Reports
Delineation of wellhead protection areas using particle tracking analysis and hydrogeologic mapping, northern Anne Arundel County, Maryland
1995, Wilson, J.M. and Achmad, G.
Report of Investigations 61
Abstract
Two computer modeling techniques are used to delineate wellhead protection areas (WHPAs) for the public-supply wells of northern Anne Arundel County, Maryland and the results of the two techniques are compared. One technique is ground-waterflow modeling using the U.S. Geological Survey MODFLOW program in combination with particle tracking using the 1989 version of the U.S. Geological Survey MODPATH program. The other technique is particle tracking using the semi analytical option of the GPTRAC module of the U.S. Environmental Protection Agency WHPA code (version 2.1). The WHPAs are for wells screened in the Lower Patapsco aquifer, a semiconfined aquifer which is overlain by the Upper Patapsco aquifer, an unconfined water-table aquifer. Both aquifers consist of unconsolidated sands, gravels and discontinuous beds of clay. The ground-water-flow model used to generate head distributions is an updated and verified version of the flow model described in Achmad (1991).
Hydrogeologic mapping is used to develop an "aquifer vulnerability map" of the study area. Four qualitative degrees of aquifer vulnerability to potential contamination are mapped.
Tritium determinations indicate that most of the water produced from the Lower Patapsco aquifer in 1991 had a component of post-1945 water. Precise dates are not determined because of mixing of ground water of different ages; the estimated tritium ages are, however, consistent with ages inferred from the particle tracking analyses.
The MODPATH and GPTRAC particle tracking techniques are used to delineate ten and twenty-year zones of transport for the public-supply wells under three different pumpage scenarios; similar aquifer parameters are input to both models. The plan views of these transport zones are compared by size and shape. Using the 1990 pumpage appropriation of 9.15 mjllion gallons per day, six of the seven ten-year zones of transport delineated by GPTRAC range from 36 to 17 percent smaller and one is 18 percent larger than the corresponding MODPATH zones; the percentage coincident area of the MODPATH and GPTRAC zones of transport ranges from 64 to 81 percent. Shapes of corresponding zones of transport are mostly similar. Increasing the simulation time to twenty years caused differences between corresponding zones of transport to increase; for two of the well fields these differences are significant.
Differences between the zones of transport described by MODPATH and GPTRAC are caused by differing capabilities of the two techniques. The zones of transport described by MODPATH are considered more realistic than those described by the semianalytical option of GPTRAC (version 2.1) because head output from MODFLOW is calibrated using field values before it is input to MODPATH. Additionally, MODFLOW can simulate heterogeneous aquifer conditions and multi-layered aquifer systems more realistically than GPTRAC. GPTRAC is limited to simulating homogeneous aquifer conditions in a single layer aquifer system and also has more limited options for treating aquifer boundary conditions than MODFLOW. In cases where the aquifer system is relati vely homogeneous, however, GPTRAC may simulate zones of transport for a semiconfined aquifer that are comparable to the MODFLOW-MODPATH technique.
MODPATH is used to backtrack particles to land surface and delineate zones of contribution (recharge areas) in the watertable aquifer. Under the model scenario that used the 1990 pumpage appropriation, the recharge areas for the well fields covered about 50 percent of the area between the well fields and the upgradient ground-water divides. In tills simulation, the time required for recharge to travel from the water table to the different pumping wells ranged from 10 to 60 years. MOD PATH is also used to compare a time-related zone of contribution with the time-related zone of transport for one of the production wells.
Overall water quality of the Lower Patapsco aquifer is good. Contaminants, however, are present in the more vulnerable regions of the aquifer. Most of the known contaminant sites are downgradient from the production wells described herein; therefore, substances released at these sites will not enter the zones of transport under the rate of actual pumpage or the 1990 pumpage appropriation. Releases from several contaminant sites located upgradient of the production wells would without remediation pose a threat to the Glendale and Sawmill well fields; at present, both of these well fields are unused.
Downloads and Data
Report of Investigations 61 (pdf, 75.3 MB)Plate 1, Report of Investigations 61 (pdf, 0.82 MB)Plate 2, Report of Investigations 61 (pdf, 0.80 MB)Plate 3, Report of Investigations 61 (pdf, 0.73 MB)Plate 4, Report of Investigations 61 (pdf, 0.64 MB)Plate 5, Report of Investigations 61 (pdf, 4.1 MB)Plate 6, Report of Investigations 61 (pdf, 4.2 MB)Plate 7, Report of Investigations 61 (pdf, 3.9 MB)