Search

The paper introduces an original permutation approach to assessment of the predictive ability of models based on artificial neural networks. Three models based on artificial neural networks (a multilayer perceptron, a network with the radial basis function, and a neural network with the generalized regression) were implemented to illustrate the permutation approach. Data on the spatial distribution of copper in the upper soil layer of Novy Urengoy (Yamalo-Nenets Autonomous Okrug, Russia) were used for modeling. To evaluate the performance of the models, three different methods were used: error indices estimation, a graphical approach (the Taylor diagram), and a randomization estimation of the probability of obtaining the divergence between the observed and predicted series under assumption that both these datasets are taken from the same population. In the permutation approach, two statistics were used: the difference in means and the correlation coefficient. The permutation approach proved to be productive, as it allowed assessing the significance of the divergence between the observed and predicted datasets. [ABSTRACT FROM AUTHOR]

A comparison was made of the main transfer functions in the hidden layer of a nonlinear autoregressive neural network with an external input (NARX) to predict changes in the methane content in the surface layer of atmospheric air. Four transfer functions were used: S-type transfer function (LS), S-type hyperbolic tangent transfer function (TS), radial basis transfer function (RB), rectified linear units (LReLU). In general, all models were good at predicting changes in surface methane concentration. The most accurate results were obtained by the model, which using the RB type transfer function. [ABSTRACT FROM AUTHOR]

The results of the prediction of a model based on an artificial neural network were compared to predict the concentration of carbon dioxide (CO2) in the surface layer of the atmosphere for different time intervals. Measurements were taken on the Arctic island of Belyy, Russia. For comparison, three time intervals were used, which differed in the dependence of carbon dioxide concentration on the time of day. A non-linear autoregressive neural network with external input (NARX) was used. The model based on NARX successfully coped with the prediction. The smallest error was for the time intervals with a strong dependence of CO2 concentration on the time of day. [ABSTRACT FROM AUTHOR]

To studying variability of atmospheric particulate matter (PM), dust content measurements in location without industrial air pollution are very useful. The first time the air dust content and the size distribution were studied at 73.1° North latitude. This research presents data of the mass concentration and size-distribution of PM in the surface layer of atmospheric air on the Arctic island Belyy, Russia. The data was collected in Belyy Island, Yamal Peninsula, Siberia, Russia, in July of the same year. The mean air dust concentration was lower instrument limit of detection. The particle size distributions had two dominant Land Breeze and Sea Breeze patterns associated with the wind direction. [ABSTRACT FROM AUTHOR]

The mass concentration, size-distribution, height stratification and season variability of atmospheric particulate matter (PM) were researched in residential quarters of the industrial city in Ekaterinburg, Russia. The air dust samples were collected from March to May 2016. The particle size distribution in Ekaterinburg showed that most particles had diameters in the range of 0.5–2.5 µm. The particulate matter size distributions had dominant pattern with right tail and maximum was near 1 µm. [ABSTRACT FROM AUTHOR]

Creating models which are able to accurately predict the distribution of pollutants based on a limited set of input data is an important task in environmental studies. In the paper two neural approaches: (multilayer perceptron (MLP)) and generalized regression neural network (GRNN)), and two geostatistical approaches: (kriging and cokriging), are using for modeling and forecasting of dust concentrations in snow cover. The area of study is under the influence of dust emissions from a copper quarry and a several industrial companies. The comparison of two mentioned approaches is conducted. Three indices are used as the indicators of the models accuracy: the mean absolute error (MAE), root mean square error (RMSE) and relative root mean square error (RRMSE). Models based on artificial neural networks (ANN) have shown better accuracy. When considering all indices, the most precision model was the GRNN, which uses as input parameters for modeling the coordinates of sampling points and the distance to the probable emissions source. The results of work confirm that trained ANN may be more suitable tool for modeling of dust concentrations in snow cover. [ABSTRACT FROM AUTHOR]