What attributes are relevant for water culverts to serve as efficient road crossing structures for mammals?
Written By Benjamin Brunen, Caroline Daguet and Jochen A. G. Jaeger
5th March 2021
Road mortality presents a substantial threat to the populations of many wildlife species without the installation of proper fences and wildlife passages. In many cases, roads create a barrier effect that results in the disruption of wildlife movement for many species. While wildlife passages have been advocated as a solution to various problems associated with roads, they are expensive and many roads still have none. However, roads usually have a series of water culverts designed to allow water to cross underneath the road, which might also be used by some mammals. In situations of financial constraint, it becomes important to explore cost effective (yet still functional) alternatives to designated wildlife passages. However, very few studies have quantified the effects that water and construction material have on the entry and full crossing of culverts by mammals. This study also employed several interesting innovations in data collection methods and statistical techniques.
What is the potential of water culverts to serve as pseudo-wildlife-passages?
Water culverts have received increasing attention for their potential to act as pseudo-wildlife-passages, facilitating animal movement under highways while serving their original purpose of water drainage. Their ubiquity shows a great potential; an even spacing of culverts under roads may already be mitigating the effects that roads have on animals to some degree, and may be improved upon with retrofitting.
Variables influencing the use of water culverts by mammals
This study aimed to (1) determine what variables influence the number of successful passages of water culverts by mammals, and to (2) parse the effects that these variables have on the entry into and subsequent full passage of water culverts by individual mammals. More specifically, the study addressed three research questions: (1) How many individuals from which species fully cross, and which culvert characteristics influence the rates of full passages? For individuals that are detected outside of the structure, what factors influence their (2) entry and (3) subsequent full passage of water culverts?
To answer these questions, the study evaluated the relationship between culvert use by mammals and 12 explanatory variables divided into two categories: structural characteristics of water culverts and surrounding environmental characteristics.
Combining animal track data with motion-sensing camera data
The study used cameras and animal track stations along a 20 kilometre stretch of Autoroute 10 in Southern Quebec (Canada). To monitor animal activity inside and outside the culverts, the researchers used motion-sensing infrared trail cameras: four cameras per culvert, two at each entrance, one directed inside and one outside (Fig. 1).
Fig. 1: Installation of trail cameras and track-boxes. A two-camera setup was employed at each entrance within each culvert. (Photo credit: Jonathan Cole, Steffy Velosa, and Mehrdokht Pourali.)
Fig. 2: Illustration of camera triggers that were used to confirm a full passage.
Full crossings were confirmed if an individual was seen on at least one camera from each entrance crossing completely from one side to the other. Any combination of camera detections were permitted to confirm full crossings, including all four cameras, both inside-facing cameras, both outward-facing cameras, or combinations of these (Fig. 2). This setup allowed the researchers to assess the attributes affecting mammals’ entry into the culverts and full crossing of culverts separately.
Animal footprint collection was employed as a tool to estimate species presence in the forest adjacent to the culvert sites. Two footprint-collecting track boxes were placed outside of each structure entrance, 40 meters away from the entrance on either side of the structure, and an additional 20 meters away from the road surface in the adjacent habitat (Fig. 3). The goal of these track boxes was to detect individuals within the adjacent habitat away from the highway as a rough index of mammal activity, including animals that may not venture out into the cleared shoulder of the road.
Fig. 3: Layout of the track-boxes in relation to each of the water culverts along Autoroute 10 in Quebec, Canada.
Noteworthy innovations used in the study
To our knowledge, this study is the first to successfully combine motion-sensing camera data from inside of water culverts with animal track-box data in the adjacent habitat to assess the relationship between mammals present in the habitat adjacent to the road versus animals within detection-range of the trail cameras. As far as we know, this research project also is the first that
(1) used a scheduled rotating motion-sensing camera study design to minimize detection bias present in individual cameras,
(2) utilized both outward- and inward-facing cameras to determine outside presence of animals as well as inside of culverts, and
(3) utilized zero-inflated negative binomial generalized linear mixed effects models to test for the presence of structural zeros in a road-related wildlife study, which is likely to become standard procedure in studies to determine factors that lead to habitat unsuitability, such as culvert water depth.
