Search

Radon potential mapping is challenging due to the limited availability of information. In this study, a new modeling process using deep learning models based on convolution neural network (CNN), long short-term memory (LSTM), and recurrent neural network (RNN) is presented to predict radon potential in the northwestern part of Gangwon Province, South Korea. The used data in this study are in two sets of dependent variables (measured soil gas radon concentrations) and independent variables (radon conditioning factors: lithology; distance from lineament; mean soil calcium oxide [Cao], potassium oxide [K2O], and ferric oxide [Fe2O3] concentrations; effective soil depth; topsoil texture; and soil drainage). The models were validated based on the area under the receiver operating curve (AUC), mean squared error (MSE), root mean square error (RMSE), and standard deviation (StD). The CNN model with AUC values of 0.906 and 0.905 in the learning and testing stages, respectively, is introduced as the optimal model. The lowest StD, MSE, and RMSE values were from the CNN, LSTM, and RNN models, respectively. Our results show that the use of deep learning models to generate radon potential maps is promising and reliable. [ABSTRACT FROM AUTHOR]

[Display omitted] • Three novel machine learning models were developed to generate radon potential map. • Soil properties and local geology significantly affected the distribution of radon. • The results indicated that the GMDH-COA model outperformed the other models. Prolonged inhalation of indoor radon and its progenies lead to severe health problems for housing occupants; therefore, housing developments in radon-prone areas are of great concern to local municipalities. Areas with high potential for radon exposure must be identified to implement cost-effective radon mitigation plans successfully or to prevent the construction of unsafe buildings. In this study, an indoor radon potential map of Chungcheongnam-do, South Korea, was generated using a group method of data handling (GMDH) algorithm based on local soil properties, geogenic, geochemical, as well as topographic factors. To optimally tune the hyper-parameters of GMDH and enhance the prediction accuracy of modelling radon distribution, the GMDH model was integrated with two metaheuristic optimization algorithms, namely the bat (BA) and cuckoo optimization (COA) algorithms. The goodness-of-fit and predictive performance of the models was quantified using the area under the receiver operating characteristic (ROC) curve (AUC), mean squared error (MSE), root mean square error (RMSE), and standard deviation (StD). The results indicated that the GMDH-COA model outperformed the other models in the training (AUC = 0.852, MSE = 0.058, RMSE = 0.242, StD = 0.242) and testing (AUC = 0.844, MSE = 0.060, RMSE = 0.246, StD = 0.0242) phases. Additionally, using metaheuristic optimization algorithms improved the predictive ability of the GMDH. The GMDH-COA model showed that approximately 7 % of the total area of Chungcheongnam-do consists of very high radon-prone areas. The information gain ratio method was used to assess the predictive ability of considered factors. As expected, soil properties and local geology significantly affected the spatial distribution of radon potential levels. The radon potential map produced in this study represents the first stage of identifying areas where large proportions of residential buildings are expected to experience significant radon levels due to high concentrations of natural radioisotopes in rocks and derived soils beneath building foundations. The generated map assists local authorities to develop urban plans more wisely towards region with less radon concentrations. [ABSTRACT FROM AUTHOR]

The adverse health effects associated with the inhalation and ingestion of naturally occurring radon gas produced during the uranium decay chain mean that there is a need to identify high-risk areas. This study detected radon-prone areas using a geographic information system (GIS)-based probabilistic and machine learning methods, including the frequency ratio (FR) model and a convolutional neural network (CNN). Ten influencing factors, namely elevation, slope, the topographic wetness index (TWI), valley depth, fault density, lithology, and the average soil copper (Cu), calcium oxide (Cao), ferric oxide (Fe 2 O 3), and lead (Pb) concentrations, were analyzed. In total, 27 rock samples with high activity concentration index values were divided randomly into training and validation datasets (70:30 ratio) to train the models. Areas were categorized as very high, high, moderate, low, and very low radon areas. According to the models, approximately 40% of the study area was classified as very high or high risk. Finally, the radon potential maps were validated using the area under the receiver operating characteristic curve (AUC) analysis. This showed that the CNN algorithm was superior to the FR method; for the former, AUC values of 0.844 and 0.840 were obtained using the training and validation datasets, respectively. However, both algorithms had high predictive power. Slope, lithology, and TWI were the best predictors of radon-affected areas. These results provide new information regarding the spatial distribution of radon, and could inform the development of new residential areas. Radon screening is important to reduce public exposure to high levels of naturally occurring radiation. [Display omitted] • Geogenic radon potential maps are generated using FR and CNN methods. • Slope, lithology, and TWI have significant influence on modelling radon level. • Approximately 20% of the study area is classified as very high risk for radon gas. [ABSTRACT FROM AUTHOR]

