The adoption of community-based water purification technology in rural communities is strongly influenced by psychological factors, yet these factors often suffer from endogeneity, leading to biased estimations of their true impact. Our study investigates this critical issue, revealing that traditional estimation methods significantly underestimate the effects of key psychological determinants. Specifically, we found that perceived ease of access and descriptive norms, when treated as exogenous, were underestimated by 175% and 76%, respectively. This oversight highlights the importance of addressing endogeneity to accurately capture the relationship between psychological factors and adoption behavior. The endogenous nature of perceived benefits and descriptive norms highlights a crucial bidirectional relationship: as adoption increases, so do positive social norms and perceived benefits, creating a reinforcing cycle that further drives adoption within the community. Interventions that fail to consider this mutual reinforcement risk undervaluing key psychological factors, potentially undermining their effectiveness. We propose that cultural factors serve as instrumental variables (IVs) to mitigate endogeneity and offer a clearer pathway through which psychological factors influence behavior. For instance, cultural traits such as "work-luck" dynamics shape individuals' proactive or passive approaches to overcoming barriers to technology access. Similarly, generalized morality, which prioritizes communal welfare over individual gain, strengthens descriptive norms that promote widespread adoption. In collectivist societies, these norms hold significant influence, compelling individuals to adopt technologies to maintain social cohesion and uphold communal values.

Our study introduces a robust theoretical framework that integrates cultural factors into the analysis of technology adoption. By leveraging cultural traits, interventions can align more closely with community values, enhancing the likelihood of sustainable adoption. This approach not only provides deeper insights into the dynamics of technology adoption but also offers practical strategies for designing culturally sensitive interventions.

In conclusion, addressing the endogeneity of psychological factors through the lens of cultural influences provides a more accurate and comprehensive understanding of the adoption process. This study advocates for the incorporation of cultural contexts in intervention strategies, ensuring they resonate with the community’s intrinsic values and beliefs. Future research could expand on this dynamic by employing system dynamic models to further explore the bidirectional feedback between psychological factors and behavior, ultimately contributing to more effective and sustainable adoption of community-based water purification technologies.

How to cite: Raj, M., Pande, S., and Ramesh, M.: Exploring Endogeneity in Psychological Determinants of Community-Based Water Purification Technology Adoption, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-20733, https://doi.org/10.5194/egusphere-egu25-20733, 2025.

The resilience of a large-scale water infrastructure system to cascading effects is
fundamentally dependent on the interdependencies of its components within the
infrastructure network. These interdependencies—which means that the states of
two or more infrastructure components are tightly interrelated through mechanisms
such as physical connection, geographical proximity, and information relay—can
cause a localized event to spread into a system-wide event. Of these, logical
interdependencies remain poorly understood. Little is known about how two
infrastructures affect the state of each other through human decisions and how such
logical connections can be detected and measured. In this study, we tackled this
gap by conducting an applied case study on the Lake Mendocino Reservoir in
California, USA. Crucially, our approach focuses on reservoir institutions (rules)
that structure human decisions around reservoir systems. Reservoir management
relies heavily on operational rules and regulations, but climate change demands
more adaptive and discretionary decision-making by operators. This may further
introduce logical interdependencies in a reservoir system. We develop a novel
framework that integrates Institutional Analysis using Large Language Models to
advance Natural Language Processing (NLP) techniques and Bayesian Network
Modeling to systematically analyze and quantify risk associated with logical
interdependencies. We aim to improve decision-making and risk management in
reservoir operations. This research provides essential insights into enhancing the
resilience of water management infrastructures, particularly in the face of climate
change.

How to cite: Gautam, S., Yu, D. J., and Hoon Cheol, S.: Enhancing Resilience in Human-Reservoir Systems with NLP and AI Frameworks, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-20066, https://doi.org/10.5194/egusphere-egu25-20066, 2025.

