Seafloor video-acoustic monitoring in a Greenlandic glacial fjord records hyperbenthos, backward-swimming fish, and narwhals

This study demonstrates that autonomous video-acoustic monitoring in a Greenlandic glacial fjord effectively characterizes a turbulent, particle-rich seafloor environment and documents diverse biodiversity, including hyperbenthos, backward-swimming fish, and narwhals.

The Gist

Imagine you want to study the secret life of a busy city square, but you can't walk there, and you can't ask the people what they're doing. Instead, you decide to hide a tiny, silent camera on the ground, pointing up at the sky, and wait to see who walks by.

That is essentially what this team of scientists did, but instead of a city square, they were looking at the deep, dark floor of a glacier fjord in Greenland, and instead of people, they were watching fish, tiny sea creatures, and even narwhals (the unicorns of the sea).

Here is the story of their underwater adventure, broken down simply:

1. The "Invisible" Spy Camera

The scientists built a small, portable device that fits into a box about the size of a large suitcase. They dropped it 260 meters (about 850 feet) down to the ocean floor.

• The Trick: Most underwater cameras use bright white lights, which can scare animals away or attract them like moths to a bug zapper. This team used red lights.
• The Analogy: Think of it like a night-vision camera at a party. Most people can't see red light well in the dark, so the "guests" (the animals) don't realize they are being watched. It allowed the scientists to be "invisible" observers.
• The Setup: The camera pointed upward. Why? Because they were hoping to catch narwhals, which are known to swim down from above and sometimes bump into things on the sea floor.

2. What They Saw: A Busy, Turbulent World

When they reviewed the footage, they found a world that was far from quiet and still.

• The "Snow" Storm: The water was full of "marine snow." This isn't actual snow, but a constant shower of tiny bits of food, dead skin, and fibers drifting down from the surface. It was like a blizzard of confetti that never stopped.
• The Tiny Dancers: The most common visitors were microscopic creatures like copepods (tiny shrimp-like bugs) and amphipods. They weren't just floating; they were jumping and darting around. The scientists saw copepods "jumping" away when they bumped into the camera's rope, like a person jumping back when they accidentally brush against a stranger in a crowd.
• The Backward Swimmer: One of the coolest finds was a snailfish (a type of fish that lives in the deep). It was seen swimming backward, just drifting with the current like a leaf in a stream, before curling its tail and freezing.
• The Narwhal Visit: They heard narwhals every day using their hydrophone (underwater microphone), but they only saw one on camera. It didn't crash into the gear; it just swam by in the background, its tusk visible for a moment. It seems the "invisible" camera worked—the narwhals didn't seem curious or scared, they just ignored it.

3. The Tides as a Conductor

The scientists noticed something fascinating about the "marine snow." The direction the particles moved wasn't random.

• The Analogy: Imagine the ocean floor is a dance floor, and the tides are the DJ. When the tide goes out, the "dance floor" spins one way. When the tide comes in, it spins the other way. The tiny particles and the little animals were all dancing to the rhythm of the tides, moving back and forth every 12 hours.

4. Why This Matters

Before this study, we knew very little about what happens right above the ocean floor in these icy, remote places.

• The Problem: We usually study the deep sea by dragging heavy nets (which crush things) or using loud sonar (which scares animals).
• The Solution: This little box is a new, gentle way to peek into the deep. It's like switching from a sledgehammer to a stethoscope. It showed us that the deep ocean floor is a bustling highway of life, full of tiny creatures that are the food for bigger animals like seals and narwhals.

The Takeaway

This paper is a proof-of-concept. It shows that we can build a small, cheap, and non-intrusive "spy camera" to explore the deepest, darkest corners of the Arctic without disturbing the residents. It's a first step toward understanding how these remote ecosystems work, proving that even in the freezing dark, life is vibrant, busy, and full of surprises.

Technical Summary

1. Problem Statement

Arctic glacial fjords are biodiversity hotspots, yet their seafloor ecosystems remain poorly documented due to remoteness and logistical challenges. Existing monitoring methods have significant limitations:

• Invasiveness: Active acoustic devices and standard lights can attract or deter megafauna (e.g., narwhals), skewing ecological data.
• Resolution Gaps: High-precision optical profilers (e.g., UVP) often lack acoustic capabilities and may miss macro-zooplankton or larger fauna.
• Sampling Bias: Traditional methods like bottom sledges or traps are destructive and do not capture animal behavior in situ.
• Data Interpretation: Passive acoustic monitoring (hydrophones) often detects sounds but cannot verify the source or behavior of the emitting organism without visual confirmation.

The study aimed to deploy a compact, non-invasive, synchronized video-acoustic system to document the hyperbenthic environment (the water layer immediately above the seafloor), assess biodiversity, and evaluate the system's interaction with local wildlife, specifically narwhals (Monodon monoceros).

