Stochastic drought quantification for the South Canadian Prairie

Chun, K; Wheater, HS; Vashchyshyn, Ilona; Khaliq, N

Abstract

For formulating robust adaptation policies under nonstationary climate, drought processes need to be characterised and modelled adequately. As a case study, the major drought episode in the early 2000s in the South Canadian Prairie is investigated using a stochastic approach driven by Global Circulation Model (GCMs) outputs, large scale climate oscillation indices and reanalysis data. The meteorological drought conditions are characterised by the Drought Severity Index (DSI). Results show that interannual drought variability cannot be modelled simply by autocorrelated structures or seasonal cycles of the precipitation series. The US National Centers for Environmental Prediction (NCEP) reanalysis data and Pacific Decadal Oscillation (PDO) Index provide interannual signals which are useful for the proposed stochastic approach to simulate realistic severe drought events. Although GCM outputs such as the Canadian Centre for Climate Modelling and Analysis (CCCma) can in principle also be used in the proposed framework to generate drought series, the simulated and historical series are less well matched. These results imply that current GCM outputs have limited information with respect to interannual signals. Finally, the possibility to extend the proposed stochastic approach to support risk-based water management is discussed.

• Presentation Conference Type: Presentation / Talk
• Conference Name: AGU Fall Meeting
• Start Date: Dec 3, 2012
• End Date: Dec 7, 2012
• Deposit Date: Feb 25, 2022
• Public URL: https://uwe-repository.worktribe.com/output/8545715
• Publisher URL: https://ui.adsabs.harvard.edu/abs/2012AGUFM.H41B1181C/abstract