ABSTRACT Methylene Blue Dye Adsorption by Peanut shell. This study aims to investigate the factors influencing the adsorption and to evaluate the potential of waste peanut shells as an effective biosorbent for wastewater treatment. The Methylene Blue (MB) dye was adsorbed out of an aqueous media using a batch technique and a cheap adsorbent made from peanut shells. X-ray diffractometer (XRD), Fourier transformed inferred spectroscopy (FTIR) and Scanning electron microscopy (SEM) analysis were used to ascertain the peanut shell's structure and properties. The effects of different parameters on dye removal performance were examined using a batch system. The outcomes of modeling investigations demonstrated that the best representation of adsorption kinetics and isotherm data was provided by pseudo-second-order kinetics and Langmuir isotherms. Response surface methodology (RSM) based on Box–Behnken design revealed a highly efficient quadratic correlation for optimizing five parameters affecting dye removal, with R2 equal to 97.6% with difference less than 0.2 between R2 adjusted and predicted 95.7% and 91.4% respectively. To estimate the percentage of dye removal, Levenberg–Marquardt (LM) was employed as the training procedure for a feed-forward back propagation neural network (FFBP-NN). The network design 5-25-30-1 was found to have the best layers and neurons after evaluating several layers and neurons.
ABSTRACT Flowchart showing optimization of 176 constructed wetlands using kinetic modelling, Taguchi design, ANOVA and SHAP, yielding operating conditions that enhance BOD, TN and TP removal kinetics. Constructed wetlands (CWs) are sustainable wastewater treatment systems, but their large land requirement limits use in urban and land-constrained areas. This study develops a data-driven kinetic–statistical framework to optimize first-order areal removal-rate coefficients (k-values) for BOD, TN, and TP in full-scale CWs. A secondary dataset comprising 176 operational CWs, including 92 horizontal-flow CWs and 84 vertical-flow CWs, was analyzed across nine design, operational, climatic, and influent-quality parameters. The k-values were estimated using the P-k-C* model standardized to 20°C and optimized using a Taguchi L32 orthogonal array, multivariate regression, sensitivity analysis, interaction analysis, and SHAP-based CatBoost interpretation. Results showed that detention time and organic loading rate mainly controlled BOD removal kinetics, while temperature and media depth strongly influenced TN and TP removal. The optimized operating envelope, including approximately 2-days detention time, 330 g/m3/day organic loading rate, 2.4 m media depth, and 24–34°C temperature, enhanced kinetic coefficients to kBOD = 0.31 m/day, kTN = 0.08 m/day, and kTP = 0.27 m/day. These results indicate that CWs can be shifted toward a high-k, low-footprint operating regime by balancing hydraulic residence time, organic loading, and reactive media capacity. The framework supports compact, hybrid, and climate-resilient CW designs by shifting systems toward a high-k, low-footprint regime, contributing to SDG-6.
ABSTRACT Water treatment facilities in rural areas often face operational and resource limitations. Coagulation and flocculation are essential processes in water treatment, with pH being a critical factor influencing the performance of coagulants. Poly-aluminium chloride (PAC) has been widely recognized for its effectiveness in removing turbidity, colour, and residual contaminants. Current research lacks sufficient focus on its use under rural treatment plant conditions where operational constraints, fluctuating water quality, and limited process control are common. This study addresses that gap by investigating the optimum pH for potable water treatment using PAC at the Tapah Water Treatment Plant. Sequential laboratory-scale jar tests were employed using a one-factor-at-a-time approach to evaluate PAC dosages (25–35 mg/L) and sodium hydroxide (NaOH) adjustments (3–18 mg/L) towards the water quality. Optimal coagulation was achieved at pH 7.0 with a PAC dosage of 31 mg/L and NaOH dosing of 9–12 mg/L, resulting in a final treated water turbidity of 3.6 NTU and residual aluminium concentrations of 0.13 mg/L. This research affirms the significance of adjusting pH for efficient water treatment to become drinkable and still meets the National Drinking Water Quality Standards. The results offer useful information for enhancing efficiency and water quality in water treatment facilities in rural and suburban settings.
