Reference evapotranspiration (ETo), a key component of the hydrological cycle, is fundamental for agriculture. Traditionally, ETo is estimated using the Penman-Monteith (PM) method, considered the standard method by the FAO due to its use of multiple climatic variables, providing a solid physical basis. This research aimed to assess machine learning techniques to estimate ETo at the Yauri meteorological station in Peru. Monthly data on air temperature (maximum, average, and minimum), wind speed, relative humidity, and extraterrestrial solar radiation were used. Two machine learning techniques, K-nearest neighbors (KNN) and artificial neural networks (ANN), were trained and tested. To verify their accuracy, scatter plots, box plots, and various performance metrics were employed. These metrics included mean absolute error (MAE), anomaly correlation coefficient (ACC), Nash--Sutcliffe efficiency (NSE), Kling--Gupta efficiency (KGE), and spectral angle (SA). The results indicate that machine learning techniques provide highly accurate estimates and can serve as viable alternatives for estimating ETo, especially in situations with limited meteorological data. The implementation of these methods can significantly improve water resource planning and management. This improvement is particularly valuable in agricultural regions with data scarcity, offering a practical tool for farmers and water managers to make informed decisions and enhance resource efficiency. The integration of machine learning in this context demonstrates its potential to address critical challenges in hydrology and agriculture.

Water is a valuable natural resource and national asset, and the primary component of ecosystems. Water sources include rivers, lakes, glaciers, rainwater, and groundwater. Water resources are essential for many economic sectors, including agriculture, animal production, forestry, industrial operations, hydropower generation, fisheries, and more. Water availability and quality are deteriorating due to factors such as population growth, industry, and urbanization. The Water Quality Index (WQI) is a useful and exclusive classification system that summarizes all aspects of water quality in a single phrase. This rating system aids in selecting the most suitable treatment method to address the challenges. The models were evaluated using key statistical factors, a dataset with six relevant parameters, and water use records. The database included electrical conductivity, pH, dissolved oxygen, nitrates, phosphates, suspended particles, and water temperature. We used three machine learning models for the classification of water namely, Support Vector Machine (SVM), K-Nearest Neighbor (KNN), and Random Forest (RF). This study was conducted on Keenjhar Lake, Karachi, Pakistan, and this work employed a dataset of 360 instances and six defining features from 1993 to 2022 on the monthly dataset. The classification algorithms were evaluated using five metrics: accuracy, recall, precision, Pearson’s correlation, and F1 score. In terms of classification, the testing results indicate that the SVM model performed the best, predicting Water Quality Classification (WQC) values with an accuracy of 99.50%. It is important to note that precise water quantity and quality predictions are vital for sustainable resource management, public health protection, and environmental preservation.

The presence of groundwater flow in a construction project affects the construction process, resulting in issues ranging from excavation stability to the redesign of foundations and infrastructure, leading to project delays and major revisions. Groundwater level drawdown, a controlled technique in Civil Engineering, mitigates this effect. Understanding the aquifer is important to ensure accurate representation in the model and to plan the appropriate dewatering technique. In Mexico, the regulation of these procedures related to laws, norms, and regulations lacks a specific bibliography. Additionally, the availability of databases with piezometric information is insufficient and limited. Methodologies for analysis have evolved worldwide to capture the system's complexity, employing numerical models to assess its behaviour. First, it was necessary to characterize the area, considering the water-table level. Subsequently, a numerical model was developed using ModFlow and its ModelMuse interface, both developed by the United States Geological Survey (USGS). This allowed the evaluation of different scenarios in response to proposed dewatering techniques and the anticipation of potential impacts, thereby avoiding the trial and error practice. Analyses of “Torre Tres Ríos”, a building in Sinaloa, Mexico, were conducted to assess water-table behaviour in the area. In July, the water-table was at 33 meters above sea level (masl). By October, it had risen to 35.74 masl, attributed to the recharge due to the rainy season and the influence of the Tamazula River. By November, it had dropped to 35.20 masl, indicating a discharge process. The steady-state model initially represented with piezometric levels close to 33 masl in July. Subsequent transient-state model outputs for October and November reported water-table elevations at 35.676 and 35.438 masl, respectively. Calibration results revealed a mean absolute error of 0.15 meters and a standard deviation of 0.174 meters, approving the model results. This information is important for informed decision-making in dewatering processes, enabling precise adjustments in pumping.

