Abstract

The Marine Corps Air Ground Combat Center (MCAGCC) Twentynine Palms, California, overlies the Surprise Spring, Deadman, Mesquite, and Mainside subbasins of the Morongo groundwater basin in the southern Mojave Desert. Historically, the MCAGCC has relied on groundwater pumped from the Surprise Spring subbasin to provide all of its potable water supply. Groundwater pumpage in the Surprise Spring subbasin has caused groundwater levels in the subbasin to decline by as much as 190 feet (ft) from 1953 through 2007. Groundwater from the other subbasins contains relatively high concentrations of fluoride, arsenic, and (or) dissolved solids, making it unsuitable for potable uses without treatment. The potable groundwater supply in Surprise Spring subbasin is diminishing because of pumping-induced overdraft and because of more restrictive Federal drinking-water standards on arsenic concentrations. The U.S. Geological Survey, in cooperation with the MCAGCC, completed this study to better understand groundwater resources in the area and to help establish a long-term strategy for regional water-resource development.

The Surprise Spring, Deadman, Mesquite, and Mainside subbasins are filled with sedimentary deposits of Tertiary age, alluvial fan deposits of Quaternary-Tertiary age, and younger alluvial and playa deposits of Quaternary age. Combined, this sedimentary sequence reaches a maximum thickness of more than 16,000 ft in the Deadman and Mesquite subbasins. The sedimentary deposits of Tertiary age yield a small amount of water to wells, and this water commonly contains high concentrations of fluoride, arsenic, and dissolved solids. The alluvial fan deposits form the principal water-bearing unit in the study area and have a combined thickness of 250 to more than 1,000 ft. The younger alluvial and playa deposits are unsaturated throughout most of the study area. Lithologic and downhole geophysical logs were used to divide the Quaternary/ Tertiary alluvial fan deposits into two aquifers (referred to as the upper and the middle aquifers) and the Tertiary sedimentary deposits into a single aquifer (referred to as the lower aquifer). In general, wells perforated in the upper aquifer yield more water than wells perforated in the middle and lower aquifers. The study area is dominated by extensive faulting and moderate to intense folding that has displaced or deformed the pre-Tertiary basement complex as well as the overlying Tertiary and Quaternary deposits. Many of these faults act as barriers to the lateral movement of groundwater flow and form many of the boundaries of the groundwater subbasins.

The principal recharge to the study area is groundwater underflow across the western and southern boundaries that originates as runoff in the surrounding mountains. Groundwater discharges naturally from the study area as spring flow, as groundwater underflow to downstream basins, and as water vapor to the atmosphere by transpiration of phreatophytes and direct evaporation from moist soil. The annual volume of water that naturally recharged to or discharged from the groundwater flow system in the study area during predevelopment conditions was estimated to be 1,010 acre-feet per year (acre-ft/yr). About 90 percent of this recharge originated as runoff from the Little San Bernardino and the Pinto Mountains to the south, and the remainder originated as runoff from the San Bernardino Mountains to the west. Evapotranspiration by phreatophytes near Mesquite Lake (dry) was the primary form of predevelopment groundwater discharge. From 1953 through 2007, approximately 139,400 acre-feet (acre-ft) of groundwater was pumped by the MCAGCC from the Surprise Spring subbasin.

A regional-scale numerical groundwater flow model was developed using MODFLOW–2000 for the Surprise Spring, Deadman, Mesquite, and Mainside subbasins. The aquifer system was simulated by using three model layers representing the upper, middle, and lower aquifers. Measured groundwater levels for predevelopment conditions (before 1953) and for the period 1953 through 2007 were used to calibrate the groundwater- flow model. The simulated steady-state (predevelopment) recharge was about 980 acre-ft/yr; about 90 percent of the recharge was in the Mesquite subbasin. Most of the simulated steady-state discharge occurred as evapotranspiration at the Mesquite Lake (dry). A total of about 145,450 acre-ft of groundwater was simulated as being pumped from the model domain during the transient simulation period (1953–2007); about 139,400 acre-ft of the total was extracted from the Surprise Spring subbasin. The transient simulation indicates that almost all of the groundwater pumped in the Surprise Spring subbasin comes from groundwater storage, which is consistent with the measured long-term declines in groundwater levels.

the measured long-term declines in groundwater levels. The calibrated groundwater model was used to evaluate the potential effects on water levels and aquifer conditions in the Surprise Spring, Deadman, Mesquite, and Mainside subbasins for water-management strategies being considered by the MCAGCC to meet the projected water demand at the base for 2008–2017. One of the main objectives of the water-management strategies is to reduce pumpage from the Surprise Spring subbasin. Reducing groundwater pumpage in the Surprise Spring subbasin by about 38 percent (about 1,345 acre-ft/yr) substantially decreased or reversed simulated hydraulic-head declines in the subbasin. Redistributing about 15 percent of the 2007 groundwater pumpage (about 550 acre-ft/yr) from the Surprise Spring to the Mainside subbasin resulted in more than 60 ft of simulated declines in hydraulic head in the Mainside subbasin by 2017; however, redistributing about 22 percent of the 2007 pumpage (about 800 acre-ft/yr) from the Surprise Spring subbasin to the Deadman subbasin resulted in 5–10 ft of simulated hydraulic-head decline in the Deadman subbasin. The water-management scenarios simulated for this study demonstrate how the calibrated regional model can be utilized to evaluate the hydrologic effects of a water-management strategy.

First posted June 28, 2011