Оn the basis of long series of observations obtained at TOR-station at the Tomsk Akademgorodok, an empirical model for prediction of average daily ozone concentrations is developed based on a multilayer neural network. A comparison with models based on multiple linear regression and autoregression was conducted. The method of neural network approach turned out to be the most successful among all others. It gives a possibility to describe 70% of the total variance and the average value of 50% of the variance of the standard deviation. In this case, the value of the mean square prediction error does not exceed the instrumental error of measurements.
atmosphere, ozone, modeling, prediction
1. Belan B.D. Ozon v troposfere. Tomsk: Izd-vo IOA SO RAN, 2010. 488 p.
2. Zvjagincev A.M. Vlijanie ozona na zdorov'e naselenija (obzor literatury po teme) // Trudy Tret'ego mezhdunarodnogo soveshhanija-seminara. M.: GMC, 2013. URL: http://cao-rhms.ru/oom/meeting.html
3. Tarasova O.A. Nabljudenija prizemnogo ozona v programme global'noj sluzhby atmosfery // Trudy Tret'ego mezhdunarodnogo soveshhanija-seminara. M.: GMC, 2013. URL: http://cao-rhms.ru/oom/meeting.html
4. Konovalov I.B., Bikmann M., Kuznecova I.N., Glazkova A.A., Vasil'eva A.V., Zaripov R.B. Ocenka vlijanija prirodnyh pozharov na zagrjaznenie vozduha v regione Moskovskogo megapolisa na osnove kombinirovannogo ispol'zovanija himichesko-transportnoj modeli i dannyh izmerenij // Izv. RAN. Fiz. atmosf. i okeana. 2011. V. 47, N 4. P. 496–507.
5. Konovalov I.B., Beekmann M., Kuznetsova I.N., Zvya-gintsev A.M., Yurova A. Atmospheric impacts of the 2010 Russian wildfires: integrating modelling and measurements of an extreme air pollution episode in the Moscow region // Atmos. Chem. Phys. 2011. V. 11, N 19. P. 10031–10056.
6. Shalygina I.Ju., Kuznecova I.N., Nahaev M.I., Glazkova A.A., Zaharova P.V., Zvjagincev A.M. Harakteristiki i metody prognoza prizemnogo ozona v Moskovskom regione // Trudy Tret'ego mezhdunarodnogo soveshhanija-seminara. M.: GMC, 2013. URL: http://cao-rhms.ru/oom/meeting.html
7. Zvjagincev A.M., Belikov I.B., Elanskij N.F., Kakadzhanova G., Kuznecova I.N., Tarasova O.A., Shalygina I.Ju. Statisticheskoe modelirovanie maksimal'nyh sutochnyh koncentracij prizemnogo ozona // Optika atmosf. i okeana. 2010. V. 23, N 2. P. 127–135.
8. Aref'ev V.N., Kashin F.V., Milehin L.I., Milehin V.L., Tereb N.V., Upjenek L.B. Koncentracija prizemnogo ozona v Obninske v 2004–2010 years. // Izv. RAN. Fiz. atmosf. i okeana. 2013. V. 49, N 1. P. 74–84.
9. Yu Feng, Wenfang Zhang, Dezhi Sun, Liqiu Zhang. Ozone concentration forecast method based on genetic algorithm optimized back propagation neural networks and support vector machine data classification // Atmos. Environ. 2011. V. 45, N 11. P. 1979–1985. DOI: 10.1016/ j.atmosenv.2011.01.022.
10. Zhang Y., Bocquet M., Mallet V., Segneur C., Baklanov A. Real-time air quality forecasting, part I; History, techniques, and current status // Atmos. Environ. 2012. V. 60. P. 632–655.
11. Cobourn W.G. Accuracy and reliability of an automated air quality forecast system for ozone in seven Kentucky metropolitan areas //Atmos. Environ. 2007. V. 41, N 28. P. 5863–5875.
12. Sousa S.I.V., Martins F.G., Alvim-Ferraz M.C.M., Pereira M.C. Multiple linear regression and artificial neural networks based on principal components to predict ozone concentrations // Environ. Modelling & Software. 2007. V. 22, N 1. P. 97–103. DOI: 10.1016/ j.envsoft.2005.12.002.