Main results: Only two species used the culverts on a regular basis
Overall, 20 species were observed outside of the drainage culverts, but only about half of them were ever detected making full crossings. Only two species highly tolerant to water, including common raccoons (Procyon lotor) and American mink (Neovison vison), were seen fully crossing the structures with regularity, whereas the number of full crossings was small (< 8) for all other species (Fig. 4). The study found that (1) the presence of water outside of and within water culverts resulted in site avoidance, (2) polyethylene culverts were strongly disfavored, (3) increasing distance from culvert entrance to habitat edge was related to more entries by the few species that actually used the culverts, (4) the luminosity of the moon had a positive effect on mammal activity and entry into culverts, and (5) that many more factors play a role in an animal’s entry into a culvert than its full crossing once it is already inside, i.e., while several variables influenced culvert entry, none had an influence on a mammal’s probability of complete passage once it had entered a culvert.
The ordinary water culverts along Autoroute 10 studied here were used much less than one would expect for designated wildlife passages, and the results suggest that they are unsuitable as substitutes for wildlife passages for mammals. It is conceivable that the water culverts might be used by more species and by a larger numbers of individuals in times of no or little water present in the culverts if wildlife fencing is installed that would guide the animals to the culverts. Therefore, we recommend the use of designated wildlife passages and fences, with culvert entrances being situated on the habitat-side of the fence (rather than the highway-side) and with sections that end at the entrances to culverts, and that in places where wildlife passages are not feasible, dry ledges be installed in existing water culverts to better allow small and medium-sized mammals to safely cross under roads while avoiding the water inside of the culverts.
Fig 4: Total numbers of detections and full crossings by species. Common raccoons and American mink showed the highest numbers of full crossings.
Limitations
The study covered 99 days at each water culvert, with cameras active at each of the 13 sites for half of the sampling period (May 21 – December 4). A longer sampling period, multiple sampling seasons, a larger number of culverts, and a larger range of culvert materials and sizes would allow for a stronger statistical analysis. The plastic smell of the polyethylene culverts may be related to the fact that those culverts were relatively new, and it is conceivable that mammals might eventually get habituated to them after some time. A study that assesses the activity of animals both before and after a predictor variable is modified would provide an excellent opportunity to test the effect of the variable on the response variable.
Author information:
Benjamin Brunen(1), Caroline Daguet(2), Jochen A. G. Jaeger(1)
1 Department of Geography, Planning and Environment, Concordia University Montreal, 1455 de Maisonneuve Blvd. West, Suite H1255, Montréal, Québec, H3G 1M8, Canada.
2 Appalachian Corridor, 37 des Pins Sud, Eastman, QC, J0E1P0, Canada.
Benjamin Brunen completed his BSc and MSc at Concordia University in Montreal, Canada. He is now working as an environmental scientist at Black Fly Environmental LTD and as a photographer in Edmonton, Alberta, Canada. Caroline Daguet is a biologist and has worked for the NGO Appalachian Corridor in Quebec, Canada. Jochen A. G. Jaeger has studied the effects of roads and traffic on wildlife and on landscapes for more than 20 years. He is a researcher and teacher at Concordia University in Montreal, Canada.
Source:
Brunen, B., Daguet, C., Jaeger, J.A.G. (2020): What attributes are relevant for drainage culverts to serve as efficient road crossing structures for mammals? Journal of Environmental Management 268: 110423 (12 pp.). https://doi.org/10.1016/j.jenvman.2020.110423
For requesting a PDF of the article, please send an email to J. Jaeger at jochen.jaeger@concordia.ca
Editor:
Tony Clevenger
Cite this summary:
Brunen, B. & Jaeger, J.A.G. 2021. What attributes are relevant for drainage culverts to serve as efficient road crossing structures for mammals? Edited by Clevenger, A. TransportEcology.info, Accessed at: https://transportecology.info/research/attributes-for-water-culverts-as-crossing-structures [Date accessed].