This abstract investigates the evolution of urban stormwater management, contrasting traditional methods with emerging approaches, emphasizing the integration of Low Impact Development (LID) strategies and Building Information Modeling (BIM). A comprehensive review of Scopus and Web of Science articles synthesizes existing research to identify trends, challenges, and opportunities in this interdisciplinary domain. Key insights include the effectiveness of LID practices such as permeable pavements, rain barrels, and the application of simulation tools like SWMM and HEC-RAS in reducing runoff and enhancing urban hydraulic modeling. The findings highlight the critical role of green infrastructure in mitigating rainfall impacts and the importance of cost-benefit analyses for evaluating LID implementation. Despite proven benefits, gaps persist in integrating LID into land-use planning, particularly in addressing future climate risks and accommodating urban growth. The study underscores the potential of 3D digital technologies to enhance stormwater management strategies, especially under extreme rainfall conditions. Additionally, the review identifies the lack of high-resolution data as a barrier to informed decision-making. It advocates for stronger collaboration between researchers and policymakers to foster sustainable urban development, improve water conservation, and minimize flooding risks. LID practices, integrated with Building Information Modeling offer a cost-effective solution to urban stormwater challenges, paving the way for resilient and sustainable cities.

How to cite: Fahimi, F. and Ostad Mirza Tehrani, M. J.: Enhancing Urban Stormwater Management: Traditional Measures versus Future Perspectives, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-18797, https://doi.org/10.5194/egusphere-egu25-18797, 2025.

Contaminant co-occurrence in water resources poses significant threats to public health and ecosystem stability, necessitating comprehensive monitoring and analysis. This study investigates the spatiotemporal distribution of arsenic, fluoride, and perfluorooctane sulfonate (PFOS) contamination in groundwater and surface water across Yorkshire from 2000 to 2023. Data for this assessment were obtained from the Environment Agency, ensuring reliable and standardised measurements across the study period. The results reveal a concerning trend of increasing arsenic and fluoride concentrations, particularly in the eastern and southern regions, with arsenic levels exceeding 10 µg/L and fluoride concentrations surpassing 1.5 mg/L in several areas by 2023. The PFOS contamination, assessed in both groundwater and surface water for 2023, highlights significant contamination in the southern regions, with concentrations exceeding 0.001 µg/L in some hotspots. The co-contamination maps indicate overlapping regions of high contaminant concentrations, suggesting potential sources of industrial pollution and agricultural runoff. This study emphasises the need for targeted mitigation strategies and continuous monitoring to protect public health and ensure water quality standards across the region.

How to cite: Agarwal, V., Kumar, M., and Saxena, A.: Spatiotemporal Assessment of Arsenic, Fluoride, and PFOS Co-Contamination in Yorkshire's Water Resources, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-20494, https://doi.org/10.5194/egusphere-egu25-20494, 2025.

Recent water demands have created immense stress on groundwater, especially in the region facing water scarcity. Hence, optimizing groundwater pumping and developing sustainable water management strategies becomes important for such areas. The traditional mesh-based methods, such as finite difference (FDM) and finite element methods (FEM) for groundwater modelling requires high-quality mesh generation. In these methods, generating a high-quality mesh for complex aquifers is a time-consuming task. Therefore, meshless methods that work with scattered field nodes and avoid mesh generation are more suitable for complex groundwater problems. The present study uses the meshless generalized finite difference method (GFDM) for modelling the groundwater flow and integrating it with particle swarm optimization (PSO) to determine the optimal pumping rates for a hypothetical aquifer system. In this work, optimal pumping rates for different groundwater withdrawal scenarios are obtained through the proposed meshless simulation-based optimization model (GFDM-PSO), indicating its application to real-world problems.

How to cite: Singh, K. G. and Pathania, T.: Optimization of groundwater pumping rates using meshless simulation-based optimization model, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-4251, https://doi.org/10.5194/egusphere-egu25-4251, 2025.

Currently, climate change and increasing water demand pose a growing threat to the future availability of water for human societies and ecosystems that depend on it. At the same time, growing evidence suggests that groundwater is playing an increasingly active role in the global water cycle, particularly in sustaining river flows worldwide (Xie et al., 2024). In this context, quantifying the water exchange between these two components of the hydrological cycle becomes essential for an integrated assessment of water availability. For this purpose, baseflow separation methods are valuable tools, though their limitations remain a subject of debate.

