Nash-Sutcliffe Efficiency Values Research Articles

An advanced machine learning-based mathematical equation for predicting compression index from readily-available soil physical properties

ABSTRACT The Compression index, known as Cc, plays a crucial role in geotechnical design. However, the conventional approach to ascertain Cc is not only expensive but also time-intensive. This paper proposes an innovative modeling approach to develop a Deep Neural Network (DNN)-based formula for the prediction of Cc incorporating readily accessible physical properties of soil, including specific gravity (G), saturation degree (Sr), liquid limit (LL), silt content (SC), and clay content (CC). To assess the predictive performance of the proposed formula, various metrics were employed, such as determination coefficient (R²), root mean square error (RMSE), mean absolute error (MAE), and Nash–Sutcliffe Efficiency (NSE). From the collected dataset, 608 samples were used as the training set for the DNN model, while the remaining 152 samples were designated as the testing set. The R², RMSE, MAE, and NSE values for the training and testing sets were 0.9821 and 0.9935; 0.126 and 0.098; 0.083 and 0.074; 0.9812 and 0.9922, respectively. These results indicate that the developed equation demonstrates outstanding predictive accuracy for estimating Cc. Furthermore, the comparative analysis underscores that the proposed DNN-based formula significantly outperforms empirical equations and multilinear regression (MLR) model in predicting Cc with higher accuracy.

Hybridizing evolutionary algorithms and multiple non-linear regression technique for stream temperature modeling

Abstract The present study hybridizes the new-generation evolutionary algorithms and the nonlinear regression technique for stream temperature modeling and compares this approach with conventional gray and black box approaches under natural flow conditions, providing a comprehensive assessment. The nonlinear equation for water temperature modeling was optimized using biogeography-based optimization (BBO) and invasive weed optimization (IWO), simulated annealing algorithm (SA) and particle swarm optimization (PSO). Two black box approaches, a feedforward neural network (FNN) and a long short-term memory (LSTM) network, were also employed for comparison. Additionally, an adaptive neuro-fuzzy inference system (ANFIS) served as a gray box model for river thermal regimes. The models were evaluated based on accuracy, complexity, generality and interpretability. Performance metrics, such as the Nash–Sutcliffe efficiency (NSE), showed that the LSTM model achieved the highest accuracy (NSE = 0.96) but required significant computational resources. In contrast, evolutionary algorithm-based models offered acceptable performance while reducing the computational complexities of LSTM, with all models achieving NSE values above 0.5. Considering interpretability, accuracy and complexity, evolutionary-based nonlinear models are recommended for general applications, such as assessing thermal river habitats. For tasks requiring very high accuracy, the LSTM model is preferred, while ANFIS provides a balanced trade-off between accuracy and interpretability, making it suitable for engineers and ecologists. While all models demonstrate similar generality, this model is developed for a specific location. For other locations, independent models with a similar architecture would need to be developed. Ultimately, the choice of model depends on specific objectives and available resources.

Machine Learning-Driven Analysis of Heat Transfer in Chemically Reactive Fluid Flow Considering Soret-Dufour Effects

ObjectiveThe study investigates tangent hyperbolic nanofluid flow across an extended media, emphasizing magnetohydrodynamic influences, mixed convection, and nanoparticle transport with the help of an innovative and advanced technique. MethodologyIt employs a sophisticated Gaussian Neural Network and Hybrid Cuckoo Search to mimic complex fluid dynamics accurately, ensuring efficient computation and solution precision. By transforming the governing system of nonlinear partial differential equations into ordinary differential equations via similarity transformation, the methodology attains computational efficiency and solution precision. Core FindingsStatistical measures confirm the model's resilience, indicating absolute error variations from E−05 to E−12 and mean squared errors continuously between E−02 − E−07, highlighting the reliability of the model, while the Error in Nash-Sutcliffe Efficiency values fall within the interval E−05 − E−15. The Theil's Inequality Coefficient values lie in the range of E−01 − 10−07. Future Direction and ApplicationsThe findings underscore the model's capacity to improve thermal management, energy efficiency, and process dependability in sectors like chemical processing, environmental engineering, and sustainable energy production. This study positions the Gaussian Neural Networks framework as a robust computational instrument, providing a foundation for subsequent investigations into intricate geometries, multi-physics phenomena, and magnetohydrodynamic systems for novel technology applications.

