Abstract
The paper forms the second part of an introduction to possible impacts of climate change on daily streamflow and extremes in the Province of Ontario, Canada. Daily streamflow simulation models developed in the companion paper (Part I) were used to project changes in frequency of future daily streamflow events. To achieve this goal, future climate information (including rainfall) at a local scale is needed. A regression-based downscaling method was employed to downscale eight global climate model (GCM) simulations (scenarios A2 and B1) to selected weather stations for various meteorological variables (except rainfall). Future daily rainfall quantities were projected using daily rainfall simulation models with downscaled future climate information. Following these projections, future daily streamflow volumes can be projected by applying daily streamflow simulation models. The frequency of future daily high-streamflow events in the warm season (May–November) was projected to increase by about 45%-55% late this century from the current condition, on average of eight-GCM A2 projections and four selected river basins. The corresponding increases for future daily low-streamflow events and future daily mean streamflow volume could be about 25%-90% and 10%-20%, respectively. In addition, the return values of annual one-day maximum streamflow volume for various return periods were projected to increase by 20%-40%, 20%-50%, and 30%-80%, respectively for the periods 2001-50, 2026-75, and 2051-2100. Inter-GCM and interscenario uncertainties of future streamflow projections were quantitatively assessed. On average, the projected percentage increases in frequency of future daily high-streamflow events are about 1.4-2.2 times greater than inter-GCM and interscenario uncertainties.
Highlights
It is widely believed that in this century climate change might result in increased flooding in many regions over the globe, based on studies conducted in the past decades
The results show that for thresholds with daily streamflow of 100, 10, 40, and 40 m3·s–1 or less in Grand, Humber, Rideau, and Upper Thames river basins, respectively, the differences between downscaled global climate model (GCM) historical runs and observations affect the future projections by about 1% - 2%
Since the antecedent precipitation index (API) calculated using rainfall data from the previous 24 days was used as a predictor in daily streamflow simulation modeling, the time period for the projection of future daily streamflow volumes is from May, rather than April, to November
Summary
The GCM projecttions have been used in the studies in different ways: 1) GCM-implied changes applied to the observed daily climate; 2) statistically downscaled GCM simulations; and 3) dynamically downscaled GCM data—regional climate model (RCM) simulations. Hypothesized scenarios, such as increases in annual mean temperature of 1 ̊C, 2 ̊C, 4 ̊C and/or changes in annual precipitation of ±5%, ±10%, ±20%, relative to the baseline climate, were used in the analysis on hydrological impacts of climate change. In addition to changes derived from GCM projections, the hypothesized scenarios were used as the input to hydrological models to project future water balance components (e.g., [7,8]). To project future climates required by hydrological models, daily and hourly observed hydrometeorological variables, such as temperature and rainfall, were modified by adding a sin-
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
