| The high latitude regions of the globe are responding to climate change at unprecedented magnitudes and rates. As the climate warms, extreme hydroclimate events are likely to change more than the mean events, and it is the extreme changes that present a risk to society, the economy and the environment of the north. The subarctic boreal forest is one of the largest ecosystems in the world and is greatly understudied with respect to hydroclimate extremes. Thus, defining a baseline for changing extremes is the first step towards planning and implementing adaptation measures to reduce risk and costs associated with the changing extremes. This thesis focuses on quantitative analysis of extreme events using historical data and future model projections of changing temperature, precipitation and streamflow in the Interior forested region of boreal Alaska.;Historically, shifts in the climate have resulted in declining magnitudes of peak flow for snow dominated and glacial Interior Alaskan basins. However, changes are variable and dependent upon watershed topography, permafrost conditions, and glacial extents. Therefore, adjacent basins respond in considerably different ways to the same climate drivers. For example, peak streamflow events in the adjacent Salcha and Chena River basins had different responses to changes in climate. In the higher elevation Salcha basin, maximum streamflow increased as spring temperatures increased but in the lower elevation Chena, winter precipitation was a control on increases in maximum streamflow, while both were influenced by the Pacific Decadal Oscillation. Analysis of hydrologic change must take this variability into account to understand extreme hydroclimate responses and correctly account for process shifts.;To examine future changes in peak streamflow, the implementation and parameterization of hydrologic models to simulate hydroclimate extremes is required. In the northern latitudes of the world, there is a sparse observational station network that may be used for evaluation and correction of hydrologic models. This presents a limitation to science in these regions of the globe and has led to a paucity of research results and consequently, a lack of understanding of the hydrology of northern landscapes. Input of observations from remote sensing and the implementation of models that contain parameterizations specific to northern regions (i.e. permafrost) is one aim of this thesis. Remote sensing of snow cover extent, an important indicator of climate change in the north, was positively validated at snow telemetry sites across Interior Alaska. Input of the snow cover extent observations into a hydrologic model used by the Alaska Pacific River Forecast Center for streamflow flood forecasting improved discharge estimates for poorly observed basins, whereas the discharge estimates in basins with good quality river discharge observations improved little. Estimates of snow water equivalent were improved compared to station results and the adaptation of the model parameters indicated that the model is more robust, particularly during the snowmelt period when model simulations are error prone.;Use of two independent hydrologic models and multiple global climate models (GCMs) and emission scenarios to simulate changes in future hydroclimate extremes indicated that large regime shifts are projected for snowmelt dominated basins of Interior Alaska. The Chena River basin, nearby Fairbanks, Alaska, is projected to be rainfall dominated by the 2080s, with smaller snowmelt peaks. Return intervals for flooding will increase by one-and-one half to double the flow volume magnitude compared to the historical return interval. Frequency of extreme streamflow events will increase five times the mean increase. These changes in extreme streamflow events necessitate further research on the implications for infrastructure, ecology and economy to constrain risk associated with the projected regime shift in boreal forested watersheds of Interior Alaska. |
