Streams hide their deepest oxygen secrets from the sensors that watch them

Every stream and river on Earth breathes. Oxygen dissolves into the water from the atmosphere, is generated by photosynthetic algae and plankton, and is consumed by microbes and animals from the water surface down into the bed sediments. For nearly seventy years, ecologists have tracked the daily rise and fall of dissolved oxygen to estimate how much carbon a stream produces and how much it consumes. But a new systematic review suggests that this familiar signal, so central to freshwater science, may be hiding far more than it reveals about where oxygen is actually being produced and consumed.

Researchers at Oregon State University, Virginia Tech, and the University of Florida synthesized 230 empirical studies published between 1964 and 2024, documenting which oxygen exchanges have been measured in streams and rivers worldwide, by what methods, and at what magnitudes. Their analysis, published in Frontiers in Water, reveals a striking imbalance: whole-stream metabolism and surface-atmosphere exchange together account for 84 percent of all reported measurements, while the individual compartments that actually generate the signal—the water column, the benthos, the hyporheic zone, and underlying sediments—remain largely unmeasured. Benthic exchanges appear in roughly 7 percent of studies, sediment oxygen demand in about 5 percent, water column metabolism in 4 percent, and hyporheic respiration in less than 1 percent.

The review builds on a framework first formalized by H.T. Odum in 1956, when he proposed an oxygen mass balance equation to describe how streams gain oxygen through photosynthesis and atmospheric reaeration and lose it through respiration. Gross primary production (GPP) reflects total oxygen generated by photosynthesis, while ecosystem respiration (ER) captures total oxygen consumed by autotrophs and heterotrophs. Their difference—net ecosystem production—has become the standard lens through which stream carbon dynamics are inferred. Yet the new synthesis demonstrates that this integrated signal emerges from the superposition of processes operating across distinct compartments, each quantified using different methods optimized for different spatial scales.

In-stream dissolved oxygen measurements dominated the methodological landscape, appearing in 81 percent of studies quantifying GPP, 79 percent quantifying ER, and nearly half of those measuring gas exchange. These approaches have evolved from labor-intensive Winkler titrations to continuous, high-frequency sensor deployments that allow reach-scale integration and Bayesian parameter estimation through tools such as streamMetabolizer. Gas exchange, the physical transfer of oxygen across the air-water interface, was most commonly quantified using tracer injections of gases such as propane, sulfur hexafluoride, and increasingly argon, which has emerged as an ideal tracer due to its similarity in solubility and diffusivity to oxygen. In contrast, water column, benthic, and sediment oxygen demand measurements relied overwhelmingly on chamber-based approaches, with field chambers accounting for 83 percent of water column studies, 64 percent of benthic studies, and 12 percent of sediment oxygen demand studies.

The magnitudes of these exchanges varied systematically with stream size and land use. Across all standardized fluxes, ecosystem respiration exceeded gross primary production on average (means of 8.0 and 4.4 grams of oxygen per square meter per day, respectively), confirming that most streams are heterotrophic, consuming more organic matter than they produce. Benthic respiration was both higher on average and more variable than benthic production, peaking in mid-sized streams at 15 grams of oxygen per square meter per day. Hyporheic respiration—the oxygen consumed along subsurface flowpaths where surface water exchanges with shallow groundwater—was markedly higher on average than sediment oxygen demand, though reported far less frequently. Gas exchange coefficients declined with stream size, from a mean of 66 per day in small streams to 7.0 per day in mid and large rivers.

Land use left a particularly strong fingerprint on subsurface exchanges. Benthic respiration peaked sharply in urban streams at 23.6 grams of oxygen per square meter per day, declining through mixed-use (7.9), agricultural (2.8), and undisturbed systems (1.1). Gas exchange coefficients showed the opposite pattern, highest in undisturbed streams at 127.9 per day and substantially lower in agricultural, mixed, and urban settings. These patterns suggest that human modification can disrupt the expected metabolic hierarchy along the river continuum, where undisturbed headwaters favor subsurface metabolism and increasing water column activity dominates downstream.