A Machine Learning Approach for Filling Long Gaps in Eddy Covariance Time Series Data in a Tropical Dry Forest

AbstractLong‐term eddy covariance (EC) data are crucial for understanding the impact of global change on ecosystem functions. However, EC data often contain long gaps, particularly in tropical dry forests (TDF) due to seasonality and El Niño‐Southern Oscillation (ENSO) phases. These factors create high variability, complex dependencies, and dynamic flux footprints. No current gap‐filling method adequately addresses long gaps in TDFs. This study introduces a novel framework for addressing this issue by (a) defining gap sizes by their relative percentages, (b) training, tuning, and evaluating two machine learning (ML) models: MissForest for short gaps and Prophet for intermediate and long gaps, and (c) predicting half‐hourly EC data from 2013 to 2022 for six EC variables, where actual gap data sets ranged from 26.6% to 28.4%, at TDF in Costa Rica. Results indicate that MissForest excelled at filling short gaps (≤5%, R2 = 0.76 and Nash‐Sutcliffe efficiency (NSE) = 0.71), while Prophet performed exceptionally well for gaps between 5% and 10% (R2 = 0.72 and NSE = 0.67). However, both models struggled with gaps between 10% and 13%. Validation showed R2 values of 0.79, 0.88, and 0.77 for CO₂ flux, sensible heat flux, and latent heat flux, respectively, with corresponding NSE values of 0.78, 0.86, and 0.72, and normalized root mean squared error (NRMSE) around 2E‐4. Additionally, to validate our results, we applied our approach at three EC sites with different ecological conditions, demonstrating robust performance. This study presents a reliable ML approach for imputing long gaps in EC data, which can be applied to sites with strong variability.

Biphasic models improved S‐metolachlor dissipation endpoint predictions

AbstractAccurate predictions of herbicide dissipation endpoints are essential for making environmental management decisions, estimating the duration of weed control, and deciding crop replant and rotation options. The performance of the single first‐order (SFO), first‐order multiple‐compartment (FOMC), bi‐exponential (BEXP), first‐order double‐exponential decay (FODED), and first‐order two‐compartment (FOTC) was evaluated to describe S‐metolachlor dissipation. Results showed that the BEXP and FOTC were not appropriate. The FODED was ranked as the top model with the lowest Akaike's information criterion (AICc) value among the other three candidate models and was followed by the FOMC model. The SFO displayed the largest root mean square error (11.43) and mean absolute error (9.16) values, and the smallest Nash–Sutcliffe efficiency value (0.88) within the set of three appropriate candidate models. The SFO predicted a time it takes the herbicide to dissipate to one‐eighth of its original concentration, 21 and 26 days shorter than that predicted by the FOMC and FODED, respectively. The SFO also predicted a time for S‐metolachlor to dissipate to one‐tenth of its original concentration, 33 days shorter than that predicted by both FOMC and FODED. Therefore, even though the SFO is effective at predicting half‐life, caution might be required when making inferences about DT90 or the time required for the dissipation of 90% of the initial concentration of pesticides in the soil using the SFO.

