0

Here's a curly one!

EDIT: As an alternative the modal raster value inside each bounding box would work, and might possibly be faster. Still unsure how to do it though. Perhaps Zonal statistics using Majority. Then I'll need to filter by the maximum count.

Background Info
Currently using ArcGIS Pro.

I currently have a grid, and a map of vegetation in raster format.
I want to find the most common vegetation by area, in each grid square.
The dataset is probably a couple of hundred thousand rows long - and likely isn't nicely sorted.

Current Process
Vegetation Raster to Vegetation Polygons (Raster to Polygon)
Pairwise Intersect Vegetation Polygons with Grid Polygons to create VegetationGrid_Polygons
Calculate Geometry Attributes on VegetationGrid_Polygons to give Area
Stuck.

Starting Table
Now I should theoretically have a table that looks like:
VegetationID, GridID, Area
1, 10, 2
2, 10, 8 <-- Notice this, there are two VegetationID's #2
2, 10, 4 <--
5, 10, 4
1, 5, 10
3, 5, 46
5, 5, 4

What tool(s) can I use to create a table that has the sum area of each VegetationID by each GridID?
VegetationID, GridID, Area
1, 10, 2
2, 10, 12 <-- This is now 12
5, 10, 4
1, 5, 10
3, 5, 46
5, 5, 4

Then, how to find the VegetationID with the largest area value by GridID?
VegetationID, GridID, Area
2, 10, 12
3, 5, 46

1
  • 1
    Try search this site for "dominant area"
    – FelixIP
    Apr 22, 2020 at 3:20

1 Answer 1

0

It seems like this approach has worked:

Zonal Statistics using "Majority".
New Raster with the majority to Polygon using Raster to Polygon.
Take my original bounding grid and turn it into points using "Feature to Point". Do this to avoid slithers during intersection.
Pairwise Intersect - pairwise because it's faster. Intersect to join attributes.
Table to Table - to get it to CSV.

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