Las Vegas Valley is experiencing a problem familiar to residents of Venice, Mexico City, Shanghai, Houston, Long Beach and many other cities around the world -- subsidence due to fluid withdrawal. Scores of residents of North Las Vegas are being forced to move from their homes in the Windsor Park area at great personal and public expense after years of fruitlessly repairing twisted homes, schools, fences, roads, and pipelines. In response to tens of millions of dollars of direct and indirect damage, city planners, land developers and lending institutions join together in asking: When will the surface expression of subsurface deformation stop occurring? Where will it be focused next? Can it be controlled?
Public awareness of subsidence has been heightened in Las Vegas by articles in the local newspapers regarding subsidence-related development of fissures throughout the valley and the necessary moving of residents of Windsor Park to new homes. As mentioned above, many homes in Windsor Park have either been destroyed or are in the process of being destroyed by subsidence-related slow movement and slow rupture of the earth.
NBMG is responding to this gradually increasing disaster in southern Nevada. John Bell compiled and interpreted land subsidence data for Las Vegas Valley up to 1980 (NBMG Bulletin 95). Supported by a consortium of eleven federal, state, county, and city agencies, John Bell and Jonathan Price initiated and supervised a thorough updating of this database. This two-year study of the problem commenced in November 1989 and included a new geodetic survey of the valley, the addition of new vertical control lines across more faults, the mapping of fissures, and a tabulation of areas of structural damage. The results of this study have been reported by NBMG in Subsidence in Las Vegas Valley, 1980-91: Final Project Report (NBMG Open-File Report 93-4). Possible mitigation measures that appear in this final project report are also listed at the end of the present article. Contributors to this unpublished report are J. W. Bell, J. G. Price, M. D. Mifflin, O. A. Adenle, R. J. Johnson, A. R. Ramelli, D. C. Helm, E. M. Fini, S. R. Waichler, and G. F. Cochran. Much of the following discussion is based on Bulletin 95 and the final project report.
As reported previously in Nevada Geology (no. 3, Summer 1989), land subsidence in Las Vegas Valley is caused primarily by groundwater withdrawal. Pumpage from aquifers in the valley has exceeded the estimated natural recharge of between 25,000 and 35,000 acre-feet per year since the mid-1940s. Groundwater withdrawal reached a maximum annual rate of about 88,000 acre-feet in 1968. Since then the annual rate has gradually been reduced due to water being imported to the valley from Lake Mead and has reached an average of about 68,000 acre-feet per year. Starting in 1987, the Las Vegas Valley Water District initiated an artificial recharge program to store water temporarily in the principal aquifer during periods of lower water use, namely from October through April. The amount of water injected annually has been increasing; annual volume was 10,360 acre-feet in 1990 and exceeded 20,000 acre-feet last year.
One result of overdraft of groundwater is a continued decline of water levels and the corresponding reduction of artesian pressure throughout most of the valley. This, in turn, causes an increase in effective stress at depth which means that the portion of the total overburden weight that is carried by grain-to-grain contacts increases while the portion carried by the interstitial water within the pore spaces decreases. As effective stresses increase in the sediments from which water is withdrawn, the porous structure yields and compresses. Land surface subsides as the cumulative effect of this gradual compression at depth migrates upward.
The annual rate of subsidence since 1980 has remained close to what it was prior to 1980. Injection of groundwater during the winter months may slow the future rate of subsidence at some locations, but this possibility had not manifested itself in 1986-87, when subsidence measurements were made and the artificial recharge program was initiated. At a few locations some subsidence was recorded to have occurred before 1950. For example, the downtown area surrounding the Post Office had subsided about 6 inches. At these sites, subsidence has continued at a fairly steady rate during the ensuing years from 1950 through the most recent measurements in 1987. The elevations of enough benchmarks were measured in 1963 to form a base for a valley-wide map of subsequent cumulative subsidence. A map showing cumulative subsidence between 1963 and 1980 was printed in the summer 1989 issue of Nevada Geology (no. 3) and a map showing subsidence between 1963 and 1987 appears on page 2 of this issue. Since 1987 so many benchmarks have been destroyed due to development and widening of roads that future maps using this historical base will probably be impossible. For this reason two geodetic surveys (GPS) were run during 1990-91 in an attempt to establish a new base.