GIS-based application of hydrological modelling in the Borkena Catchment, Ethiopia

ABSTRACT Hydrologic models have been used in different river basins across the world for a better understanding of the hydrological cycle and the water resource availability. This study tested the applicability of the Hydraulic Engineering Center Hydrologic Modelling System (HEC-HMS) in the Borkena Catchment of Awash Basin. The HEC-HMS system is used to estimate peak flood, time to peak, and runoff volume. Model calibration and validation were carried out using the observed streamflow data from 1989 to1997 and from 1998 to 2002, respectively. The observed and simulated peak streamflows for calibration were 165.3 and 122.6 m3/s and for validation were 197.3 and 155.2 m3/s, respectively, whereas the average volumes were 1,223 and 1,380 mm for calibration and 1,647 and 1,847 mm for validation, respectively. From these results, it is observed that HEC-HMS slightly underestimated peak flow and overestimated runoff volume in both calibration and validation periods. Nash–Sutcliffe Efficiency values of 0.65 and 0.70, Coefficient of Determination values of 0.69 and 0.71, and Percent Bias of 12.84 and 12.16% for calibration and validation, respectively, were obtained. It can be concluded that in areas with limited water resources and ungauged watersheds like Borkena Catchment, simulated results of this study can be used for the design of different infrastructures for flood control and prevention in the study area and adjacent catchments.

Optimizing irrigation and planting strategies to prevent non-point source pollution in the Hetao Irrigation District using SWAT-MODFLOW-RT3D model

Non-point source (NPS) pollution is a significant issue in the Hetao Irrigation District (HID), affecting local water environment and ecological safety. To explore solutions for managing NPS pollution in the HID, this study establishes the SWAT-MODFLOW-RT3D (SMFR) model based on data from recent years. The results show that the determination coefficient (R2) of monthly drainage simulation for both the calibration and validation periods are >0.6, the Nash-Sutcliffe efficiency (NSE) values are >0.7, and the percent bias (PBIAS) values are <10 %. For pollutant transport simulation (total nitrogen and total phosphorus), the R2 values for both calibration and validation periods exceed 0.6, the NSEs are >0.75, and the PBIAS values are below 20 %. The model demonstrates good performance in simulating hydrology and pollutant transport in the study area, indicating strong applicability. Building on this, the spatial and temporal distribution characteristics of nitrogen and phosphorus losses in the HID were analyzed, and the impacts of different agricultural management practices, such as irrigation and fertilization regimes, drip irrigation under film (DIUF), and crop structure adjustments, on NPS pollution were assessed. Extensive scenario testing indicates that deficit irrigation and DIUF reduce total nitrogen (TN) in agricultural NPS pollution by 7.74 % and 9.34 %, respectively, and total phosphorus (TP) by 35.12 % and 40.99 %, respectively. In contrast, crop structure adjustments have a relatively moderate reduction effect on TN and TP, ranging from 2.66 % to 6.80 %. Additionally, the study examines the NPS pollution control effects of various combined scenarios. Through economic benefit evaluation, it was found that the scenario combining deficit irrigation with crop structure adjustments has a lower cost-effectiveness (CE) ratio and a higher cost-benefit (CB) ratio, making it a more favorable measure for controlling NPS pollution in the HID.

A comparative analysis of rainfall data from IMERG early run and ground-based rain gauges on Bali Island

Abstract Ground-based precipitation measurements encounter challenges in various parts of Bali due to the limited number of gauge stations. Therefore, it is essential to identify dependable alternatives like satellite-derived precipitation data, which offer continuous precipitation time series with high spatial resolution. This study assessed the effectiveness of near real-time global satellite precipitation products, specifically the Integrated Multi-satellite Retrievals for Global Precipitation Measurement-Early Run (IMERG-E) compared to gauge data from 43 stations across Bali Province. To evaluate IMERG-E datasets, traditional point-to-pixel comparison techniques were employed, alongside statistical metrics including correlation coefficient (CC), mean absolute error (MAE), relative bias (RB), and Nash-Sutcliffe efficiency (NSE). The comparative analysis showed that the daily IMERG-E dataset performs moderately well, as evidenced by weak to moderate correlation and low MAE. IMERG-E showed evidence of underestimating rainfall, as indicated by the RB value. Conversely, IMERG-E demonstrates poor accuracy according to the NSE value. It is necessary to explore effective correction methods for IMERGE-E to establish it as a viable alternative data source.

