Habitat Suitability for the Mexican Spotted Owl

Photograph of Mexican spotted owls, courtesy of US NPS.

Project Overview

Use existing, publicly-available data sets to create a composite binary raster showing potential habitat suitable for Mexican spotted owls
Ensure certain habitat criteria are met within a given study area in southeastern UT

Tutorial and data download here from the Utah Geospatial Consortium. Data was sourced from the USGS National Map (10m DEM) and Landfire.gov (land cover classification).

All maps and calculations were done in NAD 1983 UTM Zone 12N.

Habitat Criteria and Study Area

Study area location in southeastern UT, USA.

Mexican spotted owls are native to the southwestern United States and are currently listed as a threatened species (NPS). For this project, a study area was selected in southeastern UT to evaluate suitable habitat in their native range (pictured, left).

Habitat modeling is essential to understanding where populations of these owls may live and forage. Four criteria were selected to determine unsuitable, suitable, and preferred habitat for the Mexican spotted owl. Criteria for this analysis are listed in the table below.

Table adapted from Utah Geospatial Consortium.

Suitable Habitat

The maps below show areas of suitable and preferred habitat for Mexican spotted owls within the study area. Preferred areas (dark blue) are locations that met all four habitat criteria. Note that the preferred areas are located within canyon rims and facing North, which matches the criteria descriptions of steep canyons, ledges and caves, and northerly aspects. Suitable areas (light blue) are areas that meet three out of the four criteria, and tend to radiate around areas of preferred habitat.

A higher resolution image of these maps can be found here.

Map depicting suitable and preferred Mexican spotted owl habitat within the study area. Preferred habitat areas met all four habitat criteria; suitable habitat areas met three out of four criteria.

Further analysis showed that 3% of the total study area was suitable or preferred habitat for Mexican spotted owls. The study area contained 4 sq. km of preferred habitat (see table to the right).

Table of areas of unsuitable, suitable, and preferred Mexican spotted owl habitat within the study area.

Mastery Observations

I was initially suspicious after running the slope tool on the DEM when I saw a lot of high angle values near channels. I then zoomed out and noticed the study site was very close to Canyonlands National Park, which I know has many steep canyons. After looking at the data in this larger context, the slope values made a lot more sense to me. I also ended up double-checking by adjusting the transparency of the slope layer and overlaying it on a hillshade.
I used layer transparency and hillshade a lot to determine if my resampling was correct/how to interpret the new layers. For example, I used the hillshade to determine what values meant concave (negative) and convex (positive).
The first time I tried to split up the land cover data, I used 3016-3019 as the range. This only gave me small amount of area, which seemed incorrect because when I selected pinyon-juniper from the main data set, it covered a large area. I decided to split up the categories for after 3015 instead of starting at 3016 since the numbers appeared to be categorical. Changing the sorting method seemed to work much better with this data set.
Throughout the exercise, I was wondering why we had to set our reclassification values to 0 and 1. Why not 1 and 2 or 1 and 4? Once we set up the final formula for the composite layer, how we reclassified made a lot more sense since the end goal was to find "preferred" raster cells adding up to 4.
It took a while for me to decide on what basemap to use for this. I ended up using the terrain with labels option, but I do wish there was a better way to show both tree cover and terrain. As the map currently stands, it's hard to tell what criteria the "suitable" areas did not meet. Some could be areas with unsuitable vegetation; others may not have a good amount of curvature. It would be cool to dive into the "suitable" data more to see what criteria was the most missed for this study area.

Page updated

Google Sites

Report abuse