Dates: January 13-15, 2010
Location: Millennium Harvest House, Boulder, Colorado
The workshop, "Nonstationarity, Hydrologic Frequency Analysis, and Water Management" addressed whether assumptions of stationarity are valid, use of different statistical models in nonstationarity conditions, trend analyses, how to use the output from global climate models (GCM), and how to treat uncertainty in planning, design, and operations. This workshop also resulted in a special collection of journal papers in the Journal of the American Water Resources Association, June 2011, Volume 47, Issue 3. The Featured Collection on "Nonstationarity, Hydrologic Frequency Analysis, and Water Management" included 13 papers on the topic by national and international experts.
The assumption behind traditional hydrologic frequency analysis is that climate is stationary. Stationarity means that the statistical properties of hydrologic variables in future time periods will be similar to past time periods. Anthropogenic climate change and better understanding of decadal climate variability present a challenge to the validity of the assumption. The Intergovernmental Panel on Climate Change (IPCC) has said “Climate change challenges the traditional assumption that past hydrological experience provides a good guide to future conditions” (Bates et al, 2008). Although there have been academic articles criticizing the assumption of stationarity, it is not apparent what if any alternative methods should be used as a replacement. The workshop presented and discussed proposed operational alternatives to the assumption of stationarity in hydrologic frequency analysis that can be used in a transitional period by water managers and planners, as well as a new generation of methods that could be developed. Limitations of the alternatives were also presented and discussed.