Avtandil Amiranashvili

Georgia, due to its high mountains, growth of population, intensive land use, and the number of vulnerable infrastructures (roads, large engineering constructions, etc.), as well as climate change, belongs to the one of most vulnerable regions in the world, as it is prone to hazardous geological processes such as landslides (LSs). As the statistical data on LSs in Georgia are of low quality, due to political turmoils in the first decades of XXI-th century, a three-stage methodology was applied for assessing LS susceptibility. On the first stage, we perform the initial statistical assessment of LS susceptibility, using 11 weighted objective factors analysis. Then the machine learning (ML) methods were used for obtaining the final map of LS susceptibility. In conditions of Georgia, the intensive and prolonged precipitation is one of the main triggering factors for activation of LS processes. Therefore, in this study we also use hourly precipitation data, meteorological station observations, a historical landslide catalogue and hybrid machine learning approaches to construct spatial-temporal LS hazard maps under 24 h, 3-day, 5-day, and 30-day cumulative precipitation conditions. Predictive performance was evaluated retrospectively using a balanced sample of 309 landslide events and 309 susceptibility-matched controls, including a temporal train–test split. The results characterize discrimination under this sampling design; performance at the natural frequency of landslides requires independent evaluation. The results obtained show that the acceleration of global warming in recent decades affects the assessment of LS susceptibility/hazard.

The global climate change, driven by natural processes and increasing human activities, is especially significant for Georgia. The region is experiencing increases in temperature, desertification, redistribution of precipitation, and a rise in the frequency and severity of extreme weather events. Georgia’s complex topography and its proximity to the Black and Caspian seas make it essential to employ high-resolution regional climate models to evaluate future climate change risks. In this study, we examine the results of a high-resolution simulation of mean and extreme precipitation and temperature using the Abdus Salam International Centre for Theoretical Physics Regional Climate Model version 4.7.1 for the period 1985–2008, providing an initial evaluation of the model’s performance for the territory of Georgia. The model domain (1524 km; 2388 km) encompasses the entirety of Georgia’s territory and surrounding regions. The simulation, conducted at a 12 km horizontal grid spacing using ERA5 data as boundary conditions, indicates that the least discrepancy between observed and modeled average annual temperatures and precipitation, falling within a −1 to 1 °C and −200 to 200 mm range, respectively, was observed at most stations of eastern Georgia. The largest disparities between the model and observed average annual precipitation totals were noted along the Black Sea coast, in the Kolkheti Lowland, and in some high mountain stations in western Georgia. The most significant differences in average annual temperatures between the model and observations were observed in Ambrolauri, Mt. Sabueti, and Dedoplistskaro. For Georgia territory, such a long run with such a high resolution using ERA5 as boundary conditions was conducted for the first time. Overall, the modeling results are quite satisfactory, providing a solid basis for the successful utilization of the regional climate model RegCM4.7.1 with the selected parameterization for modeling monthly mean and extreme temperatures and precipitation in Georgia.