Search Results (126)

Straw fiber curtain contains a plant fiber blanket woven from crop straw, which is mainly used to protect embankment slopes from rainwater erosion. To investigate the erosion control performance of slopes covered with straw fiber curtains of different structural configurations, physical model tests were conducted in a 95 cm × 65 cm × 50 cm (length × height × width) test box with a slope ratio of 1:1.5 under controlled artificial rainfall conditions (20 mm/h, 40 mm/h, and 60 mm/h). The study evaluated the runoff characteristics, sediment yield, and key hydrodynamic parameters of slopes under the coverage of different straw fiber curtain types. The results show that the A-type straw fiber curtain (woven with strips of straw fiber) has the best effect on water retention and sediment reduction, while the B-type straw fiber curtain (woven with thicker straw strips) with vertical straw fiber has a better effect regarding water retention and sediment reduction than the B-type transverse straw fiber curtain. The flow of rainwater on a slope covered with straw fiber curtain is mainly a laminar flow. Straw fiber curtain can promote the conversion of water flow from rapids to slow flow. The Darcy-Weisbach resistance coefficient of straw fiber curtain increases at different degrees with an increase in rainfall time. Full article

Artificial neural networks (ANNs) adjust to the underlying behavior in the dataset using a training rule or optimizer. The most popular first-and second-order optimizers, Adam (AD) and Levenberg–Marquardt (LM), were compared with the aim of predicting extreme flash floods of a runoff-dominated hydrological system. A fully connected multilayer perceptron with a shallow structure was used to reduce complexity and limit overfitting. The inputs of the ANN were determined by rainfall–water level cross-correlation analysis. For each optimizer, the hyperparameters of the ANN were selected using a grid search and the cross-validation score on a novel criterion (PERS PEAK) mixing the persistency (PERS) and the quality of flood-peak restitution (PEAK). For an extreme and unseen event used as a test set, LM outperformed AD by 25% on all performance criteria. The peak water level of this event, 66% greater than that of the training set, was predicted by 92% after more training iterations were done by the LM optimizer. This shows that the ANN can predict beyond the ranges of the training set, given the right optimizer. Nevertheless, the LM training time was up to five times longer than that of AD during grid search. Full article

Atmospheric pressure gradients determine the dynamics of the southwest monsoon (SWM) and northeast monsoon (NEM), resulting in rainfall in the Indian subcontinent. Consequently, the surface salinity, mixed layer, and thermocline are impacted by the seasonal freshwater outflow and direct rainfall. Moreover, seasonally reversing monsoon gyre and associated currents govern the northern Indian Ocean surface oceanography. This study provides an overview of the impact of these dynamic changes on sea surface temperature, salinity, and productivity by integrating more than 3000 planktonic foraminiferal censuses and bulk sediment geochemical data from sediment core tops, plankton tows, and nets between 25° N and 10° S and 40° E and 110° E of the past six decades. These data were used to construct spatial maps of the five most dominant planktonic foraminifers and illuminate their underlying environmental factors. Moreover, the cured foraminiferal censuses and the modern oceanographic data were used to test the newly developed artificial neural network (ANN) algorithm to calculate the relationship with modern water column temperatures (WCTs). Furthermore, the tested relationship between the ANN derived models was applied to two foraminiferal censuses from the northern Bay of Bengal core MGS29-GC02 (13°31′59″ N; 91°48′21″ E) and the southern Bay of Bengal Ocean Drilling Program (ODP) Site 758 (5°23.05′ N; 90°21.67′ E) to reconstruct the WCTs of the past 890 ka. The reconstructed WCTs at the 10 m water depth of core GC02 suggest dramatic changes in the sea surface during the deglacial periods (i.e., Bolling–Allerǿd and Younger Dryas) compared to the Holocene. The WCTs at Site 758 indicate a shift in the mixed-layer summer temperature during the past 890 ka at the ODP Site, in which the post-Mid-Brunhes period (at 425 ka) was overall warmer than during the prior time. However, the regional alkenone-derived sea-surface temperatures (SSTs) do not show such a shift in the mixed layer. Therefore, this study hypothesizes that the divergence in regional SSTs is most likely due to differences in seasonality and depth habitats in the paleo-proxies. Full article

To understand if machine learning algorithms could be employed in agricultural landscapes to estimate N₂O emissions, multiple linear regression (MLR), random forest regression (RFR), support vector regression (SVR) and artificial neural network (ANN) algorithms are tested on an agricultural site in Ontario, Canada. Two scenarios, High Input (HI) and Low Input (LI), were used to check the performance of these algorithms’ using R², RMSE, VE, p-factor, r-factor and visual inspection indicators. The HI consisted of discrete measurements of N₂O, rainfall, temperature, fertilizer application dates, soil nitrate, ammonium content and pH values, whereas the LI scenario did not use the latter three. The results indicated that MLR was inapplicable as the data did not satisfy its fundamental assumptions. RFR, SVR and ANN under HI were able to capture 64% (66%), 59% (63%) and 94% (43%) of the variability of emissions within the training (testing) datasets. Subsequently, these models were able to capture 92%, 29% and 75% of high emissions (>10 gm/ha/day) within their predictive intervals of 95% confidence. RFR, SVR and ANN under the LI scenario captured 72% (68%), 61% (66%) and 81% (68%) of the variability in N₂O emissions within the training (testing) datasets. While these models were found to have comparable performance in both HI and LI scenarios, HI was found to be better at capturing high emissions. Based on the computational cost, ease in finetuning, capture of peak emissions and stable model performance, RFR and ANN are recommended to estimate N₂O emissions in the study area and similar agricultural landscapes in future studies. Full article

Rainfall-induced landslide mitigation remains a critical research focus in geotechnical engineering, particularly for safeguarding buildings and infrastructure in unstable terrain. This study investigates the stabilizing performance of slopes reinforced with negative Poisson’s ratio (NPR) anchor cables under rainfall conditions through physical model tests. A scaled geological model of a heavily weathered rock slope is constructed using similarity-based materials, building a comprehensive experimental setup that integrates an artificial rainfall simulation system, a model-scale NPR anchor cable reinforcement system, and a multi-parameter data monitoring system. Real-time measurements of NPR anchor cable axial forces and slope internal stresses were obtained during simulated rainfall events. The experimental results reveal distinct response times and force distributions between upper and lower NPR anchor cables in reaction to rainfall-induced slope deformation, reflecting the temporal and spatial evolution of the slope’s internal sliding surface—including its generation, expansion, and full penetration. Monitoring data on volumetric water content, earth pressure, and pore water pressure within the slope further elucidate the evolution of effective stress in the rock–soil mass under saturation. Comparative analysis of NPR cable forces and effective stress trends demonstrates that NPR anchor cables provide adaptive stress compensation, dynamically counteracting internal stress redistribution in the slope. In addition, the structural characteristics of NPR anchor cables can effectively absorb the energy released by landslides, mitigating large deformations that could endanger adjacent buildings. These findings highlight the potential of NPR anchor cables as an innovative reinforcement strategy for rainfall-triggered landslide prevention, offering practical solutions for slope stabilization near buildings and enhancing the resilience of building-related infrastructure. Full article

Pavement deterioration is often the result of intense traffic and increased runoff from storms, floods, or other environmental factors. A practical solution to this challenge involves the use of permeable pavements, such as permeable interlocking concrete pavement (PICP), which are designed to effectively manage water runoff while supporting heavy traffic. This research investigates the effectiveness of PICP in two distinct surface patterns: stretcher bond and 45° herringbone, by assessing their performance in terms of water infiltration and runoff using two different methods. The first approach has been conducted experimentally using a laboratory apparatus designed to simulate rainfall. Various conditions were applied during the performance tests, including longitudinal (L-_Slope) and transverse (T-_Slope) slopes of (0, 2, and 4%) and rainfall intensities of (40 and 80 L/min). The second approach has been implemented theoretically using Surfer 2.0 software to simulate the distribution of infiltrated water underneath the layers of PICP. Moreover, the behavior of PICP has been analyzed statistically using artificial neural networks (ANNs). The results indicated that at a rainfall intensity of 40 L/min, equal infiltration was observed in both patterns on 0% and 4% T-_Slope. However, the 45° herringbone PICP showed better infiltration on the 8% T-_Slope. Additionally, at 80 L/min rainfall, equal infiltration was observed in both patterns on 0% L-_Slope for 0% and 4% T-_Slope. The 45° herringbone PICP also demonstrated higher water infiltration on the 8% T-_Slope, and this trend continued as the L-_Slope increased. PICP with a 45° herringbone surface pattern exhibited superiority in reducing runoff compared to the stretcher bond pattern. The statistical models for the stretcher bond and 45° herringbone patterns demonstrate high accuracy, as evidenced by their correlation coefficient (R²) values of 99.97% and 97.32%, respectively, which confirms their validity. Despite the variations between the two forms of PICP, both are strongly endorsed as excellent alternatives to conventional pavement. Full article

