Geology topics

Esther M. Pischel

Publications and source records attributed to Esther M. Pischel.

3 recordsLinked to original sources

USGS2018

Deep aquifer recharge in the Columbia River Basalt Group, upper Umatilla River Basin, northeastern Oregon

Groundwater is an important component of the water resources of the upper Umatilla River Basin of northeastern Oregon. As such, understanding the capacity of the resource is vital. Past studies have estimated recharge in the study area. One recent study of the upper Umatilla River Basin indicated that about 80 percent of recharge entering the groundwater system is discharged to streams in the study area through shallow groundwater-flow paths, leaving about 20 percent of recharge to infiltrate deeper parts of the aquifer system. The purpose of this work is to quantify the spatial distribution and variability of deep aquifer recharge in the study area and to understand the reasons for a relatively low percentage of total recharge reaching the deeper parts of the groundwater-flow system. The study area is divided into two distinct physiographic regions—the highly dissected Blue Mountains and the lowland plains. Underlying both regions of the study area are basalts of the Columbia River Basalt Group (CRBG), which is the principal aquifer in the study area. Deep incision by streams in the Blue Mountains disrupts the lateral continuity of the CRBG aquifer units, and infiltrating water is more readily diverted laterally and discharged to streams and springs. In the lowland plains, incision is less pronounced. The shallow CRBG units might be disrupted, but deeper aquifer units retain their lateral continuity and enable groundwater to infiltrate deeper and flow laterally farther downgradient before discharging. Recharge to the deep basalt aquifers is estimated as the difference between total recharge and base flow. Total recharge is the portion of precipitation and applied irrigation water that infiltrates past the root zone to become groundwater recharge. Of this total recharge, a proportion discharges to springs and streams in the study area, and the remaining water infiltrates below the base level of streams and recharges the deep basalt aquifers and contributes to the regional groundwater flow system. The portion of total recharge that recharges the regional flow system is referred to as deep aquifer recharge. Total recharge is the portion of precipitation and applied irrigation water that infiltrates past the root zone to become groundwater recharge. It is the sum of recharge from precipitation and recharge from infiltration of irrigation water. Recharge from precipitation was calculated using a regression method developed for the Columbia Plateau. Recharge from infiltrating irrigation water was obtained from a water balance model developed for the Columbia Plateau. Base flow, the component of streamflow that represents groundwater discharge as opposed to runoff from the land surface, was estimated using the Base Flow Index Modified (BFI-Modified) method, an empirical hydrograph separation technique. Base flow was estimated in eight subbasins with streamgages within the study area. Five of the eight subbasins in which base flow was estimated had permitted water rights for irrigation that specified surface water as the primary source of water. Maximum surface-water withdrawal for irrigation was estimated for all subbasins in which water rights for irrigation occur. The base-flow estimate from BFI-Modified is assumed to be the minimum amount of base flow. The sum of the BFIModified base-flow estimate and the maximum permitted surface-water withdrawal estimate for each subbasin is assumed to be the maximum amount of base flow at the streamgage. These minimum and maximum estimates of base flow were used to calculate minimum and maximum values of deep aquifer recharge in each subbasin analyzed within the study area. Subbasin estimates were scaled up to the Blue Mountains and lowland plains regions, and to the entire study area. Mean annual total recharge for 1981–2010 in the subbasins, analyzed as part of this work, ranged from 6 inches (in.) in the Patawa and Wildhorse Creek subbasins in the lowland plains to as much as 20 in. in the Umatilla River above Meacham Creek subbasin. Mean annual total recharge totaled 4 in. in the lowland plains region and 14 in. in the Blue Mountains. Mean annual total recharge for the entire study area was 11 in. Mean annual base flow ranged from 1 in. in the Patawa and Wildhorse Creek subbasins in the lowland plains to as much as 14 in. in the Umatilla River above Meacham Creek subbasin in the Blue Mountains. Mean annual deep aquifer recharge ranged from 4 in. in the Patawa and Wildhorse Creek subbasins in the lowland plains to as much as 8 in. in the Isqu’ulktpe Creek subbasin in the Blue Mountains. Deep aquifer recharge was 3–4 in. in the lowland plains region and 6 in. in the Blue Mountains. Over the entire study area, mean annual deep aquifer recharge was 5 in. Most groundwater recharge (both total and deep aquifer) in the study area occurred in the Blue Mountains, which highlights the importance of the Blue Mountains as the principal source of groundwater for the study area and for aquifers farther downgradient. Total recharge in the Blue Mountains region represents 86 percent of the mean annual total recharge in the study area in an area that encompasses 65 percent of the study area. However, only 43–44 percent of the mean annual total recharge remains in the system to recharge the deeper, regional aquifer system because the rest is discharged as base flow within the Blue Mountains region. Within the lowland plains region of the study area, an estimated 67–84 percent of the mean annual total recharge remains in the system to recharge the deep, regional aquifer system. Although total recharge in the study area represents only 14 percent of the total recharge across the study area, it contributes 20–24 percent of the water to the deep aquifer. The difference in the percentage of deep groundwater recharge in the Blue Mountains and the lowland plains is attributed to differences in the degree of stream incision. Stream channels are more incised in the Blue Mountains region than they are in the lowland plains. The dissection of the landscape in the Blue Mountains disrupts the lateral continuity of the CRBG aquifer units and allows groundwater to discharge to springs and streams rather than infiltrate more deeply. In the lowland plains region, incision is much less pronounced and deeper CRBG units likely retain their lateral continuity, enabling groundwater to infiltrate more deeply than in the Blue Mountains.

Oregon↗

USGS2017

Simulation of groundwater and surface-water flow in the upper Deschutes Basin, Oregon

This report describes a hydrologic model for the upper Deschutes Basin in central Oregon developed using the U.S. Geological Survey (USGS) integrated Groundwater and Surface-Water Flow model (GSFLOW). The upper Deschutes Basin, which drains much of the eastern side of the Cascade Range in Oregon, is underlain by large areas of permeable volcanic rock. That permeability, in combination with the large annual precipitation at high elevations, results in a substantial regional aquifer system and a stream system that is heavily groundwater dominated. The upper Deschutes Basin is also an area of expanding population and increasing water demand for public supply and agriculture. Surface water was largely developed for agricultural use by the mid-20th century, and is closed to additional appropriations. Consequently, water users look to groundwater to satisfy the growing demand. The well‑documented connection between groundwater and the stream system, and the institutional and legal restrictions on streamflow depletion by wells, resulted in the Oregon Water Resources Department (OWRD) instituting a process whereby additional groundwater pumping can be permitted only if the effects to streams are mitigated, for example, by reducing permitted surface-water diversions. Implementing such a program requires understanding of the spatial and temporal distribution of effects to streams from groundwater pumping. A groundwater model developed in the early 2000s by the USGS and OWRD has been used to provide insights into the distribution of streamflow depletion by wells, but lacks spatial resolution in sensitive headwaters and spring areas. The integrated model developed for this project, based largely on the earlier model, has a much finer grid spacing allowing resolution of sensitive headwater streams and important spring areas, and simulates a more complete set of surface processes as well as runoff and groundwater flow. In addition, the integrated model includes improved representation of subsurface geology and explicitly simulates the effects of hydrologically important fault zones not included in the previous model. The upper Deschutes Basin GSFLOW model was calibrated using an iterative trial and error approach using measured water-level elevations (water levels) from 800 wells, 144 of which have time series of 10 or more measurements. Streamflow was calibrated using data from 21 gage locations. At 14 locations where measured flows are heavily influenced by reservoir operations and irrigation diversions, so called “ naturalized ” flows, with the effects of reservoirs and diversion removed, developed by the Bureau of Reclamation, were used for calibration. Surface energy and moisture processes such as solar radiation, snow accumulation and melting, and evapotranspiration were calibrated using national datasets as well as data from long-term measurement sites in the basin. The calibrated Deschutes GSFLOW model requires daily precipitation, minimum and maximum air temperature data, and monthly data describing groundwater pumping and artificial recharge from leaking irrigation canals (which are a significant source of groundwater recharge). The calibrated model simulates the geographic distribution of hydraulic head over the 5,000 ft range measured in the basin, with a median absolute residual of about 53 ft. Temporal variations in head resulting from climate cycles, pumping, and canal leakage are well simulated over the model area. Simulated daily streamflow matches gaged flows or calculated naturalized flows for streams including the Crooked and Metolius Rivers, and lower parts of the mainstem Deschutes River. Seasonal patterns of runoff are less well fit in some upper basin streams. Annual water balances of streamflow are good over most of the model domain. Model fit and overall capabilities are appropriate for the objectives of the project. The integrated model results confirm findings from other studies and models indicating that most streamflow in the upper Deschutes Basin comes directly from groundwater discharge. The integrated model provides additional insights about the components of streamflow including direct groundwater discharge to streams, interflow, groundwater discharge to the land surface (Dunnian flow), and direct runoff (Hortonian flow). The new model provides improved capability for exploring the timing and distribution of streamflow capture by wells, and the hydrologic response to changes in other external stresses such as canal operation, irrigation, and drought. Because the model uses basic meteorological data as the primary input; and simulates surface energy and moisture balances, groundwater recharge and flow, and all components of streamflow; it is well suited for exploring the hydrologic response to climate change, although no such simulations are included in this report. The model was developed as a tool for future application; however, example simulations are provided in this report. In the example simulations, the model is used to explore the influence of well location and geologic structure on stream capture by pumping wells. Wells were simulated at three locations within a 12-mi area close to known groundwater discharge areas and crossed by a regional fault zone. Simulations indicate that the magnitude and timing of stream capture from pumping is largely controlled by the geographic location of the wells, but that faults can have a large influence on the propagation of pumping stresses.

