The Ob is the third largest Eurasian river supplying heat and freshwater to the Arctic Ocean. These inputs influence water salinity, ice coverage, ocean temperatures and ocean circulation, and ultimately the global climate system. Variability of Ob River flow on long time scales is poorly understood, however, because gaged flow records are short. Eleven tree-ring width chronologies of Pinus sibirica and Larix sibirica are developed from the floodplain of the Lower Ob River, analyzed for hydroclimatic signal and applied as predictors in a regression model to reconstruct 8-month average (December-July) discharge of the Ob River at Salekhard over the interval 1705-2012 (308 yrs). Correlation analysis suggests the signal for discharge comes through air temperature: high discharge and floodplain water levels favor cool growing-season air temperature, which limits tree growth for the sampled species at these high latitudes. The reconstruction model (R-2 = 0.31, 1937-2009 calibration period) is strongly supported by cross validation and analysis of residuals. Correlation of observed with reconstructed discharge improves with smoothing. The long-term reconstruction correlates significantly with a previous Ob River reconstruction from ring widths of trees outside the Ob River floodplain and extends that record by another century. Results suggest that large multi-decadal swings in discharge have occurred at irregular intervals, that variations in the 20th and 21st centuries have been within the envelope of natural variability of the past 3 centuries, and that discharge data for 1937-2009 underestimate both the variability and persistence of discharge in the last 3 centuries. The reconstruction gives ecologists, climatologists and water resource planners a long-term context for assessment of climate change impacts.

Water reservoirs are used to regulate river discharge for a variety of reasons, such as flood mitigation, water availability for irrigation, municipal consumption and power production purposes. Recent efforts to increase the amount of renewable power production have seen an increase in intermittent climate-variable power production due to wind and solar power production. The additional variable energy production has increased the need for regulating the capacity of the electrical system, to which hydropower production is a significant contributor. The hydraulic impact on the time lags of flows between production stations have often largely been ignored in optimization planning models in favor of computational efficiency and simplicity. In this thesis, the hydrodynamics in the stream network connecting managed reservoirs were described using the kinematic-diffusive wave (KD) equation, which was implemented in optimization schemes to illustrate the effects of wave diffusion in flow stretches on the resulting production schedule. The effect of wave diffusion within a watershed on the variance of the discharge hydrograph within a river network was also analyzed using a spectral approach, illustrating that wave diffusion increases the variance of the hydrograph while the regulation of reservoirs generally increases the variance of the hydrograph over primarily short periods. Although stream hydrodynamics can increase the potential regulation capacity, the total capacity for power regulation in the Swedish reservoir system also depends significantly on the variability in climatic variables. Alternative formulations of the environmental objectives, which are often imposed as hard constraints on discharge, were further examined. The trade-off between the objectives of hydropower production and improvement of water quality in downstream areas was examined to potentially improve the ecological and aquatic environments and the regulation capacity of the network of reservoirs. / <p>QC 20170210</p>