Application of phase analysis to predict long-term dynamics of atmospheric air temperature

Long-term average values of atmospheric air temperature (annual, autumn, winter, spring and summer) are one of the key points used to make scientifically valid decisions on the adaptation of crops to changing climate and environmental conditions. Despite their widespread use for the analysis and forecasting of various meteorological parameters of the atmosphere, the nature of such series has not yet been sufficiently studied. This determines the relevance of detailed analysis and forecasting of time series of atmospheric air temperature. At the same time, they are considered to be formed under the influence of linear and cyclic factors. To identify cyclical components, one of the most powerful and adequate time series analysis tools called phase analysis is used. In this study, a phase portrait of the time series of average air temperatures for the autumn period at the Nalchik weather station from 1961 to 2022 is constructed for preforecast analysis. To select the most significant periods of quasi-cycles, the theory of fuzzy logic is used. On its basis an algorithm for the formation of a fuzzy set of quasi-cycle lengths is implemented. Further prediction of the values of the time series in the retrospective section is carried out by the least square method. Further forecasting of time series values in the retrospective section is carried out using the least squares method. According to the results of the conducted research, it was found that the proposed model makes it possible to predict the values of average autumn air temperatures with high accuracy (5%). In the time series of average autumn air temperatures, a cycle characteristic of the 11-year cycle of solar activity and its phases is traced. All the quality criteria of the proposed model meet the requirements for the quality and adequacy of forecast models. This model can be applied to the analysis and forecasting of the average values of atmospheric air temperatures for the spring, summer and winter periods, as well as the average annual temperatures in general. The research has shown that the time series characterizing the temperature regime of atmospheric air, which is very complex in nature, can be predicted using phase analysis.

1. Aldoshkina ES. Experience in using the apparatus of neural networks for the analysis and forecast of the time series of air temperature.Uchenyye zapiski Rossiyskogo gosudarstvennogo gidrometeorologicheskogo universiteta = Scientific Notes of the Russian State Hydrometeorological University. 2009;(11):91–100. (In Russ.).

2. Bischokov RM. Ways to solve the problems of climate forecasting in the KBR.Vestnik Kurganskoy GSKHA = Bulletin of the Kurgan State Agricultural Academy. 2019;3(31):54–58. (In Russ.).

3. Efremenko DS, Kuznetsov AD, Seroukhova OS. On one algorithm for identifying local trends in the analysis of meteorological time series.Uchenyye zapiski Rossiyskogo gosudarstvennogo gidrometeorologicheskogo universiteta = Scientific Notes of the Russian State Hydrometeorological University. 2016;(7):132–141. (In Russ.).

4. Parovik RI, Firstov PP. Phase analysis of time series of geophysical fields. Vestnik KRAUNTS. Fiziko-matematicheskiye nauki = Vestnik KRAUNC. Physical and mathematical sciences. 2013;6(1):23–29. (In Russ.).

5. Perepelitsa VA, Popova EV. Fractal analysis of the behavior of natural time series.Sovremennyye aspekty ekonomiki = Modern aspects of economics. 2002;9(22):185–200. (In Russ.).

6. Box GEP, Jenkins GM. Time series analysis, forecasting and control. San Francisco: Holden-Day Publ.; 1970. 575 p. (In Russ.).

7. Golyandina NE. The Caterpillar-SSA Method: Time Series Analysis: A Study Guide. St. Petersburg: St. Petersburg State University; 2004. 76 p. (In Russ.).

8. Golyandina NE. Caterpillar-SSA method: time series forecasting: educational. St. Petersburg: St. Petersburg State University; 2004. 50 p. (In Russ.).

9. Alberg J, Nilson E, Walt J. Theory of splines and its applications. New York: Academic Press; 1967. 304 p. (In Russ.).

10. Vager BG, Serkov NK Splines in solving applied problems of meteorology and hydrology. St. Petersburg: Gidrometeoizdat; 1987. 160 p. (In Russ.).

11. Zade L. Fuzzy sets. Information and control. 1965;8:338–358.

12. Kofman A. Introduction to the theory of fuzzy sets. Mosсow: Radio and communication; 1982. 432 p. (In Russ.).

13. Rivin YuR. Cycles of the earth and the sun. Mosсow: Nauka; 1989. 162 p. (In Russ.).

14. Byuyul A, Tsefel P. SPSS: the Art of Information Processing. Analysis of Statistical Data and Restoration of Hidden Patterns. St. Petersburg: DiaSoftYup Publ.; 2005. 608 p. (In Russ.).