San Joaquin-Sacramento Basin Rain Flood Duration Skew Study

Science Center Objects

The U.S. Army Corps of Engineers (USACE) is conducting a hydrologic analysis of floodplain areas protected by the federal-state levee system within the San Joaquin-Sacramento drainage basins. 

Map of California showing Army Corps of Engineer sites

Map of California showing Army Corps of Engineer sites (Public domain.)

This study is funded by the California Department of Water Resources (DWR) as part of FloodSAFE California, their flood-hazard mitigation and floodplain mapping effort throughout California. As part of the their hydrologic analysis, USACE will require rain flood frequency curves for unregulated flow conditions at key stream sites for durations of 1-day, 3-days, 7-days, and 15-days. 

Many of the stream sites for which USACE requires flood hydrology information are larger streams whose flows are partly or heavily regulated. To enable flood-frequency analyses on these regulated-flow stream sites, USACE intends to synthesize natural (unregulated) daily discharges. USACE will inspect the daily records, and, on a year-by-year basis, extract the rain flood events from the series. Using the resulting rain flood-only series and the rain-only annual maximums, USACE will compute the annual maximum 1-day, 3-day, 7-day, and 15-day discharges for each water year.

USACE will develop its rain flood flow-frequency curves following the frequency analysis procedures detailed in the in Bulletin 17B: Guidelines for Determining Flood Flow Frequency (Interagency Advisory Committee on Water Data, 1982) and in EM 1110-2-1417. The procedure requires that the adopted skew value for each frequency curve be based on the at-site skew and appropriate regional skew value. 1-day, 3-day, 7-day, and 15-day duration regional skew values are therefore needed, but do not exist or are antiquated. Thus, the development of new regional duration skews for locations within the DWR study area are needed. 

To this end, a cooperative project between the U.S. Geological Survey (USGS) and USACE will be pursued to implement a method to develop regional rain flood duration skew values for locations within the San Joaquin-Sacramento River Basins.

Specific objectives of the study are to provide duration regional skew values for rainfall flooding of 1-day, 3-day, 7-day, and 15-day durations within the San Joaquin-Sacramento River Basins. A discussion and hydrologic rationale for these values will also be provided, and the study will be published in a USGS Open-File Report. 

USACE will define the study region and provide USGS the boundary of the region and the locations for the 38 sites. USACE will compute for each location, for the period of record, the maximum 1-day, 3-day, 7-day, and 15-day duration flows. For each location, these values will be tabulated by water year and duration. At locations subject to reservoir regulation, changes in reservoir storage will be used by USACE to estimate the unregulated series. USACE will then extract from each unregulated series, for each water year, the annual maximum rain flood event. Annual maximum flows for durations of interest will be extracted from the rain flood season flow data. 

Task 1. USGS will develop GIS dataset from this information and provide the data to the USACE for review and conformation of locations. 

Task 2. USGS will work with Jery Stedinger, Cornell University, to develop a methodology to estimate the regional rainfall flood skew values and its precision for 1-day, 3-day, 7-day, and 15-day duration flows using key variables by extending statistical methods that have recently been developed for use with annual maximum series (Reis et al., 2005; Gruber and Stedinger, 2008; Gotvald et al., 2009; Parrett et al., in progress). This method will rely upon and be consistent with methods used in the current USGS statewide annual peak flood-frequency study (Parrett et al., in progress). The analysis will explore the full range of physiographic variables provided for all the stations. In the recent analysis of the skew for annual peaks in California, elevation turned out to be important; with elevation included in the model, other variables were not statistically significant. As this study is focused solely on rainfall flood events, elevation may drop out as an important explanatory variable for the duration skew equations. There will be an explicit attempt to use the same explanatory variables in the regional skew models for all durations so as to yield the best skew estimators and maintain the precision of those estimators across all temporal durations. This will facilitate the development of consistent frequency relationships. 

Task 3. USGS and Cornell will use this methodology to compute regional skew values for each basin, and will create a graphical and numerical tool for calculating region skew for other basins within the study region using the basin centroid and/or any other appropriate basin characteristics. 

Task 4. USACE, USGS, and Cornell University will participate in monthly conference calls to discuss progress and challenges. A draft Open File Report describing the methods used to develop the predictive relationship, a description of the relationship itself, and example computations will be provided for review by the USACE by September 1, 2010. USACE will provide comments within 15 days of receipt of the draft OFR, and USGS will provide an approved OFR by Nov 1, and the final report by Dec 1, 2010.