Roadkill and Resilience: Why We Need More Wildlife Crossings and Less Habitat Fragmentation

by Galen Guerrero-Murphy on August 21, 2012

Deer Fawn

I was recently driving between Boston and New York and saw the sad remnants of a road-killed fawn in the highway. Just a few miles later on the Merritt Parkway (a nicely wooded, scenic highway dense with four lanes of Connecticut traffic and a Jersey barrier divider) I sped by another fawn, this time alive at the edge of the road with a spark of determination in its eye. It would surely bolt across the highway any second, thankfully not in front of me. There’s nothing like a really cute animal in imminent danger to awaken the senses.

Surely, we’ve all seen our share of roadkill. And often the big kills are unnoticed by the casual driver. I’ve seen hundreds of hatchling painted turtles, no more than 2 to 3 inches in diameter, flattened like pancakes on a two-laner while migrating from their sandy, upland nest site to the large wetland on the other side of the road. I’ve seen the soggy remains of countless salamanders, frogs, and toads after being caught on a road during their orchestrated emergence under the early rains of spring, migrating en masse from their hibernation sites to their wetland breeding grounds. I’ve seen dazed turkeys stuck on the narrow strip of pavement between traffic and a road divider, a fox in the headlights, and so on.

We build “walls” to species movement all of the time. Fish can’t swim through steep or perched culverts (culverts that outlet with a waterfall) installed under roads. Turtles can’t get over curbs. Small animals of all sorts are unable to cross railroad tracks. And for the animals that can make it into a road crossing pointed in the right direction, few stand a fighting chance of success.

Snapping Turtle Wildlife Crossing

This snapping turtle is out of luck with that curb up ahead.

On my way to work on I-95 northwest of Boston, I decided to get intimate with the animal crapshoot of crossing a road. Check out the video below, captured during a non-rush hour time of day. I decided not to play Frogger with my own life.

Sadly, we’ve carved up wildlife habitat with our pavement, our tracks, our fences, and our curbs. We’ve created extreme habitat fragmentation in many places of the globe, significantly affecting species’ life cycles and resilience, providing little recourse for their migrations.

Barriers to Migration

Many reptiles and amphibians, such as salamanders and frogs, hibernate each year in dry, upland burrows, crevices or other covered areas, and they emerge each season to travel to breed in distant vernal pools, wetlands or other suitable habitat. Once the job is done, they return to their hibernating areas, or “hibernacula.” Often a road intersects the breeding grounds and hibernacula.

And let me tell you, salamanders and frogs are terrible road crossers. Don’t even get me started on turtles.

Here’s another example. Many species of fish, including salmon and trout, attempt to swim upstream to lay their eggs but are often unable to pass poorly-designed culvert crossings. Each bad culvert–too steep, too shallow, perched outlets–effectively cuts off organism passage within a stream. Fishermen have felt the pain of diminished fisheries, which are often a result of carved up habitat.

The stories go on and on and on.

And not only are normal migratory patterns affected by our barriers, we’ve reduced the abilities of wildlife to randomly wander and recolonize habitat areas–behavior that creates exceptional resilience among dynamic populations. Let me explain.

Metapopulations and Resilience

The presence of redundant, interconnected nodes is a common pattern found in systems–this contributes to their resilience. From our electric transmission grid to our food supply network, we intentionally design redundancy into our human-built systems. For instance, if a single power plant goes offline we can easily tap into another power source so long as we’re connected to it with transmission lines. (In many cases our systems are not so redundant and, thus, not so resilient–a topic for another day.)

Similar patterns are found among species populations. Consider several populations of salamanders spread across a region. Perhaps a single, large population inhabits a large, productive wetland, and the population is growing. And maybe several smaller populations occur nearby in less expansive wetlands, and their populations are slowly decreasing in size.

