Camera trapping has become an increasingly widespread tool for wildlife ecologists with large numbers of studies relying on photo capture rates or presence/absence information. It is increasingly clear that camera placement can directly impact this kind of data, yet these biases are poorly understood. We used a paired camera design to investigate the effect of small-scale habitat features on species richness estimates, and capture rate and detection probability of mammal species in the Shenandoah Valley of Virginia, USA. Cameras were deployed at either log features or on game trails with a paired camera at a nearby random location. Overall capture rates were significantly higher at trail and log cameras compared to their paired random cameras, and some species showed capture rate increases as high as 9.7 times greater at feature-based cameras. We recorded more species at both log (17) and trail features (15) than at their paired control sites (13 and 12 species, respectively) yet richness estimates were indistinguishable after 659 and 385 camera nights of survey effort, respectively, for log and trail features. We detected significant increases (ranging from 11-33%) in detection probability for 5 species resulting from the presence of game trails. Detection probability was also influenced by the presence of a log feature for six species. Bias was most pronounced for the three rodents investigated, where in all cases detection probability was substantially higher (24.9-38.2%) at log cameras. Our results indicate that small-scale factors, including the presence of game trails as well as other features, can have significant impacts on the frequency and probability of species detection when camera traps are employed. Significant biases may result if the presence and quality of these features are not documented and either incorporated into analytical procedures, or controlled for in study design.
This file contains a row for each photo image recorded on camera traps during this study. The study was conducted during Summer and Fall months of 2013 and 2014 on the grounds of the Smithsonian Conservation Biology Institute in Front Royal, Virginia USA. All records of birds and humans have been removed. Cameras were established in pairs with a treatment camera (set up with a log in view, or on a game trail) and a nearby random location. End Date refers to the date after which at last one camera in the pair stopped functioning. All photo records from BOTH cameras in the pair taken after this date were removed. The grounds of the study area were divided into grids (500m by 500m) and grids not containing forest were not used. Grid codes are included with each image as is the UTM coordinate of the camera station (UTM Zone 17). "UnderCat" is a three level descriptor for level of understory vegetation at the site. "LogD" refers to log diameter in centimeters, and "TrailQ" is a scale of trail quality, with 1 being the highest quality. "PairDist_m" is the distance between the two cameras in the pair, in meters. All other details are explained in the manuscript. NOTE: Photograph capture times for deployment SCBI2013.4T are 7 hrs ahead of the actual photo time. The Date and Time listed for photos from Deployment SCBI2013.37C are incorrect and could be rectified. Photos all occurred between the deployment and pull dates, but times and actual dates are incorrect for each photo.
This file contains all the vegetation measurements taken at each camera station pair. Cameras were set up in the Summer and Fall seasons of 2013 and 2014 on the grounds of the Smithsonian Conservation Biology Institute in Front Royal, VA USA. Cameras were only established in forest habitat. Camera stations were made up of pairs of cameras, one at a log feature or game trail, the other at a random nearby location. Vegetation measurements were taken at the midpoint between the two cameras. In addition to the PairID, Grid location, and vegetation sampling date, 4 vegetation variables are included. "CovAll" is the overall level vegetative cover percentage recorded as the average percent of a 2-m high cover pole obscured from 10 m away in each cardinal direction. Cover pole was divided in 20 1-dm sections. Data were recorded as the percentage of 1-dm sections obscured ( >50%) by vegetative or structural cover. "CovLow" is the average percent of low portion (<0.5 m) of cover pole obscured by vegetative or structural cover. "UnderStem_D" is the number of woody stems >1.5 m in height and <7.5 cm dbh counted within 3 arm's-length transects, divided by 33 square meters. Transect were parallel, 20-m transects extending perpendicular to the aspect of the plot. Transects were placed both through the actual sampling location as well as 10 m up and down slope of location."OverStem_D" is the number of woody stems >7.5 cm dbh counted within 3 arm's-length transects, divided by 33 square meters. Transects used were same as for "UnderStem_D".