In the study of the aquatic environment, determining the chemical makeup and governing elements of groundwater is crucial. This study provides a thorough physico-chemical examination of the groundwater in Uzbekistan's Koson district. A vital resource for the area, the groundwater supports agricultural practices and supplies the locals with drinkable water. Key physico-chemical parameters, such as pH, electrical conductivity, total dissolved solids (TDSs), and hardness, as well as the concentrations of major ions like calcium, magnesium, sodium, potassium, bicarbonates, chlorides, sulfates, and nitrates, were measured in samples taken from various sites throughout the district in order to evaluate the water quality. The results show that both anthropogenic activities and naturally occurring geological formations significantly influence the geographical variability in groundwater quality. Elevated TDS and nitrate levels were found in a number of samples, suggesting that insufficient waste management techniques and agricultural runoff may have contaminated the area. This study also pinpointed regions where problems with water hardness exist, creating difficulties for both industrial and household uses. Through the distribution mapping of various indicators, we offer a comprehensive comprehension of the Koson groundwater quality. The results emphasize the necessity of focused management plans to safeguard and enhance groundwater supplies. Implementing sustainable farming methods, improving the infrastructure for waste treatment, and conducting routine monitoring are among the suggestions made to guarantee the long-term security and supply of groundwater for the district's requirements. This study supports Uzbekistan's efforts to protect its water resources and offers insightful information on regional water quality challenges.

The accurate quantification of soil infiltration rates is paramount for effective water resource management. However, obtaining precise field-based measurements remains a challenge due to the lack of comprehensive datasets. Understanding soil infiltration is crucial for optimizing water use, enhancing groundwater recharge, and mitigating water-related issues such as runoff and soil erosion.

This study addresses the critical need for field-based measurements of soil infiltration rates in Gaya district, Bihar. Utilizing a mini disc infiltrometer, infiltration characteristics were assessed across all 24 blocks, encompassing diverse soil types and land uses. The mini disc infiltrometer measures the rate at which water enters the soil through a small disc placed on the surface, providing insights into soil permeability and water absorption dynamics.

The results revealed significant variability in infiltration rates, with average cumulative rates ranging from 0.38 to 2.20 cm/min. Notably, the initial infiltration rates were uniformly high across all blocks, gradually decreasing with successive readings. Moreover, forested areas exhibited higher cumulative infiltration rates compared to urban and grassland regions. Approximately one-third of the blocks demonstrated infiltration rates exceeding the area average, indicating favorable conditions for groundwater recharge and emphasizing the importance of implementing recharge structures.

Further analysis identified reduced infiltration rates in inundated areas, attributed to elevated soil water table levels. To address the lack of comprehensive datasets, a district-wide infiltration rate map was developed, serving as a valuable resource for decision-making in water resource management. These findings underscore the critical role of field-based infiltration measurements in informing sustainable water management practices. By bridging the gap in data availability and offering insights into soil–water dynamics, this study contributes to the enhancement of water resource management and resilience in the face of changing environmental conditions.

The present work focuses on optimizing solar stills to address global water scarcity, impacting 2.2 billion people, aligning with UN Sustainable Development Goal 6 for sustainable water management. Solar still desalination is particularly suited to off-grid applications due to its integration with renewable energy sources.
In the scientific literature, efforts to employ AI for predicting solar still performance are hindered by the scarcity of experimental data. To overcome this, we introduce an innovative open-source Python algorithm designed to optimize solar still designs. Validated with a precise 4% error margin, this model accurately forecasts performance and addresses data scarcity by generating a comprehensive dataset for enhanced machine learning training.
The algorithm employs the 4th-order Runge–Kutta (RK4) method to solve differential equations, calculating temperatures (water, cover, absorber, and insulation), cumulative condensed water flow, efficiency, and cost. It adjusts computations based on ambient temperature and solar irradiation data, utilizing interpolation techniques for increased precision.
Additionally, the algorithm provides a visualization of device configurations and includes detailed technical descriptions. This encompasses geometric features, meteorological conditions, environmental factors, and materials data stored in an adjustable dataframe. It calculates thermodynamic properties using equations of state from the IAPWS association for each iteration. Moreover, hydraulic considerations such as the Colebrook–White equation approximation via Newton’s method for turbulent regimes are integrated to estimate the Darcy friction factor for inclined, cascade, and stepped solar still configurations.
By optimizing parameters and materials, the algorithm enhances solar still efficiency while balancing cost-effectiveness. It minimizes resource expenditures and enriches machine learning training data, demonstrating potential for innovative, economically viable solar desalination solutions.

This study was carried out by combining numerical modeling and experimental measurements to investigate the hydraulic characteristics of ungated and gated ogee spillways with high head ratios. The primary objective was to validate the use of a numerical model as a complementary approach to the experimental model for simulating the hydraulic behavior of these spillways, providing a more comprehensive understanding of their hydraulic properties under varying conditions of head ratios and relative gate openings. An Acoustic Doppler Velocimeter (ADV) was used to measure the vertical flow velocity distributions, and ultrasonic sensor wave gauges were used to obtain the time history of the water level. The results of the measurements were compared with the simulation results using a model fitted with three different turbulence models (realizable k-ε, RNG k-ε, k-ω SST). The numerical model was developed using OpenFOAM. With respect to the ungated spillway, three different head ratios ranging from 1.4 to 4.6, which correspond to high head ratios, were investigated. Similarly, three different relative gate openings ranging from 0.5 to 2 were investigated for the gated spillway. The results of water surface profiles and velocity profiles suggest that the numerical and experimental models achieve a good agreement for sections located further away from the spillway. For the ungated spillway, the simulation results for the near-spillway sections are enhanced when the head ratio increases. Considering the velocity profiles and error analysis, the realizable k–ε model was found to best predict the results of the experimental model. A discussion about the discharge equation, velocity fields, pressure fields, and the corner separation zone is also included in this study.