ABSTRACT CeO2-Bi2O3 nanocomposite powder is used as a photocatalyst to treat crystal violet dye solution under visible light. The process begins with the photocatalyst powder, which is added to the purple crystal violet dye solution. This mixture is then placed on a magnetic stirrer for the experimental setup, allowing the photocatalyst to act on the dye. Over time, the treated dye solution changes color, shown in a series of test tubes with decreasing purple intensity from left to right, indicating the dye’s gradual degradation. This demonstrates the photocatalyst’s effectiveness in breaking down the dye under visible light. Discharging toxic industrial waste, including dyes, into the environment contaminates water bodies, causing hazardous effects on humans as well as on aquatic life. It highlights the necessity of degrading these dyes before discharging them into water bodies. The photodegradation of dyes using semiconductor metal oxides in the presence of natural visible light is gaining momentum. To improve the harnessing of visible light, rare metal oxides are doped into the photocatalyst. In this work, Bi2O3, an efficient photocatalyst, was used for the crystal violet (CV) dye degradation. It was observed that the degradation efficiency was enhanced by doping it with CeO2. So, a nanocomposite of 6% CeO2-doped Bi2O3 was used as a photocatalyst, and its efficiency was compared with the pure sample. FE-SEM, EDX, XRD, and FTIR analysis indicate effective doping, and the degradation of CV was 92% when 1 g/L of the photocatalyst was used. The biotoxicity measurement using Bacillus subtilis has confirmed that the degradation by-products of CV dye are not more toxic than the parent CV dye compound.
ABSTRACT Graphical abstract of variability and trends of rainfall and its extremes. This study investigates rainfall trends, variability, and extreme precipitation events across 11 districts in the northwestern region of Uttar Pradesh, India, over the period 1970–2021. Utilizing daily rainfall data, the Mann–Kendall trend test and Sen's slope estimator were applied to detect temporal trends. The results indicated a widespread decline in rainfall and rainy days across the regions. The strongest declines were observed at Moradabad, Meerut, and Gautam Budh Nagar, where annual rainfall decreased by –5.93 to –7.58 mm year−1, accompanied by significant reductions in monsoon and post-monsoon rainy days. The analysis of rainfall extreme indices revealed significant decreasing trends for consecutive wet days (CWD) and total annual rainfall (PRCPTOT) at nearly 63.6% (7 out of 11) and 45.5% (5 out of 11) of districts, respectively. Heavy precipitation indices also showed declining patterns, with significant decreases recorded at 36.4% (4 out of 11) of districts for R10, 45.5% (5 out of 11) for R20, and 36.4% (4 out of 11) for RX5day. District-wise analysis indicated significant reductions in extreme rainfall events at Baghpat (RX1day: Q = –0.70; RX5day: Q = –1.26), Meerut (R20: Q = –0.11; R95p: Q = –3.35), and Moradabad (R20: Q = –0.10; RX5day: Q = –1.02). Significant declining trends in total annual rainfall (PRCPTOT) were further observed at Baghpat (Q = –4.42), Gautam Budh Nagar (Q = –4.86), Ghaziabad (Q = –4.95), Meerut (Q = –6.13), and Moradabad (Q = –4.86), highlighting an overall reduction in rainfall intensity and frequency across several locations. These findings underscore the growing impact of climate variability on regional hydrological regimes and highlight the importance of adaptive water resource management strategies in the face of changing precipitation extremes.
ABSTRACT In arid and semi-arid areas, soil erosion is a significant environmental issue, as it causes land degradation, transportation of sediments, and loss of reservoir storage capacity. The present study aimed to estimate the spatial distribution of soil erosion in the Adhaim Dam watershed in Iraq by applying the Revised Universal Soil Loss Equation (RUSLE) coupled with geographic information system (GIS) and remote sensing techniques in the watershed of 11,905 km2. Rainfall erosivity (R), soil erodibility (K), slope length and steepness (LS), cover-management (C), and support-practice (P) factors were computed using CHIRPS rainfall data, FAO soil data, a 30-m Copernicus DEM, and Sentinel-2 imagery. The estimated gross annual soil loss was about 39.98 million t/year (or 33.31 million m3/year assuming a sediment bulk density of 1.2 t/m3). The sediment delivery ratio is approximately 32% as determined by comparison with the sediment reported delivered to Adhaim Reservoir (10.4 million m3/year). Spatially, 65.62% of the watershed was determined to be very low erosion risk, 18.62% low, 6.38% moderate, 2.78% high, 0.62% very high, and about 6% no data. The results indicate priority areas for erosion that can be mitigated through the implementation of check dams, revegetation, and better land management practices that will result in less sediment being delivered to the reservoir and ensure long-term storage sustainability.
