Fine-tuning inflow prediction models: integrating optimization algorithms and TRMM data for enhanced accuracy.

This research explores machine learning algorithms for reservoir inflow prediction, including long short-term memory (LSTM), random forest (RF), and metaheuristic-optimized models. The impact of feature engineering techniques such as discrete wavelet transform (DWT) and XGBoost feature selection is investigated. LSTM shows promise, with LSTM-XGBoost exhibiting strong generalization from 179.81 m3/s RMSE (root mean square error) in training to 49.42 m3/s in testing. The RF-XGBoost and models incorporating DWT, like LSTM-DWT and RF-DWT, also perform well, underscoring the significance of feature engineering. Comparisons illustrate enhancements with DWT: LSTM and RF reduce training and testing RMSE substantially when using DWT. Metaheuristic models like MLP-ABC and LSSVR-PSO benefit from DWT as well, with the LSSVR-PSO-DWT model demonstrating excellent predictive accuracy, showing 133.97 m3/s RMSE in training and 47.08 m3/s RMSE in testing. This model synergistically combines LSSVR, PSO, and DWT, emerging as the top performers by effectively capturing intricate reservoir inflow patterns.

Application of data-mining technique and hydro-chemical data for evaluating vulnerability of groundwater in Indo-Gangetic Plain.

Vulnerability of groundwater is critical for the sustainable development of groundwater resources, especially in freshwater-limited coastal Indo-Gangetic plains. Here, we intend to develop an integrated novel approach for delineating groundwater vulnerability using hydro-chemical analysis and data-mining methods, i.e., Decision Tree (DT) and K-Nearest Neighbor (KNN) via k-fold cross-validation (CV) technique. A total of 110 of groundwater samples were obtained during the dry and wet seasons to generate an inventory map. Four K-fold CV approach was used to delineate the vulnerable region from sixteen vulnerability causal factors. The statistical error metrics i.e., receiver operating characteristic-area under the curve (AUC-ROC) and other advanced metrices were adopted to validate model outcomes. The results demonstrated the excellent ability of the proposed models to recognize the vulnerability of groundwater zones in the Indo-Gangetic plain. The DT model revealed higher performance (AUC = 0.97) followed by KNN model (AUC = 0.95). The north-central and north-eastern parts are more vulnerable due to high salinity, Nitrate (NO3-), Fluoride (F-) and Arsenic (As) concentrations. Policy-makers and groundwater managers can utilize the proposed integrated novel approach and the outcome of groundwater vulnerability maps to attain sustainable groundwater development and safeguard human-induced activities at the regional level.

Species, tissue and gender-related metal and element accumulation in fish species in a large reservoir (Turkey) and health risks and nutritional benefits for consumers.

Concentrations of 18 metals and elements (MEs) in the gills, skin, muscle and liver of Carasobarbus luteus and Cyprinus carpio from the Atatürk Reservoir in Turkey were investigated. The results revealed that variations in the ME contents between fish species can be attributed to different diets. The highest contents of most MEs were recorded in the gills. Gender had no significant effect on the contents of most MEs in the tissues of fish species investigated. The relations between MEs in tissues and fish size were not clear and consistent. Health risk assessment methods indicated that consumption of the studied fish species is safe. It was estimated that daily consumption of 140 g of C. carpio or 170 g of C. luteus would not be expected to cause any health risks. Furthermore, it was found that fish species would provide significant benefits in terms of intake of essential MEs.

COVID-19 pandemic, socioeconomic crisis and human stress in resource-limited settings: A case from Bangladesh.

Considering the population density, healthcare capacity, limited resources and existing poverty, environmental factors, social structure, cultural norms, and already more than 18,863 people infected, the community transmission of COVID-19 is happening fast. These exacerbated a complex fear among the public. The aim of this article is, therefore, to understand the public perception of socioeconomic crisis and human stress in resource-limited settings of Bangladesh during the COVID-19 outbreak. The sample comprised of 1066 Bangladeshi participants. Principal component analysis (PCA) was considered to design a standardized scale to measure the mental stress and socioeconomic crisis, one-way ANOVA and t-test were conducted to perceive different demographic risk groups; multiple linear regression was applied to estimate the statistically significant association between each component, and classical test theory (CTT) analysis was applied to examine the reliability of each item according to the components to develop a composite score. Without safeguarding the fundamental needs for the vulnerable ultra-poor group can undeniably cause the socioeconomic crisis and mental stress due to the COVID-19 lockdown. It has further created unemployment, deprivation, hunger, and social conflicts. The weak governance in the fragile healthcare system exacerbates the general public's anxiety as the COVID-19 testing facilities are centered around in the urban areas, a long serial to be tested, minimum or no treatment facilities in the dedicated hospital units for COVID-19 patients are the chief observations hampered along with the disruption of other critical healthcare services. One-way ANOVA and t-test confirmed food and nutritional deficiency among the vulnerable poorest section due to loss of livelihood. Also, different emergency service provider professions such as doctors, healthcare staff, police forces, volunteer organizations at the frontline, and bankers are at higher risk of infection and subsequently mentally stressed. Proper risk assessment of the pandemic and dependable risk communications to risk groups, multi-sectoral management taskforce development, transparency, and good governance with inter-ministerial coordination is required along with strengthening healthcare capacity was suggested to reduce mental and social stress causing a socioeconomic crisis of COVID-19 outbreak. Moreover, relief for the low-income population, proper biomedical waste management through incineration, and preparation for the possible natural disasters such as flood, cyclones, and another infectious disease such as dengue was suggested. Finally, this assessment process could help the government and policymakers to judge the public perceptions to deal with COVID-19 pandemic in densely populated lower-middle-income and limited-resource countries like Bangladesh.

