Real time image-based air quality forecasts using a 3D-CNN approach with an attention mechanism.

This study presented an image-based deep learning method to improve the recognition of air quality from images and produce accurate multiple horizon forecasts. The proposed model was designed to incorporate a three-dimensional convolutional neural network (3D-CNN) and the gated recurrent unit (GRU) with an attention mechanism. This study included two novelties; (i) the 3D-CNN model structure was built to extract the hidden features of multiple dimensional datasets and recognize the relevant environmental variables. The GRU was fused to extract the temporal features and improve the structure of fully connected layers. (ii) An attention mechanism was incorporated into this hybrid model to adjust the influence of features and avoid random fluctuations in particulate matter values. The feasibility and reliability of the proposed method were verified through the site images of the Shanghai scenery dataset with relevant air quality monitoring data. Results showed that the proposed method has the highest forecasting accuracy over other states of art methods. The proposed model can provide multi-horizon predictions based on efficient feature extraction and good denoising ability, which is helpful in giving reliable early warning guidelines against air pollutants.

Evaluation of environmental impact of red tide around Pearl River Estuary, Guangdong, China.

This paper presents a case study of red tide hazards around the Pearl River Estuary (PRE). Red tide hazards, meteorological data, and seawater monitoring data were collected from 1996 to 2020 at different locations around the PRE to investigate the internal and external factors influencing the occurrence of red tides. The enhancement of the assessment of estuarine trophic status (ASSETS) method enables us to evaluate the effects of meteorological factors and seawater eutrophication status on the red tide risk level. Using ASSETS, we established a framework for red tide risk assessment of the Pearl River Estuary. We analysed the external and internal factors causing the red tide based on meteorological data and seawater monitoring data in the PRE. The results show that the temperature was higher than the annual monthly average temperature of 1.265 °C, and east and north winds at velocities of 3-4 m/s could result in the formation of red tides. However, precipitation inhibits the formation of the red tide in PRE.

Prediction of long-term water quality using machine learning enhanced by Bayesian optimisation.

Water quality assessment is critical to better recognise the importance of water in human society. In this study, a new framework to predict long-term water quality is proposed by using Bayesian-optimised machine learning methods and key pollution indicators collected from monitoring stations in the Pearl River Estuary, Guangdong, China. The optimised stacked generalisation (SG-op) model achieved the best performance with the highest accuracy (0.992) and Kappa coefficient (0.987). Feature importance of the prediction model was consistent with key pollution indicators. The Spearman rank correlation coefficient was used to determine the significance level of the variation trends of different pollution indicators. The results show that the total phosphorus (TOP), dissolved oxygen (DO), chemical oxygen demand (COD), and petroleum (PET) among the key pollution indicators were on an upward trend in the study area. This framework can be applied to efficiently predict future water quality and to provide technical support for emergency pollution control.

Spatiotemporal air quality forecasting and health risk assessment over smart city of NEOM.

Modeling and predicting air pollution concentrations is important to provide early warnings about harmful atmospheric substances. However, uncertainty in the dynamic process and limited information about chemical constituents and emissions sources make air-quality predictions very difficult. This study proposed a novel deep-learning method to extract high levels of abstraction in data and capture spatiotemporal features at hourly and daily time intervals in NEOM City, Saudi Arabia. The proposed method integrated a residual network (ResNet) with the convolutional long short-term memory (ConvLSTM). The ConvLSTM method was boosted by a ResNet model for deeply extracting the spatial features from meteorological and pollutant data and thereby mitigating the loss of feature information. Then, health risk assessment was put forward to evaluate PM10 and PM2.5 risk sensitivity in five districts in NEOM City. Results revealed that the proposed method with effective feature extraction could greatly optimize the accuracy of spatiotemporal air quality forecasts compared to existing state-of-the-art models. For the next hour prediction tasks, the PM10 and PM2.5 of MASE were 9.13 and 13.57, respectively. The proposed method provides an effective solution to improve the prediction of air-pollution concentrations while being portable to other regions around the world.

A VIKOR-based approach to evaluate river contamination risks caused by wastewater treatment plant discharges.