Water pollution poses a significant threat to both human health and ecological systems worldwide. This abstract proposes the development of a comprehensive database that catalogues chemical agents (natural and human-made) that are found in various water bodies such as lakes, rivers, lagoons, and coastlines. The database will systematically document these pollutants alongside their associated harmful effects on human health and ecosystems. The impetus for creating this database stems from the growing need for a centralized, accessible repository of information that can facilitate research, policy-making, and public awareness. Water quality monitoring often yields fragmented data, which hinders the comprehensive understanding of the impact of pollutants. By aggregating data from various studies, reports, and monitoring programs, this database aims to bridge this gap, providing a holistic view of water contamination. Existing databases focus on specific aspects, such as certain types of pollutants or particular water bodies and may not always link the pollutants to their health and ecological impacts comprehensively. The proposed database would differ by aiming to be more comprehensive and integrative, covering not only a broader range of pollutants but also associating each chemical with its specific harmful effects. By aggregating this information into a single, accessible repository, the proposed database would provide a holistic view of water contamination, filling the gaps left by existing databases and offering a detailed and actionable resource for addressing water quality issues. It will help multiple stakeholders, such as researchers, environmental agencies, and policy-makers, to study the prevalence and impact of specific pollutants to identify priority areas for
intervention, draft evidence-based regulations, monitor the effectiveness of implemented policies, and identify and mitigate public health risks. Additionally, the database will empower the general public and advocacy groups with the necessary knowledge to engage in informed dialogue about water quality issues and advocate for safer environmental practices.

This study focuses on the application of Hydrologic Engineering Center's River Analysis System (HEC-RAS) to simulate the dynamic behavior of the Rideau River, located in eastern Ontario, Canada. The study area spans between 25 and 30 km of the river, including a bifurcation at Hogs Back Falls in Ottawa, where the river splits into the Rideau Canal and the Rideau River before merging with the Ottawa River. HEC-RAS is utilized to simulate river discharge, leveraging a substantial database of flow measurements from various gauging stations. These measurements allow for the determination of discharge rates for return periods ranging from 2 to 1000 years, calculated using the Gaussian method. The table below outlines the discharge values to be input into the HEC-RAS model. In addition to hydraulic simulations of the Rideau River, this study employs artificial intelligence (AI) to predict river discharge based on meteorological variables such as maximum temperature (Tmax), minimum temperature (Tmin), mean temperature (Tmean), and precipitation. By integrating AI with traditional hydrological modeling, the study aims to enhance the accuracy and reliability of flood predictions. By gaining a deeper understanding of the influence of temperature on the occurrence of spring floods, researchers and practitioners can improve the effectiveness and applicability of machine learning techniques in flood prediction and management. This study's findings underscore the importance of considering temperature fluctuations, precipitation levels, and historical discharge data in flood modeling. These insights, coupled with the reliable predictions provided by the AI model, empower decision-makers to make more accurate and effective decisions in flood management strategies. This leads to improved mitigation and adaptation measures in response to increasing flood risks. This integrated approach enhances flood prediction capabilities and supports the development of robust and adaptive flood management strategies for the Rideau River and similar watersheds.

Flood risk assessment is crucial for effective water resource management and flood mitigation strategies. This study focuses on detecting and classifying flood risk in the Rideau River basin, located in eastern Ontario, Canada, into four distinct classes. The most significant flood occurred in 2019, surpassing a 100-year flood event, and serves as a stark reminder of how climate change impacts our environment. Considering the limitations of machine learning (ML) models, which heavily rely on historical data used during training, they may struggle to accurately predict such “non-experienced” or “unseen” floods that were not encountered during the training process. To tackle this challenge, our study has utilized a combination of numerical modeling and ML to create an integrated methodology. A comprehensive dataset of river flow discharge was generated using a numerical model, encompassing a wide range of potential future floods. This synthetic dataset ensures that the ML model is trained on a diverse array of scenarios, enhancing its ability to predict flood risks accurately, even for events outside the historical record. The AI model classifies flood risk into four categories: low, moderate, high, and very high. These classifications are based on various factors, including river discharge, precipitation, and temperature data. The integration of AI with numerical modeling provides a robust framework for flood risk assessment, enabling more accurate predictions and better-informed decision-making. The findings highlight the importance of using advanced modeling techniques to address the limitations of traditional ML models in flood prediction. By incorporating a wide range of potential future floods into the training dataset, the study improves the model's ability to predict and classify flood risks under changing climatic conditions. The study underscores the need for adaptive measures to mitigate the increasing flood risks posed by climate change, ensuring the resilience and safety of vulnerable communities along the Rideau River.