The cover and management factor (C/B factor) in the Universal Soil Loss Equation (USLE) series models indicates the effects of vegetation cover and management practices on water erosion. Remote sensing technology provides abundant data and methods for the C/B factor estimation, but the applicability and accuracy of these methods can vary widely. More critically, they often overlook the impact of non-photosynthetic vegetation cover on soil erosion. This study aimed to evaluate and develop a more accurate and cost-effective method for calculating the C/B factor in the purple soil hilly region, focusing on typical small watersheds. A correlation analysis was conducted to compare four C/B factors derived from the remote sensing data, aiming to identify the most suitable method for the purple soil hilly region. Additionally, artificial rainfall simulation tests were performed to investigate the relationship between photosynthetic vegetation cover, non-photosynthetic vegetation cover, and soil erosion, leading to the development of a relational equation between integrated vegetation cover and C/B factors. The results indicate that the method from the technical regulations for dynamic monitoring of soil erosion is most suitable for calculating the C/B factor in purple soil hilly regions. On this basis, the integrated vegetation cover effectively accounted for the impact of non-photosynthetic vegetation on soil erosion, leading to a more comprehensive and precise estimation of the C/B factor. The newly developed method significantly improved the accuracy of the C/B factor calculation in the purple soil hilly region. This study provides a scientific and accurate algorithm for calculating the C/B factor in the purple soil hilly region, offering valuable insights and a methodological framework for similar studies in other areas. Full article

The study aims to assess future streamflow forecasts in the Godavari basin of India under climate change scenarios. The primary objective of the Coupled Model Inter-comparison Project Phase 6 (CMIP6) was to evaluate future streamflow forecasts across different catchments in the Godavari basin, India, with an emphasis on understanding the impacts of climate change. This study employed both conceptual and machine learning models to assess how changing precipitation patterns and temperature variations influence streamflow dynamics. Seven satellite precipitation products CMORPH, Princeton Global Forcing (PGF), Tropical Rainfall Measuring Mission (TRMM), Climate Prediction Centre (CPC), Infrared Precipitation with Stations (CHIRPS), and Precipitation Estimation from Remotely Sensed Information Using Artificial Neural Networks (PERSIANN-CDR) were evaluated in a gridded precipitation evaluation over the Godavari River basin. Results of Multi-Source Weighted-Ensemble Precipitation (MSWEP) had a Nash–Sutcliffe efficiency (NSE), coefficient of determination (R²), and root mean square error (RMSE) of 0.806, 0.831, and 56.734 mm/mon, whereas the Tropical Rainfall Measuring Mission had 0.768, 0.846, and 57.413 mm, respectively. MSWEP had the highest accuracy, the lowest false alarm ratio, and the highest Peirce’s skill score (0.844, 0.571, and 0.462). Correlation and pairwise correlation attribution approaches were used to assess the input parameters, which included a two-day lag of streamflow, maximum and minimum temperatures, and several precipitation datasets (IMD, EC-Earth3, EC-Earth3-Veg, MIROC6, MRI-ESM2-0, and GFDL-ESM4). CMIP6 datasets that had been adjusted for bias were used in the modeling process. R, NSE, RMSE, and R² assessed the model’s effectiveness. RF and M5P performed well when using CMIP6 datasets as input. RF demonstrated adequate performance in testing (0.4 < NSE < 0.50 and 0.5 < R² < 0.6) and extremely good performance in training (0.75 < NSE < 1 and 0.7 < R < 1). Likewise, M5P demonstrated good performance in both training and testing (0.4 < NSE < 0.50 and 0.5 < R² < 0.6). While RF was the best performer for both datasets, Indian Meteorological Department outperformed all CMIP6 datasets in streamflow modeling. Using the Indian Meteorological Department gridded precipitation, RF’s NSE, R, R², and RMSE values during training were 0.95, 0.979, 0.937, and 30.805 m³/s. The test results were 0.681, 0.91, 0.828, and 41.237 m³/s. Additionally, the Multi-Layer Perceptron (MLP) model demonstrated consistent performance across both the training and assessment phases, reinforcing the reliability of machine learning approaches in climate-informed hydrological forecasting. This study underscores the significance of incorporating climate change projections into hydrological modeling to enhance water resource management and adaptation strategies in the Godavari basin and similar regions facing climate-induced hydrological shifts. Full article

