Health risk assessment of soil trace elements using the Sequential Gaussian Simulation approach.

In this study, the performance of the Sequential Gaussian Simulation (SGS) approach was studied with the aim of accurately determining local health risk distributions associated with trace elements (V, Cr, Mn, Co, Ni, Cu, Zn, As, and Pb). This study plays a crucial role in determining the distribution of health risk levels, especially from heavy metals. In the SGS approach, health risk levels (non-carcinogenic and carcinogenic) were calculated for pixel sizes of 250 × 250 m2. Results were compared to the conventional Ordinary Kriging (OK) method. The cross-validation performances of both methods were compared. Non-carcinogenic health risks calculated according to SGS and OK for children were, respectively, ρc: 0.57 and 0.23, RMSE: 0.45 and 0.57, and MAE: 0.33 and 0.43. In the case of adults, non-carcinogenic SGS and OK results were, respectively, ρc: 0.53 and 0.24, RMSE: 0.06 and 0.07, and MAE: 0.04 and 0.05 for adults. Carcinogenic health risk estimates obtained by SGS and OK were, respectively, ρc: 0.72 and 0.31, RMSE: 4.1 × 10-5 and 5.8 × 10-5, and MAE: 3.2 × 10-5 and 4.3 × 10-5 in the case of children, and in the case of adults the results were, respectively, ρc: 0.71 and 0.30, RMSE: 5 × 10-6 and 4.3 × 10-6, and MAE: 4 × 10-6 and 5 × 10-6. These results indicated that SGS offered a more accurate approach in determining health risk distributions.

Trace element-based geochemical contamination characteristics and potential risks to human health: a case study from Northeast Turkey.

Trace elements measured in Artvin province soil samples in Northeast Turkey were assessed using pollution and health indices. The study area is positioned in one of the essential metallogenic belts in Turkey. This attempt is the first endeavor toward the study area in this context. The measured trace elements are As, Co, Cr, Cu, Mn, Ni, Pb and Zn, as they were assessed using pollution indices, enrichment factor, geo-accumulation index, contamination factor, and health risk assessment methods. According to the results of enrichment factor (EF), geo-accumulation index (Igeo), and contamination factor (CF), the soils of Artvin province show a slightly severe enrichment, moderately polluted and very high contaminated with arsenic, respectively. The pollution load index score (PLI) index (1.57) indicates that Artvin province is polluted in terms of trace elements. The hazard index (HI) calculated values for children and adults were 1.55 and 0.18, respectively. This revealed that the aforementioned metals can have non-carcinogenic effects (HI > 1). Total potential carcinogenic health risk (TCR) values for children and adults were 3.22 × 10-5 and 1.40 × 10-5, respectively. The non-carcinogenic risk level indicates that there may be a risk for children rather than adults.

Spatio-temporal estimation of the daily cases of COVID-19 in worldwide using random forest machine learning algorithm.

Novel Coronavirus pandemic, which negatively affected public health in social, psychological and economical terms, spread to the whole world in a short period of 6 months. However, the rate of increase in cases was not equal for every country. The measures implemented by the countries changed the daily spreading speed of the disease. This was determined by changes in the number of daily cases. In this study, the performance of the Random Forest (RF) machine learning algorithm was investigated in estimating the near future case numbers for 190 countries in the world and it is mapped in comparison with actual confirmed cases results. The number of confirmed cases between 23/01/2020 - 17/06/2020 were divided into 3 main sub-datasets: training sub-data, testing sub-data (interpolation data) and estimating sub-data (extrapolation data) for the random forest model. At the end of the study, it has been found that R2 values for testing sub-data of RF model estimates range between 0.843 and 0.995 (average R2= 0.959), and RMSE values between 141.76 and 526.18 (mean RMSE = 259.38); and that R2 values for estimating sub-data range between 0.690 and 0.968 (mean R2 = 0.914), and RMSE values between 549.73 and 2500.79 (mean RMSE = 909.37). These results show that the random forest machine learning algorithm performs well in estimating the number of cases for the near future in case of an epidemic like Novel Coronavirus, which outbreaks suddenly and spreads rapidly.

