Stream Invertebrate Ecology: Predation, Nutrients, and Alpine Benthos

Knowledge Graph (324 nodes, 1517 connections)

Research Primer

Background

Streams draining the mountains around the Rocky Mountain Biological Laboratory (RMBL) in Gothic, Colorado are home to a rich community of aquatic insects — mayflies, stoneflies, caddisflies, midges, and others — that live on and among the cobbles of the streambed. These benthic insects are the foundation of mountain stream food webs: they graze algae, shred leaves, and in turn feed fish, birds, and amphibians. Because the Gunnison Basin sits at the headwaters of the Colorado River, understanding what controls these insect communities has implications far beyond a single drainage. Changes in snowmelt timing, summer low flows, nutrient inputs, and land-use practices such as stream restoration all play out first in these high-elevation reaches.

Several concepts recur throughout the research. Predators shape prey populations not only by eating them (consumptive effects) but also by scaring them — triggering behavioral, physiological, and life-history changes called nonconsumptive effects. Fish and predatory stoneflies release chemical cues known as kairomones, which prey detect and respond to by hiding, drifting (being carried downstream in the current), or maturing at smaller sizes. Drift behavior often follows a diel periodicity, with many insects moving more at night to avoid visual predators like trout. Size at emergence — how large a mayfly is when it leaves the water as a winged adult — is an important fitness measure because larger females typically carry more eggs.

At the base of the food web, algae attached to rocks (periphyton) grow according to light, temperature, and nutrient limitation, especially the availability of phosphorus. When conditions are right, the stalk-forming diatom Didymosphenia geminata ("didymo") produces an algae bloom that blankets the streambed and restructures invertebrate habitat. Other key themes include recruitment limitation (whether the number of eggs laid constrains the size of later life stages), hydro-geomorphological constraints on where females can lay eggs, and the increasingly popular use of beaver dam analogues — human-built structures that mimic beaver dams — as a tool to restore degraded mountain streams. Researchers assess these dynamics using benthic community sampling, flow-through channel mesocosm experiments, and measures like ash-free dry mass of biofilms to quantify food resources.

Foundational work

The research tradition in Gunnison Basin streams was established by a series of studies in the 1970s and 1980s that documented the distribution, diet, and behavior of stream insects and their predators. J. David Allan's work in Cement Creek described how benthic insect diversity varies with substrate complexity and elevation (Allan, 1975), how brook trout feed selectively on drifting prey (Allan, 1981), and how nocturnal drift of larger mayflies functions as a predator-avoidance strategy (Allan, 1978). A four-year trout removal experiment surprisingly found little change in invertebrate densities, suggesting that prey exchange among stream patches can mask predator effects (Allan, 1982). Bobbi Peckarsky's parallel work demonstrated that mayflies detect stoneflies through chemical and tactile cues and respond with species-specific avoidance behaviors (Peckarsky, 1980).

These early findings set up a central puzzle: in enclosure experiments, predators often seem to have modest effects on prey numbers, yet prey clearly respond to them. Cooper and colleagues resolved part of this by showing that high rates of prey immigration and emigration between patches dilute measurable predator impacts (Cooper et al., 1990). Dodson and colleagues synthesized the growing literature on chemical and hydrodynamic signaling in freshwater systems, establishing that non-visual communication is central to predator-prey interactions in streams (Dodson et al., 1994).

Key findings

A major thread of research at RMBL has shown that the nonconsumptive effects of predators can be as important as direct predation. Peckarsky and colleagues demonstrated that predatory stoneflies reduce mayfly feeding, growth, and ultimately fecundity even when they cannot kill their prey (Peckarsky et al., 1993). Because mayfly egg number depends directly on adult body size, these sublethal effects translate into strong population-level consequences — a point formalized with demographic models (McPeek & Peckarsky, 1998). Whole-stream experiments confirmed the pattern in nature: adding trout chemical cues to a fishless reach reduced Baetis mayfly secondary production by 17% and caused individuals to emerge earlier and about 11% smaller (Koch et al., 2020), consistent with earlier manipulations showing roughly 20% size reductions under fish odor (pub_id:1931). Mayflies respond to specific cues including amino sugars in fish skin mucus (pub_id:614), and predator avoidance trades off against feeding — larvae only drift during daylight when food is scarce and predators are absent (pub_id:1394). Schmitz and colleagues argued more broadly that such risk-driven foraging shifts reshape nutrient cycling and food-web structure in ways qualitatively different from simple consumption (Schmitz et al., 2008).

A second thread concerns life history, recruitment, and dispersal. Female Baetis mayflies are selective ovipositors, preferring large emergent rocks in fast, splashing current (pub_id:1692), and — paradoxically — often lay more eggs in trout streams than in fishless ones (pub_id:1362). Local recruitment is driven by the timing and availability of suitable oviposition sites rather than by local adult emergence, implying that population regulation occurs primarily in the larval stage (Peckarsky et al., 2000). Downes and colleagues generalized these ideas across taxa, arguing that egg-to-juvenile transitions leave persistent imprints on population size, particularly where egg-laying habitat is limited (Downes et al., 2021). Climate-linked cues also matter: mayfly emergence in the East River tracks both water temperature and peak streamflow, and drought years bring earlier emergence (Harper & Peckarsky, 2006).

