Investigators of the Feather River levee failure will be especially interested in comparing flood hydrographs for various past floods that have caused failures on the levees near Marysville and Yuba City.

Here we see a comparison of floods for 1955, 1986, and 1997. The levee-threatening rise in water shows a similar profile for each year. The time of levee failure is shown by the red dot.

Note that though the 1955 and 1997 breaks occurred while the water was still rising, the 1986 failure was some time after the floodwaters began to recede. This provides an important clue; namely, that the pressures causing failure lag behind the immediate flood pressures on the levee. In 1986 the failure was a day after the peak in flood, and many of the flood fighting efforts had been called off.

The levees were not overtopped. In each case, sand boils had been observed at the toe (or outside of the toe) of the levee before failure.

We can conclude from this that some time may elapse between pressure on the levee and pressure at the boil. This lag tiome can be analyzed using an engineering model.

Other links:

Lance Williams' San Francisco Examiner lead article on levee failures, January 12, 1997

Some geological history of the failure area, leading to prediction of failure hot spots.

Some information on sand boils.

A helicopter inspection on January 17 Feather River flood hydrography Why levees may fail after the flood peak

Comparison with Pajaro River valley

Questions or Comments?

meehan@blume.stanford.edu; or call Dick Meehan at (415) 323-0525