Quantitative precipitation estimation using polarimetric radar in attenuation-prone frequency (X-band) in tropical regions characterized by convective rain systems with high intensities is a major challenge due to strong attenuations that can lead to total signal extinction over short distances. However, some authors have addressed this issue in Benin since 2006 in the framework of the African Monsoon Multidisciplinary Analysis program. Thus, with an experimental setup consisting of an X-band polarimetric weather radar (Xport) and a network of rain gauges, investigations have started on the subject with the aim of improving rainfall estimates. Based on simulated polarimetric variables and using a Multilayer Perceptron artificial neural network, several bi-variable and tri-variable algorithms were assessed in this study. The data used in this study are of two categories: (i) simulated polarimetric variables (Rayleigh reflectivity $Z$, horizontal attenuation $A_h$, horizontal reflectivity $Z_h$, differential reflectivity $Z_{dr}$, and specific differential phase $K_{dp}$) and rainfall intensity (R) obtained from Rain Drop Size Distribution (DSD) measurements used for algorithm evaluation (training and testing); (ii) polarimetric variables measured by the Xport radar and rainfall intensity measured by rain gauges used for algorithm validation. The simulations are performed using the T-matrix code, which leverages the scattering properties of spheroidal particles. The DSD measurements taken in northwest Benin were used as input for this code. For each spectrum, the T-matrix code simulates multiple variables. The simulated data (first category) were divided into two parts: one for training and one for testing. Subsequently, the best algorithms were validated with the second category of data. The performance of the algorithms during training, testing, and validation was evaluated using metrics. The best selected algorithms are $A1:R(Z,K_{dp})$ and $A12:R(Z_{dr},K_{dp})$ (among the bi-variable); $B2:R(Z_h,Z_{dr},K_{dp})$ and $B3:R(A_h,Z_{dr},K_{dp})$ (among the tri-variable). Tri-variable algorithms outperform bi-variable algorithms. Validation with observation data (Xport measurements and rain gauge network) showed that the algorithm $B3:R(A_h,Z_{dr},K_{dp})$ performs better than $B2:R(Z_h,Z_{dr},K_{dp})$. Full article

As we are advancing deeper into the twenty-first century, new challenges as well as technical opportunities in agriculture are rising. One of these issues is the increasing need for food, which is crucial for supporting the population’s nutritional needs, promoting regional trade, and ensuring food security. Climate change is another ongoing challenge in the shape of changing rainfall patterns, increasing temperatures due to high CO₂ concentrations, and over urbanization which ultimately negatively impact the crop yield. Therefore, for increased food production and the sustainability of agricultural growth, an accurate and timely crop yield prediction could be beneficial. In this paper, artificial intelligence (AI)-based sustainable methods for the evaluation of wheat production (W_P) using multiple linear regression (MLR), support vector machine (SVM), and artificial neural network (ANN) techniques are presented. The historical data of around 60 years, comprising of wheat area (W_A), temperature (T), rainfall (RF), carbon dioxide emissions from liquid and gaseous fusion CE (CE_LF, CE_GF), arable land (AL), credit disbursement (CD), and fertilizer offtake (FO) were used as potential indicators/input parameters to forecast wheat production (W_P). To further support the performance efficiency of computed prediction models, a variety of statistical tests were used, such as R-square (R²), root means square error (RMSE), and mean absolute error (MAE). The results demonstrate that all acceptance standards relating to accuracy are satisfied by the proposed models. However, the SVM outperforms MLR and ANN approaches. Additionally, parametric and sensitivity tests were performed to assess the specific influence of the input parameters. Full article