Effectiveness of three machine learning models for prediction of daily streamflow and uncertainty assessment.

This study evaluates three Machine Learning (ML) models-Temporal Kolmogorov-Arnold Networks (TKAN), Long Short-Term Memory (LSTM), and Temporal Convolutional Networks (TCN)-focusing on their capabilities to improve prediction accuracy and efficiency in streamflow forecasting. We adopt a data-centric approach, utilizing large, validated datasets to train the models, and apply SHapley Additive exPlanations (SHAP) to enhance the interpretability and reliability of the ML models. The results show that TKAN outperforms LSTM but slightly lags behind TCN in streamflow forecasting. TKAN demonstrated strong alignment with observed statistical parameters, achieving a Mean Absolute Error (MAE) of 5.799 m³/s and a Nash-Sutcliffe Efficiency (NSE) of 0.958, compared to MAE and NSE values of 8.865 m³/s and 0.942 for LSTM, and 5.706 m³/s and 0.961 for TCN, respectively. Multi-step forecasting revealed TKAN's robust performance up to a three-day forecast horizon, with a slight decline in accuracy as the forecast period extended. Uncertainty analysis indicated reasonable variance levels, with a mean 3-day forecast uncertainty of 35.02% at a 95% confidence level for TKAN, compared to 39.95% for LSTM and 28.46% for TCN. For a 7-day forecast, TKAN showed a mean uncertainty of 40.97%, compared to 45.01% for LSTM and 36.22% for TCN. By enhancing model transparency and improving datasets, this study significantly advances the integration of machine learning into hydrological forecasting, offering robust methods for developing adaptive water management systems in response to changing climate conditions.

Impacts of land use changes on water conservation in the Songhuajiang River basin in Northeast China using the SWAT model

The Songhuajiang River basin (SRB) in Northeast China underwent obvious changes in land use in recent years, which might lead to intense changes in basin hydrology and water conservation. Therefore, to evaluate the effects of land use changes on the hydrological cycle and associated water conservation in the SRB, a calibrated Soil and Water Assessment Tool (SWAT) model was used. The model auto-calibration was conducted using the SWAT-CUP (SWAT Calibration and Uncertainty Procedures) tool. Results revealed that calibration of three crop yields of corn (Zea mays L.), rice (Oryza sativa L.), and soybean (Glycine max L.) during the study period (2001–2018) had the Percent BIAS (PBIAS) values of −8.7 %∼3.0 %, 6.2 %∼19.7 %, and −8.7 %∼7.2 %, respectively in 15 zones. Good agreements were obtained for simulated results of streamflow in comparison with measured data from 12 hydrological stations, in which the overall Nash-Sutcliffe Efficiency (NSE) values for the calibration (2009–2012) and validation (2013–2015) periods were 0.60 and 0.64, respectively, and the overall PBIAS values were 7.7 % and 15.4 %. For the land use changes, the hydrological cycle of corn to rice (all hydrological variables change by more than 18 %) was more complex and intense than that of corn to soybean (all hydrological variables change not exceed 5 %). According to the simulation results of water conservation, this study found that the implementation of the policy-driven replacement of corn with soybean in the SRB was feasible. As for the economy-driven conversion from corn to rice, it is necessary to consider the substantial impacts on the hydrological cycle and the decline in water conservation. Based on simulation results, it is not recommended that producers in the Western SRB promote corn to rice conversion.

