Space-time Interpolation by Combining Air Pollution and Meteorologic Variables

Abstract

In this paper, a technique is proposed in order to study triple time series. It combines the variable of interest, sulfur dioxide (SO 2 ) with two related meteorological variables. Hence, three variables measured at the same time points are jointly analyzed. Instead of using classical multiple time series analysis, it is suggested to consider the measurements of the two meteorological variables as coordinates of a two-dimensional space and the simultaneous observation of the third variable (associated SO 2 concentrations) at each pair of coordinates. Subsequently, well-known optimum interpolation is used for predicting the SO 2 concentrations on the basis of six meteorological variables. All the variables of the study are measured at the same times (all days in 2000) around the city of Istanbul, Turkey. The triple diagrams, in the form of contour maps, help to answer various questions concerning the SO 2 concentration variability with respect to meteorological variables. The same diagrams also provide a basis for the prediction of SO 2 concentrations. It is shown that the relative prediction error is less than 10%, which is acceptable for the practical studies.

Multivariate Interpolation of Wind Field Based on Gaussian Process Regression

“…A significant advantage of combining kernel functions is the capability of extracting multi-scale information. By combining the kernels mentioned above according to Equation (8), the new kernel will be capable of extracting multiple pieces of information: periodical information, the diversity in different directions, or rough and localized features. The length-scale hyperparameters of kernels imply the influencing radius of each specific feature.…”

“…Conventional surface fitting methods, such as bilinear and spline methods, are only applicable to specific cases [8]. They are deterministic without taking the randomness of natural weather events into account.…”

Classification of the Entities Represented by Samples from Gaussian Distribution

“…Conversely temperature, rainfalls, humidity, wind speed, and sunshine do show the difference especially between the low-pollution class and the highpollution class. According to Şen et al (2006) [9], the pollution episodes in large cities are often related to high atmospheric pressure situations. The latter represent the ideal conditions for the gathering of pollutants in the air.…”