| Autor: |
Gocheva-Ilieva, S., Voynikova, D., Ivanov, A., Stoimenova, M. |
| Předmět: |
|
| Zdroj: |
AIP Conference Proceedings; 2019, Vol. 2164 Issue 1, p120005-1-120005-9, 9p, 2 Diagrams, 3 Charts, 4 Graphs |
| Abstrakt: |
Air pollution is a major problem in many urban areas in Bulgaria, harmful to the human health. In this study, based on a large number of observations for particulate matter 10 micrometers or less in diameter (PM10) and concomitant meteorological conditions, the mathematical modeling of the time series and predicting the level of future concentrations is presented. As a case study, data for the town of Nessebar, a typical coastal city, were used. The collected data are daily averaged for the period from February 2015 to March 2019. Using the autoregression moving average (ARIMA) method, models of the considered time series are built. To obtain more realistic forecasts, the methodology is implemented in two steps. The first step is to build univariate ARIMA models for any of the meteorological variables and use them to predict future values. In the second step, the calculated forecasted values in step one are applied to construct ARIMA PM10 models and to estimate the forecasts of this pollutant for a short time of 3 days ahead. The obtained models agree well to the known observed values. The proposed approach can be applied to other type of pollutants. It does not depend on additional forecasts from other sources and allows the development of a software application to predict future levels of pollution depending on the meteorological hazards. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
