Search

International audience; Key Points: Lumped parameter modeling of hyporheic nitrate removal by applying the exposure time concept.  Exposure time distributions are derived from analytical residence time distributions and reaction kinetics for hyporheic oxygen consumption.  Using exposure times rather than residence times is likely to lead to more realistic estimates for nutrient removal in the hyporheic zone Abstract The rate of biogeochemical processing associated with natural degradation and transformation processes in the hyporheic zone (HZ) is one of the largest uncertainties in predicting nutrient fluxes. We present a lumped parameter (LPM) model that can be used to quantify the mass loss for nitrate in the HZ operating at the scale of river reaches to entire catchments. The model is based on using exposure times (ET) to account for the effective timescales of reactive transport in the HZ. Reach scale ET distributions are derived by removing the portion of hyporheic residence times (RT) associated with flow through the oxic zone. The model was used to quantify nitrate removal for two scenarios: 1) a 100 m generic river reach and 2) a small agricultural catchment in Brittany (France). For the field site hyporheic RT are derived from measured in-stream 222 Rn activities and mass balance modelling. Simulations were carried out using different types of RT distributions (exponential, power-law and gamma type) for which ET were derived. Mass loss of nitrate in the HZ for the field site ranged from 0-0.45 kg d-1 depending on the RT distribution and the availability of oxygen in the streambed sediments. Simulations with power law ET distribution models only show very little removal of nitrate due to the heavy weighting towards shorter flow paths that are confined to the oxic sediments. Based on the simulation results, we suggest that ET likely lead to more realistic estimates for nutrient removal.

Bank filtration is a cost-effective and sustainable method of improving surface water quality for drinking water production. During aquifer transit, natural biodegradation and physiochemical filtration improve the quality of the raw water by removing sediments, pollutants, and pathogens. Strict regulations prohibit the use of substances that can be used to estimate aquifer residence times to define water protection areas for bank filtration. In this study, we present a novel measurement and modeling framework for deriving mean aquifer residence times for bank filtrate using the natural tracer radon-222. The method is intended for application in the drinking water sector, where extraction wells are screened over the entire aquifer and pumps are operated at high production rates. Mean aquifer residence times are estimated using composite residence time distributions that account for flow path mixing and non-uniform residence times with multiple components including bank filtrate, shallow groundwater, and deep groundwater. The mathematical framework is demonstrated for a drinking water production facility. Radon activities for the six monitored extraction wells ranged between 4,400 and 8,400 Bq/m³. Estimated mean aquifer residence times for the wells range from5 days to 110 days and strongly depend on i) the type of residence time distribution model (exponential, gamma or piston flow), ii) the mixing ratio between bank filtrate and local groundwater, and iii) the heterogeneity in the groundwater endmember. By accounting for mixing processes, we can show that radon can be used beyond the "5-fold half-life" (~20 days) commonly described in the literature as the upper limit for age dating purposes for radon. This method provides a simple and cost-efficient way to quantify residence times of bank filtrate on a regular basis without any addition of external substances to the aquifer.

The increasing use of plastic products and its inevitable decomposition after improper disposal has led to large numbers of nano- and microplastic in aqueous environments. There is currently a critical need to investigate the transport and retention mechanisms of nanoplastic particles in water-saturated porous media (e.g., aquifers or sediments) to better understand residence times and ecosystem exposure of these particles in aqueous environments. In this study, we performed a set of column experiments in order to investigate and understand the primary controls on the mobility of nanoplastics in a controlled laboratory environment. As part of the experiments, we used polystyrene nanoplastic particles (PS-NPs, 50 nm) in combination with iron oxyhydroxide–coated sand, which is known for its high surface reactivity and often can be found in natural systems in environmentally relevant amounts. We also adjusted pore water chemistry (pH, ionic strength, cation species) to represent non-uniform geochemical conditions in nature and to understand how these conditions quantitatively affect the transport of nanoplastics. Mobility and retention of PS-NPs were assessed by analyzing breakthrough curves. For negatively charged iron oxyhydroxide coatings (at pH > pHpzc), only little retention of PS-NPs could be observed. In contrast, positively charged iron oxyhydroxide coatings (pH < pHpzc) provided favorable deposition sites for the negatively charged PS-NPs. DLVO theory was used to show that high pH and low ionic strength increased the energy barriers between PS-NPs and the porous media. In contrast, low pH and high ionic strength decreased the barriers and thus increased retention in the columns. Finally, bridging agents, such as Ca2+ and Ba2+, resulted in the significant deposition of nanoplastics by forming bonds between O-containing functional groups on both the plastic and sediment surfaces. These findings indicate that the deposition and fate of nanoplastic particles are strongly affected by the water chemistry and soil components in subsurface environments.

