The transition area between rivers and their adjacent riparian aquifers, which may comprise the hyporheic zone, hosts important biochemical reactions, which control water quality. The rates of these reactions and metabolic processes are temperature dependent. Yet the thermal dynamics of riparian aquifers, especially during flooding and dynamic groundwater flow conditions, has seldom been studied. Thus, we investigated heat transport in riparian aquifers during 3 flood events of different magnitudes at 2 sites along the same river. River and riparian aquifer temperature and water‐level data along the Lower Colorado River in Central Texas, USA, were monitored across 2‐dimensional vertical sections perpendicular to the bank. At the downstream site, preflood temperature penetration distance into the bank suggested that advective heat transport from lateral hyporheic exchange of river water into the riparian aquifer was occurring during relatively steady low‐flow river conditions. Although a small (20‐cm stage increase) dam‐controlled flood pulse had no observable influence on groundwater temperature, larger floods (40‐cm and >3‐m stage increases) caused lateral movement of distinct heat plumes away from the river during flood stage, which then retreated back towards the river after flood recession. These plumes result from advective heat transport caused by flood waters being forced into the riparian aquifer. These flood‐induced temperature responses were controlled by the size of the flood, river water temperature during the flood, and local factors at the study sites, such as topography and local ambient water table configuration. For the intermediate and large floods, the thermal disturbance in the riparian aquifer lasted days after flood waters receded. Large floods therefore have impacts on the temperature regime of riparian aquifers lasting long beyond the flood's timescale. These persistent thermal disturbances may have a significant impact on biochemical reaction rates, nutrient cycling, and ecological niches in the river corridor.
Hydropeaking, the alternating storage and release of water from reservoirs for hydropower generation, perturbs the thermal regime of many large rivers. While its effects on river temperature have been long studied, impacts on the thermal regime of riverbeds remain mostly unknown, despite riverbed temperature affecting rates of nutrient cycling and habitat suitability for benthic organisms. This study combines detailed field observations and flow and heat transport modeling to assess the impact of hydropeaking on riverbed temperatures in a large regulated river. The field site was 12 km downstream from a dam that induces large daily flow variations. Vertical thermistor arrays were used to collect riverbed temperature data across the entire 70 m‐wide channel. The riverbed near the left bank was highly dynamic thermally, transitioning between river and groundwater temperatures over daily hydropeaking cycles. In contrast, the rest of the riverbed, including near the right bank, was similar in temperature to the river and had relatively stable temperatures. Modeling showed that the temperatures near the banks are explained by advective heat transport driven by hydrostatic changes in river level, while the temperatures over the rest of the channel can be explained mostly by conductive heating. Gaining groundwater conditions and high sediment hydraulic conductivity favor thermally dynamic zones near banks, while low hydraulic conductivity (below 1 m/d) and neutral or losing groundwater conditions result in muted temperature fluctuations, as observed at the right bank. These spatial patterns can help predict thermally sensitive processes in the riverbeds of hydropeaked or flooding rivers.
more » « less- NSF-PAR ID:
- 10449063
- Publisher / Repository:
- DOI PREFIX: 10.1029
- Date Published:
- Journal Name:
- Water Resources Research
- Volume:
- 57
- Issue:
- 4
- ISSN:
- 0043-1397
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
more » « less
Abstract -
more » « less
Abstract Dynamic hydrologic exchange flows in river beds and banks are important for many ecosystem functions throughout river corridors. Here we test whether the exchanges and the associated mixing between a flooding river and groundwater within the river’s bank can be effectively traced by Radon‐222 (222Rn), a naturally occurring, inert, radiogenic, and radioactive gas that can be analyzed and monitored in situ. The assessment was done by simulation of groundwater flow and reactive transport of222Rn in the bank following a single, relatively rapid (hours long) flood wave and auxiliary field observations of222Rn, temperature and total dissolved solids (a surrogate for any ionic conservative tracer). Results illustrate that222Rn is more effective than temperature and total dissolved solids in tracing dynamic hyporheic exchange.222Rn variations in space and time are larger than the analytical uncertainty of common measurement methods. The individual effects of aquifer hydraulic conductivity, dispersivity, river water222Rn concentration, and bank topography were analyzed through sensitivity analysis. Larger hydraulic conductivity and dispersivity, lower222Rn concentration in river water relative to groundwater, and gentler bank slopes resulted in a more prominent and traceable222Rn signal. The transport and residence time of exchanged water may be estimated and interpreted using reactive transport models such as those implemented here. However, such application is sensitive to fluctuations in river water222Rn, requiring it to be well characterized. The assessment provides guidance for using222Rn as a tracer for groundwater and surface water interactions in dynamic settings.
