No abstract

No abstract

Introduction 2 Purpose and scope 2 Description of study area 3 Previous investigations 3 Methods of study 7 Acknowledgments 7 Hydrogeology 8 Geologic setting 8 Hydrogeologic units 8 Floridan aquifer system 10 Dublin aquifer system 11 Midville aquifer system 11 Conceptualization of stream-aquifer flow system 12 Hydrologic budget 14 Predevelopment flow system 16 Modern-day (1987-92) flow system 16 Simulation of groundwater flow 18 Spatial and vertical discretization 20 Hydraulic characteristics 20 Boundary conditions 23 Pumpage 26 Model calibration 29 Steady-state simulation of predevelopment flow system 31 Simulated heads 31 Simulated water budget 33 Simulated groundwater recharge 39 Simulated groundwater discharge to streams 41 Simulated groundwater flow 43 Simulation of flow system, 1953-92 45 Testing of model for transient response to pumpage 45 Steady-state analysis of modern-day (1987-92) flow conditions 45 Simulated drawdown 55 Calibrated hydraulic properties 61 Trans-river flow beneath the Savannah River 74 Particle tracking analysis of advective groundwater flow 75 Westward trans-river flow 84 Eastward trans-river flow 84 Recharge areas to trans-river flow zones 85 Simulated time-of-travel 86 Trans-river flow, recharge areas, and time-of-travel at the Savannah River Site 88 Simulation of groundwater flow-Continued Sensitivity analysis 88 Limitations of digital simulation 90 Limitations of particle tracking 100 Summary and conclusions 100 Selected References 104 Appendix A. Mean-annual groundwater discharge to streams estimated using hydrograph separation and simulated groundwater discharge for predevelopment (prior to 1953) and modern-day (1987-92) conditions 114 Appendix B. Estimated groundwater discharge to streams during 1954 and 1986 droughts, and simulated groundwater discharge for predevelopment (prior to 1953) and modern-day (1987-92) conditions 115 Appendix C. Measured heads, simulated predevelopment (prior to 1953) and modern-day (1987-92) heads, and error criteria in wells used for model calibration 119 v ILLUSTRATIONS PLATES [in pocket in back of report] Plates 1-3. Maps showing: 1. Model grid and boundary conditions for the Upper Three Runs, Gordon, and Millers Pond aquifers 2. Model grid and boundary conditions for the Dublin and Midville aquifers 3. Estimated average potentiometric surface for the Upper Three Runs, Gordon, Dublin, and Midville aquifers FIGURES

A sparsely parameterized hydrochemical model has been developed by using data from Emerald Lake watershed, which is a 120-hectare alpine catchment in Sequoia National Park, California. Greater than 90 percent of the precipitation to this watershed is snow; hence, snowmelt is the dominant hydrologic event. A model which uses a single alkalinity-generating mechanism, primary mineral weathering, was able to capture the pattern of solute concentrations in surface waters during snowmelt. An empirical representation of the weathering reaction, which is based on rock weathering stoichiometry and which uses discharge as a measure of residence time, was included in the model. Results of the model indicate that current deposition levels would have to be increased between three-and eight-fold to exhaust the alkalinity of the lake during snowmelt if there is a mild acidic pulse in the stream at the beginning of snowmelt as was observed during the study period. The acidic pulse in the inflow stream at the onset of snowmelt was less pronounced than acidic pulses observed in the meltwater draining the snowpack at a point using snow lysimeters or in the laboratory. Sulfate concentrations in the streamwater were the most constant; chloride and nitrate concentrations increased slightly at the beginning of snowmelt. Additional field work is required to resolve whether the an acidic meltwater pulse occurs over a large area as well as at a point (implying sulfate-regulating mechanisms in the soil) or whether, due to physical and chemical processes within the snowpack, the acidic meltwater pulse is attenuated at the catchment scale. The modest data requirements of the model permit its applications to other alpine watersheds that are much less intensively studied than Emerald Lake watershed. Purpose and Scope One phase of the IWS was the development of simulation models to predict surface-water chemistry. These models were to serve as research tools to integrate information gained from the field program and to identify gaps in understanding the chemical mechanisms controlling surfacewater chemistry, and as predictive tools to provide a means for forecasting the chemical response of the watershed to different depositional loadings of atmospheric acidity. This paper presents a lumped, sparsely parameterized simulation model, called the Alpine Lake Forecaster, or ALF, that was developed from the IWS data. ALF is a model of intermediate complexity when compared with a fully distributed model (Sorooshian and others, 1989) and a regional lake acidification model (Nishida and others, 1989), which also were developed from this data set.

No abstract