Monday, January 8, 2018

A Look Back at Ten Years of Geotripping: "Surely they didn't build it there?" The 2nd biggest disaster in California history

There is a part of California's history that is not celebrated. The failure that took place at St. Francis Dam on March 12, 1928 took more than 600 lives, but few people have ever heard of it. I discuss this event on the first day of all of my classes (which is tomorrow, by coincidence) because it is a classic story of what happens when major engineering projects do not seek or ignore the expertise of geologists. As far as I am concerned, the lesson is a larger one: there can be horrific consequences to society when politicians ignore the expertise of scientists as they grapple with the problems of climate change, water use, and the exploitation of mineral and energy resources.  

I've been going through the archives, looking for my favorite posts in ten years of geoblogging. This one was published on January 29, 2012.
Hindsight is harsh.

Sometimes choices and judgements are made to save time, to save money. Sometimes choices are made in unfortunate ignorance, in a time when no one could have foreseen or recognized the right choices to be made. Sometimes there is no one there to provide perspective, to provide alternatives. And then people die. Lots of people.

Ask folks what they think was the worst disaster in California history and many will get it right. Upwards of 3,000 people died in the 1906 earthquake in San Francisco, and the event has shaped the psyche and attitude of many people in the state more than a century afterward. And it was brought about by a natural event.

The second worst disaster in the history of the state is far less known. Some might guess another earthquake, like the Long Beach quake of 1933 (115 dead) or the Loma Prieta earthquake of 1989 (63 dead). Historians might point to the Port Chicago munitions explosion of 1944 (320 dead). Few people are aware that it was the collapse of a dam, and that the collapse was the result of many poor choices. Hindsight is a harsh judge, but many of the mistakes were "before their time" so to speak. The fact that it happened maybe has prevented worse disasters in the intervening years.

Time (and a great deal of government effort) has erased much of the record of our state's second worst disaster. As far as I could see there is not a single plaque or monument, either concerning the horrific event, or commemorating those who were lost. There is a small cemetery where some of the victims were buried.

Looking at the slide area on the left side of the picture above, it is hard to believe that a 200 foot high dam was anchored there, in the incompetent mica schist. It is hard to believe that the failed slopes in the picture obscure an even deeper and bigger megaslide.
It is hard to look at the flat ridge on the right side of the picture above and realize that no one ever thought to check the effect of soaking the seemingly solid conglomerate in water. It is glued together primarily with gypsum, a mineral that dissolves in water. The rock falls apart when saturated.

Maybe the most stunning realization is that the schist and the conglomerate are separated by a fault zone. An inactive fault by all appearances, but a fault nonetheless. They built the dam on a mega-landslide, and on a fault zone.
It is difficult to envision that on the night of March 12, 1928, the recently completed dam failed so catastrophically that the floodplain in the photos above and below was inundated with 140 feet of water flowing at a rate of 1.7 million cubic feet per second (California's biggest river, the Sacramento, averages 30,000 cfs, and the record flood on the river was 650,000 cfs).

What happened?

As Ron and Randy correctly surmised, Friday's mystery photo was about the destruction of the St. Francis Dam in 1928. I consider it one of the most important geological events ever to happen in the state, not because a great many people died, but because they died as a result of a disregard or lack of knowledge concerning human construction projects and the geological foundations on which they are built. Earthquakes and volcanic eruptions are inevitable geologic events, but the events of 1928 were completely avoidable.

In the early twentieth century, Los Angeles was at a crossroads. The city was growing fast, and the water needs of the metropolis far exceeded locally available supplies (according to city officials anyway). The story of how the city stole (legally stole, but stolen nonetheless) the water from underneath the people of the Owens Valley is a legend of California history. The fact that much of the water went to irrigation in the San Fernando Valley instead of the city just added to the scandal. Having completed the Owens Valley Aqueduct, one of the largest public waterworks ever conceived, the city needed someplace to store the water locally, especially in preparation for drought conditions. William Mulholland, the superintendent of the predecessor to the Los Angeles Department of Water and Power, oversaw the design and construction of a series of reservoirs around the Los Angeles Basin. Nine were constructed, and St. Francis Dam in San Francisquito Canyon above the Santa Clarita Valley was the largest, with a storage capacity of 38,000 acre feet. The dam itself was about 200 feet high, and just over 600 feet across. It was a concrete gravity-arch dam, one that depended on the nature of the rock in the abutments to maintain stability.

