Search for: All records

Abstract Inland waters emit significant amounts of carbon dioxide (CO₂) to the atmosphere; however, the global magnitude and source distribution of inland water CO₂emissions remain uncertain. These fluxes have previously been “statistically upscaled” by independently estimating dissolved CO₂concentrations and gas exchange velocities to calculate fluxes. This scaling, while robust and defensible, has known limitations in representing carbon source limitations and spatial variability. Here, we develop and calibrate a CO₂transport model for the continental United States, simulating carbon transport and transformation in >22 million hydraulically connected rivers, lakes, and reservoirs. We estimate 25% lower CO₂fluxes compared to upscaling estimates forced by the same observational calibration data. While precise CO₂source distribution estimates are limited by the resolution of model parameterizations, our model suggests that stream corridor CO₂production dominates over groundwater inputs at the continental scale. Our results further suggest that the lack of observational networks for groundwater CO₂and scalable metabolic models of aquatic CO₂production remain the most salient barriers to further coupling of our model with other Earth system components. 

Abstract Chemical weathering in mountain critical zones controls river chemistry and regulates long‐term climate. Mountain landscapes contain diverse landforms created by geomorphic processes, including landslides, glacial moraines, and rock glaciers. These landforms generate unique flowpaths and water‐rock interactions that modify water chemistry as precipitation is transformed to streamflow. Variations in lithology and vegetation also strongly control water chemistry. Prior work has shown that landslides generate increased dissolved solute concentrations in rapidly uplifting mountains. However, there is still uncertainty regarding the magnitude which different geomorphic processes and land cover variations influence solute chemistry across tectonic and climatic regimes. We measured ion concentrations in surface water from areas that drain a variety of landforms and across land cover gradients in the East River watershed, a tributary of the Colorado River. Our results show that landslides produce higher solute concentrations than background values measured in streams draining soil‐mantled hillslopes and that elevated concentrations persist centuries to millennia after landslide occurrence. Channels with active bedrock incision also generate high solute concentrations, whereas solute concentrations in waters draining moraines and rock glaciers are comparable to background values. Solute fluxes from landslides and areas of bedrock incision are 1.6–1.8 times greater than nearby soil‐mantled hillslopes. Carbonic acid weathering dominates surface water samples from watersheds with greater vegetation coverage. Geomorphically enhanced weathering generates hotspots for net CO₂release or sequestration, depending on lithology, that are 1.5–3.5 times greater than background values, which has implications for understanding links among surface processes, chemical weathering, and carbon cycle dynamics in alpine watersheds. 

Abstract Rivers and streams play an important role within the global carbon cycle, in part through emissions of carbon dioxide (CO₂) to the atmosphere. However, the sources of this CO₂and their spatiotemporal variability are difficult to constrain. Recent work has highlighted the role of carbonate buffering reactions that may serve as a source of CO₂in high alkalinity systems. In this study, we seek to develop a quantitative framework for the role of carbonate buffering in the fluxes and spatiotemporal patterns of CO₂and the stable and radio‐ isotope composition of dissolved inorganic carbon (DIC). We incorporate DIC speciation calculations of carbon isotopologues into a stream network CO₂model and perform a series of simulations, ranging from the degassing of a groundwater seep to a hydrologically‐coupled 5th‐order stream network. We find that carbonate buffering reactions contribute >60% of emissions in high‐alkalinity, moderate groundwater‐CO₂environments. However, atmosphere equilibration timescales of CO₂are minimally affected, which contradicts hypotheses that carbonate buffering maintains high CO₂across Strahler orders in high alkalinity systems. In contrast, alkalinity dramatically increases isotope equilibration timescales, which acts to decouple CO₂and DIC variations from the isotopic composition even under low alkalinity. This significantly complicates a common method for carbon source identification. Based on similar impacts on atmospheric equilibration for stable and radio‐ carbon isotopologues, we develop a quantitative method for partitioning groundwater and stream corridor carbon sources in carbonate‐dominated watersheds. Together, these results provide a framework to guide fieldwork and interpretations of stream network CO₂patterns across variable alkalinities. 

Abstract Rivers and streams act as globally significant sources of nitrous oxide (N₂O) to the atmosphere, in part through denitrification reactions that will increase in response to ongoing anthropogenic nitrogen loading. While many factors that contribute to the release of N₂O relative to inert dinitrogen (N₂) are well described, the ability to predict N₂O yields from streams remains a fundamental challenge. Here, I revisit results from the second Lotic Intersite Nitrogen eXperiments (LINX II) in the context of turbulent hyporheic exchange. Denitrification efficiency, or the fraction of nitrate delivered to the streambed by stream turbulence that is chemically reduced, emerges as the single best predictor of N₂O yields and underpins the first statistically significant models of inter‐site N₂O yields. This mechanistic connection is supported by reactive transport modeling of hyporheic zone denitrification representing advective flowpaths, flowpath mixing, and diffusion‐dominated anoxic microzones. Simulated N₂O yields are inversely correlated with denitrification efficiency; however, advective models are unable to capture low LINX II N₂O yields at low denitrification efficiencies. Hyporheic zone mixing exacerbates this inability to capture observed N₂O yields via the promotion of N₂O release from fast, oxic flowpaths. Instead, anoxic microzones are required to account for LINX II observations through consistently low N₂O yields and the consumption of upstream‐produced N₂O. Together, these results provide a framework for controls on stream N₂O yields and suggest that stream corridor restoration designs aimed at increasing the capacity of hyporheic zones to remediate nitrate loading, as opposed to increasing hyporheic exchange, will also reduce proportional N₂O emissions. 

Abstract Inland waters are an important component of the global carbon budget. However, our ability to predict carbon fluxes from stream systems remains uncertain, aspCO₂varies within streams at scales of 1–100 m. This makes direct monitoring of large‐scale CO₂fluxes impractical. We incorporate CO₂input and output fluxes into a stream network advection‐reaction model, representing the first process‐based representation of stream CO₂dynamics at watershed scales. This model includes groundwater (GW) CO₂inputs, water column (WC), benthic hyporheic zone (BHZ) respiration, downstream advection, and atmospheric exchange. We evaluate this model against existing statistical methods including upscaling and multiple linear regressions through comparisons to high‐resolution streampCO₂data collected across the East River Watershed in the Colorado Rocky Mountains (USA). The stream network model accurately captures GW, evasion, and respiration‐drivenpCO₂variability and significantly outperforms multiple linear regressions for predictingpCO₂. Further, the model provides estimates of CO₂contributions from internal versus external sources suggesting that streams transition from GW‐ to BHZ‐dominated sources between 3rd and 4th Strahler orders, with GW, BHZ, and WC accounting for 49.3%, 50.6%, and 0.1% of CO₂fluxes from the watershed, respectively. Lastly, stream network model atmospheric CO₂fluxes are 4‐12x times smaller than upscaling technique predictions, largely due to relationships between streampCO₂and gas exchange velocities. Taken together, this stream network model improves our ability to predict stream CO₂dynamics and efflux. Furthermore, future applications to regional and global scales may result in a significant downward revision of global flux estimates.