Editorial: Hydrodynamics and Water Quality of Rivers and Lakes

1. Overview of Recent Developments in the Field

The hydrodynamics and water quality of rivers and lakes are governed by complex interactions among flow, mixing, stratification, sediment transport, and biogeochemical processes (Ji, 2017) [1]. These interactions are increasingly influenced by human activities, land-use changes, and climate-driven hydroclimatic extremes, making surface waters among the most vulnerable components of the hydrologic system (Mishra et al., 2021; Van Vliet et al., 2023) [2,3]. Over recent decades, advances in hydrodynamic and water quality modeling have substantially improved our ability to represent momentum, mass, and heat transport in rivers and lakes, as well as the fate and transformation of nutrients and pollutants (Bai et al., 2022; Ji, 2017) [1,4]. However, translating this scientific progress into predictive understanding and management-relevant tools remains a major challenge.

Recent developments in the field have been driven by progress in numerical modeling, monitoring technologies, and integrative analytical frameworks. Process-based hydrodynamic models have evolved from simplified one-dimensional formulations to sophisticated two- and three-dimensional systems capable of resolving complex flow structures, stratification, and circulation in rivers, lakes, and reservoirs (De Goede, 2020; Ishikawa et al., 2021; Zhang et al., 2016) [5,6,7]. Coupling these models with water quality modules has enabled detailed investigations of eutrophication, hypoxia, and contaminant transport under variable hydrological and meteorological forcing (Chung et al., 2009; Drago et al., 2021; Man et al., 2021; Ni and Zhang, 2025; Park et al., 1996; Vinçon-Leite and Casenave, 2019) [8,9,10,11,12,13]. In parallel, long-term hydrochemical monitoring and multivariate statistical analyses have provided valuable insights into the combined effects of climate variability and anthropogenic stressors on river and lake water quality, particularly in regions experiencing water scarcity and intensive water use (Burt et al., 2014; Bhateria and Jain, 2016; Kernan et al., 2011; Khan et al., 2018; Lutz et al., 2016) [14,15,16,17,18].

Another major development has been the rapid expansion of remote sensing and data-driven approaches for surface water quality assessment. Since the launch of high-resolution satellite missions, satellite-derived observations have become increasingly important for monitoring lake temperature, turbidity, chlorophyll-a, and suspended sediments at regional to global scales (Dörnhöfer, K., & Oppelt, 2016; Sawaya et al., 2003) [19,20]. More recently, machine learning and hybrid modeling approaches have been applied to estimate hydrodynamic and water quality variables, offering new opportunities to complement physics-based models, particularly in data-limited settings (Zhi et al., 2024) [21]. While these advances have significantly expanded observational coverage and analytical capability, they have also highlighted new challenges related to uncertainty, transferability, and physical interpretability.

2. Knowledge Gaps Addressed by This Special Issue

Despite this progress, several persistent knowledge gaps remain. One key gap concerns the linkage between fine-scale hydrodynamic processes (e.g., turbulence, secondary circulation) and system-scale transport of sediments and pollutants, particularly in geometrically complex river reaches and engineered waterways. A second gap lies in understanding the coupled dynamics of circulation, stratification, and water quality in lakes and reservoirs, especially under non-stationary climatic conditions and human interventions such as dam operation and water diversion. A third gap relates to integrated monitoring and prediction, where remote sensing and data-driven methods have outpaced the development of standardized frameworks for physically consistent integration with process-based models. Finally, a gap persists between scientific understanding and management-oriented applications, including the development of tools to support adaptive decision-making under uncertainty and extreme events.

3. How This Special Issue Addresses These Gaps

The eleven papers published in this Special Issue, Hydrodynamics and Water Quality of Rivers and Lakes, collectively address these gaps through complementary perspectives spanning hydrodynamic theory, numerical modeling, field observations, remote sensing, and data-driven analysis.

To address the first gap concerning fine-scale hydrodynamics and system-scale transport in complex waterways, several studies employ high-resolution numerical simulations. Shaheed et al. (contribution 9) investigated turbulence patterns in river bends and confluences using the k–ω Shear Stress Transport model, demonstrating how secondary circulation at junctions is influenced by discharge ratios and momentum flux. Similarly, Elkersh et al. (contribution 4) examined the engineered waterway of Dubai Creek using Delft3D Flexible Mesh, showing that a 13 km channel extension significantly improved flow renewal and pollutant flushing. Complementing these reach-scale analyses, Ámon et al. (contribution 1) compared depth-integrated models for computing overland flow in steep watersheds, highlighting how Large Eddy Simulations capture detailed turbulent behaviors that simpler models often sacrifice for computational efficiency.

