Water Resources Research

Sun, HongGuang; Lei, Dawei; Zhang, Yong; Qian, Jiazhong; Yu, Xiangnan

doi: 10.1029/2024wr038731pmid: N/A

Pollutant transport in discrete fracture networks (DFNs) exhibits complex dynamics that challenge reliable model predictions, even with detailed fracture data. To address this issue, this study derives an upscaled integral‐differential equation to predict transient anomalous diffusion in two‐dimensional (2D) DFNs. The model includes both transmissive and dead‐end fractures (DEFs), where stagnant water zones in DEFs cause non‐uniform flow and transient sub‐diffusive transport, as shown by both literature and DFN flow and transport simulations using COMSOL. The upscaled model's main parameters are quantitatively linked to fracture properties, especially the probability density function of DEF lengths. Numerical experiments show the model's accuracy in predicting the full‐term evolution of conservative tracers in 2D DFNs with power‐law distributed fracture lengths and two orientation sets. Field applications indicate that while model parameters for transient sub‐diffusion can be predicted from observed DFN distributions, predicting parameters controlling solute displacement in transmissive fractures requires additional field work, such as tracer tests. Parameter sensitivity analysis further correlates late‐time solute transport dynamics with fracture properties, such as fracture density and average length. Potential extensions of the upscaled model are also discussed. This study, therefore, proves that transient anomalous transport in 2D DFNs with DEFs can be at least partially predicted, offering an initial step toward improving model predictions for pollutant transport in real‐world fractured aquifer systems.

Fiori, Aldo; de Barros, Felipe P. J.; Bellin, Alberto

doi: 10.1029/2024wr038516pmid: N/A

Managed Aquifer Recharge (MAR) plays an important role in improving and supplementing groundwater storage. Many natural factors, ranging from climatic conditions to soil characteristics, can impact the efficiency of an infiltration basin. Other factors, such as engineered variables, will also influence the basin performance and the risks associated with groundwater contamination. The latter depends on the interplay between the hydraulic characteristics of the system and the soil and solute properties. The design of infiltration basins has been performed so far with the main objective of mitigating the tendency of the basin to reduce the infiltration rate with time due to clogging of the basin's bottom. Less attention has been paid to the risk of groundwater contamination by the infiltrating water. To understand the complex interplay between natural and engineering parameters on MAR efficiency and the contamination risk, we propose a risk‐oriented analytical framework. The framework allows to investigate the interplay between soil parameters, engineering design and climatic factors on the efficiency of an infiltration basin. Our framework relies on novel analytical solutions that relates the geometrical and hydrological features of the infiltration basin to its efficiency and groundwater contamination risk. The solutions incorporates the randomness associated with inflows (precipitation) and soil properties. We explore the trade‐off between efficiency and the risk of contamination and delineate a design procedure that balances these two opposing needs. Although the framework relies on simplifying assumptions, it provides a computationally efficient manner to obtain physical insights and relate model input parameters to decision making.

Pettersson, Per; Keilegavlen, Eirik; Sandve, Tor Harald; Gasda, Sarah E.; Krumscheid, Sebastian

doi: 10.1029/2024wr038073pmid: N/A

Luo, Manhua; Li, Hailong; Li, Gang; Wang, Wei; Yu, Shengchao; Ma, Qian; Zheng, Yan

doi: 10.1029/2024wr038231pmid: N/A

Coastal groundwater dynamics and solute transport were influenced by multiple factors including aquitards, tides, evaporation, and slope breaks in coastal aquifers. However, quantification of the impacts of these factors on groundwater flow and salinity distribution remains a challenge. In this study, both field observations and numerical modeling were applied to investigate hydraulic heads and groundwater salinity in a tidal flat with large‐scale seepage faces at Laizhou Bay, China. Results showed that seepage‐face evaporation increased groundwater salinity landward and promoted groundwater and salt exchange within the intertidal zone significantly in comparison to the case without evaporation. Seawater infiltrated the aquifer on the left of the slope break and discharged on the right, forming a groundwater circulation cell, which notably influenced leakage flow between unconfined and confined aquifers through the aquitard. The aquitard prevented approximately 85% of inland freshwater discharge near the slope break, resulting in the formation of two atypical freshwater discharge tubes in the upper and middle intertidal zones. Two additional groundwater circulation cells developed in the lower intertidal zone due to the spring‐neap tidal cycle. The outflow and inflow fluxes over a spring‐neap tidal cycle were numerically estimated to be 1.46 and 1.27 m2/d, respectively, with evaporation accounting for 45% of the outflow flux. These findings provide significant insights for further investigations on groundwater dynamics and solute transport in multi‐layered coastal aquifers, and have strong implications for biogeochemical processes within the intertidal zone.