Evapotranspiration Estimation with the Budyko Framework for Canadian Watersheds

Actual evapotranspiration (AET) estimation plays a crucial role in watershed management. Hydrological models are commonly used to simulate watershed responses and estimate AET. However, their calibration heavily depends on station-based data, which is often limited in availability and frequently inaccessible, making the process challenging and time-consuming. In this study, the Budyko model framework, which effectively utilizes remote sensing data for hydrological modeling and requires the calibration of only one parameter, is adopted for AET estimation across Ontario, Canada. Four different parameter estimation methods were developed and compared, and an attribution analysis was also conducted to investigate the impacts of climate and vegetation factors on AET changes. Results show that the developed Budyko models performed well, with the best model achieving a Nash-Sutcliffe Efficiency (NSE) value of 0.74 and a Root Mean Square Error (RMSE) value of 55.5 mm/year. The attribution analysis reveals that climate factors have a greater influence on AET changes compared to vegetation factors. This study presents the first Budyko modeling attempt for Canadian watersheds. It demonstrates the applicability and potential of the Budyko framework for future case studies in Canada and other cold regions, providing a new, straightforward, and efficient alternative for AET estimation and hydrological modeling.

An evaluation of soil carbon models and their role on finding ways to net-zero carbon in agricultural systems

The estimation of changes in soil organic carbon (SOC) is a key issue for national green-house gasses (GHG) inventories, climate change mitigation programs and the estimation of carbon footprint of farm products in life cycle assessments. Any strategy related to net-zero carbon in agricultural systems needs to quantify the SOC balance. In this way, SOC models help decision makers involved in agriculture to understand the dynamics of the SOC and the interaction between all variables related to soil, climate, land use, and management, to design the best solution to reduce emissions or enable carbon sequestration. Likewise, it is important to identify suitable models for the region. This study aims to address three main subjects: a) a discussion on the importance of SOC estimation for GHG inventories and the carbon footprint of crops, using the Intergovernmental Panel on Climate Change (IPCC) Tier 1 method and AMG model; b) an evaluation and brief description of the IPCC “Steady State Method” (SSM), using experimental data from two sites in Argentina, comparing these results to AMG and RothC models (both previously validated at those sites); and c) a brief discussion about the potential use of SOC models for what-if management scenarios, their real limitations and future research needs. The three models were consistent in predicting the impact of tillage and the long-term trends in changes in SOC stocks under different management practices. The SSM model was evaluated for the first time in Argentina and performed even better than the other two models. It was consistent with the observed values, when predicting the effect of tillage system under different crop rotations, including pasture systems. Regarding efficiencies of the models, they showed acceptable Nash-Sutcliffe Efficiency (NSE) values, and the root mean square error (RMSE) was also acceptable between 3 % and 7 %, within a range of 4–5 Mg C.ha−1. Therefore, the SSM model proved to be a valuable tool to estimate SOC trends for crop and pasture rotations under different management scenarios (i.e., tillage systems and fertilization), to identify best practices that allow for a zero or positive SOC balance, in two different soil and climate conditions of the Pampean Region of Argentina. In our study, the SSM did have a better fit to the data and, furthermore, this Tier 2 method is simpler than the Tier 3 models, and, therefore, is advantageous for conducting regional assessments and GHG inventories.

Development and Applicability Assessment of an Hourly Water Level Prediction Model for Agricultural Reservoirs Using Automated Machine Learning

Objectives:This study aims to develop and evaluate hourly water level prediction models for agricultural reservoirs using an automated machine learning (AutoML) approach, specifically, employing TPOT.Methods:The study focuses on the Baekrok and Baekryeon reservoirs using rainfall forecast data from the Korea Meteorological Administration, along with observed rainfall and reservoir water level data. TPOT, which utilizes genetic algorithms to automate the generation of optimal machine learning pipelines, was used to build models. Additionally, Random Forest (RF) was implemented as a benchmark to evaluate TPOT’s applicability. Predictions were generated with lead times of 1, 3, 6, and 12 hours, and model accuracy was evaluated using the Nash-Sutcliffe Efficiency (NSE), coefficient of determination (R2), and root-mean-square error (RMSE).Results and Discussion:The predictions from both TPOT and RF showed high accuracy for shorter lead times, with NSE values exceeding 0.99 for the 1-hour predictions. However, predictive accuracy decreased as lead time increased, likely due to greater uncertainty in rainfall forecast data, particularly for the 12-hour predictions. Despite this trend, TPOT-derived models maintained more stable and accurate performance compared to RF models across all lead times.Conclusion:This study demonstrates the applicability of TPOT-based AutoML techniques in predicting hourly water levels in agricultural reservoirs. Future work should explore strategies to mitigate the decline in accuracy for longer lead times.

