Skip Links

USGS - science for a changing world

U.S. Geological Survey Scientific Investigations Report 2010-5035

Prepared in cooperation with the State of Hawai‘i Department of Transportation

Flood-Frequency Estimates for Streams on Kaua‘i, O‘ahu, Moloka‘i, Maui, and Hawai‘i, State of Hawai‘i

By Delwyn S. Oki, Sarah N. Rosa, and Chiu W. Yeung


Thumbnail of and link to report PDF (8.4 MB)

This study provides an updated analysis of the magnitude and frequency of peak stream discharges in Hawai‘i. Annual peak-discharge data collected by the U.S. Geological Survey during and before water year 2008 (ending September 30, 2008) at stream-gaging stations were analyzed. The existing generalized-skew value for the State of Hawai‘i was retained, although three methods were used to evaluate whether an update was needed.

Regional regression equations were developed for peak discharges with 2-, 5-, 10-, 25-, 50-, 100-, and 500-year recurrence intervals for unregulated streams (those for which peak discharges are not affected to a large extent by upstream reservoirs, dams, diversions, or other structures) in areas with less than 20 percent combined medium- and high-intensity development on Kaua‘i, O‘ahu, Moloka‘i, Maui, and Hawai‘i. The generalized-least-squares (GLS) regression equations relate peak stream discharge to quantified basin characteristics (for example, drainage-basin area and mean annual rainfall) that were determined using geographic information system (GIS) methods.

Each of the islands of Kaua‘i,O‘ahu, Moloka‘i, Maui, and Hawai‘i was divided into two regions, generally corresponding to a wet region and a dry region. Unique peak-discharge regression equations were developed for each region. The regression equations developed for this study have standard errors of prediction ranging from 16 to 620 percent. Standard errors of prediction are greatest for regression equations developed for leeward Moloka‘i and southern Hawai‘i. In general, estimated 100-year peak discharges from this study are lower than those from previous studies, which may reflect the longer periods of record used in this study. Each regression equation is valid within the range of values of the explanatory variables used to develop the equation. The regression equations were developed using peak-discharge data from streams that are mainly unregulated, and they should not be used to estimate peak discharges in regulated streams. Use of a regression equation beyond its limits will produce peak-discharge estimates with unknown error and should therefore be avoided. Improved estimates of the magnitude and frequency of peak discharges in Hawai‘i will require continued operation of existing stream-gaging stations and operation of additional gaging stations for areas such as Moloka‘i and Hawai‘i, where limited stream-gaging data are available.

Last modified August 11, 2010
First posted June 21, 2010

  • This report is available only on the Web.

For additional information contact:
Center Director, Pacific Islands Water Science Center
U.S. Geological Survey
677 Ala Moana Blvd., Suite 415
Honolulu, HI 96813

Part of this report is presented in Portable Document Format (PDF); the latest version of Adobe Reader or similar software is required to view it. Download the latest version of Adobe Reader, free of charge.

Suggested citation:

Oki, D.S., Rosa, S.N., and Yeung, C.W., 2010, Flood-frequency estimates for streams on Kaua‘i, O‘ahu, Moloka‘i, Maui, and Hawai‘i, State of Hawai‘i: U.S. Geological Survey Scientific Investigations Report 2010-5035, 121 p.






Frequency Analysis at Gaged Sites

Regression Analysis

Application of Regression Equations

Comparison with Previous Estimates


References Cited

one Appendix

Accessibility FOIA Privacy Policies and Notices

Take Pride in America logo logo U.S. Department of the Interior | U.S. Geological Survey
URL: http:// /sir/2010/5035/index.html
Page Contact Information: USGS Publications Team
Page Last Modified: Thursday, 10-Jan-2013 19:12:29 EST