Several authors have suggested that commonly used baseflow separation methods should be applied with caution, since these methods often produce large estimation errors, when they are compared with results obtained using three-dimensional flow numerical models (hereafter referred to as 3D models), thereby limiting their applicability. Nevertheless, these methods remain a widely used alternative due to their lower data and resource requirements compared to 3D models. To address these limitations, we proposed a novel methodology based on baseflow separation methods for analysing the interactions between a shallow alluvial aquifer system and the overlying river network. Subsequently, we tested its performance against a 3D model.

The study area is the alluvial aquifer system located at the confluence of the Tarn, Aveyron and Garonne rivers. A 3D model was developed using the BRGM’s MARTHE software. The study area was divided into sub-zones that meet the same isolation conditions for the river network delimited for the analysis of the results to ensure a more robust validation. Time series of flow and cumulative volume for components of the water balance in the river network, as well as flow at gauging points, were analysed. Additionally, different integration periods (quarterly, half-yearly, annual, and biannual) were examined. Several baseflow separation methods were tested, including both digital filtering and graphical methods.

The results showed that the methods proposed by Chapman (1991) and Chapman and Maxwell (1996) consistently outperformed all others across the entire study area and for all integration periods. R² coefficients of determination greater than 0.8 were obtained in both cases for integration periods exceeding six months. Notably, shorter integration periods better captured the temporal variation of water exchange between the aquifer and the river network. However, longer integration periods produced more accurate overall results, likely because the filters struggled to capture flow reversals between the aquifer and river network during flood events.

Acknowledgments: Authors acknowledge the funding provided by project WaMA-WaDiT (PCI2024-153483) funded by MICIU /AEI /10.13039/501100011033/ UE

References

Chapman, T. G. (1991). Evaluation of automated techniques for base flow and recession analyses. Water Resources Research, 27(7), 1783–1784. https://doi.org/10.1029/91WR01007

Chapman, T. G., & Maxwell, A. I. (1996). Baseflow separation: Comparison of numerical methods with tracer experiments. Paper presented at the Hydrology and Water Resources Symposium: Water and the Environment, Institution of Engineers, Australia.

Xie, J., Liu, X., Jasechko, S., et al. (2024). Majority of global river flow sustained by groundwater. Nature Geoscience, 17, 770–777. https://doi.org/10.1038/s41561-024-01483-5

How to cite: García Montealegre, J. P., Caballero, Y., and Del Jesus Peñil, M.: Are baseflow separation methods suitable for assessing shallow alluvial aquifers’ contribution to streamflow?, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-15495, https://doi.org/10.5194/egusphere-egu25-15495, 2025.

Groundwater is a vital resource for domestic, agricultural, and industrial purposes in many regions. However, the increasing demand and unsustainable extraction practices have raised concerns about the long-term viability and sustainability of groundwater storage (GWS), especially in areas where groundwater is the primary source of meeting various demands. Here, we focus on GWS changes in India’s Northwestern states, including Gujarat, Rajasthan, Punjab, Haryana, Uttara Pradesh, and Delhi over two decades (2002-2023). These states encompass 875,249 km² area within the Indus and Ganges river basins, constitute approximately 59% cultivated land, and sustain 525.52 million people. Leveraging GRACE-based TWS data and GLDAS model data, our analysis reveals significant (P<0.05) declining GWS trends with a slope of −20.88 ± 0.53 mm/year, which is more acute than previously reported estimates. Some trend change points in February 2008 and June 2016 are detected that lead to segmented trends with slopes of −18.97 ± 2.45 mm/year (Jan-2002 to Feb-2008), −9.16 ± 1.96 mm/year (Feb-2008 to Jun-2016), −11.80 ± 2.51 mm/year (Jun-2016 to Dec-2023). Spatially divergent trends are found with high decreasing trends of more than 40 mm/year in Punjab, Haryana, Delhi, and some parts of Rajasthan and Uttara Pradesh. This is primarily due to anthropogenic activities like groundwater extraction for domestic and agricultural purposes. In contrast, Gujrat shows subtle positive trends, less than 10 mm/year, due to improved water management, irrigation practices, artificial recharge efforts, monsoonal rainfall, and efficient water extraction management​. Multi-decadal variability and the recent depletion across these six states may foster discussions on policy actions and enhanced multilateral cooperation for a sustainable future, especially in the face of escalating groundwater extraction and a warming climate. This highlights the critical need for immediate attention to water resource challenges in the Northwestern states of India.

