California Water Science Center

California Water Conditions

Real-Time California Streamflow Conditions

Analysis of the Subsidence and Water-Management Issues of the Santa Clara Valley, California

Problem: The Santa Clara Valley is a long narrow trough extending about 90 miles southeast from San Francisco Bay. The area has seen extensive ground-water development. In the first half of this century, the valley was intensively cultivated for fruit and truck crops. Subsequent development has included urbanization and industrialization and the area is now commonly known as "Silcon Valley".

Project Area
Study Area

From the early 1900's through the late 1960's water-level declines from ground-water pumpage induced regional subsidence in the Santa Clara Valley. Prior to the importation of surface water and the artificial recharge of this imported source, water-level declines of more than 200 ft. occurred in the Santa Clara Valley from the early 1900's to the mid 1960's (Poland and Ireland, 1988). Currently the water purveyors in the Santa Clara Valley in conjunction with Santa Clara Valley Water District would like to meet the ground water demands on the basin, yet are trying to limit subsidence to less than 0.01 feet per year. This constraint on subsidence may be too restrictive and requires additional evaluation on the basis of new subsidence data. Initial inspection of extensometer data indicates about 0.2 ft. of elastic deformation in the San Jose area and about 0.06 ft. of elastic deformation in the Sunnyvale area (F.S. Riley, USGS, oral commun., 1998). Similarly, preliminary inspection of interferograms from INSAR images indicate about 0.09 ft. of seasonal deformation occurred in the San Jose area for the period October, 1995 to May, 1996 (Galloway, 1997a, b). These data along with recovered water levels collectively suggest that the aquifer systems have predominantly elastic deformation that is typical of regional alluvial aquifer systems (Hanson, 1987, 1989; Epstein, 1987; Anderson and Hanson, 1987; Ireland and others, 1984, Galloway, 1998).

A successful management strategy will require minimizing subsidence while maximizing a reliable water supply to meet growing demands from water users. Important regional subsidence concerns include regional compaction and subsidence that may cause damage to man-made structures, reduce long-term yield to wells, and permanently alter natural and man-made surface drainage networks that results in potential flooding or lack of drainage. Localized differential compaction and subsidence can also result in earth fissures that can damage man-made structures, surface drainage, and become conduits of pollution or allow poor-quality water to directly enter the main aquifer systems (Carpenter, 1991). To protect the quantity and quality of ground-water, the SCVWD operates a comprehensive water-management program that includes artificial recharge; operating an in-lieu replacement program in which imported water is provided to pumpers to supplant ground-water use; and promoting conservation. 

With the recovery of water levels in the Santa Clara Valley since the mid-1960's, Santa Clara Valley Water District (SCVWD) would like to meet ground-water demands during drought conditions while at the same time limiting land subsidence.  The problem remains that the existing subsidence and ground-water flow are very complex and are not completely understood and the existing monitoring and modeling tools need to be upgraded and updated to provide water managers with a mechanism to physically address the water-management issues. The improvement of the subsidence monitoring network has been addressed by the proposal submitted to SCVWD by the USGS in Fiscal Year 1998 (Galloway, 1997a, b). There remains a need to integrate the proposed new subsidence information into a regional ground-water flow and subsidence model. Several studies completed since the original USGS subsidence studies (Poland and Ireland, 1988) have attempted to address the issues of regional ground-water development and the related effects of subsidence. These include the District's modeling efforts (CH2-MHILL, 1992) and the existing GIS developed by SCVWD. However none of these studies are current or include the wide variety of geochemical, and geomechanical data needed to address the complex water-resources management issues facing SCVWD.

Objectives: The objectives of the proposed study are to develop a management tool that incorporates new geohydrologic and geomechanical data that helps further development of ground-water while minimizing of additional subsidence. This will require reviewing the hydraulic, chemical, and land-subsidence data that have been collected since the 1960's and integrating these data into the SCVWD regional flow model. The incorporation of INSAR technology, GPS/leveling networks, updated extensometer operation, as well as other new wellbore measurement techniques will allow considerable refinement in the spatial distribution of subsidence and it's driving force, ground-water pumpage. These data in combination with an upgraded and updated 3-D ground-water/ surface-water flow model would provide the tools to help answer the following questions:

(1) Is subsidence still occurring and, if so, where and how much?

(2) How much of the subsidence is elastic (recoverable) and how much is inelastic (nonrecoverable)?

(3) What are the sources, movement, and age of ground water in the Santa Clara Valley? What are the renewable resource for aquifer system and what portions of aquifer system is being mined?

(4) What are the most efficient strategies for controlling subsidence?

(5) How can pumpage be optimally distributed with depth and rate to minimize the potential for permanent subsidence?