Transitions between riverine transport and the coupled dynamics found in lakes and reservoirs constitute the second knowledge gap, particularly under the influence of human-managed infrastructure. Kebedew et al. (contribution 6) explored these dynamics in Lake Tana, linking wind-induced circulation patterns to the spatial distribution of nutrients and water hyacinths while accounting for human interventions such as the Tana Beles hydropower tunnel. In artificial reservoirs in semi-arid Brazil, Lima et al. (contribution 7) modeled phosphorus decay coefficients, emphasizing the need for reliable nutrient retention models to manage eutrophication. Human interventions were further analyzed by Wang et al. (contribution 11) in the Shiyang River Basin, where over 20 years of data revealed how inter-basin water diversion projects create dilution effects that alleviate organic pollution. Additionally, Dumitran et al. (contribution 3) evaluated the impacts of floating photovoltaic systems on the Mihăilești Reservoir, quantifying how surface coverage improves water quality, conserves water, and reduces greenhouse gas emissions.

As physical processes become better understood, the third gap emerges regarding the development of integrated monitoring and prediction frameworks that combine observations with data-driven methods. Batina and Krtalić (contribution 2) provided a comprehensive review of remote sensing techniques for lake water quality, noting that although technology has advanced substantially since the 1970s, standardized frameworks for integrating these observations with process-based models remain limited. In a parallel effort to improve predictive accuracy and to cover the gap between extensively investigated indicators (e.g., dissolved oxygen, pH, nitrogen compounds) and those receiving less attention (e.g., phosphorus compounds, which are notoriously difficult to predict due to their complex chemical dynamics), Timis et al. (contribution 10) developed Artificial Neural Networks to forecast flow and phosphorus concentrations in the River Swale. Their results demonstrate that high-resolution, sub-daily simulations can outperform conventional process-based models, particularly for extreme events.

Ultimately, these scientific advancements must bridge the fourth gap: the transition from scientific understanding to management-oriented applications and adaptive decision-making. González-Díaz et al. (contribution 5) addressed this challenge by assessing heavy metal pollution in the Santiago–Guadalajara River Basin, linking environmental risk to human security and highlighting the need for sustainability indices to support management in industrial regions. Finally, Rusanov et al. (contribution 8) examined how stream intermittency acts as an environmental filter for biological indicators such as diatoms. Their findings connect climate-induced drought and variable flow regimes to a river’s self-purification capacity, providing a scientific basis for managing water quality in intermittent streams under conditions of high uncertainty.

This Special Issue contributes to the transformative shift in the fields of hydrology and water quality, suggesting that integrative approaches are their future. The integration of scales (e.g., low- and high-resolution in time and space; ground monitoring and satellite data), different techniques (e.g., refined conventional models, data-driven technologies, remote sensing), and research fields (e.g., solar energy with reservoir hydrodynamics [3], diatoms and flow patterns [7], AI for water flow and water quality spikes [10]) facilitates the development of reliable tools for sustainable water management which are aware of water resources’ natural variability and take into account climate change and human impact.

In summary, this Special Issue brings together diverse yet complementary studies that advance our understanding of river and lake hydrodynamics and water quality across a wide range of spatial and temporal scales. Collectively, the contributions demonstrate how high-resolution numerical modeling, long-term field observations, remote sensing, and data-driven approaches can be integrated to address pressing scientific and management challenges. By explicitly targeting key knowledge gaps—from fine-scale process understanding to system-scale prediction and decision support—these studies highlight pathways toward more reliable, transferable, and operational tools for water resource and ecosystem management. Together, the papers underscore the importance of interdisciplinary frameworks and standardized methodologies to enhance predictive capability, support adaptive management, and improve resilience of aquatic systems under growing anthropogenic pressures and climate extremes.

4. Future Research Directions and Emerging Opportunities

Even though this Special Issue offers an integrated approach to hydrodynamics, sediment transport, pollutant transport and transformation, water quality, and eutrophication in surface water systems, some critical areas require focused scientific attention which may be expanded in the future.

Recent advancements indicate that ensemble machine learning methods and computational fluid dynamics solvers are increasingly utilized for predicting flow patterns, shear stress, and scour hotspots in the design of hydraulic construction and systems, consistently surpassing traditional statistical methods in water quality and hydrological predictions (Aly & Khaled, 2023; Wang et al., 2025) [22,23]. In the future, hydrodynamic models based on artificial intelligence, integrated digital twins, and machine learning will be encountered more and more often when addressing issues pertaining to the connection between small-scale hydrodynamic processes and the transport of sediments (Bezak et al., 2025) [24] and pollutants at the system level, particularly in river areas with complex geometry and in waterways. Additionally, advanced sensor networks and nature-based solution design platforms, which combine real-time measured data with complex simulations and geomorphological information, can provide significant benefits for sustainable water resource management, as well as for the design of resilient hydraulic structures (Magnier et al., 2024) [25]. These platforms can be used also for erosion prediction, urban expansion management (Louarn et al., 2025) [26], and the simulation of surface runoff (Santos, 2025) [27].