Keywords: Groundwater storage (GWS); GRACE; GLDAS; Anthropogenic activities; Policy interventions.

How to cite: Anjaneyulu, R. and Abhishek, A.: A more acute and continuous decline in Groundwater in Northwest India , EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-543, https://doi.org/10.5194/egusphere-egu25-543, 2025.

Groundwater assessment in the Deccan basalt region of India is challenging due to its heterogeneous geology and complex aquifer dynamics. This study integrates hydrogeophysical methods, including DC resistivity and time domain Induced Polarization (DCIP) and slug tests, to evaluate aquifer potential near Bhopal, Madhya Pradesh. The research focuses on both shallow unconfined and deeper semi-confined aquifers within weathered and fractured basalt formations.

Electrical resistivity surveys included more than 25 DCIP profiles targeting weathered and fractured zones. Resistivity values ranged from 15–70 Ωm in weathered/fractured basalts and varied based on the degree of water saturation and fracturing, reflecting lithological heterogeneity. ERT profiles revealed low-resistivity and moderate-to-high chargeability zones, indicative of fracture porosity and groundwater retention. Fracture anisotropy and resistivity contrasts provided critical insights into aquifer connectivity and dynamics.

Slug tests conducted at a borehole with a drilled depth of 61 m validated geophysical findings. Hydraulic parameters, including hydraulic conductivity (3.9E-7 m/s), transmissivity (1.9E-5 m²/s), storativity (0.001), and specific storage (2.1E-5 m^-1), were estimated using Bouwer-Rice and Cooper-Bredehoeft-Papadopulos solutions. These localized parameters complement the spatially extensive data from geophysical surveys. Seasonal water-level fluctuations emphasize the significance of monsoonal recharge in sustaining aquifers.

This integrated approach highlights the role of fractures, weathered zones, and advanced geophysical techniques in delineating groundwater zones and assessing recharge potential. The findings contribute to effective groundwater exploration and sustainable management strategies, addressing water scarcity challenges in basaltic terrains of the Deccan Traps.

Keywords: Aquifer potential, Bouwer-Rice and Cooper-Bredehoeft-Papadopulos solution, ERT, slug test, weathered/fractured basalts, hydrogeology.

How to cite: Khalique, A., Singh, A., and Gaurav, K.: Integrating Geophysical and Hydrogeological Methods for Groundwater Assessment in the Deccan Basalt Region of India, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-17885, https://doi.org/10.5194/egusphere-egu25-17885, 2025.

Anthropogenic impact on aquifers leads to variations of groundwaters chemical content. This study is determined to describe current geochemical characteristics of springs in Shelkovo district in order to assess the quality of the water that is used for drinking purposes by residents.

The geological structure of the territory includes Devonian, Upper Carboniferous, Upper Jurassic and Lower Cretaceous terrigenous-carbonate rocks, overlapped by thin Quaternary sandy deposits. Surface sediments are permeable to polluted runoff waters, which can increase the vulnerability of groundwater and reduce its quality.

This research presents the obtained results of water parameters (COD, pH, electrical conductivity), the content of major ions (Ca²⁺, Mg²⁺, Na⁺, K⁺, NH₄⁺, HCO₃^-, Cl^-, SO₄^2-, NO₃^-)  for 12 springs. The spring waters are slightly mineralized (M=0.1-0.5 g/l), pH values vary from 5.5 to 7.5.  The total hardness is 0.63-5.7 mg-eq/l. The composition of the water is variable. Springs could be divided by the content of major anions: the dominance of HCO₃^- which is due to natural causes. In some cases the presence of Cl^- and SO₄^2- because of the use of fertilizers and deicing reagents in urban territories. 

The concentration of major ions was compared to maximum permissible concentrations in drinking water (by WHO standards). It was noted to slightly exceed the limit for nitrate ion as well as for chemical oxygen demand.  Some waters had a pH indicator lower than the standard range.

