Coastal evolution and future projections in Conde County, Brazil: A multi-decadal assessment via remote sensing and sea-level rise scenarios.

Global sea levels, having risen by approximately 20 cm since the mid-19th century, necessitate a critical examination of their impacts on shoreline dynamics. This research evaluates the historical (1985-2022) and future shoreline changes in Conde County, Paraíba State, Brazil, an area of significant touristic interest. Employing Landsat satellite imagery, the study utilized the Digital Shoreline Analysis System (DSAS) and a Kalman filter algorithm for cloud removal, while also assessing land use and land cover changes using data from the MapBiomas Project for 2000, 2010, and 2020. These analyses informed projections of potential inundation under various sea-level rise (SLR) scenarios: 1, 2, 5, and 10 m. Key findings revealed a negative average coastline change rate of -0.27 m/year from 1985 to 2022, indicative of erosive trends likely accelerated by human activities. Long-term projections for 2032 and 2042 anticipate continued erosion in areas identified as highly vulnerable. The SLR scenario analysis underscores the urgent need for adaptive climate measures; while a 1- or 2-meter SLR presents limited immediate effects, a 5-meter rise could lead to significant inundation across key sectors, including urban and agricultural landscapes. The projected severity of a 10-meter SLR necessitates immediate, comprehensive interventions to safeguard both natural and human systems.

Assessment of hybrid machine learning algorithms using TRMM rainfall data for daily inflow forecasting in Três Marias Reservoir, eastern Brazil.

This study investigates the application of the Gaussian Radial Basis Function Neural Network (GRNN), Gaussian Process Regression (GPR), and Multilayer Perceptron Optimized by Particle Swarm Optimization (MLP-PSO) models in analyzing the relationship between rainfall and runoff and in predicting runoff discharge. These models utilize autoregressive input vectors based on daily-observed TRMM rainfall and TMR inflow data. The performance evaluation of each model is conducted using statistical measures to compare their effectiveness in capturing the complex relationships between input and output variables. The results consistently demonstrate that the MLP-PSO model outperforms the GRNN and GPR models, achieving the lowest root mean square error (RMSE) across multiple input combinations. Furthermore, the study explores the application of the Empirical Mode Decomposition-Hilbert-Huang Transform (EMD-HHT) in conjunction with the GPR and MLP-PSO models. This combination yields promising results in streamflow prediction, with the MLP-PSO-EMD model exhibiting superior accuracy compared to the GPR-EMD model. The incorporation of different components into the MLP-PSO-EMD model significantly improves its accuracy. Among the presented scenarios, Model M4, which incorporates the simplest components, emerges as the most favorable choice due to its lowest RMSE values. Comparisons with other models reported in the literature further underscore the effectiveness of the MLP-PSO-EMD model in streamflow prediction. This study offers valuable insights into the selection and performance of different models for rainfall-runoff analysis and prediction.

Assessing coastal vulnerability and land use to sea level rise in Jeddah province, Kingdom of Saudi Arabia.

Sea level rise is one of the most serious outcomes of increasing temperatures, leading to coastal flooding, beach erosion, freshwater contamination, loss of coastal habitats, increased soil salinity, and risk of damage to coastal infrastructures. This study estimates the vulnerability to inundation for 2100 in coastal zones in Jeddah Province, Kingdom of Saudi Arabia, under various sea level rise (SLR) scenarios of 1, 2, 5, and 10 m. The predicted flooding was estimated using a combination of factors, including SLR, the bathtub model, digital elevation model, climate scenarios, and land use and land cover. The climate scenarios used were Representative Concentration Pathway (RCP) scenarios 1.9, 2.6, 4.5, and 8.5. The results of the SLR scenarios of 1, 2, 5, and 10 m revealed that 1.6, 4.7, 14.9, and 30.6% (or 88, 214, 679, 1398 km2) of the study area's coast could be classified as inundated areas. The various SLR scenarios can inundate 3.3 to 34% of the road area/length. The inundated built-up and road areas were estimated to range between 0.31 and 0.79 km2, accounting respectively for 1.18 to 3.01% of the total class areas for 1-meter and 2-meter SLR scenarios. In contrast, the inundated area will be significant in the situation of 5 and 10 m SLR scenarios. Regarding the case of a 10-meter SLR scenario, the inundation will negatively impact the built-up and road infrastructure areas, inundating 8.9 km2, with industrial infrastructures affected by inundation estimated at 0.21 km2, followed by green space infrastructures at 0.013 km2. The spatial information based on various SLR scenario impact mapping for Jeddah Province can be highly valuable for decision-makers to better plan future civil engineering structures within the framework of sustainable development.

Artificial intelligent systems optimized by metaheuristic algorithms and teleconnection indices for rainfall modeling: The case of a humid region in the mediterranean basin.

Characterized by their high spatiotemporal variability, rainfalls are difficult to predict, especially under climate change. This study proposes a multilayer perceptron (MLP) network optimized by Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Firefly Algorithm (FFA), and Teleconnection Pattern Indices - such as North Atlantic Oscillation (NAO), Southern Oscillations (SOI), Western Mediterranean Oscillation (WeMO), and Mediterranean Oscillation (MO) - to model monthly rainfalls at the Sebaou River basin (Northern Algeria). Afterward, we compared the best-optimized MLP to the application of the Extreme Learning Machine optimized by the Bat algorithm (Bat-ELM). Assessment of the various input combinations revealed that the NAO index was the most influential parameter in improving the modeling accuracy. The results indicated that the MLP-FFA model was superior to MLP-GA and MLP-PSO for the testing phase, presenting RMSE values equal to 33.36, 30.50, and 29.92 mm, respectively. The comparison between the best MLP model and Bat-ELM revealed the high performance of Bat-ELM for rainfall modeling at the Sebaou River basin, with RMSE reducing from 29.92 to 11.89 mm and NSE value from 0.902 to 0.985 during the testing phase. This study shows that incorporating the North Atlantic Oscillation (NAO) as a predictor improved the accuracy of artificial intelligence systems optimized by metaheuristic algorithms, specifically Bat-ELM, for rainfall modeling tasks such as filling in missing data of rainfall time series.

A cloud-integrated GIS for forest cover loss and land use change monitoring using statistical methods and geospatial technology over northern Algeria.

Over the last two decades, forest cover has experienced significant impacts from fires and deforestation worldwide due to direct human activities and climate change. This paper assesses trends in forest cover loss and land use and land cover changes in northern Algeria between 2000 and 2020 using datasets extracted from Google Earth Engine (GEE), such as the Hanssen Global Forest Change and MODIS Land Cover Type products (MCD12Q1). Classification was performed using the pixel-based supervised machine-learning algorithm called Random Forest (RF). Trends were analyzed using methods such as Mann-Kendall and Sen. The study area comprises 17 basins with high rainfall variability. The results indicated that the forest area decreased by 64.96%, from 3718 to 1266 km2, during the 2000-2020 period, while the barren area increased by 40%, from 134,777 to 188,748 km2. The findings revealed that the Constantinois-Seybousse-Mellegue hydrographic basin was the most affected by deforestation and cover loss, exceeding 50% (with an area of 1018 km2), while the Seybouse River basin experienced the highest percentage of cover loss at 40%. Nonparametric tests showed that seven river basins (41%) had significantly increasing trends of forest cover loss. According to the obtained results, the forest loss situation in Algeria, especially in the northeastern part, is very alarming and requires an exceptional and urgent plan to protect forests and the ecological system against wildfires and climate change. The study provides a diagnosis that should encourage better protection and management of forest cover in Algeria.

Improved weighted ensemble learning for predicting the daily reference evapotranspiration under the semi-arid climate conditions.

Evapotranspiration is an important quantity required in many applications, such as hydrology and agricultural and irrigation planning. Reference evapotranspiration is particularly important, and the prediction of its variations is beneficial for analyzing the needs and management of water resources. In this paper, we explore the predictive ability of hybrid ensemble learning to predict daily reference evapotranspiration (RET) under the semi-arid climate by using meteorological datasets at 12 locations in the Andalusia province in southern Spain. The datasets comprise mean, maximum, and minimum air temperatures and mean relative humidity and mean wind speed. A new modified variant of the grey wolf optimizer, named the PRSFGWO algorithm, is proposed to maximize the ensemble learning's prediction accuracy through optimal weight tuning and evaluate the proposed model's capacity when the climate data is limited. The performance of the proposed approach, based on weighted ensemble learning, is compared with various algorithms commonly adopted in relevant studies. A diverse set of statistical measurements alongside ANOVA tests was used to evaluate the predictive performance of the prediction models. The proposed model showed high-accuracy statistics, with relative root mean errors lower than 0.999% and a minimum R2 of 0.99. The model inputs were also reduced from six variables to only two for cost-effective predictions of daily RET. This shows that the PRSFGWO algorithm is a good RET prediction model for the semi-arid climate region in southern Spain. The results obtained from this research are very promising compared with existing models in the literature.

A novel comprehensive framework for analyzing and assessing water quality and failure consequences based on Bayesian networks.

The study of the effects of mixing potable water with wastewater is a complex and difficult research area. This difficulty is because water and sewage networks are subject to various physical, environmental, and operational factors. The main objective of the study was to propose a new comprehensive framework for analyzing and assessing water quality based on Bayesian networks. An intervention plan was proposed to reduce the consequences of water quality and networks failure. The proposed framework was applied to water distribution network of Mdaourouch city (Souk Ahras, Algeria) to demonstrate its effectiveness. The results indicated that the water contamination rate has reached 33.9 %, which caused severe consequences. The effectiveness of the proposed plan has been verified theoretically using simulations, and the results have proven to be very satisfactory. The proposed model is a decision support tool, which is expected to assist decision-makers and engineers in reviewing their plans and making the right decision. PRACTITIONER POINTS: This paper proposes a novel comprehensive framework for analyzing and assessing water quality and failure consequences based on Bayesian networks. This paper revisits the failure consequences. An intervention plan is proposed to reduce failure consequences. Results demonstrate that the proposed plan leads to fewer consequences probabilities. The proposed method can give the probability of failure of water and sewer network.