The response of vegetation to climate change is of special interest in regions where rapid warming is coupled with moisture deficit. This raises the question of the limits in plants’ acclimation ability and the consequent shifts of the vegetation cover. Radial growth dynamics and climatic response were studied in Scots pine (Pinus sylvestris L.), Siberian larch (Larix sibirica Ledeb.), and silver birch (Betula pendula Roth.) in the forest-steppe, and for Siberian elm (Ulmus pumila L.) in the steppe of South Siberia, as indicators of vegetation state and dynamics. Climate–growth relationships were analyzed by the following two approaches: (1) correlations between tree-ring width chronologies and short-term moving climatic series, and (2) optimization of the parameters of the Vaganov–Shashkin tree growth simulation model to assess the ecophysiological characteristics of species. Regional warming was accompanied by a slower increase of the average moisture deficit, but not in the severity of droughts. In the forest-steppe, the trees demonstrated stable growth and responded to the May–July climate. In the steppe, elm was limited by moisture deficit in May–beginning of June, during the peak water deficit. The forest-steppe stands were apparently acclimated successfully to the current climatic trends. It seems that elm was able to counter the water deficit, likely through its capacity to regulate transpiration by the stomatal morphology and xylem structure, using most of the stem as a water reservoir; earlier onset; and high growth rate, and these physiological traits may provide advantages to this species, leading to its expansion in steppes.

Regional and local climate change depends on continentality, orography, and human activities. In particular, local climate modification by water reservoirs can reach far from shore and downstream. Among the possible ecological consequences are shifts in plant performance. Tree-ring width of affected trees can potentially be used as proxies for reservoir impact. Correlation analysis and t-tests were applied to climatic data and tree-ring chronologies of Pinus sylvestris L. and Larix sibirica Ledeb. from moisture-deficit habitats in the intermontane Khakass-Minusinsk Depression, to assess modification of climate and tree growth by the Krasnoyarsk and Sayano-Shushenskoe Reservoirs on the Yenisei River. Abrupt significant cooling in May–August and warming in September-March occurred after the launch of the turbines in dams, more pronounced near the Sayano-Shushenskoe dam (up to – 0.5 °C in summer and to + 3.5 °C in winter) than near the Krasnoyarsk Reservoir headwaters (– 0.3 °C and + 1.4 °C). Significant lengthening of the warm season was also found for temperature thresholds 0–8 °C. Shifts of seasonality and intensity occurred in climatic responses of all tree-ring chronologies after development of water reservoirs. Patterns of these shifts, however, depended on species-specific sensitivity to climatic modification, distance from reservoirs, and physiographic regions. Mitigation of climate continentality and extremes by reservoirs appears to have offset possible negative effects of warming on tree growth.

A nested July–June precipitation reconstruction for the period AD 1777–2012 was developed from multi-century tree-ring records of Pinus sylvestris L. (Scots pine) for the Republic of Khakassia in Siberia, Russia. Calibration and verification statistics for the period 1948–2012 show a high level of skill and account for a significant portion of the observed variance (>50%) irrespective of which period is used to develop or verify the regression model. Split-sample validation supports our use of a reconstruction model based on the full period of reliable observational data (1948–2012). Thresholds (25th and 75th percentiles) based on the empirical cumulative distribution of 1948–2012 observed precipitation were used to delineate dry years and wet years of the long-term reconstruction. The longest reconstructed dry period, defined as consecutive years with less than 25th percentile of observed July–June precipitation, was 3 years (1861–1863). There was no significant difference in the number dry and wet periods during the 236 years of the reconstructed precipitation. Maps of geopotential height anomalies indicate that dry years differ from wet years primarily in the location of an anomalous 500-mb ridge approximately over the study area.