Intelligent temporal analysis of surface air temperature series for hydrometeorological monitoring

Abstract

Relevance: The ability to forecast weather conditions is essential for societal development, as it helps minimize damage caused by weather anomalies. At specific points in time, the state of environmental parameters is recorded by a hydrometeorological monitoring system. One of the parameters recorded is surface air temperature. Trends in changes to the current state of the natural environment cannot be determined without knowledge of its previous state and a model of meteorological processes. Therefore, the collection and intelligent analysis of data on the dynamics of temperature processes for subsequent refinement of the meteorological process model remains a relevant task. Aim and research objectives: To conduct an intelligent analysis of long-term time series of surface air temperature to obtain information on the temporal characteristics of surface air temperature variability. To assess changes in temperature trends, temperature trend patterns, and the scales of temperature variability processes. Methods: Time series analysis method. Scientific novelty: A temporal analysis was performed on long-term time series of surface air temperature observations not previously presented in research: temperature changes, the scales of these changes, and trends in temperature variability were assessed. Practical significance: The use of the obtained data in hydrometeorological monitoring systems to refine models of meteorological processes, improve the accuracy of identifying patterns in climate dynamics, and support forecasting decisions. Results: For long-term temperature time series with different properties (different lags, different durations), amplitude characteristics of air temperature changes and temperature trends were obtained, and the scales of temperature change were estimated. Conclusions: As a result of the study, changes in temperature patterns, temperature trends, and the scales of temperature variability processes were assessed over twenty years (from two thousand and five to two thousand and twenty-four) and over one hundred forty-four years (from one thousand eight hundred and eighty-one to two thousand twenty-four) based on an intelligent analysis of four surface air temperature observation series with lags of one year, one month, one day, and three hours.