Only 11 studies in the entire synthesis—roughly 5 percent—reported concurrent measurements of individual process domains alongside whole-stream metabolism, yet this sparse dataset offers rare mechanistic insight. In small streams, hyporheic respiration was the largest contributor to ecosystem respiration, with a median contribution of 63 percent, while water column production dominated gross primary production at a median of 93 percent. In mid-sized streams, benthic respiration dominated ER at a median of 73 percent, with no hyporheic measurements available. In large rivers, water column respiration approached whole-stream ER at a median of 97 percent, while benthic production contributed minimally to GPP. These patterns broadly align with ecological expectations along the river continuum but do not account for differences in land use or disturbance, as small streams were primarily agricultural, mid-sized streams predominantly mixed-use, and large rivers represented agricultural, mixed-use, and undisturbed settings.

The review identifies methodological heterogeneity and scaling barriers as central challenges to building integrated stream oxygen budgets. Compartments are typically studied in isolation using methods optimized for specific spatial scales, ranging from patch-scale chambers and incubations that quantify sediment and benthic exchanges to reach-scale measurements of metabolism and gas exchange. This fragmentation is compounded by inconsistent reporting units: gas exchange is typically reported as a rate constant or velocity, but the literature contains at least 26 distinct gas exchange coefficients, although recent years have seen a convergence toward a normalized k600. Whole-stream metabolism is typically expressed in areal units, whereas water column, benthic, and hyporheic studies are often reported as volumetric rates, further complicating direct comparison without appropriate metadata.

The hyporheic zone emerges as the most under-resolved component of the stream oxygen budget. Hyporheic respiration was simultaneously the most variable exchange in the synthesis and the least frequently reported, appearing in less than 1 percent of studies. This pattern likely reflects both methodological heterogeneity and a disciplinary disconnect between hyporheic and stream metabolism research. Sediment oxygen demand, total oxygen demand, benthic chamber measurements, and hyporheic respiration resolve overlapping processes through distinct approaches with different spatial footprints and temporal assumptions, and reported values may reflect anything from surface-only demand to an unknown mixture of benthic and hyporheic contributions depending on chamber design and study site. The authors caution against differencing whole-stream metabolism estimates with independently measured benthic and water column exchanges, as this approach can introduce statistical uncertainty and process error in streams where hyporheic respiration contributes substantially to dissolved oxygen dynamics.

Resolving oxygen exchanges across compartments is increasingly urgent as environmental change reshapes the processes governing stream oxygen dynamics. Warming temperatures accelerate respiration while reducing dissolved oxygen concentrations, hydrologic fragmentation and expanding intermittency restructure subsurface connectivity, and eutrophication amplifies water column and benthic respiration in ways that whole-stream metabolism cannot resolve alone. Intermittent streams, which comprise roughly half of global rivers, were included in only 4 percent of studies in the synthesis, leaving less known about how whole-stream and compartment-specific respiration respond to changing discharge and hydrologic connectivity. The authors propose that addressing these gaps through concurrent multi-compartment measurements, standardized reporting, and cross-disciplinary integration is essential to advancing a mechanistic understanding of stream oxygen dynamics—one that moves beyond single-flux measurements toward quantifying oxygen production, consumption, and transport along hydrologic pathways spanning the water column, benthos, hyporheic zone, sediments, and groundwater.

Subject of Research: A systematic review of the methods and magnitudes of dissolved oxygen exchanges in streams and rivers

Article Title: A systematic review of the methods and magnitudes of dissolved oxygen exchanges in streams and rivers

Article References: Krause, J., Knutson, C., Hotchkiss, E. R., Cohen, M. J., & Ward, A. S. (2026). A systematic review of the methods and magnitudes of dissolved oxygen exchanges in streams and rivers. Frontiers in Water, 8, Article 1954977. https://doi.org/10.3389/frwa.2026.1954977

Image Credits: AI Generated

DOI: 10.3389/frwa.2026.1954977

Keywords: dissolved oxygen, stream metabolism, hyporheic zone, systematic review, gross primary production, ecosystem respiration, gas exchange, sediment oxygen demand, rivers, benthic metabolism, water column, oxygen budget