Fine-scale movements and habitat use of fish in intermittent rivers: Behavioural insights from drying refuge pools

By combining high-resolution mapping with behavioral observations, this study reveals that fish in intermittent river refuge pools exhibit size-dependent microhabitat preferences and movement patterns that persist through drying and leave lasting behavioral imprints even after rewetting, highlighting the critical conservation value of these structurally complex habitats.

The Gist

Imagine a river as a busy highway. Usually, cars (fish) can drive anywhere, but when the summer heat hits, the highway dries up, leaving only a few isolated puddles. This is what happens in "intermittent rivers." For the fish stuck in these puddles, it's like being trapped in a lifeboat while the ocean recedes.

This study is like putting a high-definition camera and a GPS tracker on these fish to see exactly what they do when they are stuck in these shrinking lifeboats. Here is the story of what they found, told in simple terms:

1. The Setting: The "Island" Puddles

The researchers studied a river in Spain that dries up every summer. When the water stops flowing, the river breaks apart into separate pools. The fish can't swim to a new home; they are stuck in their specific "island" until the rain comes back.

The team didn't just guess where the fish were. They used drones to take aerial photos and created a super-detailed 3D map of the pool's bottom, like a topographic map for a swimming pool. Then, they watched the fish for hours, tracking their every move.

2. The Big Discovery: It's Not a Blank Canvas

You might think that when a pool gets small, the fish just swim around randomly, like people milling about in a crowded room. They were wrong.

The fish were incredibly organized. They treated the pool like a city with specific neighborhoods:

• The "Deep End" Club: The big fish (adults) hung out in the deepest parts of the pool, near rocks and overhangs. It's like they were sitting in the VIP section of a club, safe from the heat and predators.
• The "Shallow Shore" Zone: The tiny baby fish (fry) stayed near the shallow edges. They were like toddlers playing in the kiddie pool, sticking to the margins.
• The "Sunbathers": Interestingly, the fish loved the sunny spots. Since the water wasn't getting hot enough to cook them, they likely went to the sun to see better and catch food, much like humans sitting in a sunny spot to read a book.

3. The "Size Matters" Rule

The most important finding was that size changes your address.

• The Big Fish: They were the explorers. They swam further, moved faster, and claimed the deep, safe, rocky areas. They were like the adults in a family who know the house best and stay in the living room.
• The Small Fish: They were the nervous explorers. They stayed close to the edge, moved in tight, zig-zag patterns (like a moth fluttering around a lamp), and didn't go deep. They were like kids playing tag in the corner of the yard, afraid to wander too far.

4. The "Shrinking Room" Effect

As the summer got hotter and the water level dropped, the pool got smaller. You might expect the fish to panic and scramble to find the deepest spot.

• What actually happened? They didn't panic. They kept doing exactly what they were doing before. The big fish stayed deep, and the small fish stayed shallow. Even though the "room" was getting smaller, they didn't rearrange their furniture. They just squeezed into the same spots they always liked.

5. The "Hangover" Effect (The Surprise Ending)

This is the coolest part. After the summer, the rain finally came, and the pools filled back up to their normal size. You would think the fish would go back to normal immediately.

• The Twist: They didn't. The fish that had been through the drying phase were slower and less picky about where they swam than the fish in the pools that never dried up.
• The Analogy: It's like a person who just survived a stressful week in a cramped office. Even when they get back to their spacious, comfortable home, they still move slowly and sit in the corner for a while. The "scars" of the drought stayed with them, changing their behavior even after the water returned.

Why Does This Matter?

This study teaches us that when rivers dry up, the remaining puddles aren't just empty water; they are complex, structured worlds where fish have specific rules for living.

• Conservation: If we want to save these fish, we can't just look at the size of the pool. We need to make sure the "deep neighborhoods" and "rocky hideouts" are preserved, because that's where the fish actually live.
• Climate Change: As the world gets hotter and rivers dry up more often, understanding these tiny, detailed behaviors helps us predict which fish will survive and which won't.

In short: Fish in drying rivers aren't just swimming randomly; they are playing a very specific, size-dependent game of "hide and seek" in a shrinking room, and the memory of that game changes how they act even after the water returns.

Technical Summary

1. Problem Statement

Intermittent rivers and ephemeral streams (IRES) are ecologically vital but increasingly threatened by climate change and land-use alterations, leading to more frequent and severe drying events. While the demographic impacts of drying (e.g., mortality, species composition shifts) are well-documented, there is a significant knowledge gap regarding fine-scale behavioral responses of fish during the critical "disconnected-pool phase."

• The Gap: Most existing research focuses on broad spatial scales (e.g., movement between pools or reaches) or population-level metrics. Little is known about how fish utilize space, select microhabitats, and move within isolated refuge pools once surface flow ceases.
• The Question: How do fish behaviorally cope with habitat contraction within a single refuge pool? Does body size influence microhabitat selection and movement? Do drying events leave lasting behavioral "carry-over" effects even after water levels recover?

2. Methodology

The study employed a high-resolution, multi-scale observational approach in the Borró stream (a Mediterranean intermittent river in Catalonia, Spain), focusing on the native Mediterranean Barbel (Barbus meridionalis).