13. Chattopadhyay G., Chattopadhyay S. Autoregressive forecast of monthly total ozone concentration: A neurocomputing approach // Comput. & Geosci. 2009. V. 35, iss. 9. P. 1925–1932.
14. Yeganeh B., Shafire Pour Motlagh M., Rashidi Y., Kamalan H. Prediction of CO concentrations on a hybrid Partial Least Square and Support Vector Machine model // Atmos. Environ. 2012. V. 55, N 1. P. 357–365. DOI: 10.1016//j.atmosenv.2012.02.092.
15. Fikret Inal Artificial Neural Network Prediction of Tropospheric Ozone Concentrations in Istanbul, Turkey // CLEAN – Soil, Air, Water. 2010. V. 38, iss. 10. P. 897–908. DOI: 10.1002/clen.201000138.
16. Nastos P.T., Moustris K.P., Larissi I.K., Paliatsos A.G. Rain intensity forecast using Artifical Neural Networks in Athens, Greece // Atmos. Res. 2013. V. 119. P. 153–160. DOI: 10.1016/j.atmosres. 2011.07.020.
17. Shad R., Mesgari M.S., Abcar A., Shad A. Predicting air pollution using fuzzy generic linear membership kriging in GIS // Comput. Environ. Urban. 2009. V. 33, N 6. P. 472–481.
18. Arshinov M.Ju., Belan B.D., Davydov D.K., Kovalevskij V.K., Plotnikov A.P., Pokrovskij E.V., Skljadneva T.K., Tolmachev G.N. Avtomaticheskij post dlja kontrolja kachestva vozduha // Meteorol. i gidrol. 1999. N 3. P. 110–118.
19. Antohin P.N., Arshinov M.Ju., Belan B.D., Belan S.B., Davydov D.K., Kozlov A.V., Krasnov O.A., Pestunov D.A., Praslova O.V., Fofonov A.V., Inoue G., Machida T., Maksjutov Sh., Shimoyama K., Sutoh H. Primenenie samoleta An-2 dlja issledovanija sostava vozduha v pogranichnom sloe atmosfery // Optika atmosf. i okeana. 2012. V. 25, N 8. P. 714–720.
20. Antohin P.N., Arshinov M.Ju., Belan B.D., Skljadneva T.K., Tolmachev G.N. Prognoz izmenenija koncentracii ozona i ajerozolja na osnovanii predskazannogo v 24-m cikle urovnja solnechnoj aktivnosti // Optika atmosf. i okeana. 2012. V. 25, N 9. P. 778–783.
21. Antohin P.N., Belan B.D. Regulirovanie dinamiki troposfernogo ozona cherez stratosferu // Optika atmosf. i okeana. 2012. V. 25, N 10. P. 890–895.
22. Antohin P.N., Arshinova V.G., Arshinov M.Ju., Belan B.D., Belan S.B., Davydov D.K., Kozlov A.V., Krasnov O.A., Praslova O.V., Rasskazchikova T.M., Savkin D.E., Tolmachev G.N., Fofonov A.V. Sutochnaja dinamika vertikal'nogo raspredelenija ozona v pogranichnom sloe atmosfery v rajone Tomska // Optika atmosf. i okeana. 2013. V. 26, N 8. P. 665–672.
23. Arshinov M.Ju., Belan B.D., Davydov D.K., Savkin D.E., Skljadneva T.K., Tolmachev G.N., Fofonov A.V. Rezul'taty mnogoletnego monitoringa ozona v rajone goroda Tomska // Trudy Vtorogo mezhdunarodnogo soveshhanija-seminara. M.: IOF RAN, 2013. P. 38–49.
24. Salazar-Ruiz E., Ordieres J.B., Vergara E.P., Capuz-Rizo S.F. Development and comparative analysis of tropospheric ozone prediction models using linear and artificial intelligence-based models in Mexicali, Baja California (Mexico) and Calexico, California (US) // Environ. Modelling & Software archive. 2008. V. 23, iss. 8. P. 1056–1069. DOI: 10.1016/j.envsoft. 2007.11.009.
25. Zounemat-Kermani M. Hourly predictive Levenberg–Marquardt ANN and multi linear regression models for predicting of dew point temperature // Meteorol. and Atmos. Phys. 2012. V. 117, iss. 3–4. P. 181–192. DOI: 10.1007/s00703-012-0192-x.