Prediction on the fluoride contamination in groundwater at the Datong Basin, Northern China: Comparison of random forest, logistic regression and artificial neural network

Nafouanti, Mouigni Baraka; Li, Junxia; Mustapha, Nasiru Abba; Uwamungu, Placide; AL-Alimi, Dalal

doi:10.1016/j.apgeochem.2021.105054

Cited by 44 publications

(10 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Logistic regression simulates the probabilities of binary dependent variables. It assumes a linear relationship between the log odds of the dependent variable and the independent variable [64]. It applies to binary classification.…”

Section: Multinomial Logistic Regression (Mlr)mentioning

confidence: 99%

Dynamic Monitoring of Desertification in Ningdong Based on Landsat Images and Machine Learning

Chen

Shen

et al. 2022

Sustainability

View full text Add to dashboard Cite

The ecological stability of mining areas in Northwest China has been threatened by desertification for a long time. Remote sensing information combined with machine learning algorithms can effectively monitor and evaluate desertification. However, due to the fact that the geological environment of a mining area is easily affected by factors such as resource exploitation, it is challenging to accurately grasp the development process of desertification in a mining area. In order to better play the role of remote sensing technology and machine learning algorithms in the monitoring of desertification in mining areas, based on Landsat images, we used a variety of machine learning algorithms and feature combinations to monitor desertification in Ningdong coal base. The performance of each monitoring model was evaluated by various performance indexes. Then, the optimal monitoring model was selected to extract the long-time desertification information of the base, and the spatial-temporal characteristics of desertification were discussed in many aspects. Finally, the factors driving desertification change were quantitatively studied. The results showed that random forest with the best feature combination had better recognition performance than other monitoring models. Its accuracy was 87.2%, kappa was 0.825, Macro-F1 was 0.851, and AUC was 0.961. In 2003–2017, desertification land in Ningdong increased first and then slowly improved. In 2021, the desertification situation deteriorated. The driving force analysis showed that human economic activities such as coal mining have become the dominant factor in controlling the change of desert in Ningdong coal base, and the change of rainfall plays an auxiliary role. The study comprehensively analyzed the spatial-temporal characteristics and driving factors of desertification in Ningdong coal base. It can provide a scientific basis for combating desertification and for the construction of green mines.

show abstract

Section: Multinomial Logistic Regression (Mlr)mentioning

confidence: 99%

Dynamic Monitoring of Desertification in Ningdong Based on Landsat Images and Machine Learning

Chen

Shen

et al. 2022

Sustainability

View full text Add to dashboard Cite

show abstract

“…36,37 Nafouanti et al collected the chemical factors of 482 groundwater samples and evaluated the performance of three statistical technologies, which showed that RF outperformed ANN and logistic regression in predicting groundwater fluoride contamination and eight water chemistry indicators were screened, which attributed to the groundwater fluoride. 24 Tarasov et al examined and compared two solo neural networks as well as two combined techniques based on the real measurements of Cr contamination of the surface. 34 The findings demonstrated that combined models outperformed solo models in terms of accuracy, with an increase in the root mean squared error of up to 15.5%.…”

Section: Introductionmentioning

confidence: 99%

“…The attenuation rates observed in the microcosm study were commonly higher in order of magnitude than those in the field . The use of numerical models and contaminant concentration analyses based on in situ monitoring is constrained by the massive data set, prolonged processing time, and intricate structure required. , Biogeochemical factors and biological processes, which play essential roles in natural attenuation, were ignored in contaminant concentration analysis. , Nowadays, a compound-specific stable isotope analysis approach has been built to assess biodegradation based on the correlation between isotopic composition and contaminant concentration . However, long-term monitoring is still required, which is time-consuming.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning Modeling Based on Microbial Community for Prediction of Natural Attenuation in Groundwater

Zhang,

Long,

Deng

et al. 2023

Environ. Sci. Technol.

View full text Add to dashboard Cite

“…Both insufficient and excessive F − cause great harm to human health. Specifically, a lack of F − tends to cause dental caries, and excessive intake leads to fluorosis (Katsanou et al, 2013;Tarki et al, 2020;Nafouanti et al, 2021;Senarathne et al, 2021). The common diseases caused by excessive F − intake are dental fluorosis and skeletal fluorosis, which may lead to death in severe cases (Xie et al, 1999;Mondal et al, 2016;Mohammadi et al, 2017;Liu et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

Fluorine in shallow groundwater in China: A review of distribution, occurrence and environmental effects

Zhao

Zhang

et al. 2023

Front. Earth Sci.

View full text Add to dashboard Cite

With rapid economic development and the increasing demand for drinking water, a large amount of groundwater is exploited, resulting in a high F− content in groundwater, which is harmful to the environment and human body. In this study, 5,464 data points of fluoride in shallow groundwater were collected, and the F− content distribution, occurrence form and environmental impact of shallow groundwater were discussed. The results showed that 1) the F− content in shallow groundwater in China ranged from 0 to 60 mg/L, with a mean content of 0.90 mg/L; the lowest average F− content in shallow groundwater in Southwest China was 0.36 mg/L; South China (1.20 mg/L), Northeast China (1.25 mg/L) and Northwest China (1.25 mg/L) were considered high-fluoride areas, and North China (0.93 mg/L), East China (0.67 mg/L) and Central China (0.80 mg/L) were considered low-fluoride areas. The mean F− content in groundwater differed between provinces and cities. 2) The F− in shallow groundwater mainly occurred in ionic, complex ionic and organic fluoride molecular states. 3) The influence of a high F− content in shallow groundwater on the environment was mainly manifested in the increase in water F− concentration and soil F− and vegetable F− content. The influence of a high F− content on the human body was mainly manifested in an increase in urinary F− content in children, a high prevalence of dental fluorosis in children, an increase in skeletal fluorosis rate in adults with age, and an influence on cognitive function of older adults. These results provide a basis for F− pollution control and high-fluoride water treatment.

show abstract

Prediction on the fluoride contamination in groundwater at the Datong Basin, Northern China: Comparison of random forest, logistic regression and artificial neural network

Cited by 44 publications

References 56 publications

Dynamic Monitoring of Desertification in Ningdong Based on Landsat Images and Machine Learning

Dynamic Monitoring of Desertification in Ningdong Based on Landsat Images and Machine Learning

Machine Learning Modeling Based on Microbial Community for Prediction of Natural Attenuation in Groundwater

Fluorine in shallow groundwater in China: A review of distribution, occurrence and environmental effects

Contact Info

Product

Resources

About