ABSTRACT In this study, a novel cost-effective polypyrrole (PPy)-zeolite (PPy/Ze) composite adsorbent material was synthesized via oxidative polymerization, and the incorporation of polypyrrole on the zeolite surface was characterized and confirmed using Fourier-transform infrared spectroscopy and scanning electron microscopy. Batch adsorption experiments using the (PPy/Ze) composite adsorbent demonstrated a crude oil removal efficiency of 99%, with a maximum Langmuir adsorption capacity of 62.112 mg·g−1 achieved within 120 min. Equilibrium data were best described by the Langmuir isotherm model (R2 = 0.9206, p = 0.047), indicating monolayer adsorption on homogeneous surface sites, while kinetic data conformed to the pseudo-second-order model (R2 = 0.9954, p < 0.001). Thermodynamic analysis conducted at 298, 308, and 318 K revealed negative Gibbs free energy values (ΔG° = −7.12 to −5.92 kJ·mol−1), confirming spontaneous adsorption, while the negative enthalpy (ΔH° = −25.00 kJ·mol−1) and activation energy (Ea = 28.00 kJ·mol−1) collectively establish a predominantly physical adsorption mechanism governed by hydrophobic interactions between crude oil components and the PPy/Ze surface. These results demonstrate that PPy/Ze is a highly promising and cost-effective adsorbent for treating petroleum wastewater.
ABSTRACT Conceptual diagram of hydrothermal stratification and depth-dependent water quality in Koga Reservoir. Schematic illustration of vertical water quality gradients in Koga Reservoir, showing surface, middle, and bottom layers with seasonal differences in temperature, oxygen, and nutrients. Vertical temperature shapes vertical water quality gradients, yet depth-dependent variability in tropical highland reservoirs remains poorly documented. This study evaluated seasonal vertical water quality observations in the Koga Reservoir during the dry (February 2025) and wet (August 2025) seasons. Samples from surface, middle, and bottom layers were collected with a Van Dorn sampler, and physicochemical parameters were measured using standard field and laboratory procedures. Observations indicated that temperature decreased by about 9 °C from surface to bottom in both seasons, while dissolved oxygen declined from 5.95 to 2.75 mg/L (dry) and from 4.73 to 1.40 mg/L (wet), with bottom waters below the recommended guideline (>5 mg/L). Nitrate remained within FAO limits but decreased with depth, suggesting potential eutrophication risk. Phosphate and ammoniacal nitrogen increased toward bottom waters, exceeding FAO thresholds (up to 5.98 and 7.15 mg/L, respectively). Turbidity and suspended solids were markedly higher in the wet season (596 NTU and 166.5 mg/L) compared to the dry season (296 NTU and 85.6 mg/L). The findings provide preliminary observations of seasonal vertical thermal and chemical gradients within the Koga Reservoir and suggest that both in-reservoir processes and seasonal catchment conditions may contribute to observed water quality variability.
ABSTRACT Zambia's water utilities import chlorine gas and high-test hypochlorite (HTH) for disinfection. The imports pose service reliability and governance risks, because utilities must maintain residual chlorination despite foreign-exchange shocks, long transport corridors, and periodic stockouts. This study evaluates whether on-site electrolytic chlorination (OSEC), producing sodium hypochlorite from salt and electricity, can provide a cost-effective substitution pathway and whether local salt resources support scale-up. Choma was selected because it combines two interconnected treatment plants (Munzuma and ZESCO), an 18 km trunk main to Kalundu Station, HTH use, and a network representative of small-to-medium utility systems in Zambia. The analysis combines 16 months of plant records, engineering sizing, a 15-year discounted cash-flow analysis (8% discount rate), sensitivity testing, and an assessment of chlorine gas, HTH, and OSEC-grade salt. Results show that OSEC is most viable when deployed at critical network nodes rather than widespread rollout: Kalundu-plus-ZESCO reduces annual disinfection cost by 49% compared with HTH use. Demand uncertainty and system sizing are the main risks. Substituting HTH at suitable plants creates demand for an OSEC-grade salt value chain, but use depends on impurity removal and quality assurance. The study contributes a transferable utility-industry framework for improving disinfectant reliability in import-exposed water sectors.