Maximum subsidence of greater than 5 feet has occurred between 1963 and 1987 near the intersection of Craig Road and Rancho Drive in the northwest part of the valley. A potential effect of future subsidence, if it continues, may be a shift in the location of flooding areas during a storm.
What is important to note on the map is the steep gradient of differential subsidence that occurs along and near the Eglington fault (where the 2 and 5 foot subsidence contours occur close together). Abrupt changes in land surface subsidence can cause more damage to structures than does a large area subsiding more uniformly. The rate of vertical subsidence (though slow) tends to be greater on one side of a fault than on the other side. This introduces a potential for tilt, or possibly even vertical shear, on structures and pipelines located nearby.
The role of faults has been clearly established as a focus for vertical subsidence in Las Vegas Valley, as mentioned above, and also for subsidence-related fissures. Long, narrow cracks originate at depth, migrate upwards and eventually intercept the land surface. Erosion during a rain storm carries sand and other sediments into the deep crack so that a line of potholes and gullies appear at land surface which are called fissures. This process should not be confused with desiccation cracks that can also occur in an arid environment. Subsidence-related fissures are associated with cracks caused by horizontal aquifer movement at depth whereas desiccation cracks are associated with the drying of near-surface sediments, such as clay. Fissures that are mapped as subsidence related tend to be located near preexisting faults in Las Vegas Valley. Forty-five percent of the fissure lengths are located within 500 feet of the nearest fault and 82 percent are within 1,200 feet.
Hence, the location of most future fissures will probably be in the neighborhood of a preexisting fault. The physical process that triggers a crack at depth to occur along such a preexisting plane of weakness is associated with aquifer movement in response to groundwater withdrawal. Geological heterogeneities and groundwater hydraulics are intimately interwoven in the fissuring process.
The question is: In what way are they interwoven? As mentioned above, abrupt changes in vertical movement and fissures occur along or near old fault scarps. Interestingly, however, if a groundwater discharge center lies on the upthrown side of a fault, the upthrown side is observed to move downward relative to the downthrown side --opposite to the past geologic direction of motion. Geologic structure appears to control where fissures occur (namely, where abrupt changes in horizontal motion occurs) and where abrupt vertical differential movement occurs whereas groundwater hydraulics control the direction, magnitude and timing of these events.
The question remains: Can a practical quantitative tool be developed which can successfully predict the time-dependent interplay of geology and groundwater hydraulics? This interdisciplinary question that combines geology, groundwater hydrology, soil mechanics, rock mechanics and seismology is at the cutting edge of science. NBMG is taking the lead in trying to answer it.
With the aid of graduate students within the University of Nevada System, Donald C. Helm of NBMG has been conducting fundamental research into this previously unexplored interdisciplinary area. Dr. Helm has been conducting subsidence related research worldwide and joined NBMG in 1989. He is a pioneer in having developed a successful and widely used computational tool for predicting vertical subsidence. To do the same for horizontal movement and slow vertical rupture of sediments is an even greater challenge. The long term goal is to develop a predictive tool that will tell (1) when and where subsidence and fissures will (and will not) occur and (2) the rate and magnitude of such an occurrence given a future pumping and injection scenario within Las Vegas Valley.
Since 1990, the thesis topics of eight graduate students have involved different aspects of this problem. Significant advances are being made in developing new conceptual and computational tools aimed at predicting three-dimensional time-dependent subsidence and include: seismic wave propagation and slow rupture criteria through saturated and unsaturated sedimentary material, aquifer matrix deformation coupled to groundwater flow, and hydrogeologic creep. More work certainly needs to be done. Already research centers are paying close attention to the scientific progress being made in Nevada.
Much of the student research has been supported by NBMG, the Las Vegas Valley Water District, the State Water Research Institute, and a US Geological Survey Merit Project through the Nevada District Office of the USGS Water Resources Division.
The final project report contains the following list of options for mitigating the subsidence hazard in Las Vegas Valley:
---Donald C. Helm, Research Hydrogeologist