The effect of outdoor recreation on mammal habitat use and diversity revealed by COVID-19 closures

This study leverages COVID-19 park closures to demonstrate that increased outdoor recreation negatively impacts mammal habitat use and diversity, particularly for rare species, highlighting the need for adaptive management to balance recreation with conservation goals.

The Gist

Imagine two popular hiking trails in the Canadian mountains: Joffre Lakes and Garibaldi. For years, these trails have been packed with hikers, like a busy subway station during rush hour. But then, the world hit the brakes due to the COVID-19 pandemic.

Suddenly, one trail (Joffre Lakes) was shut down completely for over a year, while the other (Garibaldi) only closed for a few months before reopening. This accidental "pause" gave scientists a rare, natural experiment to answer a big question: What happens to the wild animals when the humans disappear?

Here is the story of what they found, told simply.

The Setup: The "Silent Year" vs. The "Busy Year"

Think of the two parks as two giant living rooms.

• Garibaldi was the living room where the party never really stopped; people kept coming and going.
• Joffre Lakes was the living room where everyone was told to leave for a year. It went quiet.

The researchers set up 39 "security cameras" (camera traps) in both parks to take photos of the wildlife. They wanted to see if the animals would come out of hiding when the humans left, and if they would go back into hiding when the humans returned.

The Cast of Characters

The scientists were looking at 15 different mammal species, ranging from the "big guys" (bears, wolves, cougars) to the "smaller guys" (deer, hares, marmots).

They had a theory based on a concept called the "Landscape of Fear." Imagine that humans are like giant, scary monsters to these animals.

• The Sensitive Ones: Big predators and large animals (like bears and wolves) usually have big territories and are easily stressed by humans. The scientists thought these animals would be the first to return to the trails when the humans left.
• The Tough Ones: Smaller, scrappier animals might not care as much about humans and might keep hanging out near the trails regardless.

What Actually Happened?

1. The "Big Guys" Didn't Come Out to Play (Much)
Surprisingly, even when Joffre Lakes was empty of humans for a year, the big, sensitive animals didn't flood the trails.

• The Result: There was only weak evidence that bears, deer, and martens avoided the trails when hikers were around.
• The Analogy: It's like if you left a party empty for a year, but the VIP guests (the bears) still didn't show up. It seems these animals might have gotten used to the noise and crowds over the years (a process called habituation), or they are just very good at hiding in the deep woods where the cameras couldn't see them.

2. The "Human Shield" Didn't Work
There is a theory called the "Human Shield Hypothesis." It suggests that prey animals (like deer) might hang out near humans because the humans scare away the predators (like cougars).

• The Result: The study found almost no proof of this. The prey animals didn't seem to use the hikers as bodyguards.

3. The Real Story: The "Rare Species Filter"
While the common animals (like deer and hares) didn't change their behavior much, the rare and elusive species told a different story.

• The Discovery: In 2020, when Joffre Lakes was closed, the scientists found a higher variety of species, including some very rare ones like the Canada Lynx and Grizzly Bear.
• The Filter: When the park reopened and hikers returned, these rare species seemed to vanish from the study area. It's as if the park became a "filter." The noisy, crowded environment filtered out the shy, sensitive species, leaving only the tough, tolerant ones behind.

The Big Takeaway: The "Slow Motion" Effect

The most important lesson from this paper is about time.

• When the humans left, the park didn't instantly transform into a wildlife paradise.
• When the humans returned, the wildlife didn't instantly vanish.

Think of it like a slow-motion movie. The changes in the animal community happen over years or decades, not days. The fact that rare species were only seen in the quiet year suggests that if we keep these parks too crowded, we might slowly lose these special animals without even noticing until it's too late.

Why Does This Matter?

This study is a wake-up call for park managers.

• The Good News: Many common animals can coexist with hikers. We don't need to close every park to save wildlife.
• The Bad News: The "quiet" species (the rare ones) are being pushed out by the noise.
• The Solution: We need to keep monitoring these parks for a long time. We can't just look at one summer and decide everything is fine. We need to find a balance where people can enjoy nature, but the shy, rare animals still have a quiet corner of the forest to call home.

In short: The pandemic gave us a glimpse of what a quiet world looks like. It showed us that while the "loud" animals can handle the crowds, the "quiet" ones need us to be careful, because they are the ones slipping through the cracks.

Technical Summary

1. Problem Statement

Protected areas (PAs) face increasing pressure from outdoor recreation, which can compromise conservation goals. While individual behavioral responses of wildlife to humans (e.g., avoidance, stress) are well-documented, there is a significant knowledge gap regarding how these individual responses translate into community-level changes in species richness, diversity, and composition. Furthermore, disentangling the effects of recreation from other covarying factors (seasonality, resource availability) is methodologically challenging. The authors sought to determine if increased recreation filters out sensitive species, leading to homogenized communities, and whether species traits (body size, trophic level, life history) predict sensitivity to human disturbance.