Comparison of the ratios Cl^-/(Cl^-+Na⁺) and Na⁺/(Na⁺+Cl^-) to total dissolved salt was applied in order to figure out the mechanism of spring waters forming (Gibbs, 1970). The results showed that chemical composition is primarily controlled by rock weathering. The ratio relationships between equivalent content Cl^-/Na⁺, HCO₃^-/Na⁺, Ca²⁺/Na⁺ indicate the type of rocks as a silicate (Gaillardet, 1999). The effect of human impact on groundwaters used to be assessed by comparing the equivalent ratios Cl^-/Na⁺ and NO₃^-/Na⁺ (Zhang et al, 2024). The calculations performed summarised anthropogenic impact, including agronomic activities. Significantly connections between various major ions were pointed out due to correlation analysis: as well as fertilizer components and pesticides, anti-icing reagents for roads in winter season and household chemicals from sewers were detected. 

The studied waters were formed by dissolving silicate rocks by atmospheric precipitation. As it was figured out by a significant role of chloride and sulfur ions, and presence of nitrogen-ions, the area of springs' feeding is located in permeable contaminated quaternary sediments. But despite anthropogenic impact, the chemical composition of springs correspond to WHO standards for drinking waters.

References

Gaillardet J., Dupre B., Louvat P., Allegre C.J. Global silicate weathering and CO₂ consumption rates deduced from the chemistry of large rivers // Chemical geology – 1999. – Т. 159. – №. 1-4. – С. 3-30.

Gibbs R. J. Mechanisms controlling world water chemistry //Science. – 1970. – Т. 170. – №. 3962. – С. 1088-1090. 

Zhang, H., Wang, Z., Wang, X. et al. Hydrochemical characterization and health risk assessment of different types of water bodies in Fenghuang Mountain Area, Northeast China. Environ Geochem Health 46, 292 (2024)

How to cite: Gusarova, D. and Yablonskaya, D.: Ecohydrogeological characteristics of spring waters in rural areas (northeast of Moscow region), EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-11817, https://doi.org/10.5194/egusphere-egu25-11817, 2025.

The relationship between groundwater and seismicity has been documented in various regions worldwide. Mexico is no exception to this phenomenon. On September 19, 2017, a magnitude 7.1 earthquake struck between the states of Puebla and Morelos, as reported by the National Seismological Service.

Approximately 50 km from the epicenter, the Agua Hedionda spring exhibited significant physical and chemical changes as a result of the earthquake. These changes highlight the dynamic interactions within the critical zone—the near-surface environment where rock, soil, water, air, and living organisms interact to shape the Earth's surface. The spring's discharge showed notable alterations, including a decrease in flow rate, reductions in major ion concentrations, and shifts in its isotopic composition, providing clear evidence of the connection between regional seismicity and the quality and availability of groundwater.

The analysis of changes in the spring's groundwater over time revealed its vulnerability to losing essential properties, either temporarily or permanently. Hydrochemical and volumetric flow rate data indicated that the spring underwent noticeable changes even before the earthquake. While the water chemistry showed gradual recovery by 2022, the flow rate only returned to approximately 25% of its pre-earthquake level.

In a country like Mexico, where groundwater is essential for numerous activities and where the interaction of five tectonic plates creates a dynamic seismic environment, studying the interplay between seismicity, groundwater, and processes within the critical zone is crucial for understanding and managing water resources sustainably.

How to cite: Sierra Garcia, B. A., Escolero, O., Olea Olea, S., and Medina Ortega, P.: Seismicity and Groundwater Dynamics: Impacts on the Critical Zone in spring of center Mexico, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-5314, https://doi.org/10.5194/egusphere-egu25-5314, 2025.

Fractured aquifer parameters are expected to have complex non-Gaussian spatial distributions. Gaussian Mixture Models, known for their effectiveness in representing non-Gaussian distributions, present a promising alternative for capturing the complex heterogeneity of fractured geologic settings however their usage in the fractured geologic settings is unexplored. In this study we extended the application of Gaussian mixtures to transient hydraulic tomography on laboratory-based fractured geologic settings using sequential Gaussian Mixture Model (GMM). We further examined the impact of the number of Gaussian components, sampling strategies and the amount of pumping data on the performance of the sequential GMM. Results demonstrate that GMM with an optimal number of Gaussian components effectively identifies high and low conductivity regions, fracture connectivity, and reasonably predicts drawdowns (R² = 0.61) pumping from validation ports. Stratified sampling of GMM parameters (R² = 0.74, average RMSE_median= 9.89 mm) outperforms other sampling strategies like random (R² = 0.61, average RMSE_median= 20.64 mm ), uniform (R² = 0.64, average RMSE_median= 11.70 mm) and quasi-random sampling (R² = 0.67, average RMSE_median= 11.40 mm) techniques in mapping the fracture connectivity and parameter distribution. Stratified sampling with reduced and information-based pumping data maintains commensurable accuracy (R² = 0.75, average RMSE_median= 11.34 mm). Overall, our findings suggest that the sequential GMM combined with stratified sampling technique effectively captures the spatial variability of aquifer parameters in fractured media.