ABSTRACT Climate change affects water resources in semi-arid basins by altering precipitation, land surface temperature (LST), evapotranspiration, and surface-water persistence. This study assessed hydroclimatic variability in the Lower Zab River Basin (LZRB), Iraq, during 2000–2021 using satellite and reanalysis datasets processed in Google Earth Engine. Linear regression, Mann-Kendall tests, Sen's slope, Spearman correlation, and partial correlation were applied to evaluate trends and climate-water relationships. Surface-water extent showed a slight decreasing tendency (Sen slope = −0.0125% yr−1; p = 0.091), while annual mean precipitation also decreased non-significantly (−0.00114 mm h−1 yr−1; p = 0.191) and LST increased non-significantly (+0.0406 °C yr−1; p = 0.341). Northern and eastern areas were wetter, with higher evapotranspiration and lower LST, whereas the southwest was warmer and drier. Water-body area was more strongly associated with precipitation (R² = 0.6398; r ≈ 0.80) than with LST (R² = 0.483; r ≈ −0.70), while rank-based and partial correlations confirmed LST as an important inverse stress factor. Overall, water-body variability reflects the combined effects of precipitation supply and thermal-evaporative stress, supporting cloud-based geospatial monitoring of climate-water interactions in semi-arid basins.
ABSTRACT Graphical summary illustrating nanobiosensor-based detection of waterborne pathogens. The figure shows nanomaterials functionalized with biorecognition elements that selectively capture pathogens, followed by signal transduction and artificial intelligence-assisted data analysis to enable rapid, sensitive, and accurate detection for water quality monitoring. Water is the inevitable and immediate source of spreading waterborne diseases caused by pathogenic microorganisms, posing significant risks to public health and water quality management. Early detection of waterborne pathogens and their corresponding biomarkers at an early stage in both clinical and environmental water samples is significant for preventing disease outbreaks and reducing the mortality rate. Conventional detection methods are time-consuming, labour-intensive, and less effective for water samples with complex components. Currently, nanotechnology-based biosensors are considered as new-generation detection techniques for rapid, sensitive, and selective detection of waterborne disease biomarkers, such as pathogen-specific nucleic acids, proteins, toxins, and metabolites found in contaminated water and clinical samples. Zinc oxide (ZnO) nanoparticles (NPs) demonstrate high biocompatibility, high surface activity, and improved sensing sensitivity, while gold (Au) NPs exhibit excellent photostability, electromagnetic field enhancement, and signal amplification. Consequently, ZnO-Au NPs have been used to create novel ZnO-Au nanobiosensors for the detection of pathogenic biomarkers and monitoring of water quality. In addition, artificial intelligence (AI) has fuelled the development of intelligent fifth-generation nanobiosensors, which automatically analyse data and recognize patterns. Hence, this review summarizes advances in ZnO-Au nanobiosensors integrated with AI for detecting waterborne disease biomarkers in both environmental and clinical water samples. It also discusses practical waterborne phenomena, challenges related to environmental water factors, and future prospects for intelligent nanobiosensing in field applications.
ABSTRACT Graphical abstract illustrating the water poverty index framework applied to rural communities in Ilam Province, Iran. The figure presents the five WPI components - resources, access, use, environment, and capacity - alongside key findings related to disparities in water access, water-use intensity, environmental conditions, and policy recommendations for improving rural water security. This study evaluates water poverty across rural communities in Ilam Province, Iran, using the water poverty index (WPI) to examine socioeconomic and environmental constraints affecting water security. Data from 251 villages across 12 counties were collected through stratified random sampling and aggregated to compute WPI component scores for resources, access, use, environment, and capacity. Multiple weighting schemes were applied to assess robustness. Results reveal substantial geographic variation across rural districts. Accessibility scores were consistently low (mean = 10.88), while water use (mean = 20.10) and resource availability (mean = 16.19) showed the greatest variability. Some districts exhibited extremely high water-use intensity (>60), indicating localized resource pressure, whereas others showed critically low utilization (<2), reflecting structural constraints. Environmental conditions also varied considerably. Overall, access and capacity were the weakest components, highlighting significant infrastructural and socioeconomic limitations. Despite these disparities, the aggregated WPI value (74.52) falls within the ‘low water poverty’ category under the Centre for Ecology and Hydrology classification. Findings indicate pronounced intra-provincial inequalities and emphasize the need to improve water access and strengthen local adaptive capacity to support equitable rural water security.