Han, Juntai; Yang, Yuting; Guo, Yuhan; Li, Changming; Liu, Ziwei; Tu, Zhuoyi; Xi, Haiyang

doi: 10.1029/2024wr037492pmid: N/A

Snow drought, characterized by an anomalous reduction in snowpack, exerts profound hydrological and socioeconomic impacts in cold regions. Despite its significance, the influence of diverse snow drought types, including warm, dry, and warm‐and‐dry variants, on streamflow remains inadequately understood. Here we present the first hemispheric‐scale, observation‐based assessment of snow drought patterns and the impacts on seasonal and annual streamflow (Q) across 3049 northern catchments over 1950–2020. Our findings reveal that catchments with a lower mean annual snowfall fraction (fs‾ $\overline{{f}_{\mathrm{s}}}$) exhibit a heightened prevalence and severity of warm and warm‐and‐dry snow droughts, whereas high‐fs‾ $\overline{{f}_{\mathrm{s}}}$ experience a more prevalent but less severe dry snow drought. This disparity arises from distinct sensitivities of snowpack to cold‐season precipitation and temperature. In addition, dry and warm‐and‐dry snow droughts induce a reduction in Q during both cold and warm seasons, culminating in a significant decrease in annual Q. Conversely, warm snow drought increases annual Q in catchments with fs‾≤0.3 $\overline{{f}_{\mathrm{s}}}\le 0.3$ but decreases annual Q in catchments with fs‾>0.4 $\overline{{f}_{\mathrm{s}}}\, > \,0.4$, attributable to a trade‐off between increased cold‐season streamflow (Qc) and decreased warm‐season streamflow (Qw). With ongoing climate warming, a continued reduction in snowfall is anticipated, which is expected to further increase the frequency and severity of warm and warm‐dry snow droughts. These circumstances, particularly impactful under low fs‾ $\overline{{f}_{\mathrm{s}}}$ conditions, are poised to present formidable challenges for water resources management in cold regions globally.

Gai, Bo; Kumar, Rohini; Hüesker, Frank; Mi, Chenxi; Kong, Xiangzhen; Boehrer, Bertram; Rinke, Karsten; Shatwell, Tom

doi: 10.1029/2023wr036808pmid: N/A

Lentic waters integrate atmosphere and catchment processes, and thus ultimately capture climate signals. However, studies of climate warming effects on lentic waters usually do not sufficiently account for a change in heat flux from the catchment through altered inflow temperature and discharge under climate change. This is particularly relevant for reservoirs, which are highly impacted by catchment hydrology and may be affected by upstream reservoirs or pre‐dams. This study explicitly quantified how the catchment and pre‐dams modify the thermal response of Rappbode Reservoir, Germany's largest drinking water reservoir system, to climate change. We established a catchment‐lake modeling chain in the main reservoir and its two pre‐dams utilizing the lake model GOTM, the catchment model mHM, and the stream temperature model Air2stream, forced by an ensemble of climate projections under RCP2.6 and 8.5 warming scenarios. Results exhibited a warming of 0.27/0.15°C decade−1 for the surface/bottom temperatures of the main reservoir, with approximately 8%/24% of this warming attributed to the catchment warming, respectively. The catchment warming amplified the deep water warming more than at the surface, contrary to the atmospheric warming effect, and advanced stratification by about 1 week, while having a minor impact on stratification intensity. On the other hand, pre‐dams reduced the inflow temperature into the main reservoir in spring, and consequently lowered the hypolimnetic temperature and postponed stratification onset. This shielded the main reservoir from climate warming, although overall the contribution of pre‐dams was minimal. Altogether, our study highlights the importance of catchment alterations and seasonality when projecting reservoir warming, and provides insights into catchment‐reservoir coupling under climate change.

Bernard, Martino; Barbini, Matteo; Berti, Matteo; Boreggio, Mauro; Simoni, Alessandro; Gregoretti, Carlo

doi: 10.1029/2023wr036887pmid: N/A

In the Dolomites, steep rocky cliffs are marked by numerous narrow gullies. When high‐intensity short‐duration precipitation occurs, these gullies concentrate and direct surface runoff to the screes at the foot of rock cliffs. Surface runoff mixes with loose sediments, creating a solid‐liquid surge that, as it moves downhill, increases its volume entraining debris material and transforms into a granular debris flow. Given the ongoing challenge of modeling the relationship between intense rainfall, surface runoff, and debris flow initiation, we take advantage of data from three monitoring stations operating in distinct debris flow active catchments in our study area to make progress. These stations, strategically positioned close to debris flows initiation zones, record videos and different types of flow‐stage data, helping us pinpoint the timing and form of incoming discharge hydrographs. Over a 15‐year period of observation, we collected a comprehensive data set on runoff and mass movement in these catchments, offering valuable insights into their hydrological behavior and the initiation of granular debris flows. To compute infiltration excess runoff generation, we refined an already existing hydrological model and calibrated it using discharge measured at one of the monitoring stations. Testing this updated model against observations from two other larger debris flow sites showed that it can reproduce the initial phases of a debris flow, when sediment concentration rapidly rises. These findings suggest that a well‐tuned hydrological model can predict the discharge from intense, short rainfall events that typically trigger debris flows, as well as the early stages of these phenomena.