Comparing Monthly Rainfall Prediction in West Sumatra Using SARIMA, ETS, LSTM, and XGBoosting Methods

&lt;p&gt;&lt;em&gt;T&lt;/em&gt;&lt;em&gt;he West Sumatra Province, serving as the trading center on the island of Sumatra, and boasting various attractive tourist destinations, is not immune to incidents of high precipitation leading to hydro-meteorological disasters such as floods and landslides. Therefore, the accurate prediction of monthly rainfall is crucial to minimize the impacts of high precipitation. This research aims to determine the best method for predicting monthly rainfall using data from 1992 to 2022, which can adequately represent its climatological conditions.&lt;/em&gt;&lt;em&gt; The results indicate that the Extreme Gradient Boosting method outperforms the Seasonal Autoregressive Integrated Moving Average (SARIMA), Exponential Smoothing (ETS), and Long Short-Term Memory (LSTM) methods in West Sumatra Province, represented by three weather observation points from the BMKG (Climatology Station of West Sumatra, Maritime Meteorology Station of Teluk Bayur, and Minangkabau Meteorology Station). This method exhibits the lowest error values and the strongest correlation between predicted and actual data. This is evident from the Nash-Sutcliffe Efficiency (NSE) values, which are 0.188214535, 0.613823746, and 0.545734162 (unsatisfactory-satisfactory), as well as the obtained correlation values of 0.472103386, 0.795586268, and 0.743002591 (moderate-strong). However, this method is unable to perfectly capture outlier values. These outliers arise as a result of unusual conditions, such as natural disasters or climate changes, and atmospheric phenomena like El Niño-Southern Oscillation (ENSO) and Indian Ocean Dipole (IOD), leading to exceptionally high or low precipitation&lt;/em&gt;&lt;em&gt;.&lt;/em&gt;&lt;/p&gt;

Runoff simulation of the Kaidu River Basin based on the GR4J-6 and GR4J-6-LSTM models

Study regionThe Kaidu River Basin originates from the southern slope of the Tienshan Mountains in the Xinjiang Uygur Autonomous Region, China. Study focusAccurate runoff simulation and prediction significantly affect flood control, drought resilience, and water resource allocation decisions. This study establishes the GR4J-6 model (modèle du Génie Rural à 4 paramètres Journalier-6, including a snowmelt module) and integrates it with the LSTM (Long Short-Term Memory) model to construct the hybrid GR4J-6-LSTM model and enhance the simulation accuracy of snowmelt runoff. A case study is conducted in the Kaidu River Basin to demonstrate the applicability of these models in cold and arid regions. The accuracy of the GR4J-6, LSTM, and GR4J-6-LSTM models is evaluated using Nash-Sutcliffe efficiency (NSE), Kling-Gupta Efficiency (KGE), and Root Mean Squared Error (RMSE) metrics. In addition, the contributions of each feature variable in the models are analyzed using the SHapley Additive exPlanations (SHAP) method to enhance the reliability of the results. New hydrological insights for the regionThe GR4J-6 model demonstrated good applicability in the Kaidu River Basin, with NSE, KGE, and RMSE values of 0.69, 0.79, and 39.39 m3/s during the validation period, respectively. The hybrid model GR4J-6-LSTM exhibited the highest comprehensive accuracy among all the models, with NSE, KGE, and RMSE values of 0.84, 0.87, and 28.79 m3/s, respectively. In the LSTM model, temperature and precipitation were found to significantly influence the simulated runoff, indicating that higher temperature and precipitation lead to increased runoff. In the GR4J-6-LSTM model, Tmin (minimum temperature) and the hydrological feature variable Qsim exhibited a strong positive correlation with simulated runoff, as Tmin and Qsim increased, they promoted stronger flow production. This study provides a framework for runoff simulation in snowmelt river basins, offering a reference for projecting extreme hydrological events under climate change.