If the populations are connected in some way (e.g., a salamander is conceivably able to walk or otherwise hitch a ride to the next population), they comprise what is known as a metapopulation. Each population is transient and dynamic, yet the metapopulation on a whole may be relatively stable. Individuals immigrate and emigrate, and populations come and go, but the region remains populated with salamanders.

In this example, the large, growing population is a source of salamanders for the metapopulation–occasionally a few may trek to the smaller populations, offsetting their natural decline. Or let’s imagine the large population experiences a particularly harsh year–the wetland becomes contaminated and all of the salamanders die. If the populations are still interconnected, it is possible that the smaller populations may become a source of salamanders that will eventually recolonize the larger population. The exact dynamics can be difficult to predict, but nevertheless we find that the general pattern of these interconnected population “nodes” contributes to substantial resilience of the overall metapopulation.

Metapopulation Network

A hypothetical metapopulation. Each node is a distinct population that is interconnected to others within the metapopulation. Some populations are sources (may import and export species) and some are sinks (only import species), and they may switch their source/sink status depending on ecological changes.

Metapopulation dynamics such as this are prevalent among species–and they clearly contribute to the resilience of biodiversity. Yet we hinder this resilience when we cut off the connections between animal populations with our roads, railways and other barriers to wildlife passage and crossing. We effectively stop metapopulation dynamics.

Reconnecting Wildlife

Fortunately, there are many ways we can support wildlife passage through our mazes of human activity. Pictured below are critter crossings installed on a new commuter rail line in Massachusetts to support movement of the spotted turtle (Clemmmys guttata) and other small animals between wetland habitats. Low fencing funnels the animals to narrow passageways that run underneath the tracks.

Railroad Turtle Crossing

Wildlife fence and crossings installed under the tracks of the Greenbush commuter line in Massachusets.

And below is an example of a wildlife crossing to accomodate small animal migration under a new, steep substation access road in Wakefield, Massachusetts, constructed of an open-bottom concrete span.

Open-bottom Wildlife Crossing

Open-bottom, concrete wildlife crossing installed under a new, steep substation access road in Wakefield, Massachusetts to accommodate small animal migration.

Certainly, these crossings can be expensive to build. But not so relative to the cost of a project. In the grand scheme of things, it is well worth the incremental investment to support wildlife passage and resilience. We spend billions of dollars building wildlife barriers (roads, railroads, etc.)–the least we can do is cough up a bit more for wildlife crossings.

Policies and regulations are beginning to require better crossing design and implementation, such as requiring wide, open-bottom culverts at all new stream crossings. This is a good sign. Still, for big change to occur, developers should consider voluntary implementation of wildlife crossings on their new projects and retrofits of existing ones. It’s an easy, no-brainer way to enhance wildlife habitat value in your community and cultivate your project’s sustainability and reputation.

Have you seen or built any cool examples of wildlife crossings? How else might we address habitat fragmentation?

Thumbnail photo by Eric Isselée via Other photos and images by Galen Guerrero-Murphy.

BioResilience PlantEmail Button     

Cimbria Badenhausen August 24, 2012 at 2:41 pm

Galen, this is a great site, I really enjoy reading your blogs. Good luck, and keep ‘em comin’! GreenMBAGirl


Galen Guerrero-Murphy August 24, 2012 at 11:07 pm

Cimbria, thank you for your kind words! Check back frequently (and consider subscribing to the newsletter) for much more biodiversity blogging on the horizon!


Rob Weingeist August 24, 2012 at 2:42 pm

Even house cats apparently travel 2+ miles to go hunting before returning to home!


Galen Guerrero-Murphy August 24, 2012 at 11:55 pm

Don’t tell my indoor city cat of this freedom she’s missing. Wow!

I spent some time several years ago working with California Red-legged Frogs (Rana draytonii) in northern California and took a wonderful training by Drs. Norman Scott and Galen Rathbun. They introduced me to a paper published by Dr. Scott, John Bulger and Richard Seymour (2003) that tracked migration patterns of the frog and found an individual had travelled 3,600 meters (2.24 miles!) between two sites 2,800 meters apart in under 38 days, hopping from pond to pond, and traversing 1,150 meters of uplands.