(6) How can artificial-recharge facilities (aquifer storage and recovery) be optimally operated to maximize recharge, minimize overdraft pumpage, minimize additional loading from perched aquifers, and minimize spreading of poor quality waters?

(7) How will variability in climate (especially drought conditions) and the related availability of natural and artificial recharge affect the distribution of subsidence?

In summary: the objective of this study is to refine the groundwater management tool used by the Dsitrict that incorporates new geohydrologic data, new geomechnaical data, and new modeling techniques into the existing SCVWD regional flow model.

Relevance and Benefits: There are four levels of benefit from this project. First, the USGS will benefit in the assessment of applying geologic, seismic, geochemical, and geophysical data in a San Francisco Bay coastal watershed and by providing scientific leadership through integration of Geologic Division seismic hazard studies with Water Resources Division (WRD) hydrologic studies to develop a comprehensive geohydrologic framework of the Santa Clara Valley. The WRD will specifically benefit through integration of new technology such as INSAR and multiple-layer well package with the upgrade and update of the regional ground-water/surface-water flow during calibration and parameter estimation. 

Second, the USGS will continue to benefit from applying this type of data to a non-USGS MODFLOW model that is already being used to manage and assess water resources on a regional scale and that is linked to a Surface-Water Delivery Model that is used to help manage imported water and artificial recharge (fig. 2). This also will help provide an additional basis to enhance these models with other potential features such as climate forcings and GIS linkage that cannot be accomplished by the water purveyors or private industry alone. 
Third, the local water purveyors will benefit from the enhanced ability to make longer-term decisions with the help of regional models that better simulate results of the water-supply options and regional development problems with additional accuracy. 

Fouth, the Geologic Division may benefit from our experience in multidisciplinary approaches to water-resource assessment.

Approach: The subsidence-related water-resource management issues could be completed in four tasks. Because the approach taken in the later tasks of the study will depend greatly on results of the earlier tasks, it would be appropriate to have several junctures at which the USGS and the SCVWD would assess the course of the study. The study could be expanded, contracted, or re-directed at those junctures. The four tasks that would be completed are data compilation, new data collection, inverse model analysis, and analysis of model results:

Task (1) DATA COMPILATION: Compile existing geographic, geohydrologic, subsidence, and geochemical data into a computerized Geographic Information System (GIS). Data on well construction and ownership, ground-water pumpage, surface-water flow and importation, water-level, and water-quality monitoring data from SCVWD and by other agencies will be compiled into GIS data bases. INSAR and benchmark data collected by the SCVWD and USGS will be compiled into the GIS so that these data can be used in conjunction with extensometer and other well data. We also will compile and digitize selected geophysical logs and digitize the 15 years of chart records from the two operational extensometers. We will use these data to develop preliminary maps of such items as water levels, water quality, pumpage, well locations, and well perforations in order to begin our analysis of the regional geohydrology and subsidence. We also will use the GIS to identify needs for new data collection, evaluating where the need for additional information on the different aquifers is greatest.Water-level maps will be used to assess the state of the aquifer system and for model comparison. The existing ground- water flow model will also be imported into the GIS. In summary this task will:

- Compile the following data into GIS format compatible with the District's existing system that is not already in the SCVWD GIS -- geographic, geohydrologic, subsidence including INSAR and leveling surveys, geochemical, well construction and ownership, pumpage, surface-water flows, water levels, and water quality data.

- Compile and digitize selected geophysical logs and 15 years of extensometer data (extensometer data is currently being done under a different agreement).

- Develop preliminary water quality maps and well perforations. Maps have already been made on water levels, pumping and well locations.

Task (2) DATA COLLECTION: Collect water-quality samples from selected existing wells in the basins and analyze them for selected chemical constituents and isotopes to help identify the sources of water. The GIS will guide the data collection. We expect to take full advantage of the current SCVWD monitoring program and water quality data sets. At six selected production or monitoring wells, we will sample and analyze the water for major ions, nutrients, selected trace elements, the stable isotopes of oxygen and hydrogen, and selected radioactive isotopes including tritium and carbon-14. The isotope analyses will be useful for identifying and dating sources of ground-water recharge. In cooperation with the water retailers, depth-dependent water- quality sampling and flowmeter analysis of three selected production wells will be performed to identify the many water bearing strata that contribute flow to production wells. If the wellbore flow and water quality data provide useful data for better defining the main water-bearing zones, then more wells could be sampled. The wellbore flowmetering would also help to determine the vertical distribution of pumpage needed to upgrade and refine the model (Hanson and Nishikawa, 1992, 1996). SCVWD monitoring program results will then be correlated with the water-quality results from depth-dependent samples obtained from the three selected production wells. Additional chemical analyses can be completed by SCVWD laboraotories but could not be added to our data base without lab certificaton. An additional set of depth-dependent water-quality samples, that includes a bottom-sample age date, will be collected at one well per year for the remaining three years of the project. If more water-quality sampling is needed for subsequent years this can be estimated separately on the basis of the results of the proposed initial sampling and analysis and on the basis of the additional focus needed for any water-quality components of the water-management issues.

