Exercise 2: Multiple Buffer Distance Analysis
How is the spatial presence of postfire woodpecker nests related to the spatial presence of salvage-logged forest stands?
- How are woodpecker nests clustered within survey units? (Exercise 1 and 3)
- How does this clustering relate to salvage treatment units within the survey units? (Exercise 2 and 3)
Multiple Ring Buffer in ArcMap
This tool creates shapefiles of concentric circles around features based on user-input values. Users can dissolve buffers to create a single feature class per buffer distance. Buffer widths overlapping with features of interest can indicate spatial relationships between target subjects. In this case, intersecting woodpecker nest buffers with salvage harvest polygons may reveal trends in woodpeckers selecting nest sites closer to or farther from salvage units. An equation producing percent area of each nest buffer intersected by a salvage unit serves as an index.
Above: Buffers created in a test analysis at 15 – 50 meter intervals around nest point features.
For this exercise I used 2016 and 2017 woodpecker nest point shapefiles. I created multiple ring buffer outputs for each shapefile to use in the analysis. I also used a polygon shapefile of 35 salvage harvest units included within the treatment woodpecker survey units. I used another polgyon shapefile of the woodpecker survey units and a WorldView-3 1 m raster for supplementary data.
Multiple Buffer Distance Analysis Steps
- Create separate point shapefiles for 2016 and 2017 nest points.
- Run the Multiple Ring Buffer tool in ArcMap on each shapefile at 50 meter intervals from 50 – 300 meters. The tool creates
- Use the Intersect tool in ArcMap to create a polygon shapefile of the buffers clipped to the salvage harvest units.
- Use the Dissolve tool in ArcMap to merge overlapping buffer polygons of the same type for the same nest.
Above: The green polygons indicate areas where the Intersect tool identified overlap between the buffers and salvage unit polygons. The tool creates a polygon shapefile of the overlapping areas.
Above: The final result of six buffers at 50 m intervals around nest points intersected with salvage harvest units. Each color represents an individual polygon in a shapefile. The polygons are given size information and used to calculate percent overlap of nest buffers with salvage units.
5. Use the XTools Pro extension to attribute size information to each intersected buffer shapefile for area in square meters.
6. Create a new field in each intersected buffer attribute table for percent area of the buffer intersected by a salvage unit.
7. Use the field calculator to create a formula for percent buffer area intersected: (Area field/Complete buffer size)*100
8. Export each intersected buffer table to Excel, including the salvage treatment type column and the percent buffer area intersected column for further analysis. Transfer the salvage treatment type and corresponding percent buffer area intersected columns to a combined Excel file with a sheet for every buffer distance.
9. Add zero values to each sheet for the nests falling within a control unit. The control nests act as a fourth treatment and should be included in the results.
10. Use ggplot2 in R to extract data from each sheet and create box plots for each buffer distance.
Above: Excel table of X and Y inputs for boxplots showing percent of the complete buffer intersected by a salvage harvest unit. Each buffer distance from 50 – 300 m has a sheet containing columns for these values.
I generated graphs for each of the six buffer intervals (50-300 m at 50 m intervals) in 2016 (pre-salvage) and 2017 (post-salvage) for 12 total graphs. I presented the 2016 and 2017 results for each buffer distance side by side below. In each graph, the unlabeled grouping of points to the left of salvage treatment type 1 represents nests in the control area, or nests 100% inside salvage treatment 0. Visual analysis reveals interesting patterns about woodpecker nest site selection when considering the silvicultural prescriptions designating salvage treatment types. Below is a reminder of the salvage treatments:
Treatment 1 harvests the most trees overall but retains the most large diameter trees with spacing for Lewis’s woodpeckers. Treatment 2 harvests a moderate number of trees, retaining less large diameter trees but more medium and small diameter trees. Treatment 3 harvests a limited number of trees but retains barely any large diameter with a heavy focus on small diameter for white-headed woodpeckers. The control treatments do not harvest any trees.
An obvious downfall of the following graphs is that as distance from the nest increases, percent of the nest buffer intersecting a salvage polygon will decrease. There is an overall decrease in mean and midrange values for intersecting area as the buffer distance increases. However, some significant points emerge:
- Nest distribution in Treatment 1 units generally increases in breadth between 2016 and 2017. Meaning, in 2016 the distribution of woodpecker nest distances from a salvage unit centered more closely around a mean distance near 30 – 50%. In 2017 the values appear to spread out with less preference towards a specific distance from a salvage unit.
- Nests in Treatments 2 and 3 units generally decrease in percent area intersected by a salvage unit from 2016 to 2017. Meaning, after the salvage harvest, woodpeckers are selecting nest sites farther from Treatments 2 and 3.
50 Meter Buffer Results
100 Meter Buffer Results
150 Meter Buffer Results
200 Meter Buffer Results
250 Meter Buffer Results
300 Meter Buffer Results
Problems and Critique
I would perform this analysis again with larger buffer sizes and greater buffer intervals. I am not convinced the buffer size I chose for this exercise captured significant information at this scale. I created 10 buffers from 100 – 1000 m for a future analysis. Buffer size should be determined by assumed travel and foraging distances for each woodpecker species. The 50 – 300 m scale may not be a great enough distance for local trends to develop in the data. I chose these distances because the belt transects for the woodpecker point count surveys are 200 – 300 m apart to avoid interfering with birds on neighboring transects. I thought this would be a comparable scale for the buffer analysis.
This analysis fails to address or quantify the control nests because they are too far away from salvage units for their buffers to intersect. In Exercise 3, a near analysis in ArcMap can quantify trends in control nest distances from salvage areas.
In the future I would like to produce a statistics table for each 2016 and 2017 buffer distance displaying mean, standard deviation, and other metrics for more than a visual analysis.