Another courageous individual travelled 3,200 meters (1.99 miles) between two pond sites, entirely on overland routes and in under 56 days.

And this is frog and cat scale, impressive as it may be.

The Porcupine Caribou (Rangifer tarandus granti) herd in northern Alaska and Canada travels 400 miles between summer and winter habitats and they have been observed to travel more than 3,000 miles a year. It’s sparse up there, but highways and pipelines create barriers to migration if not designed with suitable crossing areas.



Kathy Hipple August 24, 2012 at 9:25 pm
Galen Guerrero-Murphy August 24, 2012 at 10:57 pm

Kathy, excellent example, thanks for sharing!

There is a similar design implemented at Banff National Park, Alberta, Canada across the Trans Canada Highway. There’s a good photo of it and some other fun shots (including an alligator in a crossing) here:

And here’s a favorite from the same site of a couple badgers getting through a “badger culvert” crossing in BC:


Jono Thompson September 3, 2012 at 5:47 pm

I think this article hits the nail of the head in terms of habitat fragmentation and the barriers we have created for migrating species. Wildlife corridors in concept are fantastic but if not implemented properly can be a waste of time and money.

Often wildlife corridors are not wide enough and ‘edge effects’ come into play and often they are not designed with the animals in mind. A good example in Australia can be seen here: This rope bridge appears excellent in concept. However, studies have been conducted on these rope bridge crossings and the research suggests that they are barely used by the species in mind.

I think animal crossings should be mandatory in all new developments. However, the science must be conducted if they are to be successful.


Galen Guerrero-Murphy September 3, 2012 at 9:53 pm

Thanks for your comment, Jono. I’m sure many wildlife crossings are underutilized due to poor design, placement, etc. And on LinkedIn, a commenter brought up a wildlife crossing that had been fenced across. This highlights an interesting challenge–as our landscapes change around us, so too might the original effectiveness of crossings.

We must be pragmatic: implement (with a scientific basis), monitor, and adapt. We can nail the science up front, but the unexpected often occurs. Through ongoing monitoring, we can learn from what is working and what isn’t, and adapt our designs.

The importance of ongoing monitoring and responding to failure or unanticipated challenges cannot be understated. These are critical components of any pragmatic design, and they should certainly be incorporated into the long-term plans for wildlife crossings.


Jim Collinson September 8, 2012 at 7:27 am

Interesting article and comments. In Manitoba a proposed Bipole HVDC line will bisect winter woodland caribou range and put migratory birds from the Mississippi Flyway at risk of collisions in areas where they rest, feed and stage. The line will be 1386 km long and also bisect many farms in high productivity lands. An environmental hearing begins early October on this, but the Provincial Government seems determined to push it through!


Galen Guerrero-Murphy September 9, 2012 at 10:01 pm

Thanks for your comments, Jim! I’m assuming you are referring to the Manitoba Hydro Bipole III Transmission Project. There are usually many opportunities to mitigate these types of impacts, although I have not studied or worked on a project involving caribou, and I have only limited familiarity with this specific project.

Certainly, clearing forest for new or widened transmission line corridors can have an impact on species through loss of habitat, introduction of “edge effects,” creating new travel ways for predator and invasive species, etc. And construction activities can disturb species while work is underway. But generally, early successional habitats (e.g., meadow, scrub-shrub) created by new transmission line corridors can have a beneficial effect on many wildlife species (e.g., new habitat for many species of concern; increased availability of food, such as forage and berries), and a transmission line does not prohibit movement of species like roads, railroads and other projects can (except flying species…).