A third thread addresses bottom-up controls and emerging stressors. Nuisance blooms of the diatom Didymosphenia geminata are associated with low dissolved phosphorus, typically below about 2 ppb (pub_id:1090), although physical removal prevents regrowth even when phosphorus stays low (West et al., 2020). Nine years of East River surveys show that didymo blooms shift invertebrate community composition — favoring Chironomidae midges and reducing Heptageniidae mayflies — without necessarily reducing total biomass (Brogan et al., 2024), a pattern consistent with earlier comparative and experimental work (pub_id:1528, pub_id:1661).

Current frontier

Early work from the 1970s through the 1990s established the basic ecology of predator-prey and fish-invertebrate interactions in Gunnison Basin streams. Research from the 2000s and 2010s extended these findings into life-history theory, genetics, and the first detailed descriptions of didymo blooms. Since 2020, the field has pivoted toward climate change, restoration, and scaling. McIntosh and colleagues have argued for a new framework linking ecosystem size — the physical dimensions of river networks — to population and community dynamics, with implications for water management under contraction (McIntosh et al., 2024). Brogan and colleagues' multi-year didymo monitoring links low-flow years driven by climate change to community reorganization (Brogan et al., 2024), and experimental work is beginning to disentangle how temperature and nitrogen-to-phosphorus ratios jointly drive didymo growth (pub_id:148).

A rapidly growing body of recent RMBL work evaluates beaver dam analogues (BDAs) as restoration tools. Comparisons of BDA ponds, beaver-augmented BDA ponds, and natural stream pools show that beaver modification increases pond depth and surface area (pub_id:53), that temperature regimes differ substantially among habitat types, and that invertebrate communities in beaver-augmented ponds are a constrained subset of simpler BDA pond communities (pub_id:60). Studies of beaver pond age suggest that older ponds accumulate more taxa than newer ones (pub_id:253), though re-formed ponds may converge more quickly (pub_id:137). Complementing this landscape-scale work, recent oviposition studies are asking how rock size and embeddedness determine where female insects lay eggs (pub_id:44), reflecting a broader recognition that recruitment, not just larval survival, limits restoration success.

Open questions

Several questions remain open for the next decade. How will the combined pressures of warming water, earlier snowmelt, and extended low flows interact to reshape mayfly life histories, didymo dynamics, and fish-invertebrate coupling? Can beaver dam analogues be designed to reliably reproduce the thermal, hydrologic, and biological benefits of natural beaver-engineered systems, or does the absence of beavers themselves limit their restoration value? How do nonconsumptive predator effects, now well documented in the East River, propagate through nutrient cycling and energy flux between aquatic and terrestrial systems? And as river networks contract, what scaling rules govern where predators, prey, and recruitment limitation set community structure? Answering these will require continued long-term monitoring, experimental manipulations at reach and watershed scales, and tighter integration of invertebrate ecology with hydrology and restoration practice.

References

Allan, J. D. (1975). The distributional ecology and diversity of benthic insects in Cement Creek, Colorado. Ecology. →

Allan, J. D. (1978). Trout predation and the size composition of stream drift. Limnology and Oceanography. →

Allan, J. D. (1981). Determinants of diet of brook trout (Salvelinus fontinalis) in a mountain stream. Canadian Journal of Fisheries and Aquatic Science. →

Allan, J. D. (1982). The effects of reduction in trout density on the invertebrate community of a mountain stream. Ecology. →

Brogan, J. et al. (2024). Consequences of nuisance algal blooms of Didymosphenia geminata on invertebrate communities in Rocky Mountain streams. Freshwater Science. →

Cooper, S. D. et al. (1990). Prey exchange rates and the impact of predators on prey populations in streams. Ecology. →

Dodson, S. I. et al. (1994). Non-visual communication in freshwater benthos: an overview. Journal of the North American Benthological Society. →

Downes, B. J. et al. (2021). From Insects to Frogs, Egg-Juvenile Recruitment Can Have Persistent Effects on Population Sizes. Annual Review of Ecology, Evolution, and Systematics. →

Harper, M. P. & Peckarsky, B. L. (2006). Emergence cues of a mayfly in a high-altitude stream ecosystem: Potential response to climate change. Ecological Applications. →

Koch et al. (2020). Nonconsumptive effects of Brook Trout predators reduce secondary production of mayfly prey. Freshwater Science. →

McIntosh, A. R. et al. (2024). Ecosystem-size relationships of river populations and communities. Trends in Ecology & Evolution. →

McPeek, M. A. & Peckarsky, B. L. (1998). Life histories and the strengths of species interactions: combining mortality, growth, and fecundity effects. Ecology. →

Peckarsky, B. L. (1980). Predator-prey interactions between stoneflies and mayflies: behavioral observations. Ecology. →

Peckarsky, B. L. et al. (1993). Sublethal consequences of stream-dwelling predatory stoneflies on mayfly growth and fecundity. Ecology. →

Peckarsky, B. L. et al. (2000). Hydrologic and behavioral constraints on oviposition of stream insects: implications for adult dispersal. Oecologia. →

Schmitz, O. J. et al. (2008). From individuals to ecosystem function: toward an integration of evolutionary and ecosystem ecology. Ecology. →

West et al. (2020). Didymosphenia geminata blooms are not exclusively driven by low phosphorus under experimental conditions. Hydrobiologia. →

Concept (40) →

Protocol (26) →

Author (53) →

Publication (138) →

Document (2) →

Dataset (3) →