How to cite: Muraharirao, P. C. and Kbvn, P.: Sequential Gaussian Mixtures for Transient Hydraulic Tomography Inversion in Fractured Aquifers, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-4107, https://doi.org/10.5194/egusphere-egu25-4107, 2025.

Accurate estimation of the soil hydraulic conductivity function (SHCF), which describes the relationship between hydraulic conductivity and matric suction in soil, is essential for modeling flow and transport processes in the vadose zone. Traditional steady-state methods for directly determining SHCF are often laborious, time-consuming, and sometimes inadequate for capturing transient-state flow conditions. This study aims to propose a simple, quick, and accurate method for estimating SHCF that facilitates transient-state flow analysis during vadose zone modeling. The proposed method involves inverse numerical modeling using cumulative infiltration and final moisture content data from surface infiltration tests conducted with a handy mini disc infiltrometer (MDI). To validate this approach, the MDI-inverse modeling results were compared with SHCF results from another transient-state method, the instantaneous profile method (IPM), under similar initial soil conditions. The MDI infiltration tests were performed in homogeneously packed soil columns for two soils (identified as loam and silty clay loam textures) collected from nearby field sites. For each soil, separate IPM tests were conducted in soil columns equipped with soil moisture and matric suction sensors at various depths to facilitate calculation of reference SHCF. A comparison between the MDI and reference IPM results revealed a good agreement, with a low normalized RMSE (under 15%) for the estimated SHCFs and a low relative error (under 35%) for the optimized van Genuchten parameters α and n. The findings indicate that MDI-based cumulative infiltration measurements can reliably estimate SHCF via inverse simulation, providing a practical solution for field applications where traditional sensor deployment is challenging. Moreover, the results also establish MDI as a rapid, convenient, and non-invasive tool for determining SHCF for transient-state flow scenarios.

How to cite: Naik, A. P. and Pekkat, S.: Evaluating a rapid approach for estimating soil hydraulic conductivity function from near-surface infiltration measurements , EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-15178, https://doi.org/10.5194/egusphere-egu25-15178, 2025.

Moisture content available in soil, is a critical parameter for understanding the health of ecosystems, agricultural productivity, and the management of water resources. Soil moisture is an essential component in the growth of vegetation, climate regulation, and the hydrological cycle. The correct estimation of soil moisture is very crucial for optimizing irrigation, enhancing crop yields, and managing water resources. Spatial coverage limits traditional in-situ measurements, while remote sensing-based approaches, especially using SAR imagery, provide scalability to large-scale spatial coverage for soil moisture estimation. This study compares five machine learning-based models- Long Short-Term Memory (LSTM), Random Forest (RF), Multiple Linear Regression (MLR), Multi-layer Perceptron (MLP), and Support Vector Machines (SVM)-for deriving estimates of soil moisture using features based on VV and VH polarizations and incidence angle from SAR imagery. Model performance was also evaluated using in-situ measurements from Vaira Ranch in California's Central Valley, which comprises grasslands and wetlands. Meteorological data, which include precipitation and antecedent rainfall from the ERA5, were used to improve prediction. Each model was hyperparameter tuned, with LSTM adjusting layers, units, and learning rate; RF optimizing tree number, depth, and feature selection; MLR modifying regularization strength; MLP refining layers, neurons, and activation function; and SVM fine-tuning kernel type, C, and gamma. Performance metrics used for evaluation included R² and Root Mean Square Error (RMSE). The results indicated that LSTM outperformed other models with a R² of 0.89, followed by SVM at a value of 0.81 and RF at a value of 0.78. MLP and MLR values were lower at 0.67. This research focuses on the advantages of the integration of remote sensing data and meteorological information for better soil moisture estimation using machine learning and show that the advanced models such as LSTM and RF can effectively predict soil moisture, with important implications for improving agricultural management and resource planning.