Oregon↗

USGS2015

Effects of groundwater pumping on agricultural drains in the Tule Lake subbasin, Oregon and California

Since 2001, irrigators in the upper Klamath Basin have increasingly turned to groundwater to compensate for reductions in surface-water allocation caused by shifts from irrigation use to instream flows for Endangered Species Act listed fishes. The largest increase in groundwater pumping has been in and around the Bureau of Reclamation’s Klamath Irrigation Project, which includes the Tule Lake subbasin in the southern part of the upper Klamath Basin. Agricultural drains on the Klamath Project are an important source of water for downstream irrigators and for the Tule Lake and Lower Klamath Lake National Wildlife Refuges. U.S. Geological Survey regional groundwater-flow model simulations and records of irrigation-return flow pumped from the Tule Lake subbasin into the adjacent Lower Klamath Lake subbasin have indicated that water-level declines from pumping may be causing decreased flow of shallow groundwater to agricultural drains. To better define the effect of increased pumping on drain flow and on the water balance of the groundwater system, the annual water volume pumped from drains in three subareas of the Tule Lake subbasin was estimated and a fine-grid, local groundwater model of the Tule Lake subbasin was constructed. Results of the agricultural-drain flow analysis indicate that groundwater discharge to drains has decreased such that flows in 2012 were approximately 32,400 acre-ft less than the 1997–2000 average flow. This decrease was concentrated in the northern and southeastern parts of the subbasin, which corresponds with the areas of greatest groundwater pumping. Model simulation results of the Tule Lake subbasin groundwater model indicate that increased supplemental pumping is the dominant stress to the groundwater system in the subbasin. Simulated supplemental pumping and decreased recharge from irrigation between 2000 and 2010 totaled 323,573 acre-ft, 234,800 acre-ft (73 percent) of which was from supplemental pumping. The response of the groundwater system to this change in stress included about 180,500 acre-ft (56 percent) of decreased groundwater discharge to drains and a 126,000 acre-ft (39 percent) reduction in aquifer storage. The remaining 5 percent came from reduced groundwater flow to other model boundaries, including the Lost River, the Tule Lake sumps, and interbasin flow.

California, Oregon↗