The electrical resistivity tomography (ERT) method has been increasingly integrated with hydrogeological risk mitigation strategies to monitor the internal conditions and the stability of natural and artificial slopes. In this paper, we discuss a case study in which numerical simulations were essential to validate the interpretation of the resistivity images obtained from an ERT monitoring system installed on a critical slope in Italy. An initial analysis of the monitoring data after rainfall events in the study site showed that the resistivity values were decreased only in the central zone along the ERT line, but they were increased in the two sides of the profile. Opposite behaviors were observed during the drying processes following the rainfall events. Core samples show complex geology at the study site, which might justify uneven responses of the different subsurface bodies to meteorological events. However, we decided to investigate the possible inversion artifacts resulting from the individual inversion of the tomographic sections. Forward modeling simulations on simplified time-lapse models of the study site were performed to explore this problem and to compare the individual and time-lapse inversions. Synthetic tests confirmed the nature of these unexpected behaviors and assessed the absolute necessity of a time-lapse approach for a correct inversion of monitoring data in the presence of a complex geological model such as the one of this case study. By applying the time-lapse inversion approach to the real data, the inversion artifact problem was substantially solved, arriving after the proper calibration of the inversion parameters, mainly the time-lapse damping factor and the spatial and temporal roughness constraints, to a reduction in the inversion artifacts to less than 5%. Full article

The accurate assessment of water availability throughout the cassava cropping season (the initial, developmental, mid-season, and late stages) is crucial for mitigating the impacts of climate change on crop production. Using the Mann–Kendall Test, we investigated the trends in rainfall and cassava crop evapotranspiration (ETc) within the Cross River basin in Nigeria. Reference evapotranspiration (ETo) was based on two approaches, namely Artificial Neural Network (ANN) modelling and three established empirical models—the Penman–Monteith (considered the standard method), Blaney–Morin–Nigeria (BMN), and Hargreaves–Samani (HAG) models. ANN predictions were performed by using inputs from BMN and HAG parameters, denoted as BMN-ANN and HAG-ANN, respectively. The results from the ANN models were compared to those obtained from the Penman–Monteith method. Remotely sensed meteorological data spanning 39 years (1979–2017) were acquired from the Climatic Research Unit (CRU) to estimate ETc, while cassava yield data were acquired from the International Institute of Tropical Agriculture (IITA), Ibadan. The study revealed a significant upward trend in cassava crop ETc over the study period. Additionally, the ANN models outperformed the empirical models in terms of prediction accuracy. The BMN-ANN model with a Tansig activation function and a 3-3-1 architecture (number of input neurons, hidden layers, and output neurons) achieved the highest performance, with a coefficient of determination (R²) of 0.9890, a root mean square error (RMSE) of 0.000056 mm/day, and a Willmott’s index of agreement (d) of 0.9960. There is a decreasing trend in cassava yield in the region and further analysis indicated potential average daily water deficits of approximately −1.1 mm/day during the developmental stage. These deficits could potentially hinder root biomass, yield, and overall cassava yield in the Cross River basin. Our findings highlight the effectiveness of ANN modelling for irrigation planning, especially in the face of a worsening climate change scenario. Full article

Magnaporthe oryzae, the source of the rice blast, is a serious threat to the world’s rice supply, particularly in areas like Tamil Nadu, India. In this study, weather-based models were developed based on count time series and machine learning techniques like INGARCHX, Artificial Neural Networks (ANNs), and Support Vector Regression (SVR), to forecast the incidence of rice blast disease. Between 2015 and 2023, information on rice blast occurrence was gathered weekly from three locations (Thanjavur, Tirunelveli, and Coimbatore), together with relevant meteorological data like temperature, humidity, rainfall, sunshine, evaporation, and sun radiation. The associations between the occurrence of rice blast and environmental factors were investigated using stepwise regression analysis, descriptive statistics, and correlation. Mean Squared Error (MSE) and Root Mean Squared Error (RMSE) were used to assess the model’s prediction ability. The best prediction accuracy was given by the ANN, which outperformed SVR and INGARCHX in every location, according to the results. The complicated and non-linear relationships between meteorological variables and disease incidence were well-represented by the ANN model. The Diebold–Mariano test further demonstrated that ANNs are more predictive than other models. This work shows how machine learning algorithms can improve the prediction of rice blast, offering vital information for early disease management. The application of these models can help farmers make timely decisions to minimize crop losses. The findings suggest that machine learning models offer promising potential for accurate disease forecasting and improved rice management. Full article