Quantifying and upscaling chemical turnover in the hyporheic zone (HZ) is difficult due to limited reaction rate data, unknown carbon quality, and few methods for upscaling local measurements to river networks. Here we develop a method for quantifying reaction kinetics in situ in the HZ and upscaling biogeochemical turnover to catchment scales. Radon-222 was used to quantify water residence times in the HZ of the Roter Main River (RM), Germany. Residence times were then combined with O2, NO3−, CO2, DOC, and carbon quality (EEMs, SUVA) data to estimate Monod and first-order reaction rates. Monod parameters µmax and ksat for NO3− reduction were 11 µmol l−1 h−1 and 52 µmol l−1, respectively, while the first-order rate was 0.04 h−1. Carbon quality was highly bioavailable in the HZ and is unlikely to be limiting. Reaction kinetics was incorporated into the FINIFLUX model to upscale NO3− mass loss over a 32 km reach of the RM. The aims were to (1) to estimate hyporheic efficiency using Damkohler numbers (Da), and (2) calculate NO3− mass loss in the HZ over the reach. The Da analysis suggests that the hyporheic zone is inefficient for NO3− processing, however, this is somewhat misleading as the largest NO3− mass loss occurs at the shortest residence times where Da ≪ 1. This is due to the largest water flux occurring in the uppermost part of the sediment profile. Nitrate processing in the HZ accounted for 24 kg NO3− h−1 over the reach, which was 20% of the NO3− flux from the catchment.

Coastal ecosystem health and sustainability is tightly coupled to submarine groundwater discharge (SGD) and associated nutrient, carbon and pollutant fluxes. However, there are few studies that systematically analyse the interaction between the terrestrial aquifer system, catchment morphology and coastal SGD. The objective of this study was to evaluate the role of catchment morphology and how this influences the spatial distribution, timing and volume of the SGD flux to a branched tidal creek system, on the barrier island Spiekeroog, Germany. The subsurface salinity was mapped using electrical resistivity tomography (ERT) and a hydrogeochemical survey of the shallow groundwater. Temporal and spatial analysis of geochemical tracers (radon-222, chloride) in the tidal creek water was integrated into a transient 222Rn mass balance model for quantification of SGD rates during two field campaigns that encompassed spring and neap tides. The ERT mapping indicated that fresh groundwater dominated under the dune ridges down to about 15 m depth, but became progressively more brackish seawards. This is likely due to frequent tidal and storm flooding of low-lying areas. The highest groundwater fluxes into the creek were indicated by high 222Rn activities (average 468 Bq m−3) towards the dune ridge. Chloride concentrations (up to 17.3 g L−1) increased seawards showing the progressive salinization of water in the creek. The freshwater component of SGD was highly variable in time but was highest at low tide, while the total SGD flux (saline + fresh) was highest when tides changed from inflow to outflow as the rapid pressure release on the local aquifer caused a large hydraulic gradient towards the creek. Comparing the freshwater component of mean daily SGD to the creek with estimated groundwater recharge rates in the catchment (665 m3 d−1) shows that the fresh groundwater discharge exceeds fresh recharge during spring tides (~120%) but is was lower than recharge during neap tides (~27%). In this study we show that tidal creeks and their relation to catchment morphology are relevant for understanding the spatial and temporal exchange of fresh and saline water between the catchment and coastal zone.

