The city of Santa Barbara pumps most of its ground water from the Santa Barbara and Foothill ground-water basins. The Santa Barbara basin is subdivided into two storage units: Storage Unit I and Storage Unit III. The Foothill basin and Storage Unit I of the Santa Barbara basin have been studied extensively and ground-water flow models have been developed for them. In this report, the geohydrology of the Santa Barbara ground- water basin is described with a special emphasis on Storage Unit III in the southwestern part of the basin. The purposes of this study were to summarize and evaluate the geohydrology of Storage Unit III and to develop an areawide model of the Santa Barbara and Foothill basins that includes the previously unmodeled Storage Unit III.

Storage Unit III is in the southwestern part of the city of Santa Barbara. It is approximately 3.5 miles long and varies in width from about 2,000 feet in the southeast to 4,000 feet in the north-west. Storage Unit III is composed of the Santa Barbara Formation and overlying alluvium. The Santa Barbara Formation (the principal aquifer) consists of Pleistocene and Pliocene(?) unconsolidated marine sand, silt, and clay, and it has a maximum saturated thickness of about 160 feet. The alluvium that overlies the Santa Barbara Formation has a maximum saturated thickness of about 140 feet. The storage unit is bounded areally by faults and low-permeability deposits and is underlain by rocks of Tertiary age.

The main sources of recharge to Storage Unit III are seepage from Arroyo Burro and infiltration of precipitation. Most of the recharge occurs in the northwest part of the storage unit, and ground water flows toward the southeast along the unit's long axis. Lesser amounts of recharge may occur as subsurface flow from the Hope Ranch subbasin and as upwelling from the underlying Tertiary rocks. Discharge from Storage Unit III occurs as pumpage, flow to underground drains, underflow through alluvium in the vicinity of Arroyo Burro across the Lavigia Fault, evapotranspiration, and underflow to the Pacific Ocean. The faults that bound Storage Unit III generally are considered to be effective barriers to the flow of ground water. Interbasin ground-water flow occurs where deposits of younger alluvium along stream channels cross faults. Ground-water quality in Storage Unit III deposits varies with location and depth. Upward leakage of poor-quality water from the underlying Tertiary rocks occurs in the storage unit, and such leakage can be influenced by poor well construction or by heavy localized pumping. The highest dissolved-solids concentration (4,710 milligrams per liter) in ground water resulting from this upward leakage is found in the coastal part of the storage unit.

The ground-water system was modeled as two horizontal layers. In the Foothill basin and Storage Unit I the layers are separated by a confining bed. The upper layer represents the upper producing zone and the shallow zone near the coast. The lower layer represents the lower producing zone. In general, the faults in the study area were assumed to be no-flow boundaries, except for the offshore fault that forms the southeast boundary; the southeast boundary was simulated as a general-head boundary. The Storage Unit III model was combined with the preexisting Storage Unit I and Foothill basin models, using horizontal flow barriers, to form an areawide model.

The areawide model was calibrated by simulating steady-state predevelopment conditions and transient conditions for 1978-92. The nonpumping steady- state simulation was used to verify that the calibrated model yielded physically reasonable results for predevelopment conditions. The calibrated areawide model calculates water levels in Storage Unit III that are within 10 feet of measured water levels at all sites of comparison. In addition, the model adequately simulates water levels in the Storage Unit I and Foothill basin areas. A total of 33,430 acre-feet of water was pumped from the study area during the simulation period. Model results indicate that 2,833 acre-feet came from storage and 5, 332 acre-feet crossed the general-head boundary from the ocean, thus indicating that seawater intrusion could occur. A sensitivity analysis indicates that, in general, the model is most sensitive to changes in transmissivity and total recharge.