Possible Impact of Climatic Change and Anomalies to Intensity-Duration-Frequency Curves of Canadian Municipalities and Water Resources of Canadian River Basins

近几十年来，包括加拿大在内的世界范围内的极端水文事件发生得更频繁、更严重。在政府间气候变化专门委员会(IPCC)2007年的报告中，对全球气候模型的模拟一致表明，近几十年来，温室气体增加的人为强迫导致了全球范围内前所未有的气温上升。随着地球变暖，更高的温度导致更高的蒸发和降水速率，这将增加极端风暴的频率和严重程度。在市政基础设施的设计中，强度-持续时间-频率(IDF)曲线被广泛使用，假设它是平稳的。然而，全球变暖的影响可能会改变未来极端降雨事件的强度，因此基于现有IDF曲线的基础设施设计可能被低估。此外，大多数加拿大城市的IDF曲线是过时的，因为它们是从单站点降雨量数据分析和用矩量法(MOM)而不是一般极值(GeV)校准的极值I型(EVI)概率分布发展而来的。加拿大气候也可能受到厄尔尼诺南方涛动(ENSO)和太平洋年代际涛动(PDO)等气候异常的影响。结果表明，在加拿大西部，厄尔尼诺导致了很低的径流异常，而拉尼娜导致了较高的径流异常，加拿大西部冬季降水与ENSO和PDO有显著的相关性。为了解决这些重要问题，我们将：(1)基于EVI-MOM的单点分析，将过时的加拿大部分城市的IDF曲线替换为基于概率加权矩方法校准的GEV的多点分析的IDF曲线；(2)利用集合经验模式分解方法开发区域IDF曲线；(3)利用IPCC预测的气候情景驱动的中尺度大气模式模拟气候变化对选定加拿大城市IDF曲线的潜在影响，首先是独立的，然后是考虑陆地-大气反馈的耦合模式；以及(4)评估气候变化和气候异常对加拿大部分流域的综合水文影响。

In recent decades, hydrologic extremes have been occurring more frequently and in greater severity worldwide, including Canada. In the Intergovernmental Panel Climate Change (IPCC) report of 2007, simulations of global climate models unanimously demonstrate that anthropogenic forcing of increased greenhouse gases caused unprecedented temperature rise at global scale in recent decades. As the Earth warms, higher temperatures lead to higher rates of evaporation and precipitation, which will increase the frequency and severity of extreme storms. In designing municipal infrastructures, Intensity-Duration-Frequency (IDF) curves, assumed to be stationary, are used extensively. However, the effect of global warming could modify the intensity of future extreme rainfall events, and so infrastructure design based on existing IDF curves could be under-estimated. Further, IDF curves of most Canadian cities are outdated, for they were developed from single-site analysis of rainfall data and an Extreme Value Type I (EVI) probability distribution calibrated by method of moments (MOM), instead of the General Extreme Value (GEV). The Canadian climate can also be affected by climate anomalies such as El Nino Southern Oscillation (ENSO) and Pacific Decadal Oscillation (PDO). It has been shown that El Nino lead to very low streamflow anomalies, while La Nina lead to high streamflow anomalies, in western Canada; and that winter precipitation in western Canada are significantly correlated with ENSO and PDO. To address these important issues, we will: (1) Replace obsolete IDF curves of some Canadian cities based on single-site analysis of EVI-MOM, with IDF curves based on multi-site analysis of GEV calibrated by the Probability Weighted Moment method; (2) Develop regional IDF curves using the ensemble empirical mode decomposition method; (3) Simulate potential impact of climate change to the IDF curves of selected Canadian cities by a mesoscale atmospheric model driven with projected climate scenarios of IPCC, first in a stand-alone, then a coupled modes to account for the land-atmospheric feedback; and (4) Assess the combined hydrologic impact of climate change and climate anomalies to some river basins of Canada.