To gain a comprehensive understanding of the interrelated dynamics of circulation, stratification, and water quality in lakes and reservoirs influenced by variable climatic conditions and anthropogenic factors (gap 2), research methodologies can be enhanced through the application of GIS-based multi-criteria assessment tools that incorporate real-time water quality data (Luo et al., 2024) [28].

The management of water hyacinths may also encompass the transformation of harvested biomass into nano-biochar and organic fertilizers (Irewale et al., 2024) [29]. Consequently, the thermal processing of phosphorus-abundant water hyacinths yields hydroxyapatite, a slow-release fertilizer that alleviates phosphorus deficit (Ramirez et al., 2021) [30].

The tools, based on AI and site-specific, are critical for predicting the recovery time of a lake and managing eutrophication in data-scarce or climatically challenging regions (Lim & Choi, 2025) [31].

Water quality assessment and management are determined by combining high-resolution spatial data, physics-based machine learning, and real-time sensor networks. Thus, going forward, elaborate entropy models combined with GIS, with a field accuracy of more than 85%, can be successfully utilized to map and categorize the geographical distribution of water quality indices (Kumar & Augustine, 2022; Das, 2025) [32,33].

The impact of FPV on the aquatic ecosystem is determined by the lake’s morphology and coverage rate, and models such as Delft3D, CE-QUAL-W2 (Ilgen et al., 2025) [34], and MyLake (Exley et al., 2025) [35] include dedicated modules for simulating how FPVs affect surface heat exchange, wind stress, and the possibility of increased internal phosphorus loading due to bottom water anoxia. Thus, using multiple simulation models at once, together with systematic measurement campaigns of lake quality indicators (Mentzafou, et al., 2025) [36], can provide a considerably more realistic picture of the local and global effects of FPV on the ecosystem’s water mass.

The third gap’s future directions are related to integrated monitoring and prediction, where data-driven techniques and remote sensing have surpassed technologies that successfully address water quality problems by combining neural networks and machine learning algorithms. This allows for real-time monitoring and rapid response to any dangers or anomalies by boosting detection and response capabilities through the optimization of measurement locations and sample time intervals. Currently, tools based on Feedforward Backpropagation Artificial Neural Networks have changed from standalone algorithms to automated, physics-informed, energy-efficient frameworks (Xie et al., 2022) [37]. While backpropagation remains a “workhorse,” it is increasingly used in hybrid systems that handle complex nonlinear environmental and engineering data with minimum user adjustment (Peksa et al., 2025) [38]. The spatiotemporal variability of such data has motivated a shift from the conventional static feedforward neural networks towards dynamic and/or memory-capable architectures. Networks such as Nonlinear Autoregressive with Exogenous Inputs (NARX) (Sun et al., 2024; Aribarg., 2025) [39,40] or Long Short-Term Memory (LSTM) (Sahy et al., 2026; Chen et al., 2025) [41,42] are better at capturing both the system heterogeneity and temporal lag while maintaining a “memory” of system events.

The evaluation of heavy metal risks in river ecosystems has evolved from simple chemical concentration monitoring to bioavailability-based and AI-enhanced risk modelling (Shaheed et al., 2025) [43]. These techniques provide a more accurate picture of how toxins affect aquatic organisms and the health of nearby human populations. Nowadays, trait-based methods of stream ecology are evolving away from descriptive data and towards mechanistic, predictive frameworks that connect species’ functional traits directly to flow intermittence (Journiac et al., 2025) [44]. These tools give information about how drought-induced flow interruptions affect community structure and ecosystem services such as carbon cycling and secondary production (Farizo et al., 2024) [45].

In this integrative context, increasingly prominent across research fields, there is a prevailing trend in embedding physics into machine learning (Karniadakis et al., 2021) [46]. Such hybrid systems significantly outperform standard models (Lei et al., 2026) [47] as they are better informed of the process compared to conventional data-driven approaches. Therefore, they are solid candidates for digital twins and expert systems offering enhanced robustness during extreme events compared to both classical mechanistic-based and data-driven-based systems.

5. Conclusions

The eleven articles in this Special Issue show that the hydrodynamics and water quality of rivers and lakes in reaction to human activity and extreme weather events are essential and urgent issues, linked to worldwide concerns about reducing anthropogenic impacts on aquatic ecosystems. Their complementary viewpoints, which include hydrodynamic theory, numerical modelling, field observations, remote sensing, and data-based analysis, provide insights into integrated and adaptive water management. Many methods used in aquatic ecosystem research must be updated as climate change worsens in order to deal with baseline conditions that are becoming more erratic and unpredictable. Long-term water security is aided by these tools, which help communities and authorities create conservation policies, optimize water use, protect and preserve aquatic ecosystems, and strengthen climate change resilience.

From this perspective, the contributions gathered in this Special Issue constitute important achievements towards reaching this objective.