I've searched on this forum and elsewhere and I've found ways to count point features within a polygon (and I'm familiar with the model and procedure to count overlapping polygons using the "spaghetti and meatballs" method), but I'm not sure how to do this.

My initial thought was to turn all the terrestrial mammal species (TerrestMamm_diss) into points (feature to point tool) and count them that way, but I realized putting a single point within a large species range would mean that point wouldn't necessarily lie within each of the overlapping polygons.

  • If you have worked with spreadsheet pivot tables you run a union overlay of the two layers, then export the table to a dbf (or select and copy the features for paste) for use in a spreadsheet. ArcGIS does have a pivot table tool although I don't use it and can't comment on it. – johns Jan 15 '16 at 18:28
  • I don't have experience with pivot tables, but I will look into it, thank you. – MooseGirl Jan 15 '16 at 18:33
  • Just to get my head around your requirements - you would like to count all Carns1 features that overlap a TerrestMamm_diss feature of the same binomial? And the same Carns1 feature could be counted again if it also overlaps another TerrestMamm_diss feature? Is that correct? – Midavalo Jan 16 '16 at 0:23
  • Not quite--I would like to count all the TerrestMamm_diss features that overlap with each Carns1 feature. My ideal outcome would be an attribute table for Carns1 that was FID, binomial, # of TerrestMamm_diss features that overlap [the Carns1 binomial's range]. The Carns1 layer is a layer of Carnivora species' geographic ranges and the TerrestMamm_diss is a layer of all mammal's geographic ranges and I want to know how many mammals live in each Carnivora species's range. – MooseGirl Jan 17 '16 at 3:08

With some crowd sourcing, the way to do this is pretty straight-forward, although it took almost an hour of processing time on my computer (in case anyone else is frantically googling around):

Perform a one-to-many spatial join using Carns1 as the target feature (with their ranges dissolved to allow multipart features, as I did before posting the initial question) and TerrestMamm_diss (also dissolved with multipart features) as the join features. This creates a join_count field on Carns1 that counts the number of overlapping TerrestMamm_diss features.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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