Hydrogeology, brackish-water occurrence, and simulation of flow and brackish-water movement in the Aquia aquifer in the Kent Island area, Maryland
1988, Drummond, D.D.
Report of Investigations 51
A study was conducted to investigate the hydrogeology of the Aquia aquifer in the Kent Island area, Maryland. The problem of brackish-water occurrence in the Aquia aquifer was specifically addressed.
The Aquia aquifer contains sediments of Paleocene and Eocene age. The aquifer is a fine- to medium-grained glauconitic sand, containing layers of clayey sand, layers indurated by calcite cement, and abundant shell material. The upper confining bed overlies the Aquia aquifer and contains sed iments of the Nanjemoy and Calvert Formations. It is chiefly a silty clay and clayey sand and forms a leaky confining bed. The lower confining bed und erlies the Aquia aquifer and contains sediments of the Matawan and Severn Formations. It is chiefly a silty, sandy clay and forms a relatively tight confining bed. The unconfined aquifer overlies the upper confining bed and contains sediments of the Kent Island and Talbot Formations. It is chiefly a clayey silt with pockets of sand and gravel, and supplies recharge to the aquifer system below.
Brackish water is present in the Aquia aquifer a long the Chesapeake Bay shore from the northernmost tip of the island (Love Point) to at least as far south as Prices Creek. In the northern part of the brackish-water zone, the entire vertical section of the Aquia aquifer contains brackish water. In the southern part of the brackish-water zone, the bottom of the Aquia contains brackish water, but the top contains freshwater. Five major hydrogeologic controls were identified that influence the distribution and movement of brackish water in the Aquia aquifer: density-dependent flow, water pressures, the presence of calcite-cemented layers, the presence of paleochannel sediments, and the confining-bed permeability.
A quasi three-dimensional, finite-difference two-layer areal flow model was developed to simulate the response of water levels to projected pumpage in the Aquia aq uifer. The flow-model area includes Kent Island and a large part of the Eastern Shore to simulate pumping centers at the towns of Easton, Oxford, St. Michaels, and Centreville. A simulation based on the best estimate offuture pumpage indicates an add itional 5 feet of drawdown from the 1984 potentiometric surface by the year 2005 on parts of Kent Island. The greatest declines occur on the eastern part of the island near Grasonville. Simulations based on pumpage amounts 20 percent higher and lower than the best estimate indicate additional drawdowns of 6 to 8 feet and 2 to 4 feet, respectively, from the 1984 potentiometric surface. Other simulations were made to evaluate alternative pumpage conditions such as replacement of domestic pumpage by centralized supply wells and varying pumpage amounts in areas other than Kent Island. A sensitivity analysis was performed on the calibrated flow model to evaluate the effects of uncertainties in the input data on model results. The sensitivity analysis indicates that simulated water levels could be in error by 2 to 4 feet in the Kent Island area because of possible inaccuracies in the input data.
A cross-sectional solute-trans port model was developed to estimate the movement of brackish water in response to projected pumpage amounts and to evaluate the importance of the hydrogeologic controls on the distribution and movement of brackish water. Estimated distances and rates of movement should be regarded as general approximations rather than accurate predictions. Model results indicate that the freshwater/brackish-water interface will move about 440 feet inland during the 21-year simulation period (1984-2005) based on the best estimate of future pumpage. Alternative simulations based on pumpage amounts 20 percent higher and lower than the best estimate indicate interface movement of about 490 feet and about 350 feet, respectively, for that same period. Average interface velocities for the three simulations are about 21, 24, and 17 feet per year. A simulation based on continued 1984 pumpage throughout the future simulation period indicates interface movement of about 275 feet at an average velocity of about 13 feet per year. A simulation based on no pumpage in the Aquia throughout the simulation period indicates interface movement of about 40 feet in the opposite direction (bayward) at an average interface velocity of about 2 feet per year. A sensitivity analysis was run on the calibrated transport model to evaluate the effect of uncertainties in input data on model results. The sensitivity analysis indicates that projected interface movement could be in error by 125 feet for the 21-year simulation period, or about 6 feet per year, because of possible inaccuracies in the input data.
Simulations designed to evaluate the importance of the hydrogeologic controls on brackish-water movement indicate that density-dependent flow, water pressures in the Aquia aquifer, and the permeability of the upper confining bed are the most important factors. Calcite-cemented layers and paleochannel sediments provide minor controls on brackish-water movement in the Aquia aquifer in the Kent Island area.