Paper No. JAWRA-11-0020-P of the Journal of the American Water Resources Association (JAWRA). Discussions are open until six months from print publication.
Evaluation of a Regionalization Approach for Daily Flow Duration Curves in Central and Southern California Watersheds†
Article first published online: 10 OCT 2011
© 2011 American Water Resources Association
JAWRA Journal of the American Water Resources Association
Volume 48, Issue 1, pages 123–133, February 2012
How to Cite
Hope, A. and Bart, R. (2012), Evaluation of a Regionalization Approach for Daily Flow Duration Curves in Central and Southern California Watersheds. JAWRA Journal of the American Water Resources Association, 48: 123–133. doi: 10.1111/j.1752-1688.2011.00597.x
- Issue published online: 1 FEB 2012
- Article first published online: 10 OCT 2011
- Received February 14, 2011; accepted July 19, 2011.
- surface water hydrology;
- flow duration curve;
- theory of total probability
Hope, Allen and Ryan Bart, 2011. Evaluation of a Regionalization Approach for Daily Flow Duration Curves in Central and Southern California Watersheds. Journal of the American Water Resources Association (JAWRA) 48(1): 123-133. DOI:10.1111/j.1752-1688.2011.00597.x
Abstract: Discontinuities in flow regimes of ephemeral rivers in California make the modeling of daily flow duration curves (FDCs) in ungauged watersheds challenging. A regionalization approach previously developed for perennial and ephemeral watersheds in Portugal was tested in central and southern California. This approach, which is based on the theory of total probability, requires the prediction of three key flow variables in ungauged watersheds: the percentage of time the river is dry, the nonzero flow equalled or exceeded 80% of the time, and the mean daily flow for nonzero flows. Data from 41 watersheds in California were used to develop and validate regression equations for these three metrics. The methodology included an “all possible models” regression approach, an extensive set of watershed descriptors as potential independent variables, and two different methods for constructing observed FDCs. Suitable regression models could not be identified for predicting any of the required flow metrics. The contrasting results from the studies in Portugal and California were primarily attributed to differences in the aridity of watersheds in the two samples.