Wastewater treatment plants (WWTPs) rarely eliminate emerging contaminants from effluents they discharged into waterways, and therefore, represent significant contaminations sources with deleterious environmental risks. This paper presents a VIKOR-based model to assess the contamination risk posed by a cluster of WWTPs. A risk index was defined via building a membership function embodying the performance degrees of WWTPs and risks levels within the framework of fuzzy set theory. The proposed approach was tested using a case study of WWTPs cluster along the Pearl River. Sensitivity analyses were carried out to investigate the robustness of the model. The results confirmed the ability of the proposed approach to reveal the risk level of a given treatment point. Further, the comparison with a TOPSIS scheme as well as sensitivity analysis results substantiate the consistency, accuracy, and reliability of the proposed approach. It is therefore bounds to improve the decentralized management of WWTPs-induced river contamination.

Assessment of red tide risk by integrating CRITIC weight method, TOPSIS-ASSETS method, and Monte Carlo simulation.

This study proposes a red tide risk assessment method based on intercriteria correlation (CRITIC), technique for order preference by similarity to an ideal solution (TOPSIS), assessment of estuarine trophic status (ASSETS) methods and Monte Carlo simulation (MCS) to calculate the probability of each risk level. The integrated TOPSIS-ASSETS method is used to calculate the risk levels of each year, where index weight is determined by CRITIC method. MCS method is employed to calculate the probability of each risk level. The results showed that level III to level V indicates high possibility of red tides in the case study area (Tolo Harbor). The highest risk rating was level V in 1988. The change of the risk level of red tide is consistent with the real situation of the occurrence of red tide. Another case of the east part of Skagerrak Strait shows that the results of this method are consistent with field situation. When there is an error between the evaluation results and the real situation, MCS can further suggest the probability of error in the evaluation results. Meanwhile, sensitivity analysis was used to test the performance of the evaluation model and two comparative methods. The results show that the proposed risk assessment method has better performance than other methods and can provide an effective risk evaluation for red tide management.

Indices and models of surface water quality assessment: Review and perspectives.

Many technologies have been designed to monitor, evaluate, and improve surface water quality, as high-quality water is essential for human activities including agriculture, livestock, and industry. As such, in this study, we investigated water quality indices (WQIs), trophic status indices (TSIs), and heavy metal indices (HMIs) for assessing surface water quality. Based on these indices, we summarised and compared water assessment models using expert system (ES) and machine learning (ML) methods. We also discussed the current status and future perspectives of water quality management. The results of our analyses showed that assessment indices can be used in three aspects of surface water quality assessment: WQIs are aggregated from multiple parameters and commonly used in surface water quality classification; TSIs are calculated from the concentrations of different nutrients required for algae and bacteria, and employed to evaluate the eutrophication levels of lakes and reservoirs; HMIs are mainly applied for human health risk assessment and the analysis of correlation of heavy metal sources. ES- and ML-based assessment models have been developed to efficiently generate assessment indices and predict water quality status based on big data obtained from new techniques. By implementing dynamic monitoring and analysis of water quality, we designed a next-generation water quality management system based on the above indices and assessment models, which shows promise for improving the accuracy of water quality assessment.

Method for a new risk assessment of urban water quality: IFN-SPA.

Water quality is one of the most essential factors to influence human daily life and environment health. Risk assessment of water quality has critical significance to sustainable development of human society and natural systems. Set pair analysis (SPA) methods are widely used in risk assessment, especially in water resources. The essence of SPA is to classify assessment samples consider the uncertainties exist in risk assessment system based on the viewpoints of unity, difference, and opposition. The existing SPA methods are classified into two types, including (i) original SPA and (ii) comprehensive SPA. Both the original and comprehensive SPA methods have the following limitations: (i) it is need to judge whether the assessment factor belongs to type I or type II; (ii) it is need to judge whether the assessment factor is positive or negative. This method article gives a detailed description of the application of the existing SPA method. The method article is a companion paper with the original article [1]. • Description of SPA methods. • Application of SAP methods in risk assessment of water quality. • Calculate the weights of assessment factors.

Method for lake eutrophication levels evaluation: TOPSIS-MCS.