How to cite: Sharma, V., Barbhuiya, S., and Gupta, V.: Enhancing Soil Moisture Estimation through Machine Learning Models and Remote Sensing Data, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-20070, https://doi.org/10.5194/egusphere-egu25-20070, 2025.

The sustainable management of soil moisture and salinity is a critical challenge for semi-arid regions like the Mahabad Plain in northwestern Iran. This study applies the HYDRUS-1D model, calibrated using sensor-based data, to simulate water and salt dynamics in a 4 HA sugar beet field. The Mahabad Plain, covering 249 km², experiences annual precipitation of 402 mm and evaporation rates of 1,560 mm. Despite its fertile soils, the region faces persistent challenges such as waterlogging, salinity, and unsustainable irrigation practices, exacerbated by agricultural expansion and climate variability. Sensor data were collected every other day from four soil depths (0–25 cm, 25–50 cm, 50–75 cm, and 75–100 cm) in a single sugar beet field between late June 2024 and late July 2024. These measurements were used to calibrate the HYDRUS-1D model, optimizing parameters such as residual and saturated water content, hydraulic conductivity, and dispersion coefficients. Calibration metrics, including RMSE and Nash-Sutcliffe efficiency, confirmed the reliability of the simulations in replicating observed conditions. The results revealed critical inefficiencies in irrigation practices. Over-irrigation was observed, particularly in deeper soil layers, where moisture levels exceeded the optimal range of 18–25% for sugar beet cultivation. Surface layers (0–25 cm) also exhibited frequent waterlogging after irrigation events, with moisture levels surpassing 25%. Electrical conductivity (EC) levels, however, remained within the safe range of 0.6–1.3 dS/m, indicating effective salt leaching and no immediate risk to crop health. Simulations demonstrated that increasing irrigation intervals by 1–2 days could reduce water consumption by 15–30%, prevent excessive soil saturation, and promote healthier root growth. This approach ensures that soil moisture remains within the optimal range while maintaining crop yield and quality. This study is the first of its kind for the Mahabad Plain, offering a novel application of sensor-calibrated HYDRUS-1D modeling. It provides actionable recommendations for addressing water scarcity and improving agricultural sustainability. By integrating field observations with advanced modeling, the research bridges gaps in water resource management and offers replicable solutions for semi-arid agricultural systems worldwide. The findings are especially relevant as the region faces increasing agricultural demands and environmental challenges, including efforts to restore Lake Urmia. By improving irrigation efficiency and reducing agricultural water consumption, more water can be directed toward Lake Urmia, contributing to its restoration and the broader ecological balance of the region.

How to cite: Hossaini Baheri, M. and Tajrishy, M.: Simulation of Salt and Moisture Dynamics in Agricultural Fields Using HYDRUS: Insights from a Sensor-Based Calibration, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-2901, https://doi.org/10.5194/egusphere-egu25-2901, 2025.

The prediction of soil moisture movement remains challenging due to the complexity of underground flow processes and the availability of accurate soil parameters. There have been attempts to overcome this issue with parametric models and inverse modeling, but it remains challenging because it requires knowledge of initial and boundary conditions. While deep learning offers a solution, the one significant constraint remains not to violate the physical constraints. I present a novel physics-informed neural network (PINN) framework that integrates the soil moisture movement governing equation constraints with deep learning to predict soil moisture dynamics. The new approach follows mass conservation principles and soil hydraulic properties into the neural network's loss function. The model ensures physically consistent predictions. The framework simultaneously learns soil hydraulic parameters and water content distributions, adapting to heterogeneous soil conditions through a hybrid optimization strategy. The model incorporates the Van Genuchten parameterization within the physics-informed architecture to ensure consistency and accuracy. This methodology bridges the gap between computationally intensive traditional numerical solutions and pure data-driven approaches, offering a new paradigm for modeling soil water dynamics.

How to cite: Pathak, V. S.: Physics-Informed Deep Learning for Soil Water Dynamics, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-20587, https://doi.org/10.5194/egusphere-egu25-20587, 2025.