2. Methodology

Study Site and Deployment:

• Location: Inglefield Bredning, northwest Greenland (77°28.120'N, 66°21.411'W), at a depth of ~260 m.
• Duration: Deployed August 1, 2025, and recovered August 9, 2025 (approx. 8 days total; ~3 days of active video recording).
• Hardware Setup: A compact mooring system (~2.5 m long, <15 kg excluding anchor) consisting of:
  • Video: A LoggCAM synchronized with a hydrophone, equipped with two red LEDs (~660 nm) to minimize visual disturbance to cetaceans. The camera faced upward to capture animals approaching from above.
  • Acoustics: A SoundTrap ST600 hydrophone for continuous recording (96 kHz sampling rate).
  • Sensors: An acoustic release (Vemco) with internal sensors for temperature, pressure, tilt, and noise.
  • Deployment: Dropped from a boat to 260 m; recovered via acoustic release.

Data Processing:

• Visual Analysis: 223 video files (37 hours total) were manually reviewed. Taxonomic identification was limited to high-level taxa due to red-light resolution constraints.
• Automated Image Analysis:
  • Particle Tracking: Background subtraction and binarization were used to count "marine snow" particles and calculate the area covered by suspended matter.
  • Flow Dynamics: Particle Image Velocimetry (PIVlab) was applied to 10 frames per video file to determine flow speed and direction.
  • Color Analysis: Mean RGB intensity was tracked to assess light penetration and particle reflectivity.
• Acoustic Analysis: Long-term spectrograms were generated from the continuous SoundTrap data to correlate sounds with visual events.

3. Key Results

Environmental Conditions:

• The environment was highly turbulent with abundant suspended particles ("marine snow") and fibers.
• Water temperature averaged -0.18°C.
• Particle flow direction and speed were strongly modulated by tides (12-hour oscillation), with a correlation coefficient of $r=0.54$ between sea-level change and particle motion angle.
• Marine snow density was highly variable, changing up to twofold within hours.

Biodiversity and Behavior:

• Total Detections: 478 organisms were identified across at least 11 taxa.
• Dominant Taxa:
  • Amphipoda: 47% of detections.
  • Copepoda: 26% (identified as Calanoida; observed jumping away from the mooring line, likely an escape response to mechanical contact).
  • Hydrozoa & Chaetognatha: 8% each.
  • Others: Decapoda (shrimp), Liparidae (snailfish), Pterotracheoidea (heteropods), and Ctenophora.
• Notable Behaviors:
  • Backward Swimming: A Liparidae (snailfish) was observed drifting backward with the current, curling its tail, and remaining motionless for 16 seconds.
  • Escape Response: Copepods exhibited "jumpy propulsion" when colliding with the mooring line, retracting antennulae along their bodies.
  • Narwhals: Acoustically present daily (except one day). Visually, a narwhal was seen only once (August 5), where a tusk tip appeared in the background ~tens of cm from the buoy. No physical interaction or disturbance of the mooring was observed.

System Performance:

• The red light limited the observation range to <25–100 cm but appeared non-invasive; no avoidance behavior was noted.
• The upward-facing camera successfully captured animals approaching from above, though it risked sediment accumulation (which was minimal in this specific deployment).

4. Key Contributions

• Methodological Innovation: Demonstrated the efficacy of a compact, portable, video-acoustic mooring system for Arctic seafloor exploration. It bridges the gap between passive acoustics (which detect presence) and visual confirmation (which identifies species and behavior).
• Non-Invasive Monitoring: Validated the use of red light (660 nm) and passive acoustics as a method that does not deter or attract megafauna like narwhals, addressing a critical concern in Arctic research.
• Ecological Insights:
  • Provided the first direct visual documentation of the hyperbenthos in this Greenlandic fjord.
  • Observed rare behaviors, such as backward swimming in Liparidae and mechanical escape responses in deep-water copepods.
  • Confirmed that tidal currents significantly drive particle transport and likely influence the distribution of prey for higher trophic levels (seals, birds, narwhals).
• Data Resource: Created a labeled dataset of Arctic deep-sea organisms and behaviors, serving as a baseline for future machine vision training.

5. Significance

This study proves that portable, low-cost video-acoustic systems are viable tools for exploring the poorly understood Arctic seafloor. By capturing both the physical environment (turbulence, particle flux) and biological activity simultaneously, the approach offers a holistic view of ecosystem dynamics.

The findings suggest that the hyperbenthic layer in glacial fjords is a dynamic, high-energy environment rich in biodiversity, serving as a critical foraging ground. The lack of disturbance to narwhals indicates that such systems can be deployed in sensitive areas without altering animal behavior, paving the way for more extensive, long-term ecological monitoring in the rapidly changing Arctic. The study also highlights the potential for subglacial plumes to transport deep-water organisms upward, making them accessible prey for surface-dwelling predators.