Expected Changes of the Number of Days with Hail in Tbilisi to 2085

Abstract

Predictive estimates of the number of hail days (HD) and their moving averages (for 3, 5, 7, 9 and 11 years – HD_3…HD_11) per warm period of year to 2050 and 2085 an example of Tbilisi was performed. Forecasting was carried out using the AAA version of the exponential smoothing (ETS) algorithm taking into account the periodicity in the pre-forecast time series. In particular, the following results were obtained. For the time series of the measured number of days with hail and HD_11 years, no pronounced peak in periodicity is observed. For time series HD_3, the periodicity is 14 years, HD_5 – 32 years, HD_7 and HD_9 – 31 years. In the period from 2022 to 2050, the range of variability of the average values of the central points of the forecast for the number of days with hail and the values of their 95% upper level is as follows: HD - from 0.9 to 3.8, HD_3 - from 1.0 to 3.0, HD_11 – from 1.0 to 1.6. In the period from 2022 to 2085, the range of variability of the average values of the central points of the forecast for the number of days with hail and the values of their 95% upper level is as follows: HD_5 - from 0.4 to 3.0, HD_7 - from 0.7 to 1.8, HD_9 - from 0.5 to 3.