Ground‐water recharge to a shallow, unconfined, fractured dolomite aquifer underlying agricultural land in Lancaster County, Pennsylvania occurs by two mechanisms. Direct recharge occurs through pathways such as near‐surface bedrock fractures and sinkholes, and affects dissolved nitrate concentration of ground water within two to three days; its effects last only about one week. Gradual recharge occurs through small channels and pores in the unsaturated zone and affects dissolved nitrate concentration for several weeks or more after the effects of direct recharge have dissipated. Whether recharge causes an increase or decrease in dissolved nitrate concentration depends on the amount of nitrogen‐rich manure spread on the site prior to the storm. Direct recharge from a storm in March 1984, a month in which 18 tons of manure were spread, resulted in a rapid decrease in dissolved nitrate concentration of about 2.5 milligrams per liter (mg/1) as nitrogen. Direct recharge from a storm in May 1984, after 384 tons of manure had been spread in April, resulted in a rapid increase in dissolved nitrate concentration of about 3 mg/1 as nitrogen. Concentration changes caused by gradual recharge several weeks or more after the storms were of the same magnitude as those caused by direct recharge during the storm.

No abstract

Sensitivity analysis 32 Calibration of model under transient conditions 34 General procedure 34 Modeled flow components 37 Groundwater conditions 37 Surface-water conditions 37 Contents III Calibration of model under transient conditions Continued Specific yield and storage coefficient 38 Initial estimates 38 Adjustments 38 Recharge 38 Boundary conditions 41 Results 41 Sensitivity analysis 45 Digital-model evaluation of groundwater resources 45 General procedure 45 Qualitative ranking of hydrogeologic units 46 Standardized potential yield 46 Procedure 46 Results 47 Analysis by hydrogeologic unit 47 Analysis in a selected area 52 Reliability of model results 54 Suggested uses of model 58 Summary 59 References cited 61 Appendixes: A: Model program listing 64 B: Instructions for use of model program 108 C: Water-level measurement data from observation-well network 118 Metric conversion factors 128 PLATES [Plates in pocket] Contents V 42 19. Statistical comparison, by general lithology, of observed and model-generated changes in water-table altitude in calibrated transient model 45 20. Ranking of hydrogeologic units, based on effects of withdrawing 1 inch of ground water per year from modeled area 46 21. Standardized potential yield for each hydrogeologic unit, based on hypothetical withdrawal scheme 47 22. Time, since start of hypothetical withdrawal scheme, for each hydrogeologic unit to reach steady-state conditions 49 23. Percentage reduction of average annual base flow during 20-year simulation of hypothetical withdrawal scheme 49 24. Median decline of water table during 20-year simulation of hypothetical withdrawal scheme 50 25. Groundwater budget for each hydrogeologic unit after 91 days of hypothetical withdrawal scheme 51 26. Ultimate groundwater budget for each hydrogeologic unit resulting from hypothetical withdrawal scheme 52 VI Contents

A digital model of two-dimensional groundwater flow was used to predict changes in the potentiometric surface of the Floridan aquifer, 1976-2000, in a 5,938-square-mile area of west-central Florida. In 1975, ground water withdrawn from the Floridan aquifer for irrigation, phosphate mines, other industries, and municipal supplies averaged about 649 million gallons per day. Rates are projected to increase to about 840 million gallons per day by 2000. The model was calibrated under steady-state and transient conditions. Input parameters included transmissivity and storage coefficient of the Floridan aquifer; thickness, vertical hydraulic conductivity, and storage coefficient of the upper confining bed; altitudes of the water table and potentiometric surface; and groundwater withdrawals. Simulation of May 1976 to May 2000, using projected combined pumping rates for municipal supplies, irrigation, and industry (including existing and proposed phosphate mines), resulted in a rise in the potentiometric surface of about 10 feet in Polk County, and a decline of about 35 feet in parts of Manatee and Hardee Counties. The lowest simulated potentiometric level was about 30 feet below National Geodetic Vertical Datum of 1929.