Physically-based modelling of UK river flows under climate change

This study presents the model setup and results from the first calibrated, physically-based, spatially-distributed hydrological modelling of combined land cover and climate change impacts on a large sample of UK river catchments. The SHETRAN hydrological model was automatically calibrated for 698 UK catchments then driven by the 12 regional climate model projections from UKCP18, combined with urban development and natural flood management scenarios. The automatic calibration of SHETRAN produces a median Nash-Sutcliffe efficiency value of 0.82 with 581 catchments having a value greater than 0.7. 24 summary metrics were calculated to capture changes to important aspects of the flow regime. The UKCP18 realisations in SHETRAN indicate that a warming climate will cause river flows, on average, to decrease. These decreases are simulated to be greatest in the south and east of the UK, with droughts becoming longer and more severe. While high flows also decrease on average, an increased number of extremes are exhibited, implying a greater number of extreme flood events in the future, particularly in the north and west of the UK. In the urban development scenarios, for flood events there is an increase in flow with the increased urbanization, with the 1 in 3-year peak flow event showing the greatest increase. The natural flood management scenarios consider the effect of increasing woodland and adding surface water storage ponds. The inclusion of these features produces a complex response but overall, the modelling shows a reduction in low, median, and high flows, although the more extreme the flow event the smaller the percentage change in flow. Simulated timeseries and summary metric datasets are freely available on the CEDA archive.

A novel urban flood risk assessment framework based on refined numerical simulation technology

Due to urbanization and climate change, urban areas are becoming more susceptible to flooding, posing significant risks to human life and property. Numerical simulation techniques can provide valuable support in mitigating urban flood risks. The objective of this study is to establish a novel framework for flood risk assessment based on refined urban flood simulations. To achieve refined urban flood simulations, the Torricelli’s law is incorporated into a two-dimensional fluid dynamics model, and a numerical analysis model is developed to estimate the inundation within buildings and accurately simulate the entire inundation process. The model was used to reproduce a real flood event in Omihachiman City, Japan, with the Nash-Sutcliffe Efficiency (NSE) values of 0.6488 and 0.7182, demonstrating the credibility of the model. Based on the simulation results, the hydrological causes of the flood event were analyzed, and flood prevention measures were proposed. After implementing these measures, the maximum water depths at the study area were reduced to 0.14 m, 0.62 m, and 0.12 m, respectively, proving the effectiveness of the flood prevention measures. To address the issue of inundation depth not fully reflecting flood risk, the instability of human bodies in floodwaters is conducted to evaluate flood risk for adults and children in the target area. In response to increasingly frequent extreme rainfall events, the extreme precipitation events of three different return periods (10-year, 50-year, and 100-year) are used to assess the flood risk and the effectiveness of flood prevention measures. Based on the assessment results, the flood prevention measures not only reduced the flood risk levels but also decreased the flood risk area, further confirming their effectiveness.