ABSTRACT Accurate prediction of reference evapotranspiration (ET0) is essential for irrigation scheduling, water resource management, and climate-resilient agriculture, particularly in semiarid regions with increasing climatic variability and water stress. This study proposes a Physics-Informed Long Short-Term Memory (PRLSTM) framework for medium-range (15-day ahead) ET0 forecasting by integrating physical constraints derived from the FAO-56 Penman–Monteith equation into deep learning. The framework combines data-driven temporal learning with physics-based regularization to improve forecast stability and physical consistency. A 10-year hybrid meteorological dataset (2013–2022) from NASA POWER and ground-based observations in the Delhi NCR region was used for model development. Temperature, relative humidity, solar radiation, wind speed, and atmospheric pressure served as input variables. Comparative experiments against conventional LSTM across four seasons showed that PRLSTM consistently produced more accurate 15-day ET0 forecasts, with the greatest improvement during summer. Residual analysis and validation-loss behavior indicated better generalization and reduced long-horizon error drift. Embedding physical constraints into recurrent neural networks improved the reliability, interpretability, and seasonal robustness of ET0 prediction. The proposed framework can support irrigation scheduling, crop-water demand assessment, reservoir operation planning, and climate-resilient water management while improving irrigation efficiency in semiarid regions.
ABSTRACT Diagram titled flocculant feed rate adjustment in the coagulation process. On the left, a coagulation tank viewing window connects to human icons with question marks and an AI icon with a light bulb. On the right, a speech bubble from the AI displays sludge flocs and a binarized image. In the dewatering of sludge generated at wastewater treatment plants (WWTPs), setting the flocculant feed rate appropriately is vital. This study assesses operational bias in the human judgment of flocculant feed rate at a WWTP. At the studied WWTP, experienced staff observe floc size through the window of the coagulation tank and adjust the flocculant feed rate to maximize flocs with a projected area of 17–22 mm2. However, quantitative image analysis reveals that the area ratio of the target size does not increase consistently after dosage adjustments are made. In contrast, the dosage adjustment resulted in a significant decrease in the number of flocs of 7 mm2 or less and an increase in the number of flocs of 42 mm2 or more. Despite these discrepancies, no major issues occurred in the dewatering process during the study period. Thus, the dosing rate adjustment may have been influenced by observer bias in determining the size of flocs. If the experienced staff is replaced, the observational criteria can differ, and inappropriate adjustment can potentially lead to the dewatering process failing or to excessive consumption of chemicals. Therefore, incorporating quantitative image analysis into operational decisions can facilitate consistent dosing regardless of operator experience.
ABSTRACT The graphical abstract shows the location map of the study area, the stream order, the drainage density of the Infranz watershed, and the conceptual model of the study. This study examines the hydrological implications of morphometric parameters in the Infranz watershed, Lake Tana sub-basin, Northwestern Ethiopia. The study aims to examine the influence of morphometric parameters on the hydrological characteristics of the watershed. The areal, relief, and linear aspects of watersheds were quantified by using digital elevation model (DEM) data and GIS-based morphometric analysis. Results indicate an elongated, moderately dissected basin with a high compactness coefficient and a very coarse texture, implying slow runoff, low peak flow, and reduced erosion. The watershed relief (311 m) and low ruggedness (0.221) reflect limited vertical dissection and moderate energy conditions. The dendritic drainage pattern consists of four stream orders (55 streams, 152.37 km total length) with low and very coarse drainage density (0.712 km/km2) and low stream frequency (0.257 streams/km2), signifying a well-drained catchment with moderate structural control. These morphometric features collectively shape the watershed's runoff response, sediment transport, and flood regulation, thereby influencing the hydrology and ecological integrity of its riverine wetlands. Understanding these linkages supports evidence-based watershed management and wetland conservation in the Lake Tana sub-basin.
