A finite-difference ground-water flow model was used to simulate the effects of both modern-day (1988) and projected 2010 ground-water withdrawals on the Floridan aquifer system in the greater Orlando metropolitan area. This area covers about 2,500 square miles and includes all of Orange and Seminole Counties and parts of Lake, Volusia, Brevard, Osceola, and Polk Counties. The hydrogeology of the area is characterized by a thin surficial aquifer underlain by the thick, highly productive rocks of the Floridan aquifer system. Water in the Upper Floridan aquifer is brackish (chloride concentrations greater than 1,000 milligrams per liter) in discharge areas beneath and near the St. Johns and Wekiva Rivers and is freshest (chloride concentrations less than 100 milligrams per liter) inrecharge areas. A slight trend toward increasing concentrations of dissolved solids, chloride, and sulfate has been observed at Upper Floridan aquifer springs. Chloride concentrations in the Upper Floridan aquifer measured between 1966 and 1993 at the Cocoa well field have increased from 50 milligrams per liter to 120 milligrams per liter; concentrations measured in the Lower Floridan aquifer between 1966 and 1993 have increasedfrom 600 milligrams per liter to 3,000 milligrams per liter. The flow model was calibrated by comparing (a) simulated and estimated Upper Floridan aquifer predevelopment (unstressed) potentiometric surfaces, (b) simulated and measured heads at 142 Upper Floridan aquifer monitoring wells in 1988 (averageabsolute error of 1.8 feet), (c) simulated and measured discharge rates at 15 Upper Floridan aquifer springs in 1988 (306 cubic feet per second), and (d) simulated and measured drawdowns at 134 Upper Floridan aquifer monitoring wells between 1988 and May 1990 (58 and 95 percent of simulated drawdowns were within plus or minus 25and 50 percent of measured drawdowns, respectively). Relative to predevelopment conditions, model simulations indicate that about half of the 305 million gallons per day of water pumped from the Floridan aquifer system in 1988 was accounted for by increased recharge from the surficial aquifer system. About 23 cubic feet persecond was derived from increased lateral inflow. A storage coefficient of 1x10-3 provided the best comparisons of measured-to-simulated data during the transient simulation from January to May 1990. This storativity probably is greater than the true storativity of the Upper Floridan aquifer because storage contributions from the intermediateconfining unit were not accounted for during model design and development. Calibrated transmissivity ranged from 10,000 to greater than 400,000 feet squared per day in the Upper Floridan aquifer, and from 5,000 to 600,000 feet squared per day in the Lower Floridan aquifer. Calibrated intermediate confining unit leakance ranged from 1x10-5 to 4x10-3 per day and was highest in areas where the unit is thin or has been breached by numerous sinkholes. In general,calibrated transmissivity and leakance values were higher than associated aquifer-test values. Simulated recharge rates to the Upper Floridan aquifer from the surficial aquifer system ranged from less than 3 to 21 inches per year. Recharge rates of greater than 10 inches per year were simulated in areas of west Seminole, west Orange, east Lake, and southwest Volusia Counties. Recharge rates of less than 3 inches per year were simulated in east Orange and northeast Osceola Counties. The calibrated model was used to simulate the effects of increased Floridan aquifer withdrawals in the year 2010 (542 million gallons per day) on water levels and spring flow. Projected effects were simulated for both "wet" conditions (using 1988 fixed-head arrays) and for "dry" conditions (using May 1990 fixed-head arrays), thus bracketing a potential range of effects. Relative to simulated 1988 conditions, simulated 2010 spring flow decreased by 43 cubic f