Hansen, Carly H.; Iftikhar, Bilal; Pilla, Rachel M.; Griffiths, Natalie A.; Matson, Paul G.; Jager, Henriette I.

doi: 10.1029/2024wr037726pmid: N/A

Ebullitive methane (CH4) emissions in lentic ecosystems tend to concentrate at river‐lake interfaces and within shallow littoral zones. However, inconsistent definitions of the littoral zone and static representations of the lake or reservoir surface area contribute to major uncertainties in greenhouse gas (GHG) emissions estimates, particularly in reservoirs with large water‐level fluctuations. This study examines temporal variation in littoral and total surface areas of US reservoirs and demonstrates how different methods and data sources lead to discrepencies in reservoir GHG emissions at large scales and over time. We also explore variability in remotely sensed water occurrence according to maximum surface area, reservoir purposes, and hydrologic regions. Notably, the largest relative variability in surface area is exhibited by small reservoirs with a maximum surface area <1 km2 and non‐hydroelectric reservoirs. Additionally, we use a case study of measured CH4 emissions from the southeastern United States (Douglas Reservoir) to illustrate the effects of varying surface area on reservoir‐wide GHG estimates. Upscaled CH4 emissions in Douglas Reservoir differed by nearly two‐fold depending on the source of total surface area data and whether estimates accounted for seasonal fluctuations in surface area. During seasonal drawdown in Douglas Reservoir, relative littoral area varies non‐linearly; periods of lower pool elevation (and thus larger relative littoral area) likely contribute disproportionately high CH4 emission rates compared to the commonly sampled summer season when water levels are at full‐pool elevation. Improved GHG monitoring and upscaling techniques require accounting for temporal variability in reservoir surface extent and littoral area.

Solanki, Hiren; Vegad, Urmin; Kushwaha, Anuj; Mishra, Vimal

doi: 10.1029/2024wr038192pmid: N/A

Streamflow prediction is crucial for flood monitoring and early warning, which often hampered by bias and uncertainties arising from nonlinear processes, model parameterization, and errors in meteorological forecast. We examined the utility of multiple hydrological models (VIC, H08, CWatM, Noah‐MP, and CLM) and machine learning (ML) methods to improve streamflow simulations and prediction. The hydrological models (HMs) were forced with observed meteorological data from the India Meteorological Department (IMD) and meteorological forecast from the Global Ensemble Forecast System (GEFS) to simulate flood peaks and flood inundation areas. We used Multiple Linear Regression, Random Forest (RF), Extreme Gradient Boosting (XGB), and Long Short‐Term Memory (LSTM) for the post‐processing of simulated streamflow from HMs. Considering the influence of dams is crucial for the effectiveness of HMs and ML methods for improving streamflow simulations and predictions. In addition, ML‐based multi‐model ensemble streamflow from HMs performs better than individual models, highlighting the need for multi‐model‐based streamflow forecast systems. The post‐processing of streamflow simulated by the hydrological models using ML significantly improved overall streamflow simulations, with limited improvement in high‐flow conditions. The combination of physics‐based hydrological models, observed climate data, and ML methods improve streamflow predictions for flood magnitude, timing, and inundated area, which can be valuable for developing flood early warning systems in India.

Roumelis, Christian; Willert, Fabian; Scaccia, Maria; Welch, Susan; Gabor, Rachel; Carrera, Jesús; Folch, Albert; Salgot, Miquel; Sawyer, Audrey H.

doi: 10.1029/2024wr038087pmid: N/A

Coastal aquifers experience water table fluctuations that push and pull water and air through organic‐rich soils. This exchange affects the supply of oxygen, dissolved organic carbon (DOC), and nitrogen (N) to shallow aquifers and influences groundwater quality. To investigate the fate of N species, we used a meter‐long column containing a sequence of natural organic topsoil and aquifer sediments. A fluctuating head was imposed at the column bottom with local, nitrate‐rich groundwater (16.5 mg/L NO3‐N). We monitored in‐situ redox potential and collected pore water samples for analysis of inorganic N species and DOC over 16 days. Reactive processes were more complex than anticipated. The organic‐rich topsoil remained anaerobic, while mineral sediments beneath alternated between aerobic, when the water table dropped and sucked air across preferential flow paths, and anaerobic conditions, when the water table was high. A fluid flow and reactive transport model shows that when the water table rises into organic‐rich soils, it limits the flow of oxygen, while the soils release DOC, which stimulates the removal of nitrate from groundwater by denitrification. At the end of the experiment, we introduced seawater to the column to mimic a storm surge. Seawater mobilized N and DOC from shallow soil horizons, which could reach the aquifer if the surge is long enough. These processes are relevant for groundwater quality in developed coastal areas with anthropogenic N sources, as climate change and rising seas will drive changes in water table and flood dynamics.