Correction to: A study of the radiological baseline conditions around the planned Sinop (Turkey) nuclear power plant using the mapping method.

The original version of this article unfortunately contained a mistake.

A study of the radiological baseline conditions around the planned Sinop (Turkey) nuclear power plant using the mapping method.

This study makes a first attempt at a detailed estimation of the background radioactivity level and its distribution at the Sinop nuclear power plant site. The activity concentration levels of 226Ra, 232Th, 40K and 137Cs radionuclides in soil samples collected from 88 locations around Sinop Province, Turkey, in November 2016, were measured using gamma spectrometry. The distributions of radionuclide levels obtained from the results were evaluated using a geostatistical method, and the estimated radiation levels were determined using the ordinary kriging (OK) method, which is the best linear unbiased estimator (BLUE) for unmeasured points. Estimates of distribution results were evaluated using cross-validation diagrams, and it was shown that the OK method could predict radiological distributions for appropriate criteria. Finally, using the kriging parameters, distributions of radiation levels for the entire work area were mapped at a spatial resolution of 100 × 100 m2. These maps show that the natural radionuclides (226Ra, 232Th and 40K) are distributed at higher levels to the southeast of Sinop than in the other regions, and the activity of an artificial radionuclide (137Cs) is high in the interior and northern sections.

A novel hybrid approach to the mapping and prediction of the terrestrial gamma dose rate distribution in the Central Anatolia Region of Turkey.

In this study, a new approach was suggested for the estimation and mapping of Terrestrial gamma dose rate (TGDR, in nGy h-1) in the Central Anatolia Region of Turkey by using the sequential Gaussian simulation (SGS) and Artificial neural network (ANN) methods together as a hybrid. In this hybrid approach (SGS-ANN), different from the classical spatial examinations, each spatial pixel (500×500m2) were calculated separately by evaluating the activity concentrations of terrestrial radionuclides that directly affects TGDR (226Ra, 232Th and 40K, in Bq kg-1) terrestrial coordination (X and Y, in meter). Therefore, the local changes of TGDR distributions that were estimated for the study area could be determined in appropriate precision without the smoothing effect. The performance evaluation of SGS-ANN approach was conducted by comparing the results for the same study area of Ordinary kriging (OK) method which is frequently used in the literature. According to the validation diagram that was created with the observed and estimated TGDR values, the Pearson's r correlation coefficient was obtained as 0.30 and 0.65, RMSE as 31.41 nGy h-1 and 25.79 nGy h-1, MAE as 24.50 nGy h-1 and 21.29 nGy h-1 and mean error as 5.97 nGy h-1 and -1.32 nGy h-1 for the OK method and the SGS-ANN approach, respectively.

Estimation of radon flux spatial distribution in Rize, Turkey by the artificial neural networks method.

In this study, average radon flux distribution in the Rize province (Turkey) was estimated by the artificial neural networks (ANN) method. For this purpose, terrestrial gamma dose rate (TGDR), which is defined as an important proxy in determining radon flux distribution, was used. Input parameters that were used for ANN were the natural radionuclide (238U, 232Th and 40K) activity values in soil samples taken from 64 stations in Rize Province, data from ambient gamma dose rates (AGDR) directly affecting the distribution of radon flux and data of geographical coordinates. Randomly chosen 42 stations were used for ANN training and data from 22 stations were used for testing the ANN model. Performance test results gave a Pearson's r value of 0.60 (p < 0.001) and RMSE of 0.296. The area that was used for the model was divided into grids of 100 m by 100 m and a spatial distribution map was composed by using ANN predicted radon flux rates at grid nodes, whereby natural radionuclide values and Ordinary Kriging predicted values of external gamma dose rates were used for composing the map.

Interpolated mapping and investigation of environmental radioactivity levels in soils and mushrooms in the Middle Black Sea Region of Turkey.

