Select raster values within polygon extents-based on statistics per polygon extent

0
votes

I have two layers: a polygon layer defining 20 study sites and a raster layer defining depth. Using the zonal statistics tool, I calculated mean depth and standard deviation per study site. Now I wish to extract those depth pixels (per site) that exceed 1 standard deviation from the mean of the depth values for each site. The goal is to define the deepest areas of habitat per site. Any idea how to do this for all 20 sites simultaneously (since they each have a distinct mean and standard deviation value)?

1
arcgisrasterextractionarcmap
Hi @Ily and welcome to Stack Overflow! I recommend you read how to create a Minimal, Complete, and Verifiable example and also check How to Ask Good Questions so you increase your chances to get feedback and useful answers... in other words, include some code and your expected output/behavior – DarkCygnus
@DarkCygnus thank you for your feedback! I tried to upload a photo of the raster and polygon layers but don't yet have permission to do so- sorry. I am able to run most of this with R code, but I am interested in an ArcMap Desktop solution (so I added a tag for ArcMap on the original post). In regards to output/behavior-- I would like to create a new polygon feature around raster pixels that meet specific criteria. I hope that fills in the blanks, and I look forward to being a part of the Stacks community. – Ily

1 Answers

0
votes

I was able to devise a method to accomplish this task (tho likely not in the most elegant fashion): as described above I used zonal statistics to create a new raster with mean values of the raster (depth) per site (delineated by polygon layer). I then used the same tool to create a raster layer representing the standard deviation per site. Then in rater calculator I set all values to NULL which were less than 1 standard deviation from the mean depth per site. SetNull("depth"> "mean_depth" - "sd_depth", "depth")-- and this created a new raster with only those pixel values greater than 1 std from mean depth (ie the deepest habitat per site). note: because depth values were negative, we used > (greater than)