2. Methodology

The study utilized a quasi-experimental natural experiment created by the differential closure of two adjacent, highly visited PAs in British Columbia, Canada, during the COVID-19 pandemic:

• Study Sites: Joffre Lakes Park (closed for 14.5 months, April 2020–June 2021) and Garibaldi Park (closed for only ~3.5 months, reopening August 2020).
• Data Collection: 39 camera traps (CTs) were deployed across both parks (18 in Joffre, 21 in Garibaldi) from August 2020 to September 2022. Cameras were placed on and off hiking trails.
• Species Focus: 15 medium-to-large terrestrial mammal species (>250g). Seven common species met the sample size threshold for individual modeling; five rare carnivores were aggregated into a single model due to low detection rates.
• Recreation Metrics: Human detections were quantified as "Park-Month recreation intensity" (independent human detections per 100 camera days) to serve as a proxy for recreation pressure.
• Statistical Analysis:
  • Habitat Use: Bayesian Generalized Linear Mixed Models (GLMMs) with a negative binomial distribution were used to model monthly detection rates. Predictors included recreation intensity, proximity to trails (exponential decay function), NDVI (vegetation productivity), and canopy cover.
  • Community Diversity: Species accumulation curves (using the iNEXT package) were generated to estimate asymptotic species richness, Shannon diversity, and Pielou's evenness. Bootstrap hypothesis tests with Benjamini–Hochberg corrections were used to compare metrics between parks and years.

3. Key Contributions

• Natural Experiment Design: Leveraged the unique "Anthropause" (pandemic closures) to isolate the effects of recreation intensity on wildlife, providing a rare control condition (Joffre Lakes closed) against an open control (Garibaldi).
• Multi-Scale Analysis: Simultaneously analyzed individual species habitat use and community-level diversity metrics to bridge the gap between behavioral ecology and community ecology.
• Trait-Based Sensitivity: Explicitly tested the hypothesis that larger-bodied, longer-lived, and more carnivorous species are more sensitive to recreation, while also investigating the "human shield hypothesis" (where prey species benefit from predator displacement).

4. Key Results

A. Habitat Use (Individual Species):

• Weak Negative Associations: There was weak evidence that black bears, mule deer, and American martens had lower detection rates during months of high recreation intensity.
• Spatial Shifts: Bobcats and hoary marmots showed weak evidence of shifting their habitat use closer to trails during high-recreation months, suggesting a potential (though limited) "human shield" effect or tolerance of linear features.
• Environmental Drivers: Canopy cover was a stronger predictor of habitat use than recreation for most species (e.g., positive for hares, negative for martens/marmots), indicating that structural habitat features often outweigh recreational pressure in determining site use.
• Rare Species: Rare carnivores (wolves, grizzly bears, lynx) were detected almost exclusively in Joffre Lakes (the park with the longer closure/lower recreation) and only at off-trail sites with high canopy cover.

B. Community-Level Diversity:

• 2020 (Closure Period): Species richness and Shannon diversity were significantly higher in the closed Joffre Lakes compared to the open Garibaldi Park (Mean difference in richness: -5.04; Diversity: -0.33). This supports the hypothesis that sensitive species are filtered out of high-recreation areas.
• Post-Closure (2021–2022):
  • In Joffre Lakes, diversity remained relatively stable between 2020 and 2021 but was significantly lower in 2022 compared to 2020.
  • In Garibaldi, richness decreased significantly from 2020 to 2021, contrary to expectations of stability.
  • Evenness: Community evenness remained consistent across years and parks, suggesting that while rare species may be filtered out, the relative abundances of the remaining common species did not shift dramatically.

5. Significance and Implications

• Filtering Effect: The study provides evidence that high recreation intensity acts as a filter, potentially excluding sensitive, rare species (e.g., apex predators) from protected areas, leading to a loss of biodiversity even if common species persist.
• Context Dependency: The lack of strong avoidance behaviors in common species (like deer and bears) suggests that long-term exposure to recreation may lead to habituation, or that species-specific traits and local context (e.g., availability of refuge habitats) mediate responses more than recreation intensity alone.
• Management Recommendations:
  • Long-term Monitoring: Short-term studies may miss delayed or cumulative effects (e.g., the drop in diversity in Joffre Lakes was only apparent in 2022, two years after reopening).
  • Adaptive Management: Managers must prioritize the protection of low-disturbance refuges (off-trail, high-canopy areas) to support sensitive species.
  • Balancing Goals: While low-impact recreation appears compatible with the persistence of common species, the potential loss of rare species necessitates rigorous monitoring to ensure conservation objectives are met alongside recreational access.

In conclusion, the paper highlights that while common mammals may tolerate or habituate to recreation, the cumulative effect on community structure involves the subtle filtering of rare and sensitive species, a process that requires long-term data to detect and manage effectively.