Baldwin Hills Dam disaster
The Baldwin Hills Dam disaster occurred on December 14, 1963, in the Baldwin Hills neighborhood of South Los Angeles when the dam containing the Baldwin Hills Reservoir suffered a catastrophic failure and flooded the residential neighborhoods surrounding it. It began with signs of lining failure, followed by increasingly serious leakage through the dam at its East abutment. After three hours the dam breached, with a total release of, resulting in five deaths and the destruction of 277 homes. Vigorous rescue efforts averted a greater loss of life.
The reservoir was constructed on a low hilltop between 1947 and 1951 by the Los Angeles Department of Water and Power directly on an active fault line, which was subsidiary to the well known nearby Newport–Inglewood Fault. The underlying geologic strata were considered unstable for a reservoir, and the design called for a compacted soil lining meant to prevent seepage into the foundation. The fault lines were considered during planning but were deemed by some, although not all, of the engineers and geologists involved as not significant.
The former reservoir is now part of the Kenneth Hahn State Recreation Area.
Significance and diagnoses of the failure
The failure of the Baldwin Hills Reservoir received an exceptional amount of attention from the civil engineering community and remains the subject of continuing interest. The reservoir had been conceived, designed, and built during and after World War II, a time when the pace of dam building was accelerating even as some disastrous dam failures were occurring, indicating a need for safer technologies. The builders of the Baldwin Hills dam, the Los Angeles Department of Water and Power, were aware of the difficult geologic conditions presented by the site and knew from past experiences, notably the catastrophic failure of the St. Francis Dam in 1928 in which over 400 people lost their lives, the serious consequences of a failure, even of a small reservoir in an urban setting. While dams were recognized as potentially dangerous, like nuclear technologies, they were also considered by Americans as a showcase technology—a means of fending off danger and spreading progressive American technologies and associated social benefits at home and abroad.The Baldwin Hills dam designer, engineer Ralph Proctor, had also worked as an assistant civil engineer for the Los Angeles Department of Water and Power on the failed St. Francis Dam and had subsequently devised new methods of producing compacted earth fill in building its replacement. Proctor aggressively proceeded with the Baldwin Hills project even in the face of safety concerns and disagreements over important design details raised within his own department.
Late 1963, when the Baldwin Hills failure occurred, coincidentally also happened to be the time of another notable public disaster. Only two months before, at the Vajont Dam in Italy, a massive landslide into the reservoir created a seiche which overtopped the dam, thereby flooding the valley below and causing the deaths of approximately 2000 people.
The Baldwin Hills Reservoir had been built, as were others, to assure an ample supply of safe water for the people of Los Angeles in case of a catastrophe such as an earthquake, fire, or war, and its failure was a blow to engineering confidence and the subject of many writings and two professional conferences. The failure occurred shortly after the death of the authoritative Harvard engineer Karl Terzaghi whose ideas had long dominated both earth dam engineering and the engineering science of soil mechanics; Terzaghi had also made significant contributions to understanding subsidence in oilfields. This left the assessment of the Baldwin Hills failure in the hands of a new generation of engineers, some who took on conflicting roles as experts in various lawsuits.
The design and construction of the dam had been inspected and approved by the California Department of Water Resources. A meticulously documented study published by that agency in 1964—while pointing out various connections between oilfield operations in the Inglewood Oil Field and ground disturbances in the area, including beneath the reservoir and at some distance from the reservoir—concluded rather vaguely that the failure was due to "an unfortunate combination of physical factors".
The monetary damages resulting from the failure were large, and some of the investigations which followed the state study were sponsored by litigants seeking more specific conclusions relevant to legal liability. This drew attention to oilfield operations in the area. From the outset it was clear that the ground faulting and fault creep which destroyed the reservoir were probably related to the many feet of ground subsidence which had occurred a half mile west of the reservoir over decades of oil extraction in the Inglewood field. The oilfield-related subsidence in the Inglewood field, though generally denied by the oil companies as a legal policy, was documented exhaustively by the US Geological Survey in 1969. Subsidence following oil extraction from shallow deposits in unconsolidated sediments had been understood by oil industry experts since the 1920s.
Following the discovery in 1970 by geologist Douglas Hamilton of faulting and surface seepage of oilfield waste brines along the fault which traversed and extended south of the reservoir, Hamilton and Meehan concluded that oilfield injection for waste disposal and improved recovery of oil, a new technology at the time, was a significant cause of the failure, triggering hydraulic fracturing and aggravating movements on a fault traversing the reservoir even on the day of the failure. Subsequently, the US Geological Survey concluded in 1976 that displacements at the ground surface causing reservoir failure and also ground cracking in the Stocker-LaBrea area southeast of the reservoir were 90 percent or more attributable to exploitation of the Inglewood oil field, and that this faulting was likely aggravated by waterflooding with pressures exceeding hydraulic fracturing levels.