Construction was begun in 1924 and complete in 1926. During the construction Mulholland directed that the dam be made 20 feet higher than in the original plans, but he made no alterations at the base to compensate for the additional weight of the water. The filling of the dam took another two years, and was complete on March 7, 1928. On the morning of March 12, the dam keeper noted a leak of muddy water and alerted Mulholland. Small leaks of clear water from dams are usually expected; muddy leaks from a dam are very bad.  Mulholland declared that the mud was from some recent road construction and that the dam was safe. 12 hours later, the dam keeper was dead, the first victim of the collapse of the St. Francis Dam. In the hours that followed at least 600 more lives were lost.

To his credit, Mulholland took the blame for the disaster. Although he was never convicted of any crime in the matter, his career was over. He died seven years later.
Accounts at the time suggested that failure occurred as water channeled through the conglomerate along the fault contact. A reassessment of the failure by J. David Rogers finds multiple causes for the disaster, with the reactivation of the ancient landslide being the most important factor, along with hydraulic lifting of the dam which was caused by water pressing against the topmost part of the dam (which had been made higher without compensating at the base). Rogers lists many other deficiencies, including the weakness of the rocks in the dam abutments (I refer interested readers to this very fascinating pdf by Rogers that provides a blow-by-blow analysis of failure of the dam and a great deal of background information on the disaster).

Incredibly, despite the total evisceration of the dam, the central part remained standing, a 200 foot high monument to the destruction. After a sightseer fell off the top (his "friends" had tossed a rattlesnake at him), the city quarried holes in the base, filled them with five tons of dynamite, and blew up the remaining tower. Other blocks were also destroyed, as if they were trying to erase all memory of the event. One of the blocks was the "outcrop" I used in the Friday mystery photo.
The U.S. Geological Survey has a (much appreciated) photo archive from which I have gathered these photographs of the aftermath. In the photograph below, the fault line dividing the Vasquez Conglomerate from the Pelona Schist can be clearly seen (the lighter Pelona in the foreground, the dark Vasquez on upper ridge). The fault is inactive, and no earthquakes are implicated in the failure, but had the dam not failed, rising water pressure along the fault could conceivably have eventually caused renewed quake activity. The phenomenon has been noted elsewhere.
Blocks of concrete weighing thousands of tons were carried in the floodwaters nearly a half mile downstream. The magnitude of the disaster is hard to comprehend. Normal rivers have trouble moving boulders only a foot across. Besides the sheer magnitude of the flow, debris from the landslide buoyed up the blocks.
The block below was a half mile downstream. It measured approximately 63 feet long, 30 feet high, and 54 feet wide.
It is hard to find much that is positive in this disaster, but changes were made in the aftermath. The input of qualified engineering geologists became a requirement in dam-building, and much more attention was paid to the geological setting of reservoir sites. Boulder Dam on the Colorado River, one of the largest dams in existence is not in Boulder Canyon. Following the St. Francis disaster, the site of the dam was changed to Black Canyon when it was decided that the rocks that would anchor the dam were more stable there.

It would not be at all correct to say that we learned every possible lesson in dam construction. The 1963 tragedy at Vaiont Reservoir in Italy and the 1975 collapse of the Teton Reservoir, Idaho are vivid examples of unlearned lessons.

Hindsight is harsh. But it can be a teacher, too.


lyle said...

It should be noted that Mullholland was a completly self trained engineer with no college training. Civil engineering by that time was an old discipline, starting with West Point before the civil war. Of course in addition Mullholland was also the head of the water agency. and while apparently a man of vision, was unknowlegable on other things, such as adding 20 feet to the height of the dam but not making the base wider.

lyle said...

An additional comment history shows that civil engineers learn the most from failures in that they learn what not to do. Consider the Tacoma Narrows bridge failure the earlier Quebec bridge failure etc. Or the failure of the bridges in the KC Hayatt hotel.
On a side note the Pelona schist was the top layer of the subducted Farralon plate in southern Ca, so there is a 40 million year gap in the age rocks on the sides of that fault.

donna said...

Lots more at the wiki article:

Unknown said...

Ya got me to put up photos I took of the dam site when I visited with my son, then ten years old. Here it is: .