What is the cost of that fence? The impact of fences on the movements of ungulates in a hyper-arid landscape

This study demonstrates that fences and roads in the hyper-arid Greater Sossusvlei Namib Landscape significantly restrict ungulate movements and reduce home ranges, with varying degrees of impact across species like springbok and gemsbok, thereby highlighting specific barriers that require modification to restore habitat connectivity.

The Gist

Imagine the Namib Desert not as a static, empty wasteland, but as a giant, shifting puzzle. In this hyper-arid landscape, rain is like a surprise gift that appears in random spots, and food (grass) only grows where that rain falls. For the animals living there—Springbok, Gemsbok, and Hartmann's Mountain Zebras—survival means being nomads. They have to keep moving, constantly chasing these scattered "gifts" of water and grass across hundreds of miles.

Now, imagine someone decides to put up a giant, invisible wall across this puzzle. That's what fences and roads are doing to these animals.

Here is the story of the paper "What is the cost of that fence?" broken down into simple, everyday concepts:

1. The Setup: The Great Desert Chase

Think of the Greater Sossusvlei Namib Landscape as a massive, open playground. The goal for the animals is to run from the dry desert plains to the greener hills (the escarpment) whenever the weather changes.

• The Players:
  • Springbok: The agile, fast runners.
  • Gemsbok: The tough, long-distance travelers.
  • Hartmann's Mountain Zebras: The mountain climbers who are surprisingly good at jumping over obstacles.
• The Problem: In the past few decades, humans have built a web of fences (for farms and livestock) and roads. These act like giant stop signs or brick walls in the middle of the animals' running track.

2. The Experiment: Putting GPS Watches on Animals

The researchers didn't just guess; they put high-tech GPS collars (like smartwatches) on 40 animals. They tracked them for months and even years, watching exactly where they went, where they stopped, and where they got stuck.

3. The Findings: Who Gets Stuck and Why?

The "Smart" vs. The "Stuck"

• The Zebras (The Jumpers): Hartmann's Mountain Zebras are like the parkour experts of the animal world. They are strong and used to rocky terrain. They could often find gaps in the fences or jump over them. While fences slowed them down a bit, they could still get through.
• The Springbok and Gemsbok (The Trapped): These animals are like runners who hit a locked gate. They tried to cross, but the fences were too high or the gaps too small.
  • The Result: Instead of roaming hundreds of miles, these animals were forced to stay in small, fenced-in pockets. Their "home range" (the area they need to survive) shrank drastically. It's like being forced to live in a small apartment when you need a whole mansion to find enough food.

The "Wall" Effect
The study found that fences and roads didn't just block the animals; they trapped them.

• The "Fence Walk": Some animals, like a specific Gemsbok tracked for over a year, walked along a fence line for 38 kilometers (about 24 miles) trying to find a way through. They were essentially pacing back and forth, wasting precious energy, unable to reach the greener pastures on the other side.
• The Road Barrier: Major roads (like the C14 and C19) were the biggest culprits. Because these roads are often lined with fences on both sides, they act like a corridor of death. Animals get stuck in the strip of land between the road and the fence, unable to cross to the other side.

4. The Cost: Why Does This Matter?

The paper asks, "What is the cost of that fence?" The cost isn't just money; it's life and death.

• Starvation: If a drought hits the side of the fence where the animals are stuck, they can't run to the side where it's raining. They starve because the fence cut off their escape route.
• Genetic Isolation: If animals can't move between different groups, they end up breeding with their own relatives. This is like a small town where everyone is related; eventually, the population gets weak and sick.
• Overcrowding: When animals are trapped in a small area, they eat all the grass there. The land gets ruined, and the animals suffer.

5. The Solution: Cutting the Knots

The researchers aren't saying "tear down every single fence." They know farmers need fences for their livestock. However, they are saying: "Be smart about which fences stay up."

• The "Pinch Points": They identified specific roads and fence lines (like the C14, C19, and C27) that are acting as the biggest bottlenecks.
• The Fix:
  • Remove specific sections: Take down fences along low-traffic roads to let animals pass.
  • Create "Wildlife Doors": Instead of a solid wall, leave gaps or build underpasses so animals can slip through.
  • Speed Limits: On roads where fences are removed, slow down the cars so animals don't get hit.

The Bottom Line

Think of the landscape as a giant river of life. The fences are like dams blocking the flow. The animals are the water trying to reach the sea. If you block the river, the water stagnates and dries up.

This study is a plea to landowners and the government: Don't build walls that trap life. By removing just a few key fences or making them "porous" (full of holes), we can let the animals run free, find their food, and survive the harsh desert, ensuring that the Namib remains a wild, living place rather than a trapped zoo.

Technical Summary

1. Problem Statement

In hyper-arid environments like the Namib Desert, ungulate survival depends on high mobility to track ephemeral resources (rainfall, forage, surface water) that are patchily distributed. The Greater Sossusvlei Namib Landscape (GSNL) aims to restore historical wildlife movement corridors between the Namib Desert and the Great Western Escarpment. However, the landscape is fragmented by linear infrastructure, specifically fences (for livestock/game farming and legal requirements) and roads.

While the ecological need for connectivity is recognized, the specific spatial extent to which different barriers restrict the movement of key ungulate species (Springbok, Gemsbok, and Hartmann's Mountain Zebra) remains poorly quantified. The study addresses the hypothesis that fencing and roads fragment the required range of these animals, potentially leading to reduced fitness, inbreeding, and local population declines.

