The research of climatic extremeness in the Altai Mountains is relevant in connection with the presence of supporting objects for the study of soil-cover icing, the features of small water-power engineering and the existence of potential mudflow-prone areas. The study area is bounded 49-50 degrees N, 84-87 degrees E, where reference weather stations WMO 36442 Kara-Turek and WMO 36229 Ust'-Koksa are located. We selected 24 extreme indices developed by the Expert Team on Climate Change Detection, Monitoring and Indices (ETCCDMI) under the WMO: 15 for temperatures - TN10p, TX10p, TX90p, TN90p, TNn, TXn, TNx, TXx, GSL, ID, SU, FD, and 9 for precipitations - R1d, r5d, SDII, R10mm, R20mm, CDD, CWD, R95p, and PRCPTOT. Extremeness indicators were computed for 1950-2018 based on the author's algorithm. Verification of the calculation results was carried out using the RClimdex 3.2.0 package Indices related to air temperature show strong tendency to change. There has been an increase in the frequency of warm days and nights and a decrease in cool days and nights. Extreme changes in temperature are mostly stronger at night than during the day. On the other hand, for a group of indices valuable for agriculture and crop production, no significant changes have been received. Also, an annual count of days when daily precipitation is more than 10 mm and 20 mm (R10mm, R20mm), and simple daily intensity index (SDII) have no trend. Regarding the maximum 5-day precipitation total (r5d), the highest repeatability relates to July. In 70-75 % of cases, during the 5 consecutive days, there is a day on which the highest daily precipitation falls in a year; in other cases, these events are separated in time. The amount of heavy rainfall in a year is up to 26 % of the annual rainfall. Consecutive wet days index (period where the daily precipitation amount is more than 1 mm for at least consecutive 5 days) is small from April to October and about 6 days. The repeatability of such index has maximum in June-July and August-September. The length of the period with a total precipitation of not more than 1 mm per day is up to 14 days, with a maximum frequency in April. Within the area under study, at the beginning of the warm period the same predisposition to the formation of both wet and dry conditions has been noted. Such instability favors the display of extreme weather events. The computed values of r5d, r1d, SDII, R10mm, R20mm and R95p indices for both stations were determined as a feature space for clustering according to k-means iterative algorithm. As the objects, the years have been chosen. The division are separated into four groups, then stability was assessed through the Q factor "minimum variance in the group", pointed to "below average", "normal", "above average" and "anomaly". Since 1996, the inter-annual variability of values in the feature field has been increasing, changing its position from "below average" to "above average" and "anomaly", and bypassing the "normal" seen in 1995-2000 and 2006-2013. The substantial extreme indicators in the precipitation regime are the periods with/ without precipitation, defined by CWD and CDD (wet and dry spell) indices, respectively. The most probable wet months for the formation of a durable wet period with precipitation of more than 1 mm per day are April and July. In April and September there is a high probability of both dry and wet months. Consequently, the most unstable situation in the precipitation regime is related to in April.
