Some results of a study of the relationship between the mean annual sum of atmospheric precipitation and re-activated and new landslide cases in Georgia taking into account the climate change.
2022
Amiranashvili A. | Chelidze T. | Svanadze D.
Some results of statistical analysis of long-term variations of annual amount of atmospheric precipitation for 21 meteorological stations of Georgia (P) located in areas with landslides, average annual amount of precipitation for these stations (Pₐ), relationship between the Pₐ and number of re-activated and new cases of landslides (LS), and the estimated values of LS up to 2045 using predictive data on Pₐ are presented. Data from the Environmental Agency of Georgia on the P in period 1936 - 2020 and data on LS in period 1996 – 2018 are used. The forecast of Pₐ using the AAA version of the Exponential Smoothing (ETS) algorithm was carried out. In particular, the following results are obtained. The correlations between the annual amounts of P at each of the meteorological stations with averaged data for all 21 stations Pₐ are established. In 1981-2020, compared with 1936-1975, no significant variability of the mean P values is observed at 11 stations, an increase - at 6 stations, and a decrease - at 4 stations. The Pₐ value do not change during the indicated time periods. The forecast of the Pₐ value up to 2040 were estimated taking into account the periodicity of precipitation variability, which is 11 years. A cross-correlation analysis of the time series of the Pₐ and LS values showed that the best correlation between the indicated parameters is observed with a five-year advance of precipitation data. With this in mind, a linear regression equation was obtained between the five-year moving average of the Pₐ and the five-year moving average of the LS values. Using this equation and predictive Pₐ data, five-year moving averages of re-activated and new landslides cases up to 2045 were estimated. Tab. 3, Fig. 9, Ref. 17.
显示更多 [+] 显示较少 [-]