-
more » « less
Abstract Bed material transport at river bifurcations is crucial for channel stability and downstream geomorphic dynamics. However, measurements of bed material transport at bifurcations of large alluvial rivers are difficult to make, and standard estimates based on the assumption of proportional partitioning of flow and bedload transport at bifurcations may be erroneous. In this study, we employed a combined approach based on observed topographic change (erosion/deposition) and bed material transport predicted from a one‐dimensional model to investigate bed material fluxes near the engineering‐controlled Mississippi‐Atchafalaya River diversion, which is of great importance to sediment distribution and delivery to Louisiana's coast. Yang's (1973) sediment transport equation was utilized to estimate daily bed material loads upstream, downstream, and through the diversion during 2004–2013. Bathymetric changes in these channels were assessed with single beam data collected in 2004 and 2013. Results show that over the study period, 24% of the Mississippi River flow was diverted into the Atchafalaya River, while the rest remained in the mainstem Mississippi. Upstream of the diversion, the bed material yield was predicted to be 201 million metric tons (MT), of which approximately 35 MT (i.e., 17%) passed through the bifurcation channel to the Atchafalaya River. The findings from this study reveal that in the mainstem Mississippi, the percentage of bed material diversion (83%) is larger than the percentage of flow diversion (76%); Conversely, the diversion channel receives a disproportionate amount of flow (24%) relative to bed material supply (17%). Consequently, severe bed scouring occurred in the controlled Outflow Channel to the Atchafalaya River, while riverbed aggradation progressed in the mainstem Mississippi downstream of the diversion structures, implying reduced flow capacity and potential risk of a high backwater during megafloods. The study demonstrates that Yang's sediment transport equation provides plausible results of bed material fluxes for a highly complicated large river diversion, and that integration of the sediment transport equation with observed morphological changes in riverbed is a valuable approach to investigate sediment dynamics at controlled river bifurcations.
-
more » « less
Abstract Peak flow events in gaining stream/river channels cause lung model hyporheic exchange with the banks (bank storage), which fosters beneficial reactions as polluted channel water cycles through riparian groundwater. Soil pipes are common along stream/riverbanks, and enhance exchange, yet their effect on reactions such as denitrification is unknown. We used MODFLOW with the Conduit Flow Package to simulate lung model exchange during a peak flow event in a section of stream bank/riparian soil with soil pipes, and MT3D‐USGS to estimate nitrate transport and denitrification. We varied soil matrix hydraulic conductivity (
K ) and first‐order reaction constant (k ), as well as soil pipe density, length, and height above the initial channel water surface elevation (H ). The addition of soil pipes enhanced stream bank (riparian) denitrification relative to banks without pipes, e.g., a 76% increase due to adding a single 1.5 m pipe. Denitrification increased linearly with pipe density but exhibited nonlinear trends with other parameters. Sensitivity analysis revealed length and density to be most influential. Soil pipe enhancement of denitrification was governed by hyporheic volume in most cases in our study. Exceptions included (a) coarse soil (K = 10−3 m/s) and (b) lowk andH > 0. Scaling our results to the stream corridor scale estimated that five soil pipes per m cumulatively induced 3% nitrate removal along a 1 km reach. Overall, soil pipes enhanced advection of nitrate into the banks, and also increased residence times of that nitrate under certain conditions, which together enhanced denitrification. This enhancement has implications for excess nitrate management in watersheds. -
more » « less
Abstract The hyporheic zone is the ecotone between stream and river channel flow and groundwater that can process nutrients and improve water quality. Transient hyporheic zones occur in the riparian zone (bank storage or “lung model” exchange) during channel stage fluctuations. Recent studies show that soil pipes are widespread in stream banks and beneath floodplains, creating highly preferential flow between channel and riparian groundwater such that the traditional Darcy model of flow does not apply. We used MODFLOW with the conduit flow package to model a series of stream bank soil pipes and examined soil pipe density (number per m), length, diameter, height above baseflow water surface, connectivity, and matrix hydraulic conductivity on transient particle flow paths and total hyporheic exchange volume (i.e., bank storage) over the course of a peak flow (e.g., storm) event. We found that adding five soil pipes per meter more than doubled hyporheic volume. Soil pipe length was the most important control; adding one 1.5‐m‐long soil pipe caused a 73.4% increase in hyporheic volume. The effect of increasing soil pipe diameter on hyporheic volume leveled off at ~1 cm, as flow limitation switched from pipe flow to pipe‐matrix exchange. To validate our approach, we used the model to successfully reproduce trends from field studies. Our results highlight the need to consider soil pipes when modeling, monitoring, or managing bank storage, floodplain connectivity, or hyporheic exchange.
