Author(s): Tarana A. Solaiman; Slobodan P. Simonovic
Linked Author(s):
Keywords: No Keywords
Abstract: The main goal of this paper is to assess the impact of climate change on a watershed scale. The most common approaches of assessing the climate change impact are limited by the uncertainties related to the availability of proper spatial and temporal hydroclimatic data. Global circulation models provide long records but on a coarse spatial scale. The historical observed information, although has better spatial resolution, is limited temporally; hence, cannot be used as an indicator of the future climate change. One way to address these drawbacks is to use reanalysis data and scale it down to local scales. In the presented research, an analysis has been made of the correspondences and/or discrepancies between observed precipitation and temperature data, and the data from the National Centers for Environmental PredictionNational Center for Atmospheric Research (NCEP-NCAR) (a) global (NNGR) and (b) regional (NARR) reanalysis project. The following data between 1980 and 2005 has been extracted for the analysis: daily precipitation, maximum, mean and minimum temperature. The extracted data at several grid points in and around the Upper Thames River watershed in Southwestern Ontario, Canada have been used with a continuous hydrologic model to generate low flows. Both NARR and NNGR temperature (Tmax, Tmin and Tmean) data show a good synopsis of the climate conditions within the study area. The precipitation data from NNGR is less reliable than the NARR. The stream flows generated from the NARR dataset show encouraging result; however, some overestimations are also seen. The uncertainty estimations of the outputs are plotted using variation in mean and variance. The results indicate that the NARR dataset can be used as a good source for interpreting climate variation; nonetheless, a complete knowledge and careful investigations of the differences are necessary for application with the hydrologic models.
Year: 2009