The flux of water, nutrients, carbon and salt through the subsurface at the land-sea interface is an important control on coastal nutrient processes, salinization of coastal aquifers and carbon balances of the coastal zone. However, these fluxes are often spatially and temporally complex and difficult to quantify, especially in high-energy mesotidal systems. Here we use vertical temperature profiles along a morphologically complex mesotidal high-energy beachface to map and quantify water infiltration and exfiltration on the island of Spiekeroog, Germany. Water fluxes were quantified using heat transport calculations from three solutions to the 1D heat transport equation, and include 1) a steady state analytical solution, 2) a non-steady state numerical model and 3) a non-steady state analytical solution. The temperature profiles could clearly map areas of upwelling warm (up to 10 °C) groundwater during the winters of 2018 and 2019. These upwelling zones were focused on an intertidal runnel system and at the low water line, consistent with the current understanding of the site based on visual observations and hydrogeological models. The steady state model provided good fits to the measured data in the winter when the seawater temperatures were not changing significantly, but was less able to reproduce the measured profiles in spring when seawater was warming. The steady state flux rates ranged from −110 to −43 mm d−1 in the runnel and low water line to +43 mm d−1 towards the high water line. The dynamic numerical model successfully captured the propagation of the seawater temperature signal into the subsurface and was able to reproduce the temperature profiles during both seasons. The flux estimates tended to be larger with the numerical model, with up to −150 mm d−1 in the runnel and +110 mm d−1 towards the high water line. The non-steady state analytical solution could only be applied to a limited time series due to the difficulty of logging temperatures in the subsurface at this highly dynamic site. Up to 1.5 days of data suggested fluxes that were considerably higher than the other two methods with best-estimates of −400 to −900 mm d−1. Thermal Peclet numbers ranged from 0.2 to 2 suggesting that both conduction and advection of heat is important. This study demonstrates that the morphology of the beach face is an important control on spatial distribution of down-welling and upwelling zones along the beach and that temperature measurement combined with heat modelling are potentially useful methods for understanding the interactions between groundwater and the sea.

Summary The concept of Damkohler numbers have been extensively used in the discipline of chemical engineering and lately increasingly found its application into environmental science in order to describe the integrated behavior of hydrological systems with respect to their physical transport and biogeochemical transformation capabilities. Defining characteristic time scales of transport and reaction, as part of the Damkohler concept, however is not trivial especially for non-well mixed systems like catchments where physically controlled transport and biogeochemical moderated reactions can be highly variable among individual flow paths. Often, system specific residence times alone are not useful to describe the timescales of transport in the Damkohler concept, because it neglects that degradation of redox-sensitive compounds depend on dynamically changing and non-uniformly distributed hydro-biogeochemical boundary conditions that either facilitate or suppress biogeochemical reactions. In this study an approach is presented that highlights the importance to specifically distinguish between residence and exposure times if system specific transformation efficiencies are evaluated. We investigate the inter-relationship between residence and exposure time distributions for different biogeochemical processes in a virtual wetland environment that is exposed to different hydrological conditions. The relationship between exposure and residence times is mathematically described by a composition matrix that linearly relates the two identities to each other. Composition matrices for different hydrological conditions are analyzed by using the singular value decomposition technique. Results show that especially the type of couplings between the surface and subsurface flow domain control how exposure and residence times are related to each other in the wetland system and that timescales of residence and exposure typically differ by orders of magnitude. Finally, results also indicate that the assessment of system specific transformation efficiencies can be very error-prone if residence instead of exposure times are being used to derive system specific Damkohler numbers.

The interplay between precipitation and evapotranspiration determines the input into the hydrological system of a catchment. Annual values of precipitation, evapotranspiration, and runoff measured at the catchment outlet for the 2002–2009 period were available. Annual precipitation clearly surmounted the sum of evapotranspiration and runoff. Part of the observed discrepancy might be due to the heterogeneity of precipitation and evapotranspiration within the catchment which has not been studied in sufficient detail. Annual evapotranspiration fluxes were remarkably constant during this period, whereas precipitation and runoff exhibited much larger interannual variability.

Radon ( R 86 222 n ) as a hydrological tracer offers a method for studying short to medium term groundwater - surface water interactions. These high frequency processes play an important role in wetland hydrology and biogeochemistry and may influence their contribution to the global carbon cycle. Therefore, there is a definite need for robust methods to measure high resolution 222Rn time series in-situ. In this study we adapted and improved a membrane system to measure 222Rn continuously with a primary focus on a rapid response time and low power consumption. The membrane system was constructed using a hydrophobic capillary membrane and laboratory experiments were conducted to quantify the systems' response time to predefined 222Rn pulses. It was then deployed in a stream draining a riparian wetland. The new membrane system could reduce the response time by ≈ 60 % in comparison to the established silicone membrane. We could identify the behaviour of the system in response to dynamically changing 222Rn activities and suggest a new method using simple linear regression to quantify the systems’ response when the response time concept is inapplicable. Finally, we were able to measure high temporal resolution 222Rn activities reliably over an extended field deployment (68 d). We conclude that the improved system fills a gap ensuring high temporal resolution while maintaining extended maintenance intervals. This allows the user to study high frequency hydrological processes in remote areas. This new membrane system can be used to detect fast changes in 222Rn activities improving the comprehension of the underlying hydrological processes.