Dataset Information

Prediction of Dichloroethene Concentration in the Groundwater of a Contaminated Site Using XGBoost and LSTM.

ABSTRACT: Chlorinated aliphatic hydrocarbons (CAHs) are widely used in agriculture and industries and have become one of the most common groundwater contaminations. With the excellent performance of the deep learning method in predicting, LSTM and XGBoost were used to forecast dichloroethene (DCE) concentrations in a pesticide-contaminated site undergoing natural attenuation. The input variables included BTEX, vinyl chloride (VC), and five water quality indicators. In this study, the predictive performances of long short-term memory (LSTM) and extreme gradient boosting (XGBoost) were compared, and the influences of variables on models' performances were evaluated. The results indicated XGBoost was more likely to capture DCE variation and was robust in high values, while the LSTM model presented better accuracy for all wells. The well with higher DCE concentrations would lower the model's accuracy, and its influence was more evident in XGBoost than LSTM. The explanation of the SHapley Additive exPlanations (SHAP) value of each variable indicated high consistency with the rules of biodegradation in the real environment. LSTM and XGBoost could predict DCE concentrations through only using water quality variables, and LSTM performed better than XGBoost.

SUBMITTER: Xia F

PROVIDER: S-EPMC9367752 | biostudies-literature | 2022 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Prediction of Dichloroethene Concentration in the Groundwater of a Contaminated Site Using XGBoost and LSTM.

Xia Feiyang F Jiang Dengdeng D Kong Lingya L Zhou Yan Y Wei Jing J Ding Da D Chen Yun Y Wang Guoqing G Deng Shaopo S

International journal of environmental research and public health 20220730 15

Chlorinated aliphatic hydrocarbons (CAHs) are widely used in agriculture and industries and have become one of the most common groundwater contaminations. With the excellent performance of the deep learning method in predicting, LSTM and XGBoost were used to forecast dichloroethene (DCE) concentrations in a pesticide-contaminated site undergoing natural attenuation. The input variables included BTEX, vinyl chloride (VC), and five water quality indicators. In this study, the predictive performanc ...[more]

PMID: 35954730

Similar Datasets

Project description:Water pollution is a significant issue resulting from past long-term actions. The remediation projects carried out under law constraints for industrial plants, which have been the major contributors to environmental and water pollution, are currently providing a significant amount of data about contaminated soil, surface waters, and groundwater. Most of such plants worldwide are in coastal zones. Based on a significant amount of chemical and environmental data for a coastal contaminated site subject to variable groundwater salinization, this study aimed to understand the mobility of some trace elements because of coastal zone dynamics. Data concerned 688 groundwater samples, including As, Hg, Cd, Crtot, Cu, Ni, Pb, V, Se, Zn, pH, electrical conductivity, chlorides, total organic carbon and organic contaminants as quantitative variables, enhanced by additional qualitative variables such as groundwater salinity, season, water level, precipitation, and industrial activity type to make the dataset as representative as possible of the site under investigation. The study used robust multivariate statistical analyses to analyse the complex dataset and explain the relevant factors influencing contaminant behaviour under different environmental conditions. The Multivariate Statistical Analysis distinguished three clusters of trace elements with diverse reactivity to changes in groundwater salinization. The first includes Se, Cu, Crtot, V, and Ni, showing the highest correlation with electrical conductivity and chlorides because of their high affinity to form chloride or organic chloride complexes and for ion competition. Zn and Pb cluster in the second group: they are less reactive to groundwater salinization and more influenced by cation and anion competition and organic matter. The mobility of Hg and As (third cluster) significantly correlates with Fe and Mn, underlining the dominant role of reductive dissolution of trace elements-bearing minerals (Fe/Mn/Al-oxy-hydroxides) and metal-organic complexes. The correlation between the clustering of variables in groundwater and soils shows the influence of sediment structure, mineral composition, and physical and chemical soil conditions on the distribution in soils of trace elements and their transport to groundwater. The study proposes a valuable approach for assessing the effects of salinization in contaminated coastal aquifers. It supports planning multi-purpose characterization and monitoring campaigns of contaminated coastal sites and provides guidance on the correct associated remediation projects.

Project description:Fine particulate matter (PM2.5) is a major air pollutant affecting human survival, development and health. By predicting the spatial distribution concentration of PM2.5, pollutant sources can be better traced, allowing measures to protect human health to be implemented. Thus, the purpose of this study is to predict and analyze the PM2.5 concentration of stations based on the integrated deep learning of a convolutional neural network long short-term memory (CNN-LSTM) model. To solve the complexity and nonlinear characteristics of PM2.5 time series data problems, we adopted the CNN-LSTM deep learning model. We collected the PM2.5data of Qingdao in 2020 as well as meteorological factors such as temperature, wind speed and air pressure for pre-processing and characteristic analysis. Then, the CNN-LSTM deep learning model was integrated to capture the temporal and spatial features and trends in the data. The CNN layer was used to extract spatial features, while the LSTM layer was used to learn time dependencies. Through comparative experiments and model evaluation, we found that the CNN-LSTM model can achieve excellent PM2.5 prediction performance. The results show that the coefficient of determination (R2) is 0.91, and the root mean square error (RMSE) is 8.216 µg/m3. The CNN-LSTM model achieves better prediction accuracy and generalizability compared with those of the CNN and LSTM models (R2 values of 0.85 and 0.83, respectively, and RMSE values of 11.356 and 14.367, respectively). Finally, we analyzed and explained the predicted results. We also found that some meteorological factors (such as air temperature, pressure, and wind speed) have significant effects on the PM2.5 concentration at ground stations in Qingdao. In summary, by using deep learning methods, we obtained better prediction performance and revealed the association between PM2.5 concentration and meteorological factors. These findings are of great significance for improving the quality of the atmospheric environment and protecting public health.

Dataset Information

Prediction of Dichloroethene Concentration in the Groundwater of a Contaminated Site Using XGBoost and LSTM.

Publications

Prediction of Dichloroethene Concentration in the Groundwater of a Contaminated Site Using XGBoost and LSTM.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets