Between the late 1940s and 1994, groundwater levels in the Warren subbasin, California, declined by as much as 300 feet because pumping exceeded sparse natural recharge. In response, the local water district, Hi-Desert Water District, implemented an artificial-recharge program in early 1995 using imported water from the California State Water Project. Subsequently, the water table rose by as much as 250 feet; however, a study done by the U.S. Geological Survey found that the rising water table entrained high-nitrate septic effluent, which caused nitrate (as nitrogen) concentrations in some wells to increase to more than the U.S. Environmental Protection Agency maximum contaminant level of 10 milligrams per liter..
A new artificial-recharge site (site 3) was constructed in 2006 and this study, which started in 2004, was done to address concerns about the possible migration of nitrates in the unsaturated zone. The objectives of this study were to: (1) characterize the hydraulic, chemical, and microbiological properties of the unsaturated zone; (2) monitor changes in water levels and water quality in response to the artificial-recharge program at site 3; (3) determine if nitrates from septic effluent infiltrated through the unsaturated zone to the water table; (4) determine the potential for nitrates within the unsaturated zone to mobilize and contaminate the groundwater as the water table rises in response to artificial recharge; and (5) determine the presence and amount of dissolved organic carbon because of its potential to react with disinfection byproducts during the treatment of water for public use. Two monitoring sites were installed and instrumented with heat-dissipation probes, advanced tensiometers, suction-cup lysimeters, and wells so that the arrival and effects of recharging water from the State Water Project through the 250 to 425 foot-thick unsaturated zone and groundwater system could be closely observed. Monitoring site YVUZ-1 was located between two recharge ponds in the middle of site 3, and YVUZ-2 was located approximately 1,200 feet down-gradient and to the southeast in an area where septic systems have been in use since about 1960. Site YVUZ-3 only went to a depth of 42 feet and was used to sample the upper part of the unsaturated zone near a golf course. Prior to the start of artificial recharge at site 3, nitrate concentrations reported as nitrogen from the soil leachate below YVUZ-1 did not exceed 1.58 milligrams per kilogram. Nitrate-reducing bacteria concentrations of 4,300 most probable number were found at about 220 feet below land surface and at the top of the water table at YVUZ-1. Nitrate concentrations at YVUZ-2 reached a maximum concentration of about 25 milligrams per kilogram between about 100 and 121 feet below land surface; concentrations of nitrate-reducing or denitrifying bacteria were as high as 21,000 most probable number at 36 feet below land surface but did not exceed 40 most probable number below about 150 feet below land surface. Between June 2006 and September 2009, more than 9,800 acre feet of water from the State Water Project was released to site 3 ponds. The infiltration of the recharge water was predominantly vertical with limited lateral spreading to a depth of about 200 feet below land surface at YVUZ-1. Lateral spreading of the recharge water with depth was caused by geologic heterogeneities within the unsaturated zone, and resulted in varied arrival times of the recharge water to the instruments and slower rates of vertical movement with depth. No abrupt changes in soil moisture were observed at YVUZ-2, indicating that the recharge water had not reached that site by September 2009. Water levels from the monitoring wells at both sites and from five production wells nearby showed that the water table rose at a mean rate of about 0.08 feet per day between June 2006 and January 2009. The arrival of the water from the State Water Project caused relatively rapid changes in the stable-isotopic ratios from the lysimeters at YVUZ-1. The estimated average rate of infiltration of the recharge water through the unsaturated zone ranged from 3.7 to 25 feet per day. The recharge water arrived at the monitoring well below the recharge ponds between August 2007 and March 2008; the rate of vertical movement to the monitoring well was between 0.6 and 0.9 feet per day. By September 2008, a production well located 375 feet west of site 3 was producing almost 100 percent infiltrated recharge water. By contrast, the stable-isotope data from the lysimeters at YVUZ-2 showed that the recharge water had not reached this site by September 2009, but that septic effluent in the unsaturated zone likely had mixed with the native pore water to at least 154 feet below land surface. Assuming vertical infiltration, the minimum rate of infiltration of septic effluent since 1960 was about 3 feet per year. The isotopic data from the lysimeters at YVUZ-3 indicated two different sources of water to the upper 43 feet–irrigation-return flow and precipitation. Nitrate concentrations of the water from the State Water Project did not exceed 1 milligram per liter. Prior to artificial recharge, nitrate concentrations of the pore water at YVUZ-1 ranged between 6 to 18.2 milligrams per liter. After the arrival of the recharge water, the nitrate concentrations from the lysimeters and well at YVUZ-1 decreased to less than 1 milligram per liter, with the exception of samples collected at 205.5 feet, which did not exceed 4.12 milligrams per liter. The decrease in nitrate concentrations after artificial recharge indicated that the rising water table did not result in an increase of nitrates below YVUZ-1. At YVUZ-2, nitrate concentrations ranged between 12 to 479 milligrams per liter. The highest nitrate concentrations were at 92 feet below land surface and were almost seven times that of samples collected from a nearby septic tank. Nitrate concentrations from the lysimeter at 273 feet below land surface increased from 6 to almost 58 milligrams per liter after it was saturated by the rising water table in December 2007. These increases could be the result of the mobilization of high-nitrate water from regional sources of septic effluent after saturation, or the result of high-nitrate water present at the top of the water table that may be diluted deeper in the aquifer. Nitrate concentrations in groundwater from five nearby production wells and from both monitoring wells were less than 5 milligrams per liter before artificial recharge started. Nitrate concentrations decreased to less than 3 milligrams per liter in three of the production wells and the monitoring well below the recharge ponds after artificial recharge. Dissolved organic carbon concentrations were measured in the recharge water and groundwater because of the potential for dissolved organic carbon to react with chlorine to form trihalomethanes during the water-treatment process. The dissolved organic carbon concentrations of the recharge water were 3.1 milligrams per liter or less, and dissolved organic carbon concentrations of the groundwater were less than 1 milligram per liter. Even though recharge water was present in some of the wells by September 2008, the concentrations of both dissolved organic carbon and trihalomethane formation potential in the groundwater did not increase. Interpretation of these data suggests that the dissolved organic carbon from the recharge water is altered or metabolized in the unsaturated zone, either by absorption to the grain particles in the soil or by microbiological processes.
Additional publication details
USGS Numbered Series
The effects of artificial recharge on groundwater levels and water quality in the west hydrogeologic unit of the Warren subbasin, San Bernardino County, California