USGS Home
Contact USGS
Search USGS

U.S. GEOLOGICAL SURVEY
Scientific Investigations Report 2007–5237

Back to Table of Contents

Summary and Conclusions

Ground-water is an important resource in the rural communities of southern Deschutes and northern Klamath County, near La Pine, Oregon. The primary aquifer, and only source of drinking water to about 14,000 residents, comprises alluvial sand and gravel deposits within 100 ft of land surface. Nearly 60 percent of residential lots are less than 1 acre and almost all homes use on-site wastewater disposal systems. Nitrate concentrations greater than the U.S. Environmental Protection Agency drinking water MCL of 10 mg/L were discovered in the oldest developed part of the area in the late 1970s and elevated concentrations have subsequently been detected in more recently developed areas. In 2000, nitrate concentrations greater than 4 mg N/L were detected in 10 percent of domestic wells sampled by Oregon Department of Environmental Quality. Because of concern for the vulnerability of the ground-water resource, the Oregon Department of Environmental Quality and Deschutes County, in cooperation with the U.S. Geological Survey, conducted a study to develop a better understanding of the hydrologic and chemical processes that affect the movement and fate of nitrogen within the shallow aquifers of the La Pine region. Simulation models were used to test the conceptual understanding of the system and were coupled with optimization methods to provide a management model that can be used to efficiently evaluate alternative approaches for managing nitrate loading from on-site wastewater systems.

The geologic, hydrologic, and geochemical frameworks for the conceptual and numerical models were developed using several data sources including previous hydrogeologic and water-quality studies in the area, an associated, large-scale field experiment evaluating advanced treatment on-site wastewater systems, literature for similar studies in other areas, and extensive field data collection for this study.

The primary aquifer in the study area is composed of complexly interbedded fluvial silt, sand, and gravel deposits. A three-dimensional hydrofacies model of the fluvial system was created with transition probability geostatistical methods using parameters derived from analysis of two-dimensional lithologic sections and lithologic data from more than 400 drillers’ logs. Five hydrofacies were included in the final model: clay-silt, sand, gravel, lacustrine clay-silt, and basalt. Mean annual ground-water recharge to the alluvial aquifer is 3.2 in/yr, primarily from infiltration of precipitation and snowmelt. Ground-water discharges to streams, springs, and wells, and by evapotranspiration. The water-table generally is within 5–20 ft of land surface and varies seasonally over a range of a few feet in response to recharge and changing stream stage.

On-site wastewater systems are the only significant source of anthropogenic nitrogen to shallow ground water in the study area. Low recharge rates and ground-water flow velocities have, for now, generally restricted nitrate occurrence to discrete plumes within 20–30 ft of the water table. Concentrations of nitrate typically are low in deeper, older ground water due to the nature and timing of nitrate loading and transport, and to loss by denitrification. Ground water in the study area evolves from oxic to increasingly reduced conditions with increasing depth below the water table. Suboxic conditions are achieved in 15–30 years, and the transition zone from oxic to suboxic ground water is narrow. Nitrate is denitrified near the oxic-suboxic boundary. Nitrate loading from residential, commercial, and other sources using on-site wastewater systems was estimated for 1960–99 using county building records, census data, monitoring data from field studies of on-site systems, and literature values. Adjusted for seasonal residency, residential loading estimates ranged from 12 to 14 lbs/yr per home between 1960 and 1999. During this period total nitrogen loading increased from 3,900 to 81,000 lb/yr. Nitrogen loading increased by 17,000 to 98,000 lb/yr between 1999 and 2005. When all approved lots are developed (projected to occur in 2019 at current building rates), nitrogen loading is estimated to reach nearly 150,000 lb/yr.

Three-dimensional numerical simulation models were constructed at transect (2.4 mi²) and study-area (247 mi²) scales to simulate the fate and transport of nitrate within the shallow ground-water system. The transect model was used to test conceptual models at the site of a detailed geochemical investigation along a 3.5-mi long flow path within the study area. The study-area model was constructed at a scale appropriate as a planning tool for prediction of average nitrate concentrations in neighborhoods and subdivisions.

Calibration of the models was constrained by data that included measured heads, measured and estimated ground-water discharge to streams, time-of-travel estimated from chlorofluorocarbon age dates, and measured ground-water nitrate concentrations. Eight scenarios representing nitrate-loading management strategies were simulated for the 140-year period, 2000–2139. A base scenario was simulated which assumed existing and future homes would continue to use conventional on-site systems and nitrogen loading would reach the projected maximum of nearly 150,000 lb/yr in 2019. Under this scenario, simulated nitrate concentrations continue to increase until the rate of nitrate loading to the aquifer system is balanced by nitrate losses to denitrification and ground-water discharge to the nearstream environment of the Deschutes and Little Deschutes Rivers. At equilibrium, average nitrate concentrations near the water table exceed 10 mg N/L over areas totaling 9,400 acres. Other scenarios were simulated that evaluated the effects of reduced loading on water quality. Scenarios in which nitrate loading was reduced by 15–94 percent overall resulted in reductions of 22–99 percent in the area where average nitrate concentrations near the water table exceed 10 mg N/L at equilibrium. Simulated ground-water ages agree with ground-water age data and show that the system is slow to respond to changes in nitrate loading due to low recharge rates and ground-water flow velocity. Consequently, reductions in nitrate loading will not immediately reduce average nitrate concentrations and the average concentration in the aquifer will continue to increase for 25–50 years depending on the amount and timing of loading reduction. The time required for average concentrations to decrease is, in part, also due to the assumption that replacement of existing on-site wastewater systems would take place over approximately 50 years.

Results of the scenario simulations showed that there is variable capacity to receive on-site wastewater system effluent. The capacity of an area to receive on-site wastewater system effluent is related to many factors, including the density of homes, presence of upgradient residential development, ground-water recharge rate, ground-water flow velocity, and thickness of the oxic part of the aquifer.

The study-area simulation model was used to develop a decision-support tool by incorporating optimization methods. The resulting model, the Nitrate Loading Management Model (NLMM), was formulated to minimize the reduction from estimated base scenario loading that would be needed to maintain ground-water nitrate concentrations or ground-water discharge of nitrate to streams below specified levels. The NLMM uses the response matrix approach to find the optimal (minimum) loading reductions in each of 97 management areas that will meet the specified water-quality constraints. The sensitivity of the optimal solutions (loading reductions) to water-quality and other constraints was evaluated by altering constraint values. The sensitivity of optimal solutions to constraints allows decision makers to assess tradeoffs between higher levels of water quality protection and the cost of reducing nitrate loading. The sensitivity analysis of the NLMM showed that optimal (minimum) loading reductions were most sensitive to the constraints on ground-water nitrate concentration in the shallow part of the oxic ground-water system, within 5–10 ft of the water table.

Back to Table of Contents