Quantifying Source Apportionment, Co-occurrence, and Ecotoxicological Risk of Metals from Upstream, Lower Midstream, and Downstream River Segments, Bangladesh.

The positive matrix factorization (PMF) receptor model was used for the first time to quantify the source contributions to heavy metal pollution of sediment on a national basin scale in the upstream, midstream, and downstream rivers (Teesta and Kortoya-Shitalakkah and Meghna-Rupsha and Pasur) of Bangladesh. The metal contamination status, co-occurrence, and ecotoxicological risk were also investigated. Sediment samples were collected from 30 sites at a depth range of 0 to 20 cm for analysis of 9 metals using inductively coupled plasma-mass spectrometry. The mean concentrations of metals varied for upstream, lower midstream, and downstream river segments. The results showed that chromium (Cr) exhibited a strong significant co-occurrence network with other metals (e.g., manganese [Mn], iron [Fe], and nickel [Ni]). Monte Carlo simulation results of the geo-accumulation index (Igeo; 63.3%) and risk indices (48.5%) showed that cadmium (Cd) was the main contributor to sediment pollution. However, the cumulative probabilities of sediments being polluted by metals were ranked as "moderate to heavily polluted" (Igeo 46.6%; risk index 16.7%). Toxicity unit results revealed that zinc (Zn) and Cd were the key toxic contributors to sediments. The PMF model predicted metal concentrations and identified 4 potential sources. The agricultural source (factor 1) mostly contributed to copper (Cu; 78.9%) and arsenic (As; 62.8%); Ni (96.9%) and Mn (83.5%) exhibited industrial point sources (factor 2), with 2 hot spots in northwestern and southwestern regions. Cadmium (93.5%) had anthropogenic point sources (factor 3), and Fe (64.3%) and Cr (53.5%) had a mixed source (factor 4). Spatially, similar patterns between PMF apportioning factors and predicted metal sources were identified, showing the efficiency of the model for river systems analysis. The degree of metal contamination in the river segments suggests an alarming condition for biotic components of the ecosystem. Environ Toxicol Chem 2020;39:2041-2054. © 2020 SETAC.

Appraising spatial variations of As, Fe, Mn and NO3 contaminations associated health risks of drinking water from Surma basin, Bangladesh.

This study aims to appraise the spatial variations and pathways of groundwater contaminations and associated health risks in the Surma basin, Bangladesh using geostatistics, Fuzzy GIS technique and health risk modelbased on ninety groundwater samples. The results show that the mean concentrations of As, Fe, Mn and NO3 are below the Bangladesh water quality standard, whereas As, Fe and Mn concentrations exceed World Health Organization guideline values in several sampling sites. The compositional study identifies weathering of source rocks, reductive dissolution of Fe and Mn-oxyhydroxide minerals and various anthropogenic inputs as the key sources of groundwater contamination. The kriged maps show the elevated risks of Fe, Mn and NO3 concentrations from the south to northern parts and As concentration from the north to southwestern parts of the Surma basin. The results of fuzzy GIS maps confirm the outcomes of kriged maps. Cross validation results show better performance of indicator kriging over probability kriging. The results also show a spatial heterogeneity with As, Fe, Mn and NO3 concentrations, indicating the low to medium risk categories. A health risk assessment is performed using hazard quotient (HQ) and hazard index (HI). The HQ values imply that the risk of contamination through oral ingestion pathway is medium to high levels for both adults and children as the trace elements show HQ values more than one. It is found that drinking water of several upazilas exhibits high contamination and that children are more susceptible to the non-carcinogenic and carcinogenic risks than adults in the study area.