Skip to main content

Vegetation and vantage point influence visibility across diverse ecosystems: implications for animal ecology

Cite this dataset

Stein, Rachel et al. (2022). Vegetation and vantage point influence visibility across diverse ecosystems: implications for animal ecology [Dataset]. Dryad.


Visual information can influence animal behavior and habitat use in diverse ways. Visibility is the property that relates 3D habitat structure to accessibility of visual information. Despite the importance of visibility in animal ecology, this property remains largely unstudied. Our objective was to assess how habitat structure from diverse environments and animal position within that structure can influence visibility. We gathered terrestrial lidar data (1 cm at 10 m) in four ecosystems (forest, shrub-steppe, prairie, and desert) to characterize viewsheds (i.e., estimates of visibility based on spatially explicit sightlines) from multiple vantage points. Both ecosystem-specific structure and animal position influenced potential viewsheds. Generally, as height of the vantage point above the ground increased, viewshed extent also increased, but the relationships were not linear.  In low-structure ecosystems (prairie, shrub-steppe, and desert), variability in viewsheds decreased as vantage points increased to heights above the vegetation canopy. In the forest, however, variation in viewsheds was highest at intermediate heights, and markedly lower at the lowest and highest vantage points. These patterns are likely linked to the amount, heterogeneity, and distribution of vegetation structure occluding sightlines. Our work is the first to apply a new method that can be used to estimate viewshed properties relevant to animals (i.e., viewshed extent and variability). We demonstrate that these properties differ across terrestrial landscapes in complex ways that likely influence many facets of animal ecology and behavior.   


We gathered terrestrial lidar data in four ecosystems in the western USA. The sites were a forest (the University of Idaho Experimental Forest), a shrub-steppe (in the Lemhi Valley of Idaho), a prairie (a remnant of Palouse prairie at the Dave Skinner Ecological Preserve and Thorn Creek Native Seed Farm near Moscow, Idaho), and a desert (Gold Butte National Monument, Nevada). In each ecosystem, 12-m radius plots were established. In the forest, 10 plots were established that covered a wide range of stand types. In the shrub-steppe and desert (n = 6, n = 5 respectively), plots were located in areas of known animal activity. Activity was determined by identification of tracks and scat. The prairie site was constrained by its small size; 6 plots were measured in areas that did not including encroaching vegetation.

At each plot, we collected lidar data using a Leica BLK360 terrestrial laser scanner set to its standard point density (1 cm at 10 m) on a tripod 1.3 m above the ground. The scanner height was occasionally lowered to accommodate uneven terrain or vegetation. Highly reflective targets were deployed in the plots. We then gathered scans at 15-25 locations. The number of scans was determined by vegetation structure where greater complexity required more scans.

Single large point clouds detailing each plot were created by stitching the individual scans together using ReCap Pro v6.0 and CloudCompare v2.11.3. First, targets were identified in ReCap Pro which created a rough alignment of the scans. The oriented scans were moved into CloudCompare and the iterative closest point algorithm tool was used for fine-scale alignment. Once aligned, the scans were merged in CloudCompare; the targets were removed using the segment tool. ReCap Pro was used to define the center ground point as 0,0,0. The program randomly assigned the X,Y orientation and Z was elevation. The final point clouds ranged in size from 20-97 million points.

In each plot, we measured 10-m viewsheds using the R package, viewshed3d at 5 X,Y locations and multiple Z locations. The first X,Y was located at the plot center (0,0). The other four were placed two meters from the center along the X,Y axes (0,2; 0, -2; 2,0; -2,0), which ensured the sightlines were not measured beyond the point cloud extent. Z positions at each X,Y were 0.25, 0.75, 1.5, 5, and 10 m from the ground representing perspectives of wildlife in the ecosystems. In the forest, additional Z perspectives were placed every 5 m (from 5-30m). The additional Z positions were not measured above 10 m in the other ecosystems because the viewsheds could not interact with the point clouds. Spherical viewsheds were estimated at these locations using viewshed3d which measures sightlines until they encounter an obstruction (i.e., lidar data point). The angular resolution of the sightlines was set to 0.6 degrees. Measurements are reported as a viewshed graph that displays the percent of unobstructed sightlines as a function of distance from the designated vantage point. We calculated the area under the curve and defined it as the viewshed coefficient for analysis. To estimate variability of the sightlines, we segmented the point cloud at each vantage point into 20 segments of 18o and estimated the viewshed coefficient of each segment. We calculated the standard deviation and mean of these 20 segments from which we calculated a coefficient of variation. Vertical structural heterogeneity was assessed by calculating roughness, the standard deviation of the canopy heights within 10 m of each X,Y position using the lidR package.

Point cloud naming convention: XXX_#.laz. Letters indicate the site (UIEF = forest, SB = shrub-steppe, PP = prairie, GBNM = desert). Numbers are plot numbers. Forest plot numbers correspond to existing plot numbers established by other researchers in the University of Idaho Experimental Forest. Plot numbers in the other ecosystems assigned in order of lidar data collection.

Usage notes

The point cloud files are stored as .laz files which can be opened by the open source softwares Program R and CloudCompare. This file type can also be opened by proprietary softwares such as Cyclone. The viewshed measurement code was written for and executable in Program R. 


Idaho Space Grant Consortium

National Science Foundation, Award: NSF 1553550