Space-Time Modeling of Climate Change Effects on the Water Balance of Southwestern Alberta

Changes and variabilities in space and time characteristics of hydro-climatic processes play a major role in the water balance of a watershed. Formulating suitable spatio-temporal models for interactive behavior of such processes is crucial in understanding and investigating potential climate change effects on hydrology of a watershed. Historical climatic and hydrologic data are herein used to calibrate a conceptual stochastic monthly water balance for southwestern Alberta watersheds in Canada for the present day climate. The inputs to the model are monthly precipitation, evapotranspiration and temperature time series and the outputs are monthly soil moisture storages and streamflow. To account for vegetation influence on the precipitation-runoff transformation, a moisture index parameter based on the Jansen-Haise formulation and HDF-EOS data is incorporated into the hydrologic model. An adaptive technique is used to describe time evolution and uncertainties in the moisture index parameter. To simulate the effects of global warming due to the doubling of CO2, outputs from the Canadian Global Coupled Model 1 (CGCM1) are used to provide inputs that are downscaled to drive the hydrologic model. Spatial estimation of unknown data are performed using a modified Kriging approach combined with Canonical Correlation analysis. Preliminary results indicate that the formulated model is statistically adequate for estimating monthly streamflow at the 95% level. It is also shown that a temperature based approach can be used to estimate the evapotranspiration time series as an immediate process when data for the latter are either unavailable or insufficient. This result is attractive in the sense that only temperature and precipitation are needed to drive the hydrologic model. Finally, the sensitivity of moisture index and streamflow to potential climate changes is briefly discussed.