Simulating streamflow response to climate scenarios in central Canada using a simple statistical downscaling method

Abstract

Despite the importance of surface water in Manitoba, Canada, there has been relatively little investigation of the impact of climate change on the hydrology of the province. This study examines streamflow characteristics in northern Manitoba basins (Taylor and Burntwood River basins) under climate scenarios generated by global climate models (GCM) from 2 agencies (the UK Hadley Centre and the Canadian Centre for Climate Modelling and Analysis). The hydrological model SLURP (Semi-distributed Land Use-based Runoff Processes) was run with perturbed meteorological data based on the GCM simulations under 2 greenhouse gas emission scenarios (A2 and B2) for 2 future periods (2041-2070 and 2071-2099). A method to incorporate the changed variability in future temperature and precipitation to the climate scenarios was also tested. All scenarios project wetter and warmer climates, and simulation results show that the mean annual runoff is likely to increase in every scenario. The largest seasonal increase occurs in late spring and the least in winter and summer. The daily flow exceeded by 80% of the total records is also projected to substantially increase, resulting in much fewer days with extreme low flow compared to the control period. Results indicate that the combination of a hydrological model with simple GCM-based scenarios can produce results comparable to those obtained from regional climate modelling. Explicit incorporation of changed variability in climate scenarios results in higher runoff than consideration of only changes in means. KEY WORDS: Climate change impact · Runoff · Statistical downscaling · SLURP · ManitobaResale or republication not permitted without written consent of the publisher Clim Res 40: 89-102, 2009 mate models (GCMs) consistently project mean temperature increases of several degrees in northern Manitoba by the second half of the 21st century (Boer et al. 2000, Johns et al. 2003, it is imperative to assess the related effects on water resources.Few studies have addressed the effect of past or potential climate change on the hydrological cycle in Manitoba. This is in contrast to other areas in Canada where extensive studies have been undertaken, including British Columbia (see Merritt et al. 2006 for a review on B.C. studies), Ontario (e.g. Cunderlik & Simonovic 2005), and Quebec (e.g. Roy et al. 2001). Westmacott & Burn (1997) and Yulianti & Burn (1998) investigated links between climatic warming and hydrologic characteristics of river basins in the Canadian Prairies, including portions of Alberta, Saskatchewan, and Manitoba, by analysing past streamflow and temperature data. While they discovered a clear relationship between stream flow and temperature, they did not provide explicit projection of future climatic and hydrological characteristics. Sushama et al. (2006) investigated the Churchill and the Nelson River basins using 2 different versions of the Canadian Regional Climate Model (v3.6 and v3.7). Both of these rivers flow through Manitoba to Hudson Bay. The simulations ...

Downscaling of Climate Information

“…All AOGCMs indicated a warmer future in both direct output and SWG derived local scenarios, and downscaled 50-year return period temperatures increased by up to 4°C in 2041-2070 relative to . Consistent with other research on possible changes in precipitation regimes over Canada (Choi et al 2009), potential changes in the 50-year return-period daily precipitation for the mid-twenty-first century downscaled from each AOGCM were almost uniformly positive and increased by up to 25 % from values during . Analysis of the downscaling results versus use of direct AOGCM output indicated (i) application of the SWG reduced bias in extreme metrics during the baseline period, (ii) climate change signals in the SWG localized projections differed markedly from the direct AOGCM output, and (iii) uncertainty in future climate projected from the four different AOGCMs is a major contributor to overall analysis uncertainty .…”

Modelling of climate-induced hydrologic changes in the Lake Winnipeg watershed

“…Changes in precipitation principally affect maximum snow accumulation and runoff volume while temperature changes mostly affect runoff timing (Barnett et al, 2005). The potential future impacts in snowmeltdominated catchments may include a reduction in snowpack volume and an earlier onset of melt (e.g., Stewart et al, 2004;Dibike and Coulibaly, 2005;Merritt et al, 2006;Rauscher et al, 2008;Choi et al, 2009b). These model-predicted changes are already evident in twentieth and early twenty-first century trends, such as a general decline of snowpack volume (e.g., Mote et al, 2005;Stewart, 2009) and earlier occurrence of snowmelt (e.g., Adam et al, 2009;Stewart et al, 2005).…”

Quantifying uncertainty in urban flooding analysis considering hydro-climatic projection and urban development effects

“…To downscale three GCM simulations with two emission scenarios, we use a simple delta method, which has widely been used in climate change impact studies (e.g., Lettenmaier et al, 1999;Wilby and Harris, 2006;Loukas et al, 2007;Graham et al, 2007;Kay et al, 2009;Choi et al, 2009). This method first calculates monthly precipitation and temperature differences between the reference and future GCM simulations.…”