The activity concentration of natural (238U, 232Th, and 40K) and artificial (137Cs) radionuclides was determined in 50 samples (obtained from the same station) from various species of mushrooms and soil collected from the Middle Black Sea Region (Turkey). The activities of 238U, 232Th, 40K, and 137Cs were found as 84 ± 16, 45 ± 14, 570 ± 28, and 64 ± 6 Bq kg-1 (dry weight), respectively, in the mushroom samples and as 51 ± 6, 41 ± 6, 201 ± 11, and 44 ± 4 Bq kg-1, respectively, in the soil samples for the entire area of study. The results of all radionuclide activity measurements, except those of 238U and 232Th in the mushroom samples, are consistent with previous studies. In the soil samples, the mean values of 238U and 232Th are above the world mean, and the activity mean of 40K is below the world mean. Finally, the activity estimation was made with both the soil and mushroom samples for unmeasured points within the study area by using the ordinary kriging method. Radiological distribution maps were generated.

Spatial interpolation and radiological mapping of ambient gamma dose rate by using artificial neural networks and fuzzy logic methods.

The aim of this study was to determine spatial risk dispersion of ambient gamma dose rate (AGDR) by using both artificial neural network (ANN) and fuzzy logic (FL) methods, compare the performances of methods, make dose estimations for intermediate stations with no previous measurements and create dose rate risk maps of the study area. In order to determine the dose distribution by using artificial neural networks, two main networks and five different network structures were used; feed forward ANN; Multi-layer perceptron (MLP), Radial basis functional neural network (RBFNN), Quantile regression neural network (QRNN) and recurrent ANN; Jordan networks (JN), Elman networks (EN). In the evaluation of estimation performance obtained for the test data, all models appear to give similar results. According to the cross-validation results obtained for explaining AGDR distribution, Pearson's r coefficients were calculated as 0.94, 0.91, 0.89, 0.91, 0.91 and 0.92 and RMSE values were calculated as 34.78, 43.28, 63.92, 44.86, 46.77 and 37.92 for MLP, RBFNN, QRNN, JN, EN and FL, respectively. In addition, spatial risk maps showing distributions of AGDR of the study area were created by all models and results were compared with geological, topological and soil structure.

Determination and mapping the spatial distribution of radioactivity of natural spring water in the Eastern Black Sea Region by using artificial neural network method.

In this study, radiological distribution of gross alpha, gross beta, (226)Ra, (232)Th, (40)K, and (137)Cs for a total of 40 natural spring water samples obtained from seven cities of the Eastern Black Sea Region was determined by artificial neural network (ANN) method. In the ANN method employed, the backpropagation algorithm, which estimates the backpropagation of the errors and results, was used. In the structure of ANN, five input parameters (latitude, longitude, altitude, major soil groups, and rainfall) were used for natural radionuclides and four input parameters (latitude, longitude, altitude, and rainfall) were used for artificial radionuclides, respectively. In addition, 75 % of the total data were used as the data of training and 25 % of them were used as test data in order to reveal the structure of each radionuclide. It has been seen that the results obtained explain the radiographic structure of the region very well. Spatial interpolation maps covering the whole region were created for each radionuclide including spots not measured by using these results. It has been determined that artificial neural network method can be used for mapping the spatial distribution of radioactivity with this study, which is conducted for the first time for the Black Sea Region.

Dose rate estimates and spatial interpolation maps of outdoor gamma dose rate with geostatistical methods; A case study from Artvin, Turkey.

In this study, compliance of geostatistical estimation methods is compared to ensure investigation and imaging natural Fon radiation using the minimum number of data. Artvin province, which has a quite hilly terrain and wide variety of soil and located in the north-east of Turkey, is selected as the study area. Outdoor gamma dose rate (OGDR), which is an important determinant of environmental radioactivity level, is measured in 204 stations. Spatial structure of OGDR is determined by anisotropic, isotropic and residual variograms. Ordinary kriging (OK) and universal kriging (UK) interpolation estimations were calculated with the help of model parameters obtained from these variograms. In OK, although calculations are made based on positions of points where samples are taken, in the UK technique, general soil groups and altitude values directly affecting OGDR are included in the calculations. When two methods are evaluated based on their performances, it has been determined that UK model (r = 0.88, p < 0.001) gives quite better results than OK model (r = 0.64, p < 0.001). In addition, as a result of the maps created at the end of the study, it was illustrated that local changes are better reflected by UK method compared to OK method and its error variance is found to be lower.