ABSTRACT Quantifying the Impact of Land Cover Change on Groundwater Recharge of Welmel Watershed in the Genale-Dawa Basin Using the WetSpass Model. This study assessed the impact of land use and land cover (LULC) change on groundwater recharge in the Welmel watershed of the Genale-Dawa Basin using the WetSpass model. Spatial data on land cover, soil texture, slope, topography, groundwater level, and long-term meteorological records were used as model inputs. Land cover maps for 1991, 2001, 2011, and 2021 together with 33 years (1987–2020) of daily meteorological data were applied to estimate long-term seasonal and annual groundwater recharge. Model inputs were prepared as 30 m resolution grid maps and calibrated using literature and expert knowledge to represent watershed conditions. The results show that the mean annual groundwater recharge decreased from 241.8 mm in 1991 to 221.1 mm in 2001, 195.1 mm in 2011, and 168.5 mm in 2021. These values correspond to 22.2, 20.3, 17.9, and 15.5% of the mean annual precipitation (1,087.5 mm), respectively. Seasonal analysis indicates that most recharge occurs during the Belg and Kiremt seasons. LULC analysis revealed significant forest reduction and expansion of agricultural land, settlements, bush land, and grassland, which increased surface runoff and reduced infiltration. Overall, mean annual groundwater recharge declined by about 6.7% between 1991 and 2021 that corresponds to 73.3 mm equivalent to 23.9 and 48.9 mm of the wet and dry season, respectively, occurred mainly due to deforestation and agricultural expansion. The findings provide important baseline information for sustainable water resource management and land use planning in the watershed.
ABSTRACT Flowchart showing the study workflow: hydrometeorological data acquisition, preprocessing, feature engineering into four input scenarios, temporal lag configurations, training of Prophet, ML, and DL models, performance evaluation, and key result of NSE approximately equal to 0.9984. River stage forecasting in data-scarce, monsoon-driven deltaic systems remains a persistent challenge for flood early warning. This study evaluates Facebook's extended Prophet framework against established machine learning (linear regression, LightGBM regression, random forest regression) and deep learning (bidirectional LSTM, gated recurrent unit) models for daily river stage prediction at multiple lead times (3, 7, 15, and 30 days) on Bangladesh's Old Brahmaputra River using 15 years of hydrometeorological data (2007–2021). Four input scenarios were assessed: (i) rainfall-only, (ii) temperature-only, (iii) combined rainfall–temperature, and (iv) river stage–rainfall–temperature integration. For meteorological-only inputs, Prophet achieved 30-day Nash–Sutcliffe efficiency (NSE) of 0.79, outperforming all alternatives – a counterintuitive result where accuracy improved with increasing lead time due to Prophet's additive seasonal decomposition. When the antecedent river stage was included, all models improved, with Prophet and linear regression achieving statistically indistinguishable accuracy (MAE ≈ 0.043 m, NSE ≈ 0.9984), demonstrating that autoregressive dependencies dominate when hydrological measurements are available. A novel finding is that temperature-only inputs sustained Prophet-based forecasting at 98.8–99.4% of rainfall-driven accuracy, providing a viable contingency for data-scarce networks. Principal component analysis–based ranking synthesized nine performance metrics, revealing systematic underprediction during extreme flood events – a critical limitation for operational deployment in vulnerable deltaic regions.
ABSTRACT The complex, nonlinear synergy between three-dimensional hydrodynamics and transient geotechnical processes governs riverbank erosion in tidal river deltas. Traditional assessments often decouple these mechanisms, limiting predictive accuracy under combined extreme forcings. This study proposes a novel hybrid deterministic-stochastic framework that integrates in-situ observations, physics-based modelling, and machine learning to quantify riverbank instability. First, a 3D RANS hydrodynamic model (Delft3D) was one-way coupled with transient seepage and limit-equilibrium models (GeoStudio) to generate model-derived reference stability states. Using the Buckingham Pi theorem, the physical variables were reduced to two critical dimensionless groups representing hydrodynamic toe scour (driven by secondary flows) and geotechnical susceptibility (driven by rapid drawdown pore-pressure lag). To formulate a generalised predictive metric, a Monte Carlo simulation was executed to generate a diverse synthetic dataset, upon which Logistic Regression was applied to calibrate a novel Composite Erosion Index (IE). The index was calibrated on a synthetic dataset comprising 500 Monte Carlo scenarios, each labelled by the coupled deterministic models. The data-driven calibration achieved a classification accuracy of 96.7% with respect to these model-derived reference labels, quantifying the fidelity of the closed-form index to the physics-based simulations rather than field-predictive skill, and isolated a decision boundary (IE ≥ 1.0) that normalises the disparate magnitudes of fluid shear and soil resistance. Application to the highly dynamic Vam Nao River in the Vietnamese Mekong Delta indicated that catastrophic mass failures are most likely to occur when severe secondary currents and rapid tidal drawdown coincide in space and time. The proposed index provides a scalable screening tool for assessing geomorphic hazards and supporting the design of Nature-based Solutions (NbS) for riverbank stabilisation in complex deltaic environments. The calibrated coefficients are site-specific, and independent validation against dated field-failure inventories remains a priority for future work.
ABSTRACT Graphical overview of the study framework and key findings. The study begins with a comparison of three major stormwater treatment test protocols (BW CoP, NJCAT, and DIBt), leading to the selection and adaptation of the DIBt protocol. The adapted protocol is applied under controlled laboratory conditions to evaluate both filter chamber performance (particle and hydrocarbon removal) and filter media performance (metals and nutrients). Results indicate strong removal of suspended solids (TSS), hydrocarbons, copper (Cu), and phosphorus (P), while zinc (Zn) shows fluctuating behaviour and nitrogen (N) exhibits negative removal. The findings highlight the influence of hydraulic conditions and cumulative loading during consecutive sub-tests, and emphasize the need for harmonized, region-specific testing frameworks for compact stormwater treatment systems. Urbanization increases impervious surfaces, generating higher stormwater volumes and pollutant loads that challenge conventional drainage systems. While nature-based solutions (NBS) are widely used for stormwater management, compact stormwater treatment systems (CSTS) provide a decentralized alternative suited to space-limited urban areas. However, their evaluation is hindered by inconsistent international testing protocols. This study compares three major CSTS performance standards, the British Water Code of Practice (BW CoP), the New Jersey Corporation for Advanced Technology (NJCAT) program, and the Deutsches Institut für Bautechnik (DIBt) protocol, to identify methodological differences and testing implications. Based on this comparison, a modified DIBt protocol is developed, expanded to include nutrient (phosphorus and nitrogen) tests, was applied to a filter chamber under controlled laboratory conditions. Results showed particle removal ranging from 25% at high flow to 97% at low flow conditions, hydrocarbon removal remaining consistently high (98–99%), and metal removal efficiencies varying with flow, copper increasing from 19 to 95% and zinc from slightly negative (release) to 88%. Nutrient tests indicated phosphorus removal from 32 to 84% and negative nitrogen removal rates. These findings highlight the need for harmonized testing frameworks reflecting regional conditions to improve CSTS evaluation and implementation.
ABSTRACT Diagram contrasting a 63.38% flooded urban area before protection with a 6.19% area after bypass canal construction, concluding with a recommendation to keep bank roughness below 0.05. Urban centers in semi-arid regions face escalating flood threats exacerbated by complex topography and intensifying climate extremes. To address limitations in existing flood assessments, this research evaluates the hydraulic performance of a future protection project covering both the Medjerda River and the bypass canal simultaneously in Bou Salem, Tunisia. Coupling high-resolution 5 m LiDAR topography and ESA WorldCover data within a fully 2D HEC-RAS hydrodynamic model enabled the simulation of flood dynamics across 10- to 100-year return periods. The integrated megaproject (bypass canal, dikes, dredging) reduces the 100-year flooded urban extent from a baseline of 63.38% to only 6.19%. Furthermore, evaluating operational resilience through Manning coefficient variations reveals high system sensitivity to riparian roughness. If vegetation maintenance is neglected, a roughness increase to n = 0.05 swells the flood extent by 73%, while reaching n = 0.09 causes a 121% increase, effectively nullifying the initial protection. Practically, this study provides quantitative evidence that long-term physical defence reliability inherently depends on continuous riverbank management.