2. Methodology

Study Area:
The research focused on the GSNL in Namibia, encompassing the Namib-Naukluft National Park (NNP), adjacent freehold farms, and private reserves (e.g., NamibRand). The area was divided into three spatial regions: North, Mid, and South.

Data Collection:

• Subjects: 40 adult ungulates were fitted with satellite telemetry collars:
  • 12 Springbok (Antidorcas marsupialis)
  • 13 Gemsbok (Oryx gazella) – all females to track herd dynamics.
  • 15 Hartmann's Mountain Zebra (Equus zebra hartmannae) – mixed sex.
• Duration: Tracking periods ranged from ~910 to ~3,124 days per species, generating tens of thousands of location points.
• Preprocessing: The first two weeks of post-capture data were removed to eliminate behavioral artifacts caused by handling and collaring.

Analytical Approach:
The study employed a spatially adaptive Binomial Generalised Additive Model (GAM) using the R package MRSea.

• Model Structure: The models partitioned animal distribution into two components:
  1. Environmental drivers: A spatially adaptive 2D smooth term capturing underlying habitat preferences.
  2. Barrier effects: A flexible term for the distance to the nearest barrier (fence or road).
• Comparison Strategy: To isolate the "barrier effect," the authors compared predictions from the full model (environment + barrier) against a simpler model (environment only).
• Identification of Critical Barriers: By calculating the difference between the two models, the study identified grid cells where animal density was significantly higher than expected near a barrier (indicating blockage/accumulation) or where crossings were absent.
• Home Range Estimation: Defined as the grid cells encompassing the top 90% of relative presence probabilities, with uncertainty estimated via parametric bootstrapping.

3. Key Results

Home Range Sizes:

• Springbok: Smallest mean home range (140 km²).
• Hartmann's Zebra: Intermediate mean home range (236 km²).
• Gemsbok: Largest mean home range (282 km²), nearly double that of springbok.
• Observation: Home ranges were generally smaller in areas with high barrier density (fences/roads) compared to open, unfenced areas (e.g., NamibRand), suggesting restricted movement.

Species-Specific Barrier Responses:

• Hartmann's Mountain Zebra: Demonstrated the highest permeability. They frequently crossed fences and were often found close to barriers (within 1 km). However, specific sections of the NNP fence (between C28 and D1962) and the C19 road acted as hard barriers.
• Gemsbok: Showed significant restriction. They tended to stay 3–5 km away from barriers but were frequently blocked by specific farm fences and the C19 road in the southern region. One tracked individual (SAT1107_1769) traveled along a 38 km road fence and 21 km of farm fence for 16 months without successfully crossing, highlighting severe movement inhibition.
• Springbok: Highly restricted. They accumulated near barriers (<1 km) but struggled to cross them. They were unable to cross the NNP fence line effectively and were blocked by the C14 and C19 main roads.

Critical Barriers Identified:
The study pinpointed specific infrastructure causing the most resistance:

• Main Roads: C14 (Coast to Maltahöhe) and C19 (Solitaire to Maltahöhe) were the most significant barriers for all three species.
• District Roads: C24, C27, D0850, D0855, and D0872.
• Fences: The NNP boundary fence (specifically the northern section) and specific livestock farm boundaries (e.g., Weltevrede farm) acted as solid barriers for springbok and gemsbok, while zebra showed more ability to traverse them.

4. Key Contributions

• Quantification of Barrier Effects: The study moves beyond qualitative observations to statistically quantify the "cost" of fences by isolating barrier effects from environmental drivers using advanced GAMs.
• Species-Specific Nuance: It highlights that barrier permeability is species-specific; Hartmann's zebra are significantly more capable of crossing fences than springbok or gemsbok in this landscape.
• Spatial Prioritization: The research provides a spatially explicit map of "choke points," identifying exactly which road segments and fence sections require modification to restore connectivity.
• Methodological Innovation: The use of a "counterfactual" modeling approach (comparing scenarios with and without barriers) allows for the identification of specific barrier locations that deviate from general landscape permeability.

5. Significance and Recommendations

Ecological Significance:
The findings confirm that linear infrastructure fragments the hyper-arid landscape, forcing ungulates into smaller home ranges and preventing access to ephemeral resources essential for survival in a changing climate. This fragmentation poses risks of inbreeding, reduced genetic diversity, and population decline.

Management Recommendations:

• Targeted Removal: A blanket removal of all fences is not recommended due to safety and agricultural needs. Instead, the study advocates for the targeted removal or modification of specific high-impact barriers (C14, C19, and associated farm fences).
• Mitigation Strategies:
  • Fence Gaps: Creating fixed fence gaps or removing specific strands in low-traffic areas.
  • Road Management: Removing fences along low-traffic roads (e.g., C27 through NamibRand) combined with reduced speed limits and warning signs has shown promise without increasing vehicle-wildlife collisions.
  • Prioritization: Focus conservation efforts on the identified "hard barriers" (C14, C19, and specific NNP fence sections) to maximize the restoration of movement corridors for the most vulnerable species (Springbok and Gemsbok).

This study provides the scientific basis for the GSNL's goal of creating a "fence-free Namib," offering a data-driven roadmap for balancing agricultural needs with critical wildlife conservation.