Integration of remotely-sensed and ground-based measurements to constrain simulations of groundwater flow and land subsidence, San Joaquin Valley, CA

Extensive groundwater withdrawal from the unconsolidated deposits in the San Joaquin Valley caused widespread aquifer-system compaction and resultant land subsidence between 1926 and 1970-locally exceeding 8 meters (m). To identify the extent of the subsidence, a monitoring network consisting of 31 extensometers was developed and maintained in the 1960s. The importation of surface water in the early 1970s resulted in decreased pumping and a steady recovery of water levels. This recovery reduced the rate of compaction, and consequently, the monitoring network deteriorated. However, lack of imported surface-water availability during 1976-77, 1986-92, and 2007-09 has caused groundwater pumping to increase, which has resulted in water levels to decline to near-historic lows and renewed compaction to occur. Land subsidence resulting from this compaction has reduced freeboard and flow capacity of the Delta-Mendota Canal, the California Aqueduct, and other canals that deliver irrigation water and transport floodwater, requiring expensive repairs. To identify existing and future subsidence, a monitoring network is being developed that includes resurrecting some of the extensometers and piezometers from the old network and augmenting these ground-based measurements with remotely-sensed measurements from Interferometric Synthetic Aperture Radar and continuous Global Positioning System stations. Preliminary results from the monitoring network indicate that subsidence is occurring in locations of known historical subsidence. These results are being used to develop groundwater-flow and subsidence models to help understand and effectively manage future subsidence. A 1-D model was developed to identify the depth of the compactable units near Oro Loma, where about 60 m of water-level decline is associated with about 3 m of subsidence. The subsurface geology is well-constrained by detailed descriptions of continuous core and by geophysical logs. Analysis indicates that the Corcoran Clay (CC) is about 27 m thick, which many think is the major compactable unit. The hydraulic parameters controlling subsidence-elastic and inelastic specific storage and vertical hydraulic conductivity-were constrained using results from consolidation tests. The model results indicate more than 90 percent of the total compaction is occurring below the CC in 18 clay layers ranging from about 0.03 to 11 m thick and totaling about 48 m, or about half of the aggregate thickness of all simulated aquitards. The model results also indicate that the CC has only compacted about 0.08 m owing to its low diffusivity and large thickness, and will ultimately compact in about 5,000 years. As the monitoring network becomes fully developed, high frequency measurements of compaction, land subsidence, and groundwater levels will enable inclusion of shorter-term elastic deformation in simulations. Results from the 1-D model, in terms of both model construction and hydraulic parameters, will be used to constrain the 3-D flow and subsidence models of the Central Valley. The measurements obtained from the monitoring network may ultimately provide managers with the subsidence information needed to manage the water conveyance systems, water-banking strategies, and other infrastructure.