By 1972, nearly a decade after the failure, the immediate legal issues had been settled out of court and the matter was reopened as a topic of discussion among investigators in a published engineering conference at Purdue University.
Engineer Thomas Leps, who had served as consultant on the 1964 state investigation, took on a role as neutral reviewer in this and most subsequent American studies of the failure. Leps concluded that there had been about 7 inches of offset on the fault beneath the reservoir during its life, about 2 inches of which had occurred in the months just before the failure. Leps associated the latter with repressurization of the oilfield. This, along with stretching of the ground due to subsidence of about 12 feet from oil extraction, had caused the lining failure which doomed the reservoir.
Some prominent consultants including those on a team led by Arthur Casagrande, Harvard successor to Karl Terzaghi, held that oilfield operations were not a significant influence at all but that the failure was the result of defective siting and design with the heavy weight of the dam and reservoir being the significant cause of the fatal foundation movement. This view exonerated the oil companies, namely Standard Oil, which had sponsored the study. Casagrande refused to acknowledge any ground movements in the area as being related to oilfield operations and argued that ground movements that affected the dam were found only beneath the reservoir, not in adjoining areas.
Most of these questions were examined once again in 1986 following investigations of a suspiciously similar major failure of the Bureau of Reclamation's Teton Dam in June 1976, and a near failure of the Department of Water and Power's Lower Van Norman Dam in the 1971 San Fernando earthquake. Professor Ronald Scott of Caltech, who had participated in the Casagrande studies, noted at a follow-up 1987 conference on Baldwin Hills that Casagrande had ignored or been unaware of ground movements clearly unrelated to the reservoir in his analysis. Another engineer, Stanley Wilson—who had also worked with Casagrande on the 1972 studies and supported the claim that oilfield subsidence was an insignificant cause—now conceded that analogous ground offsets extended well outside the reservoir area, notably in the Stocker-LaBrea area, so that the reservoir and other fault movements could not be attributed to the reservoir itself—thus tacitly attributing responsibility for the failure to oilfield operations. Hence, there appeared to be convergence of opinion on the role of oilfield subsidence and repressurization.
The issue of oilfield causation was a central theme in most of these discussions, with little attention having been directed to the details of the failure. The absolute necessity of a lining for this site was generally taken for granted in these proceedings even as it had been by Proctor himself, regardless of the fact that almost all earth dams perform satisfactorily without linings. Some suggestions as to possible preventive design and construction techniques that might have made the dam safer were raised to engineering consensus and reached a state of textbook knowledge in the late 1980s. For example, the character of the compacted earth lining was raised, if obliquely, in the suggestion made in the end that improved performance might have come from the use of a different lining material.
In 2001, a new angle on failure analysis was introduced by Mahunthan and Schofield, who concluded that overcompaction of the dam fill and lining was a significant aggravating factor in both the Baldwin Hills and Teton failures. This assertion was based on Schofield's concepts of critical-state soil mechanics, a corollary of which was that heavily compacted but lightly confined soils could be dangerously unstable where seepage forces were present. This issue had not been raised in the previous American-dominated discussions and remains in some degree contrary to American ideas in both theoretical soil mechanics and practical geotechnical engineering. In fact, the 1964 DWR failure study implied that heavy compaction was a favored technique for earth dam construction, and this assumption appeared not to have been reexamined over the twenty five years of post-failure investigation and discussion.
The failure of the reservoir has been a subject of ongoing interest in the field of dam-breach studies. A recent study examined the dam failure as a two-stage process and succeeded in modeling the flood in the urban area downstream.
Although the Baldwin Hills Reservoir site has now been dedicated as a community park, and there is no further significant hazard associated with ground movements there, the associated faults to the southeast continue to move significantly as of 2012, causing damage to private and public facilities. The current oilfield operator, Plains Exploration and Production Company, which has intensified production and development efforts in the oilfield with the rising price of petroleum, does not, unlike its predecessor Standard Oil, acknowledge any causal connection between fault movements and oilfield activities, and has retained a team of consultants who support this position or conclude that the causes of the movements are unknown.
The role of shallow hydraulic fracturing, which has recently been introduced as a means of stimulating production at depths of about in the southeast part of the Inglewood field, and at greater depths elsewhere in the field, has also generated public concern and controversy. However, oil operators, while admitting that fracture pressures are being exceeded, refuse to acknowledge a relationship between injection at fracturing pressure levels and fault movement. The PXP and PXP consultant conclusions, that adverse effects are either unknown or not present, are disputed by other reviewers.
Recent discharges of oilfield gases in the Baldwin Hills may also be related to raised pressures resulting from injection, and may be of similar origin as the gas problems in the nearby Salt Lake field.