• Some readily available spatial covariates were used to map landslide susceptibility. • The capability of the GMDH model was explored for landslide susceptibility mapping. • Five hybridized models were developed based on metaheuristic optimization approaches. • Only some hybridized models were better than random forest and boosted regression trees. • The hybridized model of particle swarm optimization algorithm had the highest accuracy. Landslide susceptibility (LS) mapping is an essential tool for landslide risk assessment. This study aimed to provide a new approach with better performance for landslide mapping and adopting readily available variables. In addition, it investigates the capability of a state-of-the-art model developed using the group method of data handling (GMDH) to spatially model LS. Furthermore, hybridized models of GMDH were developed using different metaheuristic algorithms. The study area was the Bonghwa region of South Korea, for which an accurate landslide inventory dataset is available. We considered a total of 13 spatial covariates (altitude, slope, aspect, topographic wetness index, valley depth, plan curvature, profile curvature, distance from fault, distance from river, distance from road, land use, density of forest, and lithology were chosen as independent variables). Two benchmark models—random forest and boosted regression trees—were used to compare their results with the standalone GMDH and hybridized models. We compared model accuracy using the two most robust evaluation metrics, root mean square error (RMSE) and area under the receiver operating characteristic curve (AUROC). The validation results showed that hybridized models outperformed the standalone GMDH model. Moreover, the hybridized GMDH-PSO (AUC = 0.83, RMSE = 0.108), GMDH-IWO (AUC = 0.81, RMSE = 0.111), GMDH-BBO (AUC = 0.8; RMSE = 0.12), and GMDH-ICA (AUC = 0.8; RMSE = 0.117) had a better predictive performance than both RF and BRT. Therefore, the proposed approach could successfully produce landslide susceptibility maps using relatively few readily available variables and can be repeated in data-scarce regions. [ABSTRACT FROM AUTHOR]

Managing natural hazards such as land subsidence (LS) is important because they cause large economic and human loss. LS has become a significant challenge in South Korea due to its many abandoned coal mines. Therefore, preparing LS zoning maps is vital to controlling damage caused by LS. In this study, genetic algorithm (GA) and binary whale optimization algorithm were used to select the most influential conditioning factors from the initial set of 19 factors. Subsequently, the whale optimization algorithm, and Laplacian whale optimization algorithm (LXWOA) were utilized to determine the optimal hyperparameter values for the convolutional neural network (CNN) model. The results showed that the CNN-GA-LXWOA algorithm provided a more accurate and reliable LS susceptibility map, with an area under the receiver operating characteristic curve of 0.9606, root mean square error of 0.2974 and accuracy of 91.09%. This algorithm covered 95.6% of past subsidence occurrences in the high and very high LS susceptibility classes, demonstrating its suitability for predicting future subsidence areas. We found that six main factors (elevation, drift, lineament density, slope, distance to railroads, and railroad density) control LS occurrence in Taebaek, followed by groundwater depth, lithology, and profile curvature. The proposed model may be applied for other regions with different parameters and environmental factors due to its flexible structure. The final map created in this study provides a useful tool for better LS management to mitigate the adverse effects of this natural hazard in the study area. [ABSTRACT FROM AUTHOR]