Numerical Simulation of the Groundwater-Flow System in Tributary Subbasins and Vicinity, Lower Skagit River Basin, Skagit and Snohomish Counties, Washington: Usgs Scientific Investigations Report 2010-5184

A groundwater-flow model was developed to evaluate the effects of potential groundwater withdrawals and consumptive use on streamflows in tributary subbasins of the lower portion of the Skagit River basin. The study area covers about 155 square miles along the Skagit River and its tributary subbasins (East Fork Nookachamps Creek, Nookachamps Creek, Carpenter Creek, Fisher Creek) in southwestern Skagit County and northwestern Snohomish County, Washington. The Skagit River occupies a large, relatively flat alluvial valley that extends across the northern and western margins of the study area, and is bounded to the south and east by upland and mountainous terrain. The alluvial valley and upland are underlain by unconsolidated deposits of glacial and inter- glacial origin. Bedrock underlies the alluvial valley and upland areas, and crops out throughout the mountainous terrain. Nine hydrogeologic units are recognized in the study area and form the basis of the groundwater-flow model. Groundwater flow in tributary subbasins of the lower Skagit River and vicinity was simulated using the groundwater-flow model, MODFLOW-2000. The finite-difference model grid consists of 174 rows, 156 columns, and 15 layers. Each model cell has a horizontal dimension of 500 by 500 feet. The thickness of model layers varies throughout the model area. Groundwater flow was simulated for both steady-state and transient conditions. The steady-state condition simulated average recharge, discharge, and water levels for the period, August 2006-September 2008. The transient simulation period, September 2006-September 2008, was divided into 24 monthly stress periods. Initial conditions for the transient model were developed from a 6-year ?lead-in? period that used recorded precipitation and Skagit River levels, and extrapolations of other boundary conditions. During model calibration, variables were adjusted within probable ranges to minimize differences between measured and simulated groundwater levels and stream baseflows. The final calibrated steady-state and transient models have weighted mean residual of -10.1 and -2.2 feet, respectively (negative residuals indicate that measured value is less than simulated value). Simulated inflow to the model area was about 144,000 acre-feet per year (acre-ft/yr) (81 percent of simulated inflow) from precipitation and secondary recharge, and about 32,700 acre-ft/yr (19 percent of simulated inflow) from stream and lake leakage. Simulated outflow from the model primarily was through discharge to streams and lakes (about 166,500 acre-ft/yr; 94 percent of simulated outflow), and withdrawals from wells (about 9,800 acre-ft/yr; 6 percent of simulated outflow). Model simulations were conducted to demonstrate model performance and to provide representative examples of how the model may be used to evaluate the effects of potential changes in groundwater withdrawals, consumptive use, and recharge on groundwater levels and tributary stream baseflows.