Study Design & Sites

• Location: A 2.3 km reach containing both drying pools (seasonally isolated, depth declines) and non-drying pools (perennial, stable depth).
• Subjects: Six focal pools (4 drying, 2 non-drying) monitored over two months (May–July).
• Target Species: Barbus meridionalis, categorized into five size classes (fry <1 cm to adults >12 cm).

Data Collection Techniques

1. High-Resolution Pool Mapping:
   • Orthomapping: Drones (DJI Mavic Mini) captured aerial imagery to create geo-referenced orthophotos (~1 cm accuracy).
   • Bathymetry: Researchers waded pools to measure depth every ~50 cm and mapped refuge features (rocks, overhangs).
   • Grid System: Pools were overlaid with a 1 m² grid to create digital schematics for tracking.
2. Behavioral Observations:
   • Whole-Pool Grid Observations (GO): Observers scanned the entire pool grid to record fish presence/absence and size class in each cell.
   • Focal Follows (FF): Individual fish were tracked for 1-minute intervals to record trajectories, swimming speed, path straightness, and social behavior (shoaling).
3. Environmental Monitoring:
   • Recorded sun exposure (direct vs. shade), water temperature, and daily weather data.
   • Tracked water level changes relative to a reference point to calculate "relative depth" (standardizing depth across drying stages).

Data Analysis

• Statistical Models: Linear mixed-effects models (LMM) and binomial generalized linear mixed-effects models (GLMM).
• Variables: Analyzed pool-level occupancy, within-pool habitat associations (depth, refuge, sun), individual movement metrics (displacement, speed, turning rate), and temporal changes (drying vs. rewetting phases).
• Controls: Accounted for pool identity, date, time of day, and size class interactions.

3. Key Results

A. Non-Random Habitat Use

Fish did not use the available pool area randomly. They occupied only ~18% of the wetted area, consistently concentrating in specific microhabitats.

• Depth Preference: Fish strongly preferred deeper areas. This preference was significantly stronger in drying pools than in non-drying pools.
• Refuge Association: Fish were significantly more likely to be found in grid cells containing physical refuges (rocks, overhangs).
• Sun Exposure: Contrary to thermoregulation hypotheses, fish preferred sun-exposed cells, likely due to improved prey detection visibility.

B. Strong Size-Dependent Segregation (Ontogenetic Shifts)

Body size was the primary driver of behavioral structure:

• Small Fish (Fry/Juveniles): Concentrated in shallow margins, moved shorter distances, and exhibited tortuous (zig-zag) paths with frequent turning. They were more likely to use sunlit areas.
• Large Fish: Occupied deeper, refuge-rich zones, moved farther distances, and showed more directional movement. They were more social (shoaling) than the smallest or largest extremes.
• Movement Dynamics: Larger fish sampled broader areas of the pool, while smaller fish engaged in intensive local exploration.

C. Stability During Drying vs. Carry-Over Effects

• During Drying: Despite a ~21% reduction in pool surface area and significant shallowing, fish did not alter their fine-scale behavior. They maintained their relative positions (e.g., staying in the "deepest" part relative to the pool bottom) and movement patterns. They did not retreat further as water levels dropped; they simply occupied shallower absolute depths while maintaining relative depth preferences.
• After Rewetting (Carry-Over Effects): Once pools refilled to pre-drying levels, behavioral changes were observed:
  • Reduced Activity: Fish moved significantly less after rewetting compared to before drying.
  • Weakened Depth Bias: In previously drying pools, the strong preference for deep water weakened, converging toward the patterns seen in non-drying pools. This suggests the drying experience leaves a detectable "imprint" on behavior even after physical conditions normalize.

4. Key Contributions

1. Resolution of Scale: The study moves beyond reach-scale or pool-scale analysis to resolve fine-scale (cm to m) behavioral dynamics within isolated pools, revealing that refuge pools are internally structured habitats, not homogeneous water bodies.
2. Ontogenetic Behavioral Niches: It demonstrates that size-segregation persists even under extreme spatial confinement, with different life stages utilizing distinct microhabitats within the same pool.
3. Behavioral Plasticity vs. Rigidity: The findings challenge the assumption that fish constantly reorganize as habitats shrink. Instead, fish appear to maintain stable spatial organization relative to the pool's bathymetry until the habitat is completely lost.
4. Carry-Over Effects: It provides empirical evidence that drying history influences post-drying behavior (reduced activity, altered depth preference), highlighting a lag in behavioral recovery that previous studies may have missed.

5. Significance and Implications

• Conservation: The internal structure of refuge pools (depth gradients and refuge availability) is critical for fish persistence. Conservation efforts must look beyond just "keeping water" to ensuring the structural complexity of these pools remains intact.
• Climate Change Resilience: As intermittency increases globally, understanding that fish can maintain stable microhabitat use during contraction suggests a degree of resilience. However, the "carry-over" effects indicate that even mild drying events can alter population dynamics and behavior long after water returns.
• Methodological Advance: The combination of drone orthomapping, high-resolution bathymetry, and repeated focal follows offers a robust framework for studying animal behavior in constrained, dynamic environments, applicable to other taxa and ecosystems.

In conclusion, the paper argues that refuge pools are complex, structured environments where fish behavior is finely tuned to depth, refuge, and body size. The persistence of these patterns during drying, and their modification after rewetting, underscores the importance of protecting the physical integrity of intermittent river refugia.