The slope erosion in the distribution area of completely weathered granite is often relatively severe, causing serious ecological damage and property loss. Ecological restoration is the most effective means of soil erosion control. Taking completely weathered granite backfill soil as the research object, two types of slope protection plants, Vetiver grass and Pennisetum hydridum, were selected. We analyzed these two herbaceous plants’ soil reinforcement and slope protection effects through artificial planting experiments, indoor simulated rainfall experiments, and direct shear tests. The test results showed that the runoff and sediment production rates of the two herbaceous plant slopes were significantly lower than those of the bare slope, with the order of bare slope > Vetiver grass slope > Pennisetum hydridum slope. Compared with the bare slope, the cumulative sediment production of the Vetiver grass slope at 60 min decreased by 56.73–60.09%, and the Pennisetum hydridum slope decreased by 75.97–78.45%. The indoor direct shear test results showed that soil cohesion decreases with increasing water content. As the root content of Vetiver grass roots increases, soil cohesion first increases and then decreases, reaching a maximum value when the root content is 1.44%. As the root content of Pennisetum hydridum increases, soil cohesion increases. The internal friction angle increases slightly with increasing water content, while the root content does not significantly affect the internal friction angle. Therefore, the shear strength of soil decreases when the water content increases. The shear strength of the Vetiver grass root-soil composite reaches a peak at a root content of 1.44%, while the shear strength of the giant king grass root-soil composite increases as the root content increases. At the same root content, the shear strength of the Vetiver grass root-soil composite is slightly higher than that of giant king grass. The reinforcement effect of roots on shallow soil is better than on deep soil. Both herbaceous plants have an excellent soil-fixing and slope-protecting impact on the fully weathered granite backfill slope. Pennisetum hydridum’s soil and water conservation effect is significantly better than that of the Vetiver grass. In contrast, Vetiver grass roots slightly outperform Pennisetum hydridum in enhancing the shear strength of the soil. The research results can provide a theoretical basis for the vegetation slope protection treatment of fully weathered granite backfill slopes. Full article

Wildfire susceptibility maps play a crucial role in preemptively identifying regions at risk of future fires and informing decisions related to wildfire management, thereby aiding in mitigating the risks and potential damage posed by wildfires. This study employs eXplainable Artificial Intelligence (XAI) techniques, particularly SHapley Additive exPlanations (SHAP), to map wildfire susceptibility in Izmir Province, Türkiye. Incorporating fifteen conditioning factors spanning topography, climate, anthropogenic influences, and vegetation characteristics, machine learning (ML) models (Random Forest, XGBoost, LightGBM) were used to predict wildfire-prone areas using freely available active fire pixel data (MODIS Active Fire Collection 6 MCD14ML product). The evaluation of the trained ML models showed that the Random Forest (RF) model outperformed XGBoost and LightGBM, achieving the highest test accuracy (95.6%). All of the classifiers demonstrated a strong predictive performance, but RF excelled in sensitivity, specificity, precision, and F-1 score, making it the preferred model for generating a wildfire susceptibility map and conducting a SHAP analysis. Unlike prevailing approaches focusing solely on global feature importance, this study fills a critical gap by employing a SHAP summary and dependence plots to comprehensively assess each factor’s contribution, enhancing the explainability and reliability of the results. The analysis reveals clear associations between factors such as wind speed, temperature, NDVI, slope, and distance to villages with increased fire susceptibility, while rainfall and distance to streams exhibit nuanced effects. The spatial distribution of the wildfire susceptibility classes highlights critical areas, particularly in flat and coastal regions near settlements and agricultural lands, emphasizing the need for enhanced awareness and preventive measures. These insights inform targeted fire management strategies, highlighting the importance of tailored interventions like firebreaks and vegetation management. However, challenges remain, including ensuring the selected factors’ adequacy across diverse regions, addressing potential biases from resampling spatially varied data, and refining the model for broader applicability. Full article