Based on the Environmental Impact Statement (EIS) for the Bipole III Project, the project will impact Boreal Woodland Caribou Habitat. The following mitigation measures will be implemented (not an exhaustive list): limiting construction to the winter to avoid disturbing calving females, preserving and maintaining low-growing trees in the new corridor (e.g., low-growing black spruce and tamarack), and conducting long-term monitoring of caribou and predator movement (e.g., wolves and black bear). A supplemental report on the potential caribou impacts was just released in August (2012) and is available here. It maintained that project-related impacts should not have a significant adverse effect on the caribou populations. This seems like a reasonable assessment based on what I read in the EIS and supplemental report. I’d be curious to hear from others who may have a closer relationship with this project and/or professional experience working with caribou.

I was interested to see that the EIS discussed the possible impact from increased access to habitat areas by hunters (via ATV/snowmobile along the new corridor). It seems to me that over hunting is a separate issue from the transmission line, but, nevertheless, they will attempt to restrict future ATV and snowmobile access along the corridor.

I’ve worked on a transmission project where we left young evergreens and “topped” larger trees within a new corridor to leave a somewhat forested habitat under the transmission lines. It’s worked fairly well, but this was in a very small area when compared to the Bipole III plan. I am quite curious to see how this turns out–what density of low-growing trees remain in the habitat areas after clearing and implementation of the mitigation measures. If the transmission line is high enough (as is the case with an HVDC), you can leave quite a bit of tree (subject to ongoing vegetative management) and still maintain the adequate clearances between vegetation and electrical wires (conductors).

In regards to birds and transmission lines, there is so much more that we can be doing to prevent electrocution and collision (and more and more good stuff being implemented on new projects, but we need retrofits too!). This includes installation of visual markers on the conductors (colored balls, spinning disks, streamers) and designing the line so all of the conductors are on the same horizontal plane. Resource agencies are more and more requiring these measures in high-risk areas, including on the Bipole III project.

Regarding bird electrocution, this occurs when the bird creates a circuit–so if it touches two of the conductor wires (“conductors”) or if it touches a conductor while perching/nesting on a grounded structure. Designs where conductors hang below structure cross-arms (where birds often perch and nest), and increasing distance between adjacent conductors can reduce the risk of electrocution.

Here is a great resource on designing transmission lines to minimize impacts to birds.

Really, the more we think and learn about these issues, the more we realize there are good ways to help mitigate many of the problems with a bit of added time and effort, and with a commitment to finding creative solutions for all involved stakeholders!


Tess O'Brien September 9, 2012 at 6:04 pm

Great article and really important topic that sadly seems not to get any attention. Human development has had massive impacts on wildlife migration and we’re all too familiar with the results.

Your article took me back to 2005 when, as a student studying environmental business and politics in Denmark, I took a trip to Croatia and was dumbfounded by the brand new highway – it’s cleanliness, the digital wind socks and the impressive tunnels through the hillsides. Most importantly, though, I was impressed by greenways that had been installed to facilitate wildlife migration over the highway every couple of miles. It was the first time I had seen such a thing and ever since, I’ve been looking for them here. Sadly, I have yet to see any in the US. I wonder what group would take this on as cause and try to champion it – perhaps even introduce legislation that would require any new roadway to address wildlife migration.

Thanks, Galen, great site and high-quality content.


Galen Guerrero-Murphy September 9, 2012 at 10:49 pm

Tess, thank you for the positive feedback and comment!

Here is a collection of wildlife bridge photos that include a few located in the US.

And here are details on US wildlife crossings collected through a partnership with the USGS, Utah State University and the Transportation Research Board.

Keep an eye out for underpasses, too–wildlife are often routed under roadways. Crossings are out there, but definitely not enough of them!

Check out ARC. They are working to raise awareness about the importance of wildlife crossings–they’re definitely a champion of crossings, and are up to good things.


Kylie January 15, 2013 at 6:04 pm

Please also check out the 41 wildlife crossing structures in Montana – see them at
Thanks for this great info!


Galen Guerrero-Murphy January 16, 2013 at 11:43 am

Kylie, thanks for sharing. This is a great site you linked to!


Cancel reply

Leave a Comment

Your email address will not be published. Required fields are marked *

Previous post:

Next post: