On the cascade of uncertainty in CMIP5 climate projections for scenario-led hydrologic modelling in major river basins western Canada

Transcription

1 On the cascade of uncertainty in CMIP5 climate projections for scenario-led hydrologic modelling in major river basins western Canada Asong Z.E., S. Razavi, H. Wheater June 20 th, 2017

2 MOTIVATION The Intergovernmental Panel on Climate Change (IPCC) 2013 report: highlights uncertainty quantification as one of the most pressing research issues in determining how we can adapt to uncertain climate futures Two main perspectives on climate risk assessment for adaptation: Top-down (also known as scenario-led ) Bottom-up methods focus on reducing vulnerability to past and present climate variability Climate projections have demonstrated the need to adapt to a changing climate: but have been less helpful (so far) in guiding how to effectively adapt Partly due to the cascade of uncertainty Wilby & Dessai (2010, Weather)

3 MOTIVATION Uncertainty cascade in climate change impact assessment and adaptation analysis Going from assumptions about future society and global emissions of greenhouse gases to what that means for the climate to real decisions on a local scale Source:

4 PROBLEM Conventional uses of downscaling include: production of scenarios data inputs for impacts modelling evaluation of the consequences relative to present climate and discussion of appropriate adaptation responses However: large uncertainties attached to climate model scenarios cascade into even larger uncertainties in downscaled regional climate change scenarios and impacts A bigger problem: Which climate model? Which scenario? How many realizations? Hawkins & Sutton, 2011

5 PROBLEM: CLIMATE UNCERTYAINTY CCRN: modelling and prediction of the changing Earth system and its feedbacks under future climates Each of the steps in the process contains uncertainty, but which step is the most important?

6 WESTERN CANADA Top-down or scenario-led approach Cascade of uncertainty in CMIP5 projected change in precipitation and surface air temperature over 3 major river basins Time scales Annual, DJF, JJA McKenzie Fraser Nelson-Churchill Time windows , , Relative to

7 ANN McKenzie River Basin DJF JJA The three levels of the pyramid highlight the uncertainty due to the choice of RCP, GCMs and realization of climate variability. The intersection on the top row for each time period is the multi-scenario, multi-model, multi-realization mean

8 ANN Fraser River Basin DJF JJA For the near-term ( ), the relative importance of the RCPs is far smaller than the uncertainty in the model response. However, at the end of the century, the RCP uncertainty tends to dominate more

9 Nelson-Churchill River Basin ANN DJF JJA For the near-term ( ), the relative importance of the RCPs is far smaller than the uncertainty in the model response. However, at the end of the century, the RCP uncertainty tends to dominate more

10 Some recommendations for CCRN scenarios of change Time windows to consider in a multi-scenario context? RCP scenarios diverge only after 2050 Large envelop of uncertainty in precip, RCP8.5 over McKenzie RB Use multiple realizations/models? Which RCPs to consider? RCP8.5 increasing greenhouse gas emissions RCP4.5 stabilization scenario RCP2.6 very low greenhouse gas concentration levels

11 THANK YOU