Forecasting Lake Nokoué Water Levels Using Long Short-Term Memory Network

The forecasting of hydrological flows (rainfall depth or rainfall discharge) is becoming increasingly important in the management of hydrological risks such as floods. In this study, the Long Short-Term Memory (LSTM) network, a state-of-the-art algorithm dedicated to time series, is applied to predict the daily water level of Lake Nokoué in Benin. This paper aims to provide an effective and reliable method to enable the reproduction of the future daily water level of Lake Nokoué, which is influenced by a combination of two phenomena: rainfall and river flow (runoff from the Ouémé River, the Sô River, the Porto-Novo lagoon, and the Atlantic Ocean). Performance analysis based on the forecasting horizon indicates that LSTM can predict the water level of Lake Nokoué up to a forecast horizon of t + 10 days. Performance metrics such as Root Mean Square Error (RMSE), coefficient of correlation (R2), Nash–Sutcliffe Efficiency (NSE), and Mean Absolute Error (MAE) agree on a forecast horizon of up to t + 3 days. The values of these metrics remain stable for forecast horizons of t + 1 day, t + 2 days, and t + 3 days. The values of R2 and NSE are greater than 0.97 during the training and testing phases in the Lake Nokoué basin. Based on the evaluation indices used to assess the model’s performance for the appropriate forecast horizon of water level in the Lake Nokoué basin, the forecast horizon of t + 3 days is chosen for predicting future daily water levels.

A Convolutional Neural Network-Weighted Cellular Automaton Model for the Fast Prediction of Urban Pluvial Flooding Processes

Deep learning models demonstrate impressive performance in rapidly predicting urban floods, but there are still limitations in enhancing physical connectivity and interpretability. This study proposed an innovative modeling approach that integrates convolutional neural networks with weighted cellular automaton (CNN-WCA) to achieve the precise and rapid prediction of urban pluvial flooding processes and enhance the physical connectivity and reliability of modeling results. The study began by generating a rainfall-inundation dataset using WCA and LISFLOOD-FP, and the CNN-WCA model was trained using outputs from LISFLOOD-FP and WCA. Subsequently, the pre-trained model was applied to simulate the flood caused by the 20 July 2021 rainstorm in Zhengzhou City. The predicted inundation spatial distribution and depth by CNN-WCA closely aligned with those of LISFLOOD-FP, with the mean absolute error concentrated within 5 mm, and the prediction time of CNN-WCA was only 0.8% that of LISFLOOD-FP. The CNN-WCA model displays a strong capacity for accurately predicting changes in inundation depths within the study area and at susceptible points for urban flooding, with the Nash-Sutcliffe efficiency values of most flood-prone points exceeding 0.97. Furthermore, the physical connectivity of the inundation distribution predicted by CNN-WCA is better than that of the distribution obtained with a CNN. The CNN-WCA model with additional physical constraints exhibits a reduction of around 34% in instances of physical discontinuity compared to CNN. Our results prove that the CNN model with multiple physical constraints has significant potential to rapidly and accurately simulate urban flooding processes and improve the reliability of prediction.

Climate change impacts on the hydrological behaviour of watershed in northeastern Tunisia (Oued El Abid watershed)

Climate change significantly impacts watershed hydrology, altering water supplies, natural disasters, and eco-hydrologic processes. These effects vary geographically, with some regions facing severe droughts while others experience increased precipitation and flooding. Climate change influences atmospheric evaporation demand, precipitation patterns, vegetation composition, streamflow characteristics, and groundwater dynamics. The temperature increases and rainfall changes in some watersheds can lead to increased potential evapotranspiration and decreased streamflow and soil water. Flood frequency analyses indicate an overall increasing trend in the latter half of the 21st century. The Oued El Abid catchment area, as like all catchments in northern Tunisia, is likely to be subject to major climatic and hydrological variations as a result of climate change. Advanced hydrological modelling using the HBV-Light model was set up to simulate current conditions and future scenarios based on Afric-Cordex climate models. The model demonstrates robust performance, achieving Nash-Sutcliffe Efficiency (NSE) values of 0.71 in calibration (1972-1992), and 0.66 in validation from 1993-2001. This climate change has a direct effect on the hydrological projections, which reveal significant changes, with annual flow reductions ranging from 35% to 43%, with more severe impacts under the RCP8.5 scenario for 2069-2099. Spring and summer seasons are particularly vulnerable, showing flow decreases of up to 68% and over 40%, respectively, extreme monthly variability is projected, reaching 88% reduction in July under RCP8.5. This research provides important results for water resource managers, highlighting the necessity of long-term planning and sustainable water management practices in the face of climate change.