Check that out! I have results!!! Like real ones, ones that mean something, ones that changed how I see the world, ones that will help us thinking differently about how bears and people are managed on trails. It’s not revolutionary but it’s close enough to make me do a little dance in my office and smile a lot when I think about it.
Remote Camera results
Although I will be putting remote cameras up again this year, I have now completed most of the data analysis from Fall 2013 to Summer 2014. It was a long slog through data cleaning and analysis, but I finally got some results that make sense (amazing and plenty of reason to dance around in and of itself).
The remote cameras took LOTS of pictures. In total, just under 150,000 images were entered into analysis from a total of 192 cameras on 34 different trail networks. Of these, 55 pictures were of grizzly bears, the rest were of people. So the first thing I learned is that there are a lot of people in the National Parks recreating all times of day and all days of the week. It’s astounding.
In a previous post, I talked about on a regression analysis to find which variables were the best predictors of whether or not a grizzly bear will be captured on camera. For this analysis I included the following variables:
habitat quality – because bears really want to hang out where the best food and habitat are. (Hypothesis: ↑ habitat quality; ↑ probability of bear photos).
human use category – a measure of how many people are frequenting the trail. (Hypothesis: ↑ human use; ↓ probability of bear photos).
- season – are bears more likely to be captured on human use trails during certain times of year. I didn’t have a hypothesis for this really.
time of day – when are bears using trails? (Hypothesis: ↑ bear photos during dawn/dusk).
activity type – what kinds of human activities on the trails lead to whether or not bears are captured on trails. (Hypothesis: ↑ probability of bear photos on trails where people are hiking; ↓ probability of bear photos on trails where people are biking).
distance to road – bears don’t really like roads. (Hypothesis: ↑ distance to road; ↑ probability of bear photos).
So what did I find? Sometimes my hypothesis was supported and other times it wasn’t.
- Probability of capturing a bear does increase with habitat quality.
- Probability of capturing a bear does decrease with increasing human use (high use trails have less pictures of bears).
- Probability of capturing a bear on a trail increases in the spring and decreases in the fall.
- Probability of capturing a bear on a trail is highest during the dawn/dusk, then night, then day.
- Probability of capturing a bear on a trail is related to activity type. The order of human activities that lead to capturing bears on cameras from highest to lowest is: hiker, runner, horse rider, vehicle, biker.
- Probability of capturing a bear decreased as the camera was farther from a road.
All of these results weren’t really perplexing… except the last one. What the??? Bears are more likely to be captured on camera close to roads? That’s the opposite to my hypothesis. What is going on??
Biking and Thinking
When I saw these results, I didn’t understand the distance to roads thing. I biked home in the Australian sun and talked to myself the whole way, trying to brainstorm reasons for what I was seeing, As in so much scientific research, one hypothesis led to another…
My hypothesis is that it comes down to barriers to movement. The cameras that are closer to roads are typically near towns or the transportation corridor of the Bow Valley. In this space, there are many barriers to grizzly bear movement: the Trans Canada highway, a railway, towns, the Bow River, steep slopes, and other human developments. All of these barriers combine to limit the number of options a grizzly bear has to move through the landscape. In the valley bottom, bears are literally more committed to walking on human use trails because that is the best, and sometimes only, way to move through the landscape. In the back country, bears have several options for movement. They can choose to walk on a human use trail, but if it’s busy they can walk on a wildlife trail or up a creek or on the other side of a wide open meadow.
Understanding why bears are more likely to use these trails closer to roads is my current focus and I’m using the bear GPS data to do that. My next steps will be to a least cost path and step selection function analyses in ArcGIS to test this hypothesis. Stay tuned for a blog post about that.
But there’s another perspective to consider….
Spinning it around
Why bears are on trails closer to roads doesn’t change the fact that they are. So if I just consider the fact that bears are more likely to be captured on trails closer to roads, and consider that capture probability on camera can be related to encounter probability (people running into bears on trails), then:
People have a higher chance of encountering a bear on a trail that is close to town or the highway, particularly during the spring, than they do of encountering a bear on a trail in the back country.
Once I started thinking about it this way, things shifted for me. If we are trying to minimize human-bear encounters, then we really need to think about the front country a bit more. I think that most wildlife managers are already doing this, but are we as a public thinking about it this way?
Bringing it all together
The visitor survey showed that people are not as prepared as they could be for recreating in bear country, especially on half day and full day hikes. About 45% of half day and full day hikers had bear spray with them, but over 80% of back country hikers had bear spray. This demonstrates a disconnect between trail users perceived level of risk of an encounter and actual probability of encounter.
It’s a challenge because I live in a mountain town and there are plenty of times that I just go out for a quick walk to get some fresh air and I don’t always grab my bear spray as I walk out the door. But the data shows that I actually have a higher chance of encountering a bear on the trail in this situation than I do in the back country. Will I be leaving the house without my bear spray again – not bloody likely!
I’m excited about these results because it’s a great example of why interdisciplinary research is necessary. The remote cameras provided one aspect of the story, the visitor surveys gave a human dimension to that, and my subsequent GPS analysis will add more detail to the bear perspective. At the end, it all comes together to help us understand how bears and people are using the same space at the same time – and that is why I’m doing this all in the first place.
So take your bear spray people, even if you’re just going on a little dog walk for 20 minutes. You never know what you’ll find out there! Check out this little video about how to use bear spray.