Sampling and analysis will help determine aquifers being pumped by existing wells and help refine estimates of the vertical distribution of pumpage from typical production wells. This will, in turn, provide fundamental information needed to upgrade the ground-water flow model.

To determnine the age of the water in the basin this task will collect and sample water from six production wells for major ions, nutrients, selected trace elements, stable isotopes of oxygen, hydrogen, and radioactive isotopes of tritium and carbon-14.

At three selected wells in the first year and one additional well in three subsequent years, water bearing strata will be determined and vertical pumpage distribution will be identified.

The USGS is proposing to complete high resolution sesimic imaging near the San Jose extensometer site. Recent results of our cooperatively funded study to test he viability of utilizing Intererometric Synthetic Aperature Radar (INSAR) to monitor land subsidence in the Santa Clara Valley, indicate an abrupt change in the seasonal land surface elevation near the extensometer site. The seismic imaging will help determine if this change is the result of lithologic or structural features.

The USGS is proposing to acquire seismic reflection and refraction profiles along a one-kilometer line oriented approximately east-west, that will pass near the extensometer site. Seismic sensors will be place every 5 meters to collect data from 1.5 to 350 meters (5 to 1,200 feet) below land surface. The seismic source will be shotgun shots at 18-inch depths every 5 meters along the line. The proposed seismic imaging would be collected in March. Collection of the data will require 5 trained USGS geophysicists and several technicians. The results of the seismic imaging will be summarized in an interpretive geologic cross-section, that will be prepared for review within twelve weeks of the collection of the data. An Open-file report will be completed in Federal Fiscal Year 1999 to summarize the acquisition and interpretation of the seismic imaging. The seismic imaging will be completed under the direction of Rufus Catchings of the Geologic Division of the USGS.

Task (3) INVERSE ANALYSIS: The existing regional model simulate ground-water flow using the USGS model, MODFLOW (McDonald and Harbaugh, 1988). The model employed six layers with simple aquifer properties, surface recharge, and pumpage.These models employed the period 1978-82 for a quasi-steady-state simulation and a transient simulation for the period 1970-83. This model (CH2M-HILL, 1992) simulated the upper confining layer explicitly (layer 2). A version of this model (Wilson, 1993) was utilized in study of pumping sequence and pattern and effects on subsidence and included the time-delayed simulation of compaction using the USGS Interbed Storage (IBS) package (S.A. Leake, USGS, written commun, 1991) but the surface recharge was simplified to constant head cells throughout the top model layer (layer 1) above the confining/perching fine-grained layer (layer 2). In addition all interbeds were assumed to be normally consolidated for the beginning of the simulation period 1978-82. Pumping also was grouped over one to four-square mile regions, vertically distributed equally between the lower aquifer layers (layers 3-5), and uniformly distributed over annual periods. This version of model did not calibrate well.

The USGS will upgrade and update the existing MODFLOW ground-water flow model for simulation of subsidence and water-resources management. The upgrades will be implemented on the existing model constructed for SCVWD by CH2M-HILL (1992). The model will incorporate all the new data and information from the first phase of the study. The upgrade will also include recent enhancements made to MODFLOW (Harbaugh and McDonald, 1996) and additional simulation packages to simulate streamflow routing (Prudic, 1989), variably saturated and rewetted cells (McDonald and others, 1991), faults as horizontal flow barriers (Hsieh and Freckleton, 1993), wells pumping from multiple aquifers (McDonald, 1984), and well, streamflow and subsidence hydrograph comparisons (Hanson and Leake, 1998). The model will also include the USGS subsidence package (Leake and Prudic, 1991). The model will simulate ground-water flow throughout the Santa Clara Valley; and will attempt to include the ability to simulate streamflow and canal routing, climatically variable recharge and streamflow (Hanson and Dettinger, 1996), rewetting and development of a perched ground- water system, artificial recharge, and subsidence. This allows us to evaluate how changes in one subarea (such as the Santa Theresa subbasin)--increased artificial recharge or increased pumpage, for example--may affect conditions in adjacent subareas. Simulation would descretize time on a seasonal basis that would represent a significant period of historical ground-water development. Seasonal stress periods also will facilitate the testing of water-resources management for supply and demand operations that occur in different seasons of the year.

