PROSPECTIVE FORECAST OF TEMPERATURE IN VAN PROVINCE USING MARKOV PROCESSES


Creative Commons License

Uyar B., Güzel Urcan F.

4. INTERNATIONAL NEW YORK ACADEMIC RESEARCH CONGRESS, New York, Amerika Birleşik Devletleri, 15 - 16 Ocak 2022, ss.33

  • Yayın Türü: Bildiri / Özet Bildiri
  • Basıldığı Şehir: New York
  • Basıldığı Ülke: Amerika Birleşik Devletleri
  • Sayfa Sayıları: ss.33
  • Van Yüzüncü Yıl Üniversitesi Adresli: Evet

Özet

Weather forecasting has an important place in human life. Uncertainty in the forecast, uncertain weather,

can cause confusion. Therefore, for centuries, people have made weather forecasts to avoid confusion

and uncertainty (Stephens et al., 2012). In the past, people made predictions using their own methods

and observation. Knowing the weather has the ability to facilitate many situations, from people's travels

to their vacations and daily chores (Hares et al., 2010). In this study, the Markov chain of meteorological

temperature data of the last 30 years obtained from the Regional Directorate of Meteorology in Van

province, special cases and analyzes of the Markov chain were examined and tried to be modeled with

stochastic processes. By using Matlab program, by classifying the received data according to their

ranges, it was determined which class it belongs to for each data in the data set. Transition matrix was

created over the classes, and probability transition matrix was calculated with the transition matrix. With

the obtained probability transition matrix, the weather temperature forecast of Van for the next month

was made. In the V matrix obtained at the end of the study, the average temperature expected in the first

month of the next year was the 2nd case with a probability of 63%. As a result of this, the temperature

range that is valid in the 2nd case is determined as [0°C-20°C]. Making temperature predictions for the

following months, seasons or years with the help of the graphs obtained and the predictions that can be

made with possible Markov processes, will be beneficial for our country in terms of being prepared for

expectations and struggling with many events from prospective public-private investment planning to

heat and drought-related thirst and their consequences. . Based on this result; It adds positive value to

our lives with the expectation of temperature and weather changes due to climates, making it easier to

predict the situation. This provides added value for us. In this context, Markov processes make a visible

contribution to science and, accordingly, to life.