Application of Decadal Scale Projections Based on Large Scale Climate Indices to Decision Making in the Colorado River Basin

Abstract:

Effective water resources planning and management requires skillful decisions on multi-year or decadal timeframes. In basins such as the Colorado River Basin (CRB), streamflow is not stationary but exhibits variability that reflects teleconnections with large scale climate indices such as Atlantic Multi-decadal Oscillation (AMO) and Pacific Decadal Oscillation (PDO). A recently developed stochastic streamflow simulation and projection model, the Wavelet K-Nearest Neighbor (WKNN) model, identifies and reconstructs dominant quasi-periodic signals in the AMO and PDO using wavelet analysis, simulates each using block K-Nearest Neighbor (K-NN) bootstrap, then simulates the streamflow using a K-NN bootstrap conditioned on the simulated climate forcings, and has been demonstrated to produce skillful decadal scale projections of streamflow in the CRB. The Bureau of Reclamation’s 2012 Colorado River Basin Supply and Demand Study used scenarios to explore the use of options and strategies such as infrastructure development, conservation and efficiency improvements to address supply-demand imbalances. Each year in the simulated scenarios, decision criteria such as reservoir elevations and average flows over recent years were applied to determine system vulnerability and the need to implement options and strategies to mitigate future shortages. This presentation describes the addition of the WKNN generated decadal scale flow projections to the decision criteria. In addition, periods of poor predictability are identified by using a nonlinear dynamical system based approach to recover the underlying dynamics. Time varying predictability is assessed by quantifying the divergence of trajectories in the phase space with time, using Local Lyapunov Exponents (LLE). Skillful decadal scale streamflow projections within the high predictable time epochs are used to indicate future flow conditions and improve decisions. An ensemble of projections is considered to be wet or dry based whether or not the mean exceeds some reference threshold and that information is used to constrain the decisions. Based on projections being wet, dry or unpredictable, improved decisions are shown to reduce cost or reduce shortage and are illustrated by tradeoff curves of risk of shortage vs. cost.