The implementation of the new concepts, new packages, and additional historical period (1989-1999) will require some recalibration of the model. Recent work has demonstrated that inverse modeling provides capabilities that help modelers maximize the use of numerical models and model data especially in cases where the simulated system is very complex as in the case of the Santa Clara Valley Basin. Inverse techniques can be used to determine uncertainty, sensitivity, and relative importance of conceptual components that are simulated (Yeh, 1986; Hill, 1998). The benefits of using inversing modeling include the following:

- Determining parameter values that produce the best possible fit to the available observations.
- Performing diagnostic statistics that quantify the quality of the calibration and data shortcomings and additional data needs.
- Inferential statistics to quantify the reliability of parameter estimates and predictions.
- Assessing model-parameter values, uncertainties, and sensitivities
- Identify the significance of alternate models that include additional features or significantly altered model components

In summary this task will use current USGS inverse modeling programs MODFLOWP and UCODE (Hill, 1992,1994; Poeter and Hill, 1998) to assist with calibration, sensitivity and uncertainty analysis, and with conceptual model descrimination analysis.

The displacement maps produced from the ongoing INSAR studies (Galloway, 1997a, b) will be used to constrain the simulation results and model parameters in the elastic range. Hydrographs from wells, streamflow gaging stations, and benchmarks will also be employed for comparison and calibration of the revised model. The estimation of preconsolidation heads will also be attempted to determine the amount of water- level decline available prior to inelastic subsidence.

Simulation of historical and future water-supply operations and the demand for ground water are additionally impacted by variability in climate. Evaluation of the management of the water resources should include the long-term assessment of variable climate. Variation of climate drives artificial and natural recharge, availability of imported water, and indirectly drives subsidence through pumpage. Assessment of the historical climate variations will form the basis for partitioning historical recharge and streamflow data used to recalibrate and upgrade the simulation of historical conditions. In summary this task will:

- Update the existing groundwater model with modules that will simulate streamflow routing, variably saturated and re-wetted cells, faults as horizontal flow barriers if applicable, pumping from multiple aquifers, and a subsidence package that will incorporate the information developed in tasks 1 and 2.

- Recalibrate the model with these modifications incorporated and incorporating the data from 1989 to present.

- Compare the new module to calculate subsidence, inter-bed storage, and the District's updated methodology

Task (4) ANALYSIS OF MODEL RESULTS: Refinement of such a simulation model will allow rigorous testing of concepts about the three-dimensional movement of ground water, the hydraulic properties of the different aquifers, the effect of recharge, the potential for land subsidence, and the effect of boundary conditions such as faults, recharge, and confining layers. Evaluation of hydrologic budgets would be the final synthesis of the analysis of the updated and upgraded ground-water flow and subsidence model. This task allows both managers, scientists, and engineers to understand the importance and relative magnitude of volumetric flows between subareas, aquifers, or from specific water-supply and demand components that comprise the hydrologic system within it's current state of development. Previous studies have indicated that visualization of the flows and effects, such as subsidence and water- level changes, are an important part of describing and understanding the various facets of the water-resources management. This can be as critical for the general public served by SCVWD as it is for managers and regulatory agencies. The visualization of hydrologic features would be accomplished through the use of the GIS and existing ARCVIEW extensions that are currently available.

In summary this task will:

- Test concepts about groundwater movement within the basin and identify the areas that are being mined and the areas of the basin that have renewable water supply.

- Display graphically the movement of water within the groundwater basin.

Progress and Significant Results in FY 1999: The project is in the intial phases and has completed the intial water chemistry sampling and compilation of modeling data and software.

Plans for FY 2000: The study will develop new techniques for simulating pumpage from multiple aquifers. The study also is combining existing data into a current assessment of simulation results, geohydrology, surface-water and ground-water hydrology and geographic information system database.

Reports: Two USGS reports will result from this study. The first report would describe the collection and compilation of data. The second report would describe the upgraded and updated model, simulation results. This report will also describe the evaluation of calibration, upgrades, and updates from inverse model results. It is likely that several journal articles also will be prepared on specific aspects of the study findings. The visualization and hydrologic budget tasks will be an integral part of these publications as well as GIS presentations.

Number: CA554
Cooperating Agencies: Santa Clara Valley Water District
Project Chief: Randall T. Hanson
Period of Project: April 1999 through September 2002

Accessibility FOIA Privacy Policies and Notices logo U.S. Department of the Interior | U.S. Geological Survey
For Page Information: Send Us a Message
Page Last Modified: Thursday, 05-Jan-2012 15:41:59 EST