Monte Carlo simulation (MCS) is applied in the engineering with great fuzziness and uncertainty. Technique for order preference by similarity to an ideal solution (TOPSIS) method is used to deal with multi-criteria decision-making issue. Membership function is used to determine the membership degree of evaluated index. This paper presents the method for lake eutrophication level evaluation. The developed approach merges MCS method, TOPSIS method and membership function. The evaluated results are consistent with real eutrophication level in Lake Erhai, China. Global sensitivity analysis (GSA) is conducted. Results show that potassium permanganate index (CODMn) displays the highest negative correlation with the evaluated results and Secchi disc (SD) performs the highest positive correlation under different errors in measured data. The novelty of this work are: (1) the application of TOPSIS considers Surface water environmental quality standards and measured data. Besides, the Monte Carlo simulation method is applied to generate a normal distributed dataset to overcome the errors caused by human and equipment in data collection. The approach is utilized in the article, titled "Approach based on TOPSIS and Monte Carlo simulation methods to evaluate lake eutrophication levels" (Lin et al., 2020) [1].•Developed approach merges TOPSIS and MCS method.•It can increase the reliability of evaluated result.

Data on performance and variation index for shield tunnelling through soft deposit.

The data presented in this article pertain to field records of EPB shield machine in Metro Line No. 5 in Tianjin, China. Field performance of shield machine (cutterhead, screw machine, and shield advancing) are shown in the figures. Specifically, the database consists of the main parameters for shield tunnelling including cutterhead rotation speed, cutterhead torque, screw machine rotation speed, screw machine torque, shield thrust, and shield advance rate. In addition, the calculation process of energy consumption and variation index R2 during the tunnelling are displayed. The value of the dataset is the consideration of silt or clay soil encountered in the shield tunnelling area including the proportion of soils, grain gradation, and effects on performance and energy consumption of different parts in shield machine. These field data are applied to evaluate the construction efficiency in the article titled "Construction efficiency of shield tunnelling through soft deposit in Tianjin" [1].

Assessment and management of lake eutrophication: A case study in Lake Erhai, China.

Some wastewater sources, such as agricultural waste and runoff, and industrial sewage, can degrade water quality. This study summarises the sources and corresponding mechanisms that trigger eutrophication in lakes. Additionally, the trophic status index and water quality index (WQI) which are effective tools for evaluating the degree of eutrophication of lakes, have been discussed. This study also explores the main nutrients (nitrogen and phosphorus) driving transformations in the water body and sediment. Lake Erhai was used as a case study, and it was found to be in a mesotrophic state, with N and P co-limitation before 2006, and only P limitation since 2006. Finally, effective measures to maintain sustainable development in the watershed are proposed, along with a framework for an early warning system adopting the latest technologies (geographic information systems (GIS), remote sensing (RS)) for preventing eutrophication.

Data in intelligent approach for estimation of disc cutter life using hybrid metaheuristic algorithm.

This data in brief presents the monitoring data measured during shield tunnelling of Guangzhou-Shenzhen intercity railway project. The monitoring data includes shield operational parameters, geological conditions, and geometry at the site. The presented data were arbitrarily split into two subsets including the training and testing datasets. The field observations are compared to the forecasting values of the disc cutter life assessed using a hybrid metaheuristic algorithm proposed for "Prediction of disc cutter life during shield tunnelling with artificial intelligent via incorporation of genetic algorithm into GMDH-type neural network" [1]. The presented data can provide a guidance for cutter exchange in shield tunnelling.

A novel method to calculate pressure on the twin-tunnel in layered strata.

Tunnel pressure from the surrounding rocks plays a critical role for the safety of tunnel. The existing methods for calculate twin-tunnel pressure supposed that the tunnel is buried in a uniform soil layer. This work presents detailed equations of an analytical method to calculate the twin-tunnel pressure in layered strata, which can consider the effects from soil layers. The proposed method is applied to analyse the pressure of the metro twin-tunnels in Chongqing. To demonstrate the efficiency of the proposed analytical method, both the tunnel pressure in layered strata and single strata were calculated. The method article is a companion paper with the original article [1]. • Analyses of the soil parameters; • Determine the failure pattern A/B; • Calculate the vertical and horizontal pressure.

Data on evolutionary hybrid neural network approach to predict shield tunneling-induced ground settlements.

The dataset presented in this article pertains to records of shield tunneling-induced ground settlements in Guangzhou Metro Line No. 9. Field monitoring results obtained from both the two tunnel lines are put on display. In total, 17 principal variables affecting ground settlements are tabulated, which can be divided into two categories: geological condition parameters and shield operation parameters. Shield operation parameters are specifically provided in time series. Another value of the dataset is the consideration of karst encountered in the shield tunnel area including the karst cave height, the distance between karst cave and tunnel invert, and the karst cave treatment scheme. The dataset can be used to enrich the database of settlement caused by shield tunneling as well as to train artificial intelligence-based ground settlement prediction models. The dataset presented herein were used for the article titled "Evolutionary hybrid neural network approach to predict shield tunneling-induced ground settlements" (Zhang et al., 2020).

Approach based on TOPSIS and Monte Carlo simulation methods to evaluate lake eutrophication levels.

This study presents an approach for eutrophication evaluation based on the technique for order preference by similarity to an ideal solution (TOPSIS) method and Monte Carlo simulation (MCS). The MCS is employed to produce a normally distributed dataset based on the observed data while the TOPSIS method and membership function are used to evaluate the level of eutrophication. Herein, a eutrophication problem in Lake Erhai is evaluated to check the performance of the proposed approach. The evaluation results were consistent with the real situation when the coefficient P in the membership function is equal to 1. Moreover, the developed approach is able to (i) deal with evaluation items with inherent fuzziness and uncertainties, (ii) improve the reliability of evaluation results via MCS, and (iii) raise the tolerance to errors in measured data. A global sensitivity analysis indicated that the potassium permanganate index (CODMn) and Secchi disc (SD) are the most sensitive factors in the developed approach. Finally, a range for the coefficient P value in the membership function was recommended.

Experimental data of water swelling characteristics of polymer materials for tunnel sealing gasket.

This article provides comprehensive experimental data of two polymers for shield tunnel sealing gasket: i) water swelling polyurethane (WSP) and ii) a mixture (WSRP) of WSP and water swelling rubber (WSR). Water-swelling tests are conducted to investigate the microstructural changes and properties of both WSP and WSRP during water swelling. These data can be useful for the quantitative evaluation of water swelling performance of WSP and WSRP. The data presented herein was used for the article, titled "Experimental investigation of water swelling characteristics of polymer materials for tunnel sealing gasket" [1].

Data on a coupled ENN / t-SNE model for soil liquefaction evaluation.

The data presented in this paper pertain to case records of liquefaction potential surveys in earthquake prone areas. Field performances of 219 sites obtained from various regions (U.S.A, Japan, Turkey, China, Canada, etc …) are put on display. Specifically, this database consists of 253 cone penetration test (CPT) field records, among which 72 cases that did not liquefied and 181 cases that liquefied. In total, 10 principal variables are tabulated including the earthquake magnitude, maximum ground surface acceleration, depth, water depth, total overburden stress, effective overburden stress, Cone Penetration Test (CPT) tip resistance, CPT friction ratio, fines content, shear stress ratio. These data were arbitrarily split into a testing set of 53 cases and a training set of 200 cases. These field observations are compared to prediction values of liquefaction potential assessed using the evolutionary neural network proposed for "Evaluation of soil liquefaction with AI technology incorporating a coupled ENN/t-SNE model" [1].

Data in risk assessment of mega-city infrastructures related to land subsidence using improved trapezoidal FAHP.

Land subsidence caused serious damages of mage-city infrastructures. This data in brief presents a new questionnaire to establish judgment matrix during the risk assessment of land subsidence. The data source of the assessment factors is provided. The analytical hierarchy process (AHP) and interval fuzzy AHP (FAHP) are used to calibrate the weights of assessment factors. The new questionnaire is used to collect the viewpoints from experts. Based on the viewpoints of experts, the judgment matrix can be established using pairwise comparison. The data presented herein was used for the article, titled "Risk assessment of mega-city infrastructures related to land subsidence using improved trapezoidal FAHP" Lyu et al. (2019) [1].

Risk assessment of mega-city infrastructures related to land subsidence using improved trapezoidal FAHP.

This study presents an improved trapezoidal fuzzy analytic hierarchy process (FAHP) to assess the risk of mega-city infrastructures related to land subsidence. The trapezoidal fuzzy numbers are used to express the relative importance between assessment factors. A new questionnaire is proposed in this study to collect judgements from consulting experts. Both the original AHP and the trapezoidal FAHP with the new questionnaire are applied to assess the risk of infrastructures in relation to land subsidence in Shanghai. The risks assessed using the trapezoidal FAHP at locations with significant infrastructures are higher than those assessed using the original AHP. This indicates that the trapezoidal FAHP method with the new questionnaire can be used to effectively capture the high risks for significant industrial infrastructures related to land subsidence. Moreover, the obtained results were compared with the current land subsidence prevention zone, and it was observed that the existing land subsidence prevention zone in